Annotation of imach/src/imach.c, revision 1.286
1.286 ! brouard 1: /* $Id: imach.c,v 1.285 2018/04/21 21:02:16 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.286 ! brouard 4: Revision 1.285 2018/04/21 21:02:16 brouard
! 5: Summary: Some bugs fixed, valgrind tested
! 6:
1.285 brouard 7: Revision 1.284 2018/04/20 05:22:13 brouard
8: Summary: Computing mean and stdeviation of fixed quantitative variables
9:
1.284 brouard 10: Revision 1.283 2018/04/19 14:49:16 brouard
11: Summary: Some minor bugs fixed
12:
1.283 brouard 13: Revision 1.282 2018/02/27 22:50:02 brouard
14: *** empty log message ***
15:
1.282 brouard 16: Revision 1.281 2018/02/27 19:25:23 brouard
17: Summary: Adding second argument for quitting
18:
1.281 brouard 19: Revision 1.280 2018/02/21 07:58:13 brouard
20: Summary: 0.99r15
21:
22: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
23:
1.280 brouard 24: Revision 1.279 2017/07/20 13:35:01 brouard
25: Summary: temporary working
26:
1.279 brouard 27: Revision 1.278 2017/07/19 14:09:02 brouard
28: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
29:
1.278 brouard 30: Revision 1.277 2017/07/17 08:53:49 brouard
31: Summary: BOM files can be read now
32:
1.277 brouard 33: Revision 1.276 2017/06/30 15:48:31 brouard
34: Summary: Graphs improvements
35:
1.276 brouard 36: Revision 1.275 2017/06/30 13:39:33 brouard
37: Summary: Saito's color
38:
1.275 brouard 39: Revision 1.274 2017/06/29 09:47:08 brouard
40: Summary: Version 0.99r14
41:
1.274 brouard 42: Revision 1.273 2017/06/27 11:06:02 brouard
43: Summary: More documentation on projections
44:
1.273 brouard 45: Revision 1.272 2017/06/27 10:22:40 brouard
46: Summary: Color of backprojection changed from 6 to 5(yellow)
47:
1.272 brouard 48: Revision 1.271 2017/06/27 10:17:50 brouard
49: Summary: Some bug with rint
50:
1.271 brouard 51: Revision 1.270 2017/05/24 05:45:29 brouard
52: *** empty log message ***
53:
1.270 brouard 54: Revision 1.269 2017/05/23 08:39:25 brouard
55: Summary: Code into subroutine, cleanings
56:
1.269 brouard 57: Revision 1.268 2017/05/18 20:09:32 brouard
58: Summary: backprojection and confidence intervals of backprevalence
59:
1.268 brouard 60: Revision 1.267 2017/05/13 10:25:05 brouard
61: Summary: temporary save for backprojection
62:
1.267 brouard 63: Revision 1.266 2017/05/13 07:26:12 brouard
64: Summary: Version 0.99r13 (improvements and bugs fixed)
65:
1.266 brouard 66: Revision 1.265 2017/04/26 16:22:11 brouard
67: Summary: imach 0.99r13 Some bugs fixed
68:
1.265 brouard 69: Revision 1.264 2017/04/26 06:01:29 brouard
70: Summary: Labels in graphs
71:
1.264 brouard 72: Revision 1.263 2017/04/24 15:23:15 brouard
73: Summary: to save
74:
1.263 brouard 75: Revision 1.262 2017/04/18 16:48:12 brouard
76: *** empty log message ***
77:
1.262 brouard 78: Revision 1.261 2017/04/05 10:14:09 brouard
79: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
80:
1.261 brouard 81: Revision 1.260 2017/04/04 17:46:59 brouard
82: Summary: Gnuplot indexations fixed (humm)
83:
1.260 brouard 84: Revision 1.259 2017/04/04 13:01:16 brouard
85: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
86:
1.259 brouard 87: Revision 1.258 2017/04/03 10:17:47 brouard
88: Summary: Version 0.99r12
89:
90: Some cleanings, conformed with updated documentation.
91:
1.258 brouard 92: Revision 1.257 2017/03/29 16:53:30 brouard
93: Summary: Temp
94:
1.257 brouard 95: Revision 1.256 2017/03/27 05:50:23 brouard
96: Summary: Temporary
97:
1.256 brouard 98: Revision 1.255 2017/03/08 16:02:28 brouard
99: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
100:
1.255 brouard 101: Revision 1.254 2017/03/08 07:13:00 brouard
102: Summary: Fixing data parameter line
103:
1.254 brouard 104: Revision 1.253 2016/12/15 11:59:41 brouard
105: Summary: 0.99 in progress
106:
1.253 brouard 107: Revision 1.252 2016/09/15 21:15:37 brouard
108: *** empty log message ***
109:
1.252 brouard 110: Revision 1.251 2016/09/15 15:01:13 brouard
111: Summary: not working
112:
1.251 brouard 113: Revision 1.250 2016/09/08 16:07:27 brouard
114: Summary: continue
115:
1.250 brouard 116: Revision 1.249 2016/09/07 17:14:18 brouard
117: Summary: Starting values from frequencies
118:
1.249 brouard 119: Revision 1.248 2016/09/07 14:10:18 brouard
120: *** empty log message ***
121:
1.248 brouard 122: Revision 1.247 2016/09/02 11:11:21 brouard
123: *** empty log message ***
124:
1.247 brouard 125: Revision 1.246 2016/09/02 08:49:22 brouard
126: *** empty log message ***
127:
1.246 brouard 128: Revision 1.245 2016/09/02 07:25:01 brouard
129: *** empty log message ***
130:
1.245 brouard 131: Revision 1.244 2016/09/02 07:17:34 brouard
132: *** empty log message ***
133:
1.244 brouard 134: Revision 1.243 2016/09/02 06:45:35 brouard
135: *** empty log message ***
136:
1.243 brouard 137: Revision 1.242 2016/08/30 15:01:20 brouard
138: Summary: Fixing a lots
139:
1.242 brouard 140: Revision 1.241 2016/08/29 17:17:25 brouard
141: Summary: gnuplot problem in Back projection to fix
142:
1.241 brouard 143: Revision 1.240 2016/08/29 07:53:18 brouard
144: Summary: Better
145:
1.240 brouard 146: Revision 1.239 2016/08/26 15:51:03 brouard
147: Summary: Improvement in Powell output in order to copy and paste
148:
149: Author:
150:
1.239 brouard 151: Revision 1.238 2016/08/26 14:23:35 brouard
152: Summary: Starting tests of 0.99
153:
1.238 brouard 154: Revision 1.237 2016/08/26 09:20:19 brouard
155: Summary: to valgrind
156:
1.237 brouard 157: Revision 1.236 2016/08/25 10:50:18 brouard
158: *** empty log message ***
159:
1.236 brouard 160: Revision 1.235 2016/08/25 06:59:23 brouard
161: *** empty log message ***
162:
1.235 brouard 163: Revision 1.234 2016/08/23 16:51:20 brouard
164: *** empty log message ***
165:
1.234 brouard 166: Revision 1.233 2016/08/23 07:40:50 brouard
167: Summary: not working
168:
1.233 brouard 169: Revision 1.232 2016/08/22 14:20:21 brouard
170: Summary: not working
171:
1.232 brouard 172: Revision 1.231 2016/08/22 07:17:15 brouard
173: Summary: not working
174:
1.231 brouard 175: Revision 1.230 2016/08/22 06:55:53 brouard
176: Summary: Not working
177:
1.230 brouard 178: Revision 1.229 2016/07/23 09:45:53 brouard
179: Summary: Completing for func too
180:
1.229 brouard 181: Revision 1.228 2016/07/22 17:45:30 brouard
182: Summary: Fixing some arrays, still debugging
183:
1.227 brouard 184: Revision 1.226 2016/07/12 18:42:34 brouard
185: Summary: temp
186:
1.226 brouard 187: Revision 1.225 2016/07/12 08:40:03 brouard
188: Summary: saving but not running
189:
1.225 brouard 190: Revision 1.224 2016/07/01 13:16:01 brouard
191: Summary: Fixes
192:
1.224 brouard 193: Revision 1.223 2016/02/19 09:23:35 brouard
194: Summary: temporary
195:
1.223 brouard 196: Revision 1.222 2016/02/17 08:14:50 brouard
197: Summary: Probably last 0.98 stable version 0.98r6
198:
1.222 brouard 199: Revision 1.221 2016/02/15 23:35:36 brouard
200: Summary: minor bug
201:
1.220 brouard 202: Revision 1.219 2016/02/15 00:48:12 brouard
203: *** empty log message ***
204:
1.219 brouard 205: Revision 1.218 2016/02/12 11:29:23 brouard
206: Summary: 0.99 Back projections
207:
1.218 brouard 208: Revision 1.217 2015/12/23 17:18:31 brouard
209: Summary: Experimental backcast
210:
1.217 brouard 211: Revision 1.216 2015/12/18 17:32:11 brouard
212: Summary: 0.98r4 Warning and status=-2
213:
214: Version 0.98r4 is now:
215: - displaying an error when status is -1, date of interview unknown and date of death known;
216: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
217: Older changes concerning s=-2, dating from 2005 have been supersed.
218:
1.216 brouard 219: Revision 1.215 2015/12/16 08:52:24 brouard
220: Summary: 0.98r4 working
221:
1.215 brouard 222: Revision 1.214 2015/12/16 06:57:54 brouard
223: Summary: temporary not working
224:
1.214 brouard 225: Revision 1.213 2015/12/11 18:22:17 brouard
226: Summary: 0.98r4
227:
1.213 brouard 228: Revision 1.212 2015/11/21 12:47:24 brouard
229: Summary: minor typo
230:
1.212 brouard 231: Revision 1.211 2015/11/21 12:41:11 brouard
232: Summary: 0.98r3 with some graph of projected cross-sectional
233:
234: Author: Nicolas Brouard
235:
1.211 brouard 236: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 237: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 238: Summary: Adding ftolpl parameter
239: Author: N Brouard
240:
241: We had difficulties to get smoothed confidence intervals. It was due
242: to the period prevalence which wasn't computed accurately. The inner
243: parameter ftolpl is now an outer parameter of the .imach parameter
244: file after estepm. If ftolpl is small 1.e-4 and estepm too,
245: computation are long.
246:
1.209 brouard 247: Revision 1.208 2015/11/17 14:31:57 brouard
248: Summary: temporary
249:
1.208 brouard 250: Revision 1.207 2015/10/27 17:36:57 brouard
251: *** empty log message ***
252:
1.207 brouard 253: Revision 1.206 2015/10/24 07:14:11 brouard
254: *** empty log message ***
255:
1.206 brouard 256: Revision 1.205 2015/10/23 15:50:53 brouard
257: Summary: 0.98r3 some clarification for graphs on likelihood contributions
258:
1.205 brouard 259: Revision 1.204 2015/10/01 16:20:26 brouard
260: Summary: Some new graphs of contribution to likelihood
261:
1.204 brouard 262: Revision 1.203 2015/09/30 17:45:14 brouard
263: Summary: looking at better estimation of the hessian
264:
265: Also a better criteria for convergence to the period prevalence And
266: therefore adding the number of years needed to converge. (The
267: prevalence in any alive state shold sum to one
268:
1.203 brouard 269: Revision 1.202 2015/09/22 19:45:16 brouard
270: Summary: Adding some overall graph on contribution to likelihood. Might change
271:
1.202 brouard 272: Revision 1.201 2015/09/15 17:34:58 brouard
273: Summary: 0.98r0
274:
275: - Some new graphs like suvival functions
276: - Some bugs fixed like model=1+age+V2.
277:
1.201 brouard 278: Revision 1.200 2015/09/09 16:53:55 brouard
279: Summary: Big bug thanks to Flavia
280:
281: Even model=1+age+V2. did not work anymore
282:
1.200 brouard 283: Revision 1.199 2015/09/07 14:09:23 brouard
284: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
285:
1.199 brouard 286: Revision 1.198 2015/09/03 07:14:39 brouard
287: Summary: 0.98q5 Flavia
288:
1.198 brouard 289: Revision 1.197 2015/09/01 18:24:39 brouard
290: *** empty log message ***
291:
1.197 brouard 292: Revision 1.196 2015/08/18 23:17:52 brouard
293: Summary: 0.98q5
294:
1.196 brouard 295: Revision 1.195 2015/08/18 16:28:39 brouard
296: Summary: Adding a hack for testing purpose
297:
298: After reading the title, ftol and model lines, if the comment line has
299: a q, starting with #q, the answer at the end of the run is quit. It
300: permits to run test files in batch with ctest. The former workaround was
301: $ echo q | imach foo.imach
302:
1.195 brouard 303: Revision 1.194 2015/08/18 13:32:00 brouard
304: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
305:
1.194 brouard 306: Revision 1.193 2015/08/04 07:17:42 brouard
307: Summary: 0.98q4
308:
1.193 brouard 309: Revision 1.192 2015/07/16 16:49:02 brouard
310: Summary: Fixing some outputs
311:
1.192 brouard 312: Revision 1.191 2015/07/14 10:00:33 brouard
313: Summary: Some fixes
314:
1.191 brouard 315: Revision 1.190 2015/05/05 08:51:13 brouard
316: Summary: Adding digits in output parameters (7 digits instead of 6)
317:
318: Fix 1+age+.
319:
1.190 brouard 320: Revision 1.189 2015/04/30 14:45:16 brouard
321: Summary: 0.98q2
322:
1.189 brouard 323: Revision 1.188 2015/04/30 08:27:53 brouard
324: *** empty log message ***
325:
1.188 brouard 326: Revision 1.187 2015/04/29 09:11:15 brouard
327: *** empty log message ***
328:
1.187 brouard 329: Revision 1.186 2015/04/23 12:01:52 brouard
330: Summary: V1*age is working now, version 0.98q1
331:
332: Some codes had been disabled in order to simplify and Vn*age was
333: working in the optimization phase, ie, giving correct MLE parameters,
334: but, as usual, outputs were not correct and program core dumped.
335:
1.186 brouard 336: Revision 1.185 2015/03/11 13:26:42 brouard
337: Summary: Inclusion of compile and links command line for Intel Compiler
338:
1.185 brouard 339: Revision 1.184 2015/03/11 11:52:39 brouard
340: Summary: Back from Windows 8. Intel Compiler
341:
1.184 brouard 342: Revision 1.183 2015/03/10 20:34:32 brouard
343: Summary: 0.98q0, trying with directest, mnbrak fixed
344:
345: We use directest instead of original Powell test; probably no
346: incidence on the results, but better justifications;
347: We fixed Numerical Recipes mnbrak routine which was wrong and gave
348: wrong results.
349:
1.183 brouard 350: Revision 1.182 2015/02/12 08:19:57 brouard
351: Summary: Trying to keep directest which seems simpler and more general
352: Author: Nicolas Brouard
353:
1.182 brouard 354: Revision 1.181 2015/02/11 23:22:24 brouard
355: Summary: Comments on Powell added
356:
357: Author:
358:
1.181 brouard 359: Revision 1.180 2015/02/11 17:33:45 brouard
360: Summary: Finishing move from main to function (hpijx and prevalence_limit)
361:
1.180 brouard 362: Revision 1.179 2015/01/04 09:57:06 brouard
363: Summary: back to OS/X
364:
1.179 brouard 365: Revision 1.178 2015/01/04 09:35:48 brouard
366: *** empty log message ***
367:
1.178 brouard 368: Revision 1.177 2015/01/03 18:40:56 brouard
369: Summary: Still testing ilc32 on OSX
370:
1.177 brouard 371: Revision 1.176 2015/01/03 16:45:04 brouard
372: *** empty log message ***
373:
1.176 brouard 374: Revision 1.175 2015/01/03 16:33:42 brouard
375: *** empty log message ***
376:
1.175 brouard 377: Revision 1.174 2015/01/03 16:15:49 brouard
378: Summary: Still in cross-compilation
379:
1.174 brouard 380: Revision 1.173 2015/01/03 12:06:26 brouard
381: Summary: trying to detect cross-compilation
382:
1.173 brouard 383: Revision 1.172 2014/12/27 12:07:47 brouard
384: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
385:
1.172 brouard 386: Revision 1.171 2014/12/23 13:26:59 brouard
387: Summary: Back from Visual C
388:
389: Still problem with utsname.h on Windows
390:
1.171 brouard 391: Revision 1.170 2014/12/23 11:17:12 brouard
392: Summary: Cleaning some \%% back to %%
393:
394: The escape was mandatory for a specific compiler (which one?), but too many warnings.
395:
1.170 brouard 396: Revision 1.169 2014/12/22 23:08:31 brouard
397: Summary: 0.98p
398:
399: Outputs some informations on compiler used, OS etc. Testing on different platforms.
400:
1.169 brouard 401: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 402: Summary: update
1.169 brouard 403:
1.168 brouard 404: Revision 1.167 2014/12/22 13:50:56 brouard
405: Summary: Testing uname and compiler version and if compiled 32 or 64
406:
407: Testing on Linux 64
408:
1.167 brouard 409: Revision 1.166 2014/12/22 11:40:47 brouard
410: *** empty log message ***
411:
1.166 brouard 412: Revision 1.165 2014/12/16 11:20:36 brouard
413: Summary: After compiling on Visual C
414:
415: * imach.c (Module): Merging 1.61 to 1.162
416:
1.165 brouard 417: Revision 1.164 2014/12/16 10:52:11 brouard
418: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
419:
420: * imach.c (Module): Merging 1.61 to 1.162
421:
1.164 brouard 422: Revision 1.163 2014/12/16 10:30:11 brouard
423: * imach.c (Module): Merging 1.61 to 1.162
424:
1.163 brouard 425: Revision 1.162 2014/09/25 11:43:39 brouard
426: Summary: temporary backup 0.99!
427:
1.162 brouard 428: Revision 1.1 2014/09/16 11:06:58 brouard
429: Summary: With some code (wrong) for nlopt
430:
431: Author:
432:
433: Revision 1.161 2014/09/15 20:41:41 brouard
434: Summary: Problem with macro SQR on Intel compiler
435:
1.161 brouard 436: Revision 1.160 2014/09/02 09:24:05 brouard
437: *** empty log message ***
438:
1.160 brouard 439: Revision 1.159 2014/09/01 10:34:10 brouard
440: Summary: WIN32
441: Author: Brouard
442:
1.159 brouard 443: Revision 1.158 2014/08/27 17:11:51 brouard
444: *** empty log message ***
445:
1.158 brouard 446: Revision 1.157 2014/08/27 16:26:55 brouard
447: Summary: Preparing windows Visual studio version
448: Author: Brouard
449:
450: In order to compile on Visual studio, time.h is now correct and time_t
451: and tm struct should be used. difftime should be used but sometimes I
452: just make the differences in raw time format (time(&now).
453: Trying to suppress #ifdef LINUX
454: Add xdg-open for __linux in order to open default browser.
455:
1.157 brouard 456: Revision 1.156 2014/08/25 20:10:10 brouard
457: *** empty log message ***
458:
1.156 brouard 459: Revision 1.155 2014/08/25 18:32:34 brouard
460: Summary: New compile, minor changes
461: Author: Brouard
462:
1.155 brouard 463: Revision 1.154 2014/06/20 17:32:08 brouard
464: Summary: Outputs now all graphs of convergence to period prevalence
465:
1.154 brouard 466: Revision 1.153 2014/06/20 16:45:46 brouard
467: Summary: If 3 live state, convergence to period prevalence on same graph
468: Author: Brouard
469:
1.153 brouard 470: Revision 1.152 2014/06/18 17:54:09 brouard
471: Summary: open browser, use gnuplot on same dir than imach if not found in the path
472:
1.152 brouard 473: Revision 1.151 2014/06/18 16:43:30 brouard
474: *** empty log message ***
475:
1.151 brouard 476: Revision 1.150 2014/06/18 16:42:35 brouard
477: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
478: Author: brouard
479:
1.150 brouard 480: Revision 1.149 2014/06/18 15:51:14 brouard
481: Summary: Some fixes in parameter files errors
482: Author: Nicolas Brouard
483:
1.149 brouard 484: Revision 1.148 2014/06/17 17:38:48 brouard
485: Summary: Nothing new
486: Author: Brouard
487:
488: Just a new packaging for OS/X version 0.98nS
489:
1.148 brouard 490: Revision 1.147 2014/06/16 10:33:11 brouard
491: *** empty log message ***
492:
1.147 brouard 493: Revision 1.146 2014/06/16 10:20:28 brouard
494: Summary: Merge
495: Author: Brouard
496:
497: Merge, before building revised version.
498:
1.146 brouard 499: Revision 1.145 2014/06/10 21:23:15 brouard
500: Summary: Debugging with valgrind
501: Author: Nicolas Brouard
502:
503: Lot of changes in order to output the results with some covariates
504: After the Edimburgh REVES conference 2014, it seems mandatory to
505: improve the code.
506: No more memory valgrind error but a lot has to be done in order to
507: continue the work of splitting the code into subroutines.
508: Also, decodemodel has been improved. Tricode is still not
509: optimal. nbcode should be improved. Documentation has been added in
510: the source code.
511:
1.144 brouard 512: Revision 1.143 2014/01/26 09:45:38 brouard
513: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
514:
515: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
516: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
517:
1.143 brouard 518: Revision 1.142 2014/01/26 03:57:36 brouard
519: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
520:
521: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
522:
1.142 brouard 523: Revision 1.141 2014/01/26 02:42:01 brouard
524: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
525:
1.141 brouard 526: Revision 1.140 2011/09/02 10:37:54 brouard
527: Summary: times.h is ok with mingw32 now.
528:
1.140 brouard 529: Revision 1.139 2010/06/14 07:50:17 brouard
530: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
531: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
532:
1.139 brouard 533: Revision 1.138 2010/04/30 18:19:40 brouard
534: *** empty log message ***
535:
1.138 brouard 536: Revision 1.137 2010/04/29 18:11:38 brouard
537: (Module): Checking covariates for more complex models
538: than V1+V2. A lot of change to be done. Unstable.
539:
1.137 brouard 540: Revision 1.136 2010/04/26 20:30:53 brouard
541: (Module): merging some libgsl code. Fixing computation
542: of likelione (using inter/intrapolation if mle = 0) in order to
543: get same likelihood as if mle=1.
544: Some cleaning of code and comments added.
545:
1.136 brouard 546: Revision 1.135 2009/10/29 15:33:14 brouard
547: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
548:
1.135 brouard 549: Revision 1.134 2009/10/29 13:18:53 brouard
550: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
551:
1.134 brouard 552: Revision 1.133 2009/07/06 10:21:25 brouard
553: just nforces
554:
1.133 brouard 555: Revision 1.132 2009/07/06 08:22:05 brouard
556: Many tings
557:
1.132 brouard 558: Revision 1.131 2009/06/20 16:22:47 brouard
559: Some dimensions resccaled
560:
1.131 brouard 561: Revision 1.130 2009/05/26 06:44:34 brouard
562: (Module): Max Covariate is now set to 20 instead of 8. A
563: lot of cleaning with variables initialized to 0. Trying to make
564: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
565:
1.130 brouard 566: Revision 1.129 2007/08/31 13:49:27 lievre
567: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
568:
1.129 lievre 569: Revision 1.128 2006/06/30 13:02:05 brouard
570: (Module): Clarifications on computing e.j
571:
1.128 brouard 572: Revision 1.127 2006/04/28 18:11:50 brouard
573: (Module): Yes the sum of survivors was wrong since
574: imach-114 because nhstepm was no more computed in the age
575: loop. Now we define nhstepma in the age loop.
576: (Module): In order to speed up (in case of numerous covariates) we
577: compute health expectancies (without variances) in a first step
578: and then all the health expectancies with variances or standard
579: deviation (needs data from the Hessian matrices) which slows the
580: computation.
581: In the future we should be able to stop the program is only health
582: expectancies and graph are needed without standard deviations.
583:
1.127 brouard 584: Revision 1.126 2006/04/28 17:23:28 brouard
585: (Module): Yes the sum of survivors was wrong since
586: imach-114 because nhstepm was no more computed in the age
587: loop. Now we define nhstepma in the age loop.
588: Version 0.98h
589:
1.126 brouard 590: Revision 1.125 2006/04/04 15:20:31 lievre
591: Errors in calculation of health expectancies. Age was not initialized.
592: Forecasting file added.
593:
594: Revision 1.124 2006/03/22 17:13:53 lievre
595: Parameters are printed with %lf instead of %f (more numbers after the comma).
596: The log-likelihood is printed in the log file
597:
598: Revision 1.123 2006/03/20 10:52:43 brouard
599: * imach.c (Module): <title> changed, corresponds to .htm file
600: name. <head> headers where missing.
601:
602: * imach.c (Module): Weights can have a decimal point as for
603: English (a comma might work with a correct LC_NUMERIC environment,
604: otherwise the weight is truncated).
605: Modification of warning when the covariates values are not 0 or
606: 1.
607: Version 0.98g
608:
609: Revision 1.122 2006/03/20 09:45:41 brouard
610: (Module): Weights can have a decimal point as for
611: English (a comma might work with a correct LC_NUMERIC environment,
612: otherwise the weight is truncated).
613: Modification of warning when the covariates values are not 0 or
614: 1.
615: Version 0.98g
616:
617: Revision 1.121 2006/03/16 17:45:01 lievre
618: * imach.c (Module): Comments concerning covariates added
619:
620: * imach.c (Module): refinements in the computation of lli if
621: status=-2 in order to have more reliable computation if stepm is
622: not 1 month. Version 0.98f
623:
624: Revision 1.120 2006/03/16 15:10:38 lievre
625: (Module): refinements in the computation of lli if
626: status=-2 in order to have more reliable computation if stepm is
627: not 1 month. Version 0.98f
628:
629: Revision 1.119 2006/03/15 17:42:26 brouard
630: (Module): Bug if status = -2, the loglikelihood was
631: computed as likelihood omitting the logarithm. Version O.98e
632:
633: Revision 1.118 2006/03/14 18:20:07 brouard
634: (Module): varevsij Comments added explaining the second
635: table of variances if popbased=1 .
636: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
637: (Module): Function pstamp added
638: (Module): Version 0.98d
639:
640: Revision 1.117 2006/03/14 17:16:22 brouard
641: (Module): varevsij Comments added explaining the second
642: table of variances if popbased=1 .
643: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
644: (Module): Function pstamp added
645: (Module): Version 0.98d
646:
647: Revision 1.116 2006/03/06 10:29:27 brouard
648: (Module): Variance-covariance wrong links and
649: varian-covariance of ej. is needed (Saito).
650:
651: Revision 1.115 2006/02/27 12:17:45 brouard
652: (Module): One freematrix added in mlikeli! 0.98c
653:
654: Revision 1.114 2006/02/26 12:57:58 brouard
655: (Module): Some improvements in processing parameter
656: filename with strsep.
657:
658: Revision 1.113 2006/02/24 14:20:24 brouard
659: (Module): Memory leaks checks with valgrind and:
660: datafile was not closed, some imatrix were not freed and on matrix
661: allocation too.
662:
663: Revision 1.112 2006/01/30 09:55:26 brouard
664: (Module): Back to gnuplot.exe instead of wgnuplot.exe
665:
666: Revision 1.111 2006/01/25 20:38:18 brouard
667: (Module): Lots of cleaning and bugs added (Gompertz)
668: (Module): Comments can be added in data file. Missing date values
669: can be a simple dot '.'.
670:
671: Revision 1.110 2006/01/25 00:51:50 brouard
672: (Module): Lots of cleaning and bugs added (Gompertz)
673:
674: Revision 1.109 2006/01/24 19:37:15 brouard
675: (Module): Comments (lines starting with a #) are allowed in data.
676:
677: Revision 1.108 2006/01/19 18:05:42 lievre
678: Gnuplot problem appeared...
679: To be fixed
680:
681: Revision 1.107 2006/01/19 16:20:37 brouard
682: Test existence of gnuplot in imach path
683:
684: Revision 1.106 2006/01/19 13:24:36 brouard
685: Some cleaning and links added in html output
686:
687: Revision 1.105 2006/01/05 20:23:19 lievre
688: *** empty log message ***
689:
690: Revision 1.104 2005/09/30 16:11:43 lievre
691: (Module): sump fixed, loop imx fixed, and simplifications.
692: (Module): If the status is missing at the last wave but we know
693: that the person is alive, then we can code his/her status as -2
694: (instead of missing=-1 in earlier versions) and his/her
695: contributions to the likelihood is 1 - Prob of dying from last
696: health status (= 1-p13= p11+p12 in the easiest case of somebody in
697: the healthy state at last known wave). Version is 0.98
698:
699: Revision 1.103 2005/09/30 15:54:49 lievre
700: (Module): sump fixed, loop imx fixed, and simplifications.
701:
702: Revision 1.102 2004/09/15 17:31:30 brouard
703: Add the possibility to read data file including tab characters.
704:
705: Revision 1.101 2004/09/15 10:38:38 brouard
706: Fix on curr_time
707:
708: Revision 1.100 2004/07/12 18:29:06 brouard
709: Add version for Mac OS X. Just define UNIX in Makefile
710:
711: Revision 1.99 2004/06/05 08:57:40 brouard
712: *** empty log message ***
713:
714: Revision 1.98 2004/05/16 15:05:56 brouard
715: New version 0.97 . First attempt to estimate force of mortality
716: directly from the data i.e. without the need of knowing the health
717: state at each age, but using a Gompertz model: log u =a + b*age .
718: This is the basic analysis of mortality and should be done before any
719: other analysis, in order to test if the mortality estimated from the
720: cross-longitudinal survey is different from the mortality estimated
721: from other sources like vital statistic data.
722:
723: The same imach parameter file can be used but the option for mle should be -3.
724:
1.133 brouard 725: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 726: former routines in order to include the new code within the former code.
727:
728: The output is very simple: only an estimate of the intercept and of
729: the slope with 95% confident intervals.
730:
731: Current limitations:
732: A) Even if you enter covariates, i.e. with the
733: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
734: B) There is no computation of Life Expectancy nor Life Table.
735:
736: Revision 1.97 2004/02/20 13:25:42 lievre
737: Version 0.96d. Population forecasting command line is (temporarily)
738: suppressed.
739:
740: Revision 1.96 2003/07/15 15:38:55 brouard
741: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
742: rewritten within the same printf. Workaround: many printfs.
743:
744: Revision 1.95 2003/07/08 07:54:34 brouard
745: * imach.c (Repository):
746: (Repository): Using imachwizard code to output a more meaningful covariance
747: matrix (cov(a12,c31) instead of numbers.
748:
749: Revision 1.94 2003/06/27 13:00:02 brouard
750: Just cleaning
751:
752: Revision 1.93 2003/06/25 16:33:55 brouard
753: (Module): On windows (cygwin) function asctime_r doesn't
754: exist so I changed back to asctime which exists.
755: (Module): Version 0.96b
756:
757: Revision 1.92 2003/06/25 16:30:45 brouard
758: (Module): On windows (cygwin) function asctime_r doesn't
759: exist so I changed back to asctime which exists.
760:
761: Revision 1.91 2003/06/25 15:30:29 brouard
762: * imach.c (Repository): Duplicated warning errors corrected.
763: (Repository): Elapsed time after each iteration is now output. It
764: helps to forecast when convergence will be reached. Elapsed time
765: is stamped in powell. We created a new html file for the graphs
766: concerning matrix of covariance. It has extension -cov.htm.
767:
768: Revision 1.90 2003/06/24 12:34:15 brouard
769: (Module): Some bugs corrected for windows. Also, when
770: mle=-1 a template is output in file "or"mypar.txt with the design
771: of the covariance matrix to be input.
772:
773: Revision 1.89 2003/06/24 12:30:52 brouard
774: (Module): Some bugs corrected for windows. Also, when
775: mle=-1 a template is output in file "or"mypar.txt with the design
776: of the covariance matrix to be input.
777:
778: Revision 1.88 2003/06/23 17:54:56 brouard
779: * 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.
780:
781: Revision 1.87 2003/06/18 12:26:01 brouard
782: Version 0.96
783:
784: Revision 1.86 2003/06/17 20:04:08 brouard
785: (Module): Change position of html and gnuplot routines and added
786: routine fileappend.
787:
788: Revision 1.85 2003/06/17 13:12:43 brouard
789: * imach.c (Repository): Check when date of death was earlier that
790: current date of interview. It may happen when the death was just
791: prior to the death. In this case, dh was negative and likelihood
792: was wrong (infinity). We still send an "Error" but patch by
793: assuming that the date of death was just one stepm after the
794: interview.
795: (Repository): Because some people have very long ID (first column)
796: we changed int to long in num[] and we added a new lvector for
797: memory allocation. But we also truncated to 8 characters (left
798: truncation)
799: (Repository): No more line truncation errors.
800:
801: Revision 1.84 2003/06/13 21:44:43 brouard
802: * imach.c (Repository): Replace "freqsummary" at a correct
803: place. It differs from routine "prevalence" which may be called
804: many times. Probs is memory consuming and must be used with
805: parcimony.
806: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
807:
808: Revision 1.83 2003/06/10 13:39:11 lievre
809: *** empty log message ***
810:
811: Revision 1.82 2003/06/05 15:57:20 brouard
812: Add log in imach.c and fullversion number is now printed.
813:
814: */
815: /*
816: Interpolated Markov Chain
817:
818: Short summary of the programme:
819:
1.227 brouard 820: This program computes Healthy Life Expectancies or State-specific
821: (if states aren't health statuses) Expectancies from
822: cross-longitudinal data. Cross-longitudinal data consist in:
823:
824: -1- a first survey ("cross") where individuals from different ages
825: are interviewed on their health status or degree of disability (in
826: the case of a health survey which is our main interest)
827:
828: -2- at least a second wave of interviews ("longitudinal") which
829: measure each change (if any) in individual health status. Health
830: expectancies are computed from the time spent in each health state
831: according to a model. More health states you consider, more time is
832: necessary to reach the Maximum Likelihood of the parameters involved
833: in the model. The simplest model is the multinomial logistic model
834: where pij is the probability to be observed in state j at the second
835: wave conditional to be observed in state i at the first
836: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
837: etc , where 'age' is age and 'sex' is a covariate. If you want to
838: have a more complex model than "constant and age", you should modify
839: the program where the markup *Covariates have to be included here
840: again* invites you to do it. More covariates you add, slower the
1.126 brouard 841: convergence.
842:
843: The advantage of this computer programme, compared to a simple
844: multinomial logistic model, is clear when the delay between waves is not
845: identical for each individual. Also, if a individual missed an
846: intermediate interview, the information is lost, but taken into
847: account using an interpolation or extrapolation.
848:
849: hPijx is the probability to be observed in state i at age x+h
850: conditional to the observed state i at age x. The delay 'h' can be
851: split into an exact number (nh*stepm) of unobserved intermediate
852: states. This elementary transition (by month, quarter,
853: semester or year) is modelled as a multinomial logistic. The hPx
854: matrix is simply the matrix product of nh*stepm elementary matrices
855: and the contribution of each individual to the likelihood is simply
856: hPijx.
857:
858: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 859: of the life expectancies. It also computes the period (stable) prevalence.
860:
861: Back prevalence and projections:
1.227 brouard 862:
863: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
864: double agemaxpar, double ftolpl, int *ncvyearp, double
865: dateprev1,double dateprev2, int firstpass, int lastpass, int
866: mobilavproj)
867:
868: Computes the back prevalence limit for any combination of
869: covariate values k at any age between ageminpar and agemaxpar and
870: returns it in **bprlim. In the loops,
871:
872: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
873: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
874:
875: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 876: Computes for any combination of covariates k and any age between bage and fage
877: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
878: oldm=oldms;savm=savms;
1.227 brouard 879:
1.267 brouard 880: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 881: Computes the transition matrix starting at age 'age' over
882: 'nhstepm*hstepm*stepm' months (i.e. until
883: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 884: nhstepm*hstepm matrices.
885:
886: Returns p3mat[i][j][h] after calling
887: p3mat[i][j][h]=matprod2(newm,
888: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
889: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
890: oldm);
1.226 brouard 891:
892: Important routines
893:
894: - func (or funcone), computes logit (pij) distinguishing
895: o fixed variables (single or product dummies or quantitative);
896: o varying variables by:
897: (1) wave (single, product dummies, quantitative),
898: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
899: % fixed dummy (treated) or quantitative (not done because time-consuming);
900: % varying dummy (not done) or quantitative (not done);
901: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
902: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
903: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
904: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
905: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 906:
1.226 brouard 907:
908:
1.133 brouard 909: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
910: Institut national d'études démographiques, Paris.
1.126 brouard 911: This software have been partly granted by Euro-REVES, a concerted action
912: from the European Union.
913: It is copyrighted identically to a GNU software product, ie programme and
914: software can be distributed freely for non commercial use. Latest version
915: can be accessed at http://euroreves.ined.fr/imach .
916:
917: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
918: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
919:
920: **********************************************************************/
921: /*
922: main
923: read parameterfile
924: read datafile
925: concatwav
926: freqsummary
927: if (mle >= 1)
928: mlikeli
929: print results files
930: if mle==1
931: computes hessian
932: read end of parameter file: agemin, agemax, bage, fage, estepm
933: begin-prev-date,...
934: open gnuplot file
935: open html file
1.145 brouard 936: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
937: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
938: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
939: freexexit2 possible for memory heap.
940:
941: h Pij x | pij_nom ficrestpij
942: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
943: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
944: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
945:
946: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
947: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
948: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
949: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
950: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
951:
1.126 brouard 952: forecasting if prevfcast==1 prevforecast call prevalence()
953: health expectancies
954: Variance-covariance of DFLE
955: prevalence()
956: movingaverage()
957: varevsij()
958: if popbased==1 varevsij(,popbased)
959: total life expectancies
960: Variance of period (stable) prevalence
961: end
962: */
963:
1.187 brouard 964: /* #define DEBUG */
965: /* #define DEBUGBRENT */
1.203 brouard 966: /* #define DEBUGLINMIN */
967: /* #define DEBUGHESS */
968: #define DEBUGHESSIJ
1.224 brouard 969: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 970: #define POWELL /* Instead of NLOPT */
1.224 brouard 971: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 972: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
973: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 974:
975: #include <math.h>
976: #include <stdio.h>
977: #include <stdlib.h>
978: #include <string.h>
1.226 brouard 979: #include <ctype.h>
1.159 brouard 980:
981: #ifdef _WIN32
982: #include <io.h>
1.172 brouard 983: #include <windows.h>
984: #include <tchar.h>
1.159 brouard 985: #else
1.126 brouard 986: #include <unistd.h>
1.159 brouard 987: #endif
1.126 brouard 988:
989: #include <limits.h>
990: #include <sys/types.h>
1.171 brouard 991:
992: #if defined(__GNUC__)
993: #include <sys/utsname.h> /* Doesn't work on Windows */
994: #endif
995:
1.126 brouard 996: #include <sys/stat.h>
997: #include <errno.h>
1.159 brouard 998: /* extern int errno; */
1.126 brouard 999:
1.157 brouard 1000: /* #ifdef LINUX */
1001: /* #include <time.h> */
1002: /* #include "timeval.h" */
1003: /* #else */
1004: /* #include <sys/time.h> */
1005: /* #endif */
1006:
1.126 brouard 1007: #include <time.h>
1008:
1.136 brouard 1009: #ifdef GSL
1010: #include <gsl/gsl_errno.h>
1011: #include <gsl/gsl_multimin.h>
1012: #endif
1013:
1.167 brouard 1014:
1.162 brouard 1015: #ifdef NLOPT
1016: #include <nlopt.h>
1017: typedef struct {
1018: double (* function)(double [] );
1019: } myfunc_data ;
1020: #endif
1021:
1.126 brouard 1022: /* #include <libintl.h> */
1023: /* #define _(String) gettext (String) */
1024:
1.251 brouard 1025: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1026:
1027: #define GNUPLOTPROGRAM "gnuplot"
1028: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1029: #define FILENAMELENGTH 132
1030:
1031: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1032: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1033:
1.144 brouard 1034: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1035: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1036:
1037: #define NINTERVMAX 8
1.144 brouard 1038: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1039: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1040: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1041: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1042: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1043: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1044: #define MAXN 20000
1.144 brouard 1045: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1046: /* #define AGESUP 130 */
1047: #define AGESUP 150
1.268 brouard 1048: #define AGEINF 0
1.218 brouard 1049: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1050: #define AGEBASE 40
1.194 brouard 1051: #define AGEOVERFLOW 1.e20
1.164 brouard 1052: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1053: #ifdef _WIN32
1054: #define DIRSEPARATOR '\\'
1055: #define CHARSEPARATOR "\\"
1056: #define ODIRSEPARATOR '/'
1057: #else
1.126 brouard 1058: #define DIRSEPARATOR '/'
1059: #define CHARSEPARATOR "/"
1060: #define ODIRSEPARATOR '\\'
1061: #endif
1062:
1.286 ! brouard 1063: /* $Id: imach.c,v 1.285 2018/04/21 21:02:16 brouard Exp $ */
1.126 brouard 1064: /* $State: Exp $ */
1.196 brouard 1065: #include "version.h"
1066: char version[]=__IMACH_VERSION__;
1.283 brouard 1067: char copyright[]="April 2018,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2018";
1.286 ! brouard 1068: char fullversion[]="$Revision: 1.285 $ $Date: 2018/04/21 21:02:16 $";
1.126 brouard 1069: char strstart[80];
1070: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1071: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1072: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1073: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1074: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1075: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1076: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1077: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1078: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1079: int cptcovprodnoage=0; /**< Number of covariate products without age */
1080: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1081: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1082: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1083: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1084: int nsd=0; /**< Total number of single dummy variables (output) */
1085: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1086: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1087: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1088: int ntveff=0; /**< ntveff number of effective time varying variables */
1089: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1090: int cptcov=0; /* Working variable */
1.218 brouard 1091: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1092: int npar=NPARMAX;
1093: int nlstate=2; /* Number of live states */
1094: int ndeath=1; /* Number of dead states */
1.130 brouard 1095: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1096: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1097: int popbased=0;
1098:
1099: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1100: int maxwav=0; /* Maxim number of waves */
1101: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1102: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1103: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1104: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1105: int mle=1, weightopt=0;
1.126 brouard 1106: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1107: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1108: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1109: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1110: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1111: int selected(int kvar); /* Is covariate kvar selected for printing results */
1112:
1.130 brouard 1113: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1114: double **matprod2(); /* test */
1.126 brouard 1115: double **oldm, **newm, **savm; /* Working pointers to matrices */
1116: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1117: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1118:
1.136 brouard 1119: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1120: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1121: FILE *ficlog, *ficrespow;
1.130 brouard 1122: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1123: double fretone; /* Only one call to likelihood */
1.130 brouard 1124: long ipmx=0; /* Number of contributions */
1.126 brouard 1125: double sw; /* Sum of weights */
1126: char filerespow[FILENAMELENGTH];
1127: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1128: FILE *ficresilk;
1129: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1130: FILE *ficresprobmorprev;
1131: FILE *fichtm, *fichtmcov; /* Html File */
1132: FILE *ficreseij;
1133: char filerese[FILENAMELENGTH];
1134: FILE *ficresstdeij;
1135: char fileresstde[FILENAMELENGTH];
1136: FILE *ficrescveij;
1137: char filerescve[FILENAMELENGTH];
1138: FILE *ficresvij;
1139: char fileresv[FILENAMELENGTH];
1.269 brouard 1140:
1.126 brouard 1141: char title[MAXLINE];
1.234 brouard 1142: char model[MAXLINE]; /**< The model line */
1.217 brouard 1143: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1144: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1145: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1146: char command[FILENAMELENGTH];
1147: int outcmd=0;
1148:
1.217 brouard 1149: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1150: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1151: char filelog[FILENAMELENGTH]; /* Log file */
1152: char filerest[FILENAMELENGTH];
1153: char fileregp[FILENAMELENGTH];
1154: char popfile[FILENAMELENGTH];
1155:
1156: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1157:
1.157 brouard 1158: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1159: /* struct timezone tzp; */
1160: /* extern int gettimeofday(); */
1161: struct tm tml, *gmtime(), *localtime();
1162:
1163: extern time_t time();
1164:
1165: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1166: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1167: struct tm tm;
1168:
1.126 brouard 1169: char strcurr[80], strfor[80];
1170:
1171: char *endptr;
1172: long lval;
1173: double dval;
1174:
1175: #define NR_END 1
1176: #define FREE_ARG char*
1177: #define FTOL 1.0e-10
1178:
1179: #define NRANSI
1.240 brouard 1180: #define ITMAX 200
1181: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1182:
1183: #define TOL 2.0e-4
1184:
1185: #define CGOLD 0.3819660
1186: #define ZEPS 1.0e-10
1187: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1188:
1189: #define GOLD 1.618034
1190: #define GLIMIT 100.0
1191: #define TINY 1.0e-20
1192:
1193: static double maxarg1,maxarg2;
1194: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1195: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1196:
1197: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1198: #define rint(a) floor(a+0.5)
1.166 brouard 1199: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1200: #define mytinydouble 1.0e-16
1.166 brouard 1201: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1202: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1203: /* static double dsqrarg; */
1204: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1205: static double sqrarg;
1206: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1207: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1208: int agegomp= AGEGOMP;
1209:
1210: int imx;
1211: int stepm=1;
1212: /* Stepm, step in month: minimum step interpolation*/
1213:
1214: int estepm;
1215: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1216:
1217: int m,nb;
1218: long *num;
1.197 brouard 1219: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1220: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1221: covariate for which somebody answered excluding
1222: undefined. Usually 2: 0 and 1. */
1223: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1224: covariate for which somebody answered including
1225: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1226: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1227: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1228: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1229: double *ageexmed,*agecens;
1230: double dateintmean=0;
1231:
1232: double *weight;
1233: int **s; /* Status */
1.141 brouard 1234: double *agedc;
1.145 brouard 1235: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1236: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1237: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1238: double **coqvar; /* Fixed quantitative covariate nqv */
1239: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1240: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1241: double idx;
1242: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1243: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1244: /*k 1 2 3 4 5 6 7 8 9 */
1245: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1246: /* Tndvar[k] 1 2 3 4 5 */
1247: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1248: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1249: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1250: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1251: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1252: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1253: /* Tprod[i]=k 4 7 */
1254: /* Tage[i]=k 5 8 */
1255: /* */
1256: /* Type */
1257: /* V 1 2 3 4 5 */
1258: /* F F V V V */
1259: /* D Q D D Q */
1260: /* */
1261: int *TvarsD;
1262: int *TvarsDind;
1263: int *TvarsQ;
1264: int *TvarsQind;
1265:
1.235 brouard 1266: #define MAXRESULTLINES 10
1267: int nresult=0;
1.258 brouard 1268: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1269: int TKresult[MAXRESULTLINES];
1.237 brouard 1270: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1271: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1272: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1273: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1274: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1275: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1276:
1.234 brouard 1277: /* 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 1278: 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 */
1279: 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 */
1280: 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 */
1281: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1282: 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 */
1283: 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 1284: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1285: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1286: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1287: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1288: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1289: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1290: 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 */
1291: 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 */
1292:
1.230 brouard 1293: int *Tvarsel; /**< Selected covariates for output */
1294: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1295: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1296: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1297: 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 1298: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1299: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1300: int *Tage;
1.227 brouard 1301: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1302: 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 1303: 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*/
1304: 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 1305: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1306: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1307: int **Tvard;
1308: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1309: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1310: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1311: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1312: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1313: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1314: double *lsurv, *lpop, *tpop;
1315:
1.231 brouard 1316: #define FD 1; /* Fixed dummy covariate */
1317: #define FQ 2; /* Fixed quantitative covariate */
1318: #define FP 3; /* Fixed product covariate */
1319: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1320: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1321: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1322: #define VD 10; /* Varying dummy covariate */
1323: #define VQ 11; /* Varying quantitative covariate */
1324: #define VP 12; /* Varying product covariate */
1325: #define VPDD 13; /* Varying product dummy*dummy covariate */
1326: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1327: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1328: #define APFD 16; /* Age product * fixed dummy covariate */
1329: #define APFQ 17; /* Age product * fixed quantitative covariate */
1330: #define APVD 18; /* Age product * varying dummy covariate */
1331: #define APVQ 19; /* Age product * varying quantitative covariate */
1332:
1333: #define FTYPE 1; /* Fixed covariate */
1334: #define VTYPE 2; /* Varying covariate (loop in wave) */
1335: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1336:
1337: struct kmodel{
1338: int maintype; /* main type */
1339: int subtype; /* subtype */
1340: };
1341: struct kmodel modell[NCOVMAX];
1342:
1.143 brouard 1343: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1344: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1345:
1346: /**************** split *************************/
1347: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1348: {
1349: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1350: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1351: */
1352: char *ss; /* pointer */
1.186 brouard 1353: int l1=0, l2=0; /* length counters */
1.126 brouard 1354:
1355: l1 = strlen(path ); /* length of path */
1356: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1357: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1358: if ( ss == NULL ) { /* no directory, so determine current directory */
1359: strcpy( name, path ); /* we got the fullname name because no directory */
1360: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1361: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1362: /* get current working directory */
1363: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1364: #ifdef WIN32
1365: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1366: #else
1367: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1368: #endif
1.126 brouard 1369: return( GLOCK_ERROR_GETCWD );
1370: }
1371: /* got dirc from getcwd*/
1372: printf(" DIRC = %s \n",dirc);
1.205 brouard 1373: } else { /* strip directory from path */
1.126 brouard 1374: ss++; /* after this, the filename */
1375: l2 = strlen( ss ); /* length of filename */
1376: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1377: strcpy( name, ss ); /* save file name */
1378: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1379: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1380: printf(" DIRC2 = %s \n",dirc);
1381: }
1382: /* We add a separator at the end of dirc if not exists */
1383: l1 = strlen( dirc ); /* length of directory */
1384: if( dirc[l1-1] != DIRSEPARATOR ){
1385: dirc[l1] = DIRSEPARATOR;
1386: dirc[l1+1] = 0;
1387: printf(" DIRC3 = %s \n",dirc);
1388: }
1389: ss = strrchr( name, '.' ); /* find last / */
1390: if (ss >0){
1391: ss++;
1392: strcpy(ext,ss); /* save extension */
1393: l1= strlen( name);
1394: l2= strlen(ss)+1;
1395: strncpy( finame, name, l1-l2);
1396: finame[l1-l2]= 0;
1397: }
1398:
1399: return( 0 ); /* we're done */
1400: }
1401:
1402:
1403: /******************************************/
1404:
1405: void replace_back_to_slash(char *s, char*t)
1406: {
1407: int i;
1408: int lg=0;
1409: i=0;
1410: lg=strlen(t);
1411: for(i=0; i<= lg; i++) {
1412: (s[i] = t[i]);
1413: if (t[i]== '\\') s[i]='/';
1414: }
1415: }
1416:
1.132 brouard 1417: char *trimbb(char *out, char *in)
1.137 brouard 1418: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1419: char *s;
1420: s=out;
1421: while (*in != '\0'){
1.137 brouard 1422: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1423: in++;
1424: }
1425: *out++ = *in++;
1426: }
1427: *out='\0';
1428: return s;
1429: }
1430:
1.187 brouard 1431: /* char *substrchaine(char *out, char *in, char *chain) */
1432: /* { */
1433: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1434: /* char *s, *t; */
1435: /* t=in;s=out; */
1436: /* while ((*in != *chain) && (*in != '\0')){ */
1437: /* *out++ = *in++; */
1438: /* } */
1439:
1440: /* /\* *in matches *chain *\/ */
1441: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1442: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1443: /* } */
1444: /* in--; chain--; */
1445: /* while ( (*in != '\0')){ */
1446: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1447: /* *out++ = *in++; */
1448: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1449: /* } */
1450: /* *out='\0'; */
1451: /* out=s; */
1452: /* return out; */
1453: /* } */
1454: char *substrchaine(char *out, char *in, char *chain)
1455: {
1456: /* Substract chain 'chain' from 'in', return and output 'out' */
1457: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1458:
1459: char *strloc;
1460:
1461: strcpy (out, in);
1462: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1463: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1464: if(strloc != NULL){
1465: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1466: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1467: /* strcpy (strloc, strloc +strlen(chain));*/
1468: }
1469: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1470: return out;
1471: }
1472:
1473:
1.145 brouard 1474: char *cutl(char *blocc, char *alocc, char *in, char occ)
1475: {
1.187 brouard 1476: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1477: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1478: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1479: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1480: */
1.160 brouard 1481: char *s, *t;
1.145 brouard 1482: t=in;s=in;
1483: while ((*in != occ) && (*in != '\0')){
1484: *alocc++ = *in++;
1485: }
1486: if( *in == occ){
1487: *(alocc)='\0';
1488: s=++in;
1489: }
1490:
1491: if (s == t) {/* occ not found */
1492: *(alocc-(in-s))='\0';
1493: in=s;
1494: }
1495: while ( *in != '\0'){
1496: *blocc++ = *in++;
1497: }
1498:
1499: *blocc='\0';
1500: return t;
1501: }
1.137 brouard 1502: char *cutv(char *blocc, char *alocc, char *in, char occ)
1503: {
1.187 brouard 1504: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1505: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1506: gives blocc="abcdef2ghi" and alocc="j".
1507: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1508: */
1509: char *s, *t;
1510: t=in;s=in;
1511: while (*in != '\0'){
1512: while( *in == occ){
1513: *blocc++ = *in++;
1514: s=in;
1515: }
1516: *blocc++ = *in++;
1517: }
1518: if (s == t) /* occ not found */
1519: *(blocc-(in-s))='\0';
1520: else
1521: *(blocc-(in-s)-1)='\0';
1522: in=s;
1523: while ( *in != '\0'){
1524: *alocc++ = *in++;
1525: }
1526:
1527: *alocc='\0';
1528: return s;
1529: }
1530:
1.126 brouard 1531: int nbocc(char *s, char occ)
1532: {
1533: int i,j=0;
1534: int lg=20;
1535: i=0;
1536: lg=strlen(s);
1537: for(i=0; i<= lg; i++) {
1.234 brouard 1538: if (s[i] == occ ) j++;
1.126 brouard 1539: }
1540: return j;
1541: }
1542:
1.137 brouard 1543: /* void cutv(char *u,char *v, char*t, char occ) */
1544: /* { */
1545: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1546: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1547: /* gives u="abcdef2ghi" and v="j" *\/ */
1548: /* int i,lg,j,p=0; */
1549: /* i=0; */
1550: /* lg=strlen(t); */
1551: /* for(j=0; j<=lg-1; j++) { */
1552: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1553: /* } */
1.126 brouard 1554:
1.137 brouard 1555: /* for(j=0; j<p; j++) { */
1556: /* (u[j] = t[j]); */
1557: /* } */
1558: /* u[p]='\0'; */
1.126 brouard 1559:
1.137 brouard 1560: /* for(j=0; j<= lg; j++) { */
1561: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1562: /* } */
1563: /* } */
1.126 brouard 1564:
1.160 brouard 1565: #ifdef _WIN32
1566: char * strsep(char **pp, const char *delim)
1567: {
1568: char *p, *q;
1569:
1570: if ((p = *pp) == NULL)
1571: return 0;
1572: if ((q = strpbrk (p, delim)) != NULL)
1573: {
1574: *pp = q + 1;
1575: *q = '\0';
1576: }
1577: else
1578: *pp = 0;
1579: return p;
1580: }
1581: #endif
1582:
1.126 brouard 1583: /********************** nrerror ********************/
1584:
1585: void nrerror(char error_text[])
1586: {
1587: fprintf(stderr,"ERREUR ...\n");
1588: fprintf(stderr,"%s\n",error_text);
1589: exit(EXIT_FAILURE);
1590: }
1591: /*********************** vector *******************/
1592: double *vector(int nl, int nh)
1593: {
1594: double *v;
1595: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1596: if (!v) nrerror("allocation failure in vector");
1597: return v-nl+NR_END;
1598: }
1599:
1600: /************************ free vector ******************/
1601: void free_vector(double*v, int nl, int nh)
1602: {
1603: free((FREE_ARG)(v+nl-NR_END));
1604: }
1605:
1606: /************************ivector *******************************/
1607: int *ivector(long nl,long nh)
1608: {
1609: int *v;
1610: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1611: if (!v) nrerror("allocation failure in ivector");
1612: return v-nl+NR_END;
1613: }
1614:
1615: /******************free ivector **************************/
1616: void free_ivector(int *v, long nl, long nh)
1617: {
1618: free((FREE_ARG)(v+nl-NR_END));
1619: }
1620:
1621: /************************lvector *******************************/
1622: long *lvector(long nl,long nh)
1623: {
1624: long *v;
1625: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1626: if (!v) nrerror("allocation failure in ivector");
1627: return v-nl+NR_END;
1628: }
1629:
1630: /******************free lvector **************************/
1631: void free_lvector(long *v, long nl, long nh)
1632: {
1633: free((FREE_ARG)(v+nl-NR_END));
1634: }
1635:
1636: /******************* imatrix *******************************/
1637: int **imatrix(long nrl, long nrh, long ncl, long nch)
1638: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1639: {
1640: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1641: int **m;
1642:
1643: /* allocate pointers to rows */
1644: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1645: if (!m) nrerror("allocation failure 1 in matrix()");
1646: m += NR_END;
1647: m -= nrl;
1648:
1649:
1650: /* allocate rows and set pointers to them */
1651: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1652: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1653: m[nrl] += NR_END;
1654: m[nrl] -= ncl;
1655:
1656: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1657:
1658: /* return pointer to array of pointers to rows */
1659: return m;
1660: }
1661:
1662: /****************** free_imatrix *************************/
1663: void free_imatrix(m,nrl,nrh,ncl,nch)
1664: int **m;
1665: long nch,ncl,nrh,nrl;
1666: /* free an int matrix allocated by imatrix() */
1667: {
1668: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1669: free((FREE_ARG) (m+nrl-NR_END));
1670: }
1671:
1672: /******************* matrix *******************************/
1673: double **matrix(long nrl, long nrh, long ncl, long nch)
1674: {
1675: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1676: double **m;
1677:
1678: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1679: if (!m) nrerror("allocation failure 1 in matrix()");
1680: m += NR_END;
1681: m -= nrl;
1682:
1683: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1684: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1685: m[nrl] += NR_END;
1686: m[nrl] -= ncl;
1687:
1688: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1689: return m;
1.145 brouard 1690: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1691: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1692: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1693: */
1694: }
1695:
1696: /*************************free matrix ************************/
1697: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1698: {
1699: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1700: free((FREE_ARG)(m+nrl-NR_END));
1701: }
1702:
1703: /******************* ma3x *******************************/
1704: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1705: {
1706: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1707: double ***m;
1708:
1709: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1710: if (!m) nrerror("allocation failure 1 in matrix()");
1711: m += NR_END;
1712: m -= nrl;
1713:
1714: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1715: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1716: m[nrl] += NR_END;
1717: m[nrl] -= ncl;
1718:
1719: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1720:
1721: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1722: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1723: m[nrl][ncl] += NR_END;
1724: m[nrl][ncl] -= nll;
1725: for (j=ncl+1; j<=nch; j++)
1726: m[nrl][j]=m[nrl][j-1]+nlay;
1727:
1728: for (i=nrl+1; i<=nrh; i++) {
1729: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1730: for (j=ncl+1; j<=nch; j++)
1731: m[i][j]=m[i][j-1]+nlay;
1732: }
1733: return m;
1734: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1735: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1736: */
1737: }
1738:
1739: /*************************free ma3x ************************/
1740: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1741: {
1742: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1743: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1744: free((FREE_ARG)(m+nrl-NR_END));
1745: }
1746:
1747: /*************** function subdirf ***********/
1748: char *subdirf(char fileres[])
1749: {
1750: /* Caution optionfilefiname is hidden */
1751: strcpy(tmpout,optionfilefiname);
1752: strcat(tmpout,"/"); /* Add to the right */
1753: strcat(tmpout,fileres);
1754: return tmpout;
1755: }
1756:
1757: /*************** function subdirf2 ***********/
1758: char *subdirf2(char fileres[], char *preop)
1759: {
1760:
1761: /* Caution optionfilefiname is hidden */
1762: strcpy(tmpout,optionfilefiname);
1763: strcat(tmpout,"/");
1764: strcat(tmpout,preop);
1765: strcat(tmpout,fileres);
1766: return tmpout;
1767: }
1768:
1769: /*************** function subdirf3 ***********/
1770: char *subdirf3(char fileres[], char *preop, char *preop2)
1771: {
1772:
1773: /* Caution optionfilefiname is hidden */
1774: strcpy(tmpout,optionfilefiname);
1775: strcat(tmpout,"/");
1776: strcat(tmpout,preop);
1777: strcat(tmpout,preop2);
1778: strcat(tmpout,fileres);
1779: return tmpout;
1780: }
1.213 brouard 1781:
1782: /*************** function subdirfext ***********/
1783: char *subdirfext(char fileres[], char *preop, char *postop)
1784: {
1785:
1786: strcpy(tmpout,preop);
1787: strcat(tmpout,fileres);
1788: strcat(tmpout,postop);
1789: return tmpout;
1790: }
1.126 brouard 1791:
1.213 brouard 1792: /*************** function subdirfext3 ***********/
1793: char *subdirfext3(char fileres[], char *preop, char *postop)
1794: {
1795:
1796: /* Caution optionfilefiname is hidden */
1797: strcpy(tmpout,optionfilefiname);
1798: strcat(tmpout,"/");
1799: strcat(tmpout,preop);
1800: strcat(tmpout,fileres);
1801: strcat(tmpout,postop);
1802: return tmpout;
1803: }
1804:
1.162 brouard 1805: char *asc_diff_time(long time_sec, char ascdiff[])
1806: {
1807: long sec_left, days, hours, minutes;
1808: days = (time_sec) / (60*60*24);
1809: sec_left = (time_sec) % (60*60*24);
1810: hours = (sec_left) / (60*60) ;
1811: sec_left = (sec_left) %(60*60);
1812: minutes = (sec_left) /60;
1813: sec_left = (sec_left) % (60);
1814: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1815: return ascdiff;
1816: }
1817:
1.126 brouard 1818: /***************** f1dim *************************/
1819: extern int ncom;
1820: extern double *pcom,*xicom;
1821: extern double (*nrfunc)(double []);
1822:
1823: double f1dim(double x)
1824: {
1825: int j;
1826: double f;
1827: double *xt;
1828:
1829: xt=vector(1,ncom);
1830: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1831: f=(*nrfunc)(xt);
1832: free_vector(xt,1,ncom);
1833: return f;
1834: }
1835:
1836: /*****************brent *************************/
1837: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1838: {
1839: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1840: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1841: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1842: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1843: * returned function value.
1844: */
1.126 brouard 1845: int iter;
1846: double a,b,d,etemp;
1.159 brouard 1847: double fu=0,fv,fw,fx;
1.164 brouard 1848: double ftemp=0.;
1.126 brouard 1849: double p,q,r,tol1,tol2,u,v,w,x,xm;
1850: double e=0.0;
1851:
1852: a=(ax < cx ? ax : cx);
1853: b=(ax > cx ? ax : cx);
1854: x=w=v=bx;
1855: fw=fv=fx=(*f)(x);
1856: for (iter=1;iter<=ITMAX;iter++) {
1857: xm=0.5*(a+b);
1858: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1859: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1860: printf(".");fflush(stdout);
1861: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1862: #ifdef DEBUGBRENT
1.126 brouard 1863: 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);
1864: 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);
1865: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1866: #endif
1867: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1868: *xmin=x;
1869: return fx;
1870: }
1871: ftemp=fu;
1872: if (fabs(e) > tol1) {
1873: r=(x-w)*(fx-fv);
1874: q=(x-v)*(fx-fw);
1875: p=(x-v)*q-(x-w)*r;
1876: q=2.0*(q-r);
1877: if (q > 0.0) p = -p;
1878: q=fabs(q);
1879: etemp=e;
1880: e=d;
1881: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1882: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1883: else {
1.224 brouard 1884: d=p/q;
1885: u=x+d;
1886: if (u-a < tol2 || b-u < tol2)
1887: d=SIGN(tol1,xm-x);
1.126 brouard 1888: }
1889: } else {
1890: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1891: }
1892: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1893: fu=(*f)(u);
1894: if (fu <= fx) {
1895: if (u >= x) a=x; else b=x;
1896: SHFT(v,w,x,u)
1.183 brouard 1897: SHFT(fv,fw,fx,fu)
1898: } else {
1899: if (u < x) a=u; else b=u;
1900: if (fu <= fw || w == x) {
1.224 brouard 1901: v=w;
1902: w=u;
1903: fv=fw;
1904: fw=fu;
1.183 brouard 1905: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1906: v=u;
1907: fv=fu;
1.183 brouard 1908: }
1909: }
1.126 brouard 1910: }
1911: nrerror("Too many iterations in brent");
1912: *xmin=x;
1913: return fx;
1914: }
1915:
1916: /****************** mnbrak ***********************/
1917:
1918: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1919: double (*func)(double))
1.183 brouard 1920: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1921: the downhill direction (defined by the function as evaluated at the initial points) and returns
1922: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1923: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1924: */
1.126 brouard 1925: double ulim,u,r,q, dum;
1926: double fu;
1.187 brouard 1927:
1928: double scale=10.;
1929: int iterscale=0;
1930:
1931: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1932: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1933:
1934:
1935: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1936: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1937: /* *bx = *ax - (*ax - *bx)/scale; */
1938: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1939: /* } */
1940:
1.126 brouard 1941: if (*fb > *fa) {
1942: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1943: SHFT(dum,*fb,*fa,dum)
1944: }
1.126 brouard 1945: *cx=(*bx)+GOLD*(*bx-*ax);
1946: *fc=(*func)(*cx);
1.183 brouard 1947: #ifdef DEBUG
1.224 brouard 1948: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1949: 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 1950: #endif
1.224 brouard 1951: 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 1952: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1953: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1954: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1955: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1956: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1957: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1958: fu=(*func)(u);
1.163 brouard 1959: #ifdef DEBUG
1960: /* f(x)=A(x-u)**2+f(u) */
1961: double A, fparabu;
1962: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1963: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1964: 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);
1965: 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 1966: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1967: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1968: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1969: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1970: #endif
1.184 brouard 1971: #ifdef MNBRAKORIGINAL
1.183 brouard 1972: #else
1.191 brouard 1973: /* if (fu > *fc) { */
1974: /* #ifdef DEBUG */
1975: /* printf("mnbrak4 fu > fc \n"); */
1976: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1977: /* #endif */
1978: /* /\* 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 *\\/ *\/ */
1979: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1980: /* dum=u; /\* Shifting c and u *\/ */
1981: /* u = *cx; */
1982: /* *cx = dum; */
1983: /* dum = fu; */
1984: /* fu = *fc; */
1985: /* *fc =dum; */
1986: /* } else { /\* end *\/ */
1987: /* #ifdef DEBUG */
1988: /* printf("mnbrak3 fu < fc \n"); */
1989: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1990: /* #endif */
1991: /* dum=u; /\* Shifting c and u *\/ */
1992: /* u = *cx; */
1993: /* *cx = dum; */
1994: /* dum = fu; */
1995: /* fu = *fc; */
1996: /* *fc =dum; */
1997: /* } */
1.224 brouard 1998: #ifdef DEBUGMNBRAK
1999: double A, fparabu;
2000: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2001: fparabu= *fa - A*(*ax-u)*(*ax-u);
2002: 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);
2003: 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 2004: #endif
1.191 brouard 2005: dum=u; /* Shifting c and u */
2006: u = *cx;
2007: *cx = dum;
2008: dum = fu;
2009: fu = *fc;
2010: *fc =dum;
1.183 brouard 2011: #endif
1.162 brouard 2012: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2013: #ifdef DEBUG
1.224 brouard 2014: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2015: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2016: #endif
1.126 brouard 2017: fu=(*func)(u);
2018: if (fu < *fc) {
1.183 brouard 2019: #ifdef DEBUG
1.224 brouard 2020: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2021: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2022: #endif
2023: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2024: SHFT(*fb,*fc,fu,(*func)(u))
2025: #ifdef DEBUG
2026: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2027: #endif
2028: }
1.162 brouard 2029: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2030: #ifdef DEBUG
1.224 brouard 2031: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2032: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2033: #endif
1.126 brouard 2034: u=ulim;
2035: fu=(*func)(u);
1.183 brouard 2036: } else { /* u could be left to b (if r > q parabola has a maximum) */
2037: #ifdef DEBUG
1.224 brouard 2038: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2039: 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 2040: #endif
1.126 brouard 2041: u=(*cx)+GOLD*(*cx-*bx);
2042: fu=(*func)(u);
1.224 brouard 2043: #ifdef DEBUG
2044: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2045: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2046: #endif
1.183 brouard 2047: } /* end tests */
1.126 brouard 2048: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2049: SHFT(*fa,*fb,*fc,fu)
2050: #ifdef DEBUG
1.224 brouard 2051: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2052: 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 2053: #endif
2054: } /* 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 2055: }
2056:
2057: /*************** linmin ************************/
1.162 brouard 2058: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2059: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2060: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2061: the value of func at the returned location p . This is actually all accomplished by calling the
2062: routines mnbrak and brent .*/
1.126 brouard 2063: int ncom;
2064: double *pcom,*xicom;
2065: double (*nrfunc)(double []);
2066:
1.224 brouard 2067: #ifdef LINMINORIGINAL
1.126 brouard 2068: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2069: #else
2070: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2071: #endif
1.126 brouard 2072: {
2073: double brent(double ax, double bx, double cx,
2074: double (*f)(double), double tol, double *xmin);
2075: double f1dim(double x);
2076: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2077: double *fc, double (*func)(double));
2078: int j;
2079: double xx,xmin,bx,ax;
2080: double fx,fb,fa;
1.187 brouard 2081:
1.203 brouard 2082: #ifdef LINMINORIGINAL
2083: #else
2084: double scale=10., axs, xxs; /* Scale added for infinity */
2085: #endif
2086:
1.126 brouard 2087: ncom=n;
2088: pcom=vector(1,n);
2089: xicom=vector(1,n);
2090: nrfunc=func;
2091: for (j=1;j<=n;j++) {
2092: pcom[j]=p[j];
1.202 brouard 2093: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2094: }
1.187 brouard 2095:
1.203 brouard 2096: #ifdef LINMINORIGINAL
2097: xx=1.;
2098: #else
2099: axs=0.0;
2100: xxs=1.;
2101: do{
2102: xx= xxs;
2103: #endif
1.187 brouard 2104: ax=0.;
2105: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2106: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2107: /* 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)) */
2108: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2109: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2110: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2111: /* 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 2112: #ifdef LINMINORIGINAL
2113: #else
2114: if (fx != fx){
1.224 brouard 2115: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2116: printf("|");
2117: fprintf(ficlog,"|");
1.203 brouard 2118: #ifdef DEBUGLINMIN
1.224 brouard 2119: 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 2120: #endif
2121: }
1.224 brouard 2122: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2123: #endif
2124:
1.191 brouard 2125: #ifdef DEBUGLINMIN
2126: 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 2127: 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 2128: #endif
1.224 brouard 2129: #ifdef LINMINORIGINAL
2130: #else
2131: if(fb == fx){ /* Flat function in the direction */
2132: xmin=xx;
2133: *flat=1;
2134: }else{
2135: *flat=0;
2136: #endif
2137: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2138: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2139: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2140: /* fmin = f(p[j] + xmin * xi[j]) */
2141: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2142: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2143: #ifdef DEBUG
1.224 brouard 2144: 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);
2145: 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);
2146: #endif
2147: #ifdef LINMINORIGINAL
2148: #else
2149: }
1.126 brouard 2150: #endif
1.191 brouard 2151: #ifdef DEBUGLINMIN
2152: printf("linmin end ");
1.202 brouard 2153: fprintf(ficlog,"linmin end ");
1.191 brouard 2154: #endif
1.126 brouard 2155: for (j=1;j<=n;j++) {
1.203 brouard 2156: #ifdef LINMINORIGINAL
2157: xi[j] *= xmin;
2158: #else
2159: #ifdef DEBUGLINMIN
2160: if(xxs <1.0)
2161: printf(" before xi[%d]=%12.8f", j,xi[j]);
2162: #endif
2163: 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) */
2164: #ifdef DEBUGLINMIN
2165: if(xxs <1.0)
2166: 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 );
2167: #endif
2168: #endif
1.187 brouard 2169: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2170: }
1.191 brouard 2171: #ifdef DEBUGLINMIN
1.203 brouard 2172: printf("\n");
1.191 brouard 2173: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2174: 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 2175: for (j=1;j<=n;j++) {
1.202 brouard 2176: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2177: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2178: if(j % ncovmodel == 0){
1.191 brouard 2179: printf("\n");
1.202 brouard 2180: fprintf(ficlog,"\n");
2181: }
1.191 brouard 2182: }
1.203 brouard 2183: #else
1.191 brouard 2184: #endif
1.126 brouard 2185: free_vector(xicom,1,n);
2186: free_vector(pcom,1,n);
2187: }
2188:
2189:
2190: /*************** powell ************************/
1.162 brouard 2191: /*
2192: Minimization of a function func of n variables. Input consists of an initial starting point
2193: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2194: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2195: such that failure to decrease by more than this amount on one iteration signals doneness. On
2196: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2197: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2198: */
1.224 brouard 2199: #ifdef LINMINORIGINAL
2200: #else
2201: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2202: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2203: #endif
1.126 brouard 2204: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2205: double (*func)(double []))
2206: {
1.224 brouard 2207: #ifdef LINMINORIGINAL
2208: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2209: double (*func)(double []));
1.224 brouard 2210: #else
1.241 brouard 2211: void linmin(double p[], double xi[], int n, double *fret,
2212: double (*func)(double []),int *flat);
1.224 brouard 2213: #endif
1.239 brouard 2214: int i,ibig,j,jk,k;
1.126 brouard 2215: double del,t,*pt,*ptt,*xit;
1.181 brouard 2216: double directest;
1.126 brouard 2217: double fp,fptt;
2218: double *xits;
2219: int niterf, itmp;
1.224 brouard 2220: #ifdef LINMINORIGINAL
2221: #else
2222:
2223: flatdir=ivector(1,n);
2224: for (j=1;j<=n;j++) flatdir[j]=0;
2225: #endif
1.126 brouard 2226:
2227: pt=vector(1,n);
2228: ptt=vector(1,n);
2229: xit=vector(1,n);
2230: xits=vector(1,n);
2231: *fret=(*func)(p);
2232: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2233: rcurr_time = time(NULL);
1.126 brouard 2234: for (*iter=1;;++(*iter)) {
1.187 brouard 2235: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2236: ibig=0;
2237: del=0.0;
1.157 brouard 2238: rlast_time=rcurr_time;
2239: /* (void) gettimeofday(&curr_time,&tzp); */
2240: rcurr_time = time(NULL);
2241: curr_time = *localtime(&rcurr_time);
2242: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2243: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2244: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2245: for (i=1;i<=n;i++) {
1.126 brouard 2246: fprintf(ficrespow," %.12lf", p[i]);
2247: }
1.239 brouard 2248: fprintf(ficrespow,"\n");fflush(ficrespow);
2249: printf("\n#model= 1 + age ");
2250: fprintf(ficlog,"\n#model= 1 + age ");
2251: if(nagesqr==1){
1.241 brouard 2252: printf(" + age*age ");
2253: fprintf(ficlog," + age*age ");
1.239 brouard 2254: }
2255: for(j=1;j <=ncovmodel-2;j++){
2256: if(Typevar[j]==0) {
2257: printf(" + V%d ",Tvar[j]);
2258: fprintf(ficlog," + V%d ",Tvar[j]);
2259: }else if(Typevar[j]==1) {
2260: printf(" + V%d*age ",Tvar[j]);
2261: fprintf(ficlog," + V%d*age ",Tvar[j]);
2262: }else if(Typevar[j]==2) {
2263: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2264: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2265: }
2266: }
1.126 brouard 2267: printf("\n");
1.239 brouard 2268: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2269: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2270: fprintf(ficlog,"\n");
1.239 brouard 2271: for(i=1,jk=1; i <=nlstate; i++){
2272: for(k=1; k <=(nlstate+ndeath); k++){
2273: if (k != i) {
2274: printf("%d%d ",i,k);
2275: fprintf(ficlog,"%d%d ",i,k);
2276: for(j=1; j <=ncovmodel; j++){
2277: printf("%12.7f ",p[jk]);
2278: fprintf(ficlog,"%12.7f ",p[jk]);
2279: jk++;
2280: }
2281: printf("\n");
2282: fprintf(ficlog,"\n");
2283: }
2284: }
2285: }
1.241 brouard 2286: if(*iter <=3 && *iter >1){
1.157 brouard 2287: tml = *localtime(&rcurr_time);
2288: strcpy(strcurr,asctime(&tml));
2289: rforecast_time=rcurr_time;
1.126 brouard 2290: itmp = strlen(strcurr);
2291: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2292: strcurr[itmp-1]='\0';
1.162 brouard 2293: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2294: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2295: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2296: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2297: forecast_time = *localtime(&rforecast_time);
2298: strcpy(strfor,asctime(&forecast_time));
2299: itmp = strlen(strfor);
2300: if(strfor[itmp-1]=='\n')
2301: strfor[itmp-1]='\0';
2302: 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);
2303: 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 2304: }
2305: }
1.187 brouard 2306: for (i=1;i<=n;i++) { /* For each direction i */
2307: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2308: fptt=(*fret);
2309: #ifdef DEBUG
1.203 brouard 2310: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2311: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2312: #endif
1.203 brouard 2313: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2314: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2315: #ifdef LINMINORIGINAL
1.188 brouard 2316: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2317: #else
2318: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2319: flatdir[i]=flat; /* Function is vanishing in that direction i */
2320: #endif
2321: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2322: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2323: /* because that direction will be replaced unless the gain del is small */
2324: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2325: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2326: /* with the new direction. */
2327: del=fabs(fptt-(*fret));
2328: ibig=i;
1.126 brouard 2329: }
2330: #ifdef DEBUG
2331: printf("%d %.12e",i,(*fret));
2332: fprintf(ficlog,"%d %.12e",i,(*fret));
2333: for (j=1;j<=n;j++) {
1.224 brouard 2334: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2335: printf(" x(%d)=%.12e",j,xit[j]);
2336: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2337: }
2338: for(j=1;j<=n;j++) {
1.225 brouard 2339: printf(" p(%d)=%.12e",j,p[j]);
2340: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2341: }
2342: printf("\n");
2343: fprintf(ficlog,"\n");
2344: #endif
1.187 brouard 2345: } /* end loop on each direction i */
2346: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2347: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2348: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2349: for(j=1;j<=n;j++) {
1.225 brouard 2350: if(flatdir[j] >0){
2351: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2352: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2353: }
2354: /* printf("\n"); */
2355: /* fprintf(ficlog,"\n"); */
2356: }
1.243 brouard 2357: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2358: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2359: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2360: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2361: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2362: /* decreased of more than 3.84 */
2363: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2364: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2365: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2366:
1.188 brouard 2367: /* Starting the program with initial values given by a former maximization will simply change */
2368: /* the scales of the directions and the directions, because the are reset to canonical directions */
2369: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2370: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2371: #ifdef DEBUG
2372: int k[2],l;
2373: k[0]=1;
2374: k[1]=-1;
2375: printf("Max: %.12e",(*func)(p));
2376: fprintf(ficlog,"Max: %.12e",(*func)(p));
2377: for (j=1;j<=n;j++) {
2378: printf(" %.12e",p[j]);
2379: fprintf(ficlog," %.12e",p[j]);
2380: }
2381: printf("\n");
2382: fprintf(ficlog,"\n");
2383: for(l=0;l<=1;l++) {
2384: for (j=1;j<=n;j++) {
2385: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2386: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2387: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2388: }
2389: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2390: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2391: }
2392: #endif
2393:
1.224 brouard 2394: #ifdef LINMINORIGINAL
2395: #else
2396: free_ivector(flatdir,1,n);
2397: #endif
1.126 brouard 2398: free_vector(xit,1,n);
2399: free_vector(xits,1,n);
2400: free_vector(ptt,1,n);
2401: free_vector(pt,1,n);
2402: return;
1.192 brouard 2403: } /* enough precision */
1.240 brouard 2404: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2405: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2406: ptt[j]=2.0*p[j]-pt[j];
2407: xit[j]=p[j]-pt[j];
2408: pt[j]=p[j];
2409: }
1.181 brouard 2410: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2411: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2412: if (*iter <=4) {
1.225 brouard 2413: #else
2414: #endif
1.224 brouard 2415: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2416: #else
1.161 brouard 2417: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2418: #endif
1.162 brouard 2419: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2420: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2421: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2422: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2423: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2424: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2425: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2426: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2427: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2428: /* Even if f3 <f1, directest can be negative and t >0 */
2429: /* mu² and del² are equal when f3=f1 */
2430: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2431: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2432: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2433: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2434: #ifdef NRCORIGINAL
2435: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2436: #else
2437: 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 2438: t= t- del*SQR(fp-fptt);
1.183 brouard 2439: #endif
1.202 brouard 2440: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2441: #ifdef DEBUG
1.181 brouard 2442: 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);
2443: 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 2444: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2445: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2446: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2447: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2448: 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);
2449: 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);
2450: #endif
1.183 brouard 2451: #ifdef POWELLORIGINAL
2452: if (t < 0.0) { /* Then we use it for new direction */
2453: #else
1.182 brouard 2454: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2455: 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 2456: 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 2457: 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 2458: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2459: }
1.181 brouard 2460: if (directest < 0.0) { /* Then we use it for new direction */
2461: #endif
1.191 brouard 2462: #ifdef DEBUGLINMIN
1.234 brouard 2463: printf("Before linmin in direction P%d-P0\n",n);
2464: for (j=1;j<=n;j++) {
2465: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2466: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2467: if(j % ncovmodel == 0){
2468: printf("\n");
2469: fprintf(ficlog,"\n");
2470: }
2471: }
1.224 brouard 2472: #endif
2473: #ifdef LINMINORIGINAL
1.234 brouard 2474: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2475: #else
1.234 brouard 2476: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2477: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2478: #endif
1.234 brouard 2479:
1.191 brouard 2480: #ifdef DEBUGLINMIN
1.234 brouard 2481: for (j=1;j<=n;j++) {
2482: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2483: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2484: if(j % ncovmodel == 0){
2485: printf("\n");
2486: fprintf(ficlog,"\n");
2487: }
2488: }
1.224 brouard 2489: #endif
1.234 brouard 2490: for (j=1;j<=n;j++) {
2491: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2492: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2493: }
1.224 brouard 2494: #ifdef LINMINORIGINAL
2495: #else
1.234 brouard 2496: for (j=1, flatd=0;j<=n;j++) {
2497: if(flatdir[j]>0)
2498: flatd++;
2499: }
2500: if(flatd >0){
1.255 brouard 2501: printf("%d flat directions: ",flatd);
2502: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2503: for (j=1;j<=n;j++) {
2504: if(flatdir[j]>0){
2505: printf("%d ",j);
2506: fprintf(ficlog,"%d ",j);
2507: }
2508: }
2509: printf("\n");
2510: fprintf(ficlog,"\n");
2511: }
1.191 brouard 2512: #endif
1.234 brouard 2513: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2514: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2515:
1.126 brouard 2516: #ifdef DEBUG
1.234 brouard 2517: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2518: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2519: for(j=1;j<=n;j++){
2520: printf(" %lf",xit[j]);
2521: fprintf(ficlog," %lf",xit[j]);
2522: }
2523: printf("\n");
2524: fprintf(ficlog,"\n");
1.126 brouard 2525: #endif
1.192 brouard 2526: } /* end of t or directest negative */
1.224 brouard 2527: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2528: #else
1.234 brouard 2529: } /* end if (fptt < fp) */
1.192 brouard 2530: #endif
1.225 brouard 2531: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2532: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2533: #else
1.224 brouard 2534: #endif
1.234 brouard 2535: } /* loop iteration */
1.126 brouard 2536: }
1.234 brouard 2537:
1.126 brouard 2538: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2539:
1.235 brouard 2540: 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 2541: {
1.279 brouard 2542: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2543: * (and selected quantitative values in nres)
2544: * by left multiplying the unit
2545: * matrix by transitions matrix until convergence is reached with precision ftolpl
2546: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2547: * Wx is row vector: population in state 1, population in state 2, population dead
2548: * or prevalence in state 1, prevalence in state 2, 0
2549: * newm is the matrix after multiplications, its rows are identical at a factor.
2550: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2551: * Output is prlim.
2552: * Initial matrix pimij
2553: */
1.206 brouard 2554: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2555: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2556: /* 0, 0 , 1} */
2557: /*
2558: * and after some iteration: */
2559: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2560: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2561: /* 0, 0 , 1} */
2562: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2563: /* {0.51571254859325999, 0.4842874514067399, */
2564: /* 0.51326036147820708, 0.48673963852179264} */
2565: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2566:
1.126 brouard 2567: int i, ii,j,k;
1.209 brouard 2568: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2569: /* double **matprod2(); */ /* test */
1.218 brouard 2570: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2571: double **newm;
1.209 brouard 2572: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2573: int ncvloop=0;
1.169 brouard 2574:
1.209 brouard 2575: min=vector(1,nlstate);
2576: max=vector(1,nlstate);
2577: meandiff=vector(1,nlstate);
2578:
1.218 brouard 2579: /* Starting with matrix unity */
1.126 brouard 2580: for (ii=1;ii<=nlstate+ndeath;ii++)
2581: for (j=1;j<=nlstate+ndeath;j++){
2582: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2583: }
1.169 brouard 2584:
2585: cov[1]=1.;
2586:
2587: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2588: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2589: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2590: ncvloop++;
1.126 brouard 2591: newm=savm;
2592: /* Covariates have to be included here again */
1.138 brouard 2593: cov[2]=agefin;
1.187 brouard 2594: if(nagesqr==1)
2595: cov[3]= agefin*agefin;;
1.234 brouard 2596: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2597: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2598: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2599: /* 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 2600: }
2601: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2602: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2603: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2604: /* 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 2605: }
1.237 brouard 2606: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2607: if(Dummy[Tvar[Tage[k]]]){
2608: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2609: } else{
1.235 brouard 2610: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2611: }
1.235 brouard 2612: /* 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 2613: }
1.237 brouard 2614: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2615: /* 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 2616: if(Dummy[Tvard[k][1]==0]){
2617: if(Dummy[Tvard[k][2]==0]){
2618: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2619: }else{
2620: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2621: }
2622: }else{
2623: if(Dummy[Tvard[k][2]==0]){
2624: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2625: }else{
2626: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2627: }
2628: }
1.234 brouard 2629: }
1.138 brouard 2630: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2631: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2632: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2633: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2634: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2635: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2636: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2637:
1.126 brouard 2638: savm=oldm;
2639: oldm=newm;
1.209 brouard 2640:
2641: for(j=1; j<=nlstate; j++){
2642: max[j]=0.;
2643: min[j]=1.;
2644: }
2645: for(i=1;i<=nlstate;i++){
2646: sumnew=0;
2647: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2648: for(j=1; j<=nlstate; j++){
2649: prlim[i][j]= newm[i][j]/(1-sumnew);
2650: max[j]=FMAX(max[j],prlim[i][j]);
2651: min[j]=FMIN(min[j],prlim[i][j]);
2652: }
2653: }
2654:
1.126 brouard 2655: maxmax=0.;
1.209 brouard 2656: for(j=1; j<=nlstate; j++){
2657: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2658: maxmax=FMAX(maxmax,meandiff[j]);
2659: /* 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 2660: } /* j loop */
1.203 brouard 2661: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2662: /* 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 2663: if(maxmax < ftolpl){
1.209 brouard 2664: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2665: free_vector(min,1,nlstate);
2666: free_vector(max,1,nlstate);
2667: free_vector(meandiff,1,nlstate);
1.126 brouard 2668: return prlim;
2669: }
1.169 brouard 2670: } /* age loop */
1.208 brouard 2671: /* After some age loop it doesn't converge */
1.209 brouard 2672: printf("Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. \n\
1.208 brouard 2673: Earliest 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);
1.209 brouard 2674: /* 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); */
2675: free_vector(min,1,nlstate);
2676: free_vector(max,1,nlstate);
2677: free_vector(meandiff,1,nlstate);
1.208 brouard 2678:
1.169 brouard 2679: return prlim; /* should not reach here */
1.126 brouard 2680: }
2681:
1.217 brouard 2682:
2683: /**** Back Prevalence limit (stable or period prevalence) ****************/
2684:
1.218 brouard 2685: /* 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) */
2686: /* 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 2687: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2688: {
1.264 brouard 2689: /* 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 2690: matrix by transitions matrix until convergence is reached with precision ftolpl */
2691: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2692: /* Wx is row vector: population in state 1, population in state 2, population dead */
2693: /* or prevalence in state 1, prevalence in state 2, 0 */
2694: /* newm is the matrix after multiplications, its rows are identical at a factor */
2695: /* Initial matrix pimij */
2696: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2697: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2698: /* 0, 0 , 1} */
2699: /*
2700: * and after some iteration: */
2701: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2702: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2703: /* 0, 0 , 1} */
2704: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2705: /* {0.51571254859325999, 0.4842874514067399, */
2706: /* 0.51326036147820708, 0.48673963852179264} */
2707: /* If we start from prlim again, prlim tends to a constant matrix */
2708:
2709: int i, ii,j,k;
1.247 brouard 2710: int first=0;
1.217 brouard 2711: double *min, *max, *meandiff, maxmax,sumnew=0.;
2712: /* double **matprod2(); */ /* test */
2713: double **out, cov[NCOVMAX+1], **bmij();
2714: double **newm;
1.218 brouard 2715: double **dnewm, **doldm, **dsavm; /* for use */
2716: double **oldm, **savm; /* for use */
2717:
1.217 brouard 2718: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2719: int ncvloop=0;
2720:
2721: min=vector(1,nlstate);
2722: max=vector(1,nlstate);
2723: meandiff=vector(1,nlstate);
2724:
1.266 brouard 2725: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2726: oldm=oldms; savm=savms;
2727:
2728: /* Starting with matrix unity */
2729: for (ii=1;ii<=nlstate+ndeath;ii++)
2730: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2731: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2732: }
2733:
2734: cov[1]=1.;
2735:
2736: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2737: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2738: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2739: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2740: ncvloop++;
1.218 brouard 2741: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2742: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2743: /* Covariates have to be included here again */
2744: cov[2]=agefin;
2745: if(nagesqr==1)
2746: cov[3]= agefin*agefin;;
1.242 brouard 2747: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2748: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2749: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2750: /* 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 2751: }
2752: /* for (k=1; k<=cptcovn;k++) { */
2753: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2754: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2755: /* /\* 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])]); *\/ */
2756: /* } */
2757: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2758: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2759: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2760: /* 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]); */
2761: }
2762: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2763: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2764: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2765: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2766: for (k=1; k<=cptcovage;k++){ /* For product with age */
2767: if(Dummy[Tvar[Tage[k]]]){
2768: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2769: } else{
2770: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2771: }
2772: /* 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]); */
2773: }
2774: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2775: /* 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]); */
2776: if(Dummy[Tvard[k][1]==0]){
2777: if(Dummy[Tvard[k][2]==0]){
2778: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2779: }else{
2780: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2781: }
2782: }else{
2783: if(Dummy[Tvard[k][2]==0]){
2784: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2785: }else{
2786: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2787: }
2788: }
1.217 brouard 2789: }
2790:
2791: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2792: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2793: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2794: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2795: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2796: /* ij should be linked to the correct index of cov */
2797: /* age and covariate values ij are in 'cov', but we need to pass
2798: * ij for the observed prevalence at age and status and covariate
2799: * number: prevacurrent[(int)agefin][ii][ij]
2800: */
2801: /* 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 *\/ */
2802: /* 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 *\/ */
2803: 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 2804: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2805: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2806: /* for(i=1; i<=nlstate+ndeath; i++) { */
2807: /* printf("%d newm= ",i); */
2808: /* for(j=1;j<=nlstate+ndeath;j++) { */
2809: /* printf("%f ",newm[i][j]); */
2810: /* } */
2811: /* printf("oldm * "); */
2812: /* for(j=1;j<=nlstate+ndeath;j++) { */
2813: /* printf("%f ",oldm[i][j]); */
2814: /* } */
1.268 brouard 2815: /* printf(" bmmij "); */
1.266 brouard 2816: /* for(j=1;j<=nlstate+ndeath;j++) { */
2817: /* printf("%f ",pmmij[i][j]); */
2818: /* } */
2819: /* printf("\n"); */
2820: /* } */
2821: /* } */
1.217 brouard 2822: savm=oldm;
2823: oldm=newm;
1.266 brouard 2824:
1.217 brouard 2825: for(j=1; j<=nlstate; j++){
2826: max[j]=0.;
2827: min[j]=1.;
2828: }
2829: for(j=1; j<=nlstate; j++){
2830: for(i=1;i<=nlstate;i++){
1.234 brouard 2831: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2832: bprlim[i][j]= newm[i][j];
2833: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2834: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2835: }
2836: }
1.218 brouard 2837:
1.217 brouard 2838: maxmax=0.;
2839: for(i=1; i<=nlstate; i++){
2840: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2841: maxmax=FMAX(maxmax,meandiff[i]);
2842: /* 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 2843: } /* i loop */
1.217 brouard 2844: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2845: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2846: if(maxmax < ftolpl){
1.220 brouard 2847: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2848: free_vector(min,1,nlstate);
2849: free_vector(max,1,nlstate);
2850: free_vector(meandiff,1,nlstate);
2851: return bprlim;
2852: }
2853: } /* age loop */
2854: /* After some age loop it doesn't converge */
1.247 brouard 2855: if(first){
2856: first=1;
2857: 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\
2858: 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);
2859: }
2860: 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 2861: 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);
2862: /* 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); */
2863: free_vector(min,1,nlstate);
2864: free_vector(max,1,nlstate);
2865: free_vector(meandiff,1,nlstate);
2866:
2867: return bprlim; /* should not reach here */
2868: }
2869:
1.126 brouard 2870: /*************** transition probabilities ***************/
2871:
2872: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2873: {
1.138 brouard 2874: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2875: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2876: model to the ncovmodel covariates (including constant and age).
2877: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2878: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2879: ncth covariate in the global vector x is given by the formula:
2880: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2881: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2882: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2883: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2884: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2885: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2886: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2887: */
2888: double s1, lnpijopii;
1.126 brouard 2889: /*double t34;*/
1.164 brouard 2890: int i,j, nc, ii, jj;
1.126 brouard 2891:
1.223 brouard 2892: for(i=1; i<= nlstate; i++){
2893: for(j=1; j<i;j++){
2894: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2895: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2896: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2897: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2898: }
2899: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2900: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2901: }
2902: for(j=i+1; j<=nlstate+ndeath;j++){
2903: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2904: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2905: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2906: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2907: }
2908: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2909: }
2910: }
1.218 brouard 2911:
1.223 brouard 2912: for(i=1; i<= nlstate; i++){
2913: s1=0;
2914: for(j=1; j<i; j++){
2915: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2916: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2917: }
2918: for(j=i+1; j<=nlstate+ndeath; j++){
2919: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2920: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2921: }
2922: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2923: ps[i][i]=1./(s1+1.);
2924: /* Computing other pijs */
2925: for(j=1; j<i; j++)
2926: ps[i][j]= exp(ps[i][j])*ps[i][i];
2927: for(j=i+1; j<=nlstate+ndeath; j++)
2928: ps[i][j]= exp(ps[i][j])*ps[i][i];
2929: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2930: } /* end i */
1.218 brouard 2931:
1.223 brouard 2932: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2933: for(jj=1; jj<= nlstate+ndeath; jj++){
2934: ps[ii][jj]=0;
2935: ps[ii][ii]=1;
2936: }
2937: }
1.218 brouard 2938:
2939:
1.223 brouard 2940: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2941: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2942: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2943: /* } */
2944: /* printf("\n "); */
2945: /* } */
2946: /* printf("\n ");printf("%lf ",cov[2]);*/
2947: /*
2948: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2949: goto end;*/
1.266 brouard 2950: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2951: }
2952:
1.218 brouard 2953: /*************** backward transition probabilities ***************/
2954:
2955: /* 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 ) */
2956: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2957: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2958: {
1.266 brouard 2959: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2960: * 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 2961: */
1.218 brouard 2962: int i, ii, j,k;
1.222 brouard 2963:
2964: double **out, **pmij();
2965: double sumnew=0.;
1.218 brouard 2966: double agefin;
1.268 brouard 2967: double k3=0.; /* constant of the w_x diagonal matrixe (in order for B to sum to 1 even for death state) */
1.222 brouard 2968: double **dnewm, **dsavm, **doldm;
2969: double **bbmij;
2970:
1.218 brouard 2971: doldm=ddoldms; /* global pointers */
1.222 brouard 2972: dnewm=ddnewms;
2973: dsavm=ddsavms;
2974:
2975: agefin=cov[2];
1.268 brouard 2976: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2977: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2978: the observed prevalence (with this covariate ij) at beginning of transition */
2979: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2980:
2981: /* P_x */
1.266 brouard 2982: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2983: /* outputs pmmij which is a stochastic matrix in row */
2984:
2985: /* Diag(w_x) */
2986: /* Problem with prevacurrent which can be zero */
2987: sumnew=0.;
1.269 brouard 2988: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2989: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2990: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2991: sumnew+=prevacurrent[(int)agefin][ii][ij];
2992: }
2993: if(sumnew >0.01){ /* At least some value in the prevalence */
2994: for (ii=1;ii<=nlstate+ndeath;ii++){
2995: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2996: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2997: }
2998: }else{
2999: for (ii=1;ii<=nlstate+ndeath;ii++){
3000: for (j=1;j<=nlstate+ndeath;j++)
3001: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3002: }
3003: /* if(sumnew <0.9){ */
3004: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3005: /* } */
3006: }
3007: k3=0.0; /* We put the last diagonal to 0 */
3008: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3009: doldm[ii][ii]= k3;
3010: }
3011: /* End doldm, At the end doldm is diag[(w_i)] */
3012:
3013: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
3014: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
3015:
3016: /* Diag(Sum_i w^i_x p^ij_x */
3017: /* 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 3018: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3019: sumnew=0.;
1.222 brouard 3020: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3021: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3022: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3023: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3024: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3025: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3026: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3027: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3028: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3029: /* }else */
1.268 brouard 3030: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3031: } /*End ii */
3032: } /* 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 */
3033:
3034: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
3035: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3036: /* end bmij */
1.266 brouard 3037: return ps; /*pointer is unchanged */
1.218 brouard 3038: }
1.217 brouard 3039: /*************** transition probabilities ***************/
3040:
1.218 brouard 3041: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3042: {
3043: /* According to parameters values stored in x and the covariate's values stored in cov,
3044: computes the probability to be observed in state j being in state i by appying the
3045: model to the ncovmodel covariates (including constant and age).
3046: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3047: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3048: ncth covariate in the global vector x is given by the formula:
3049: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3050: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3051: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3052: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3053: Outputs ps[i][j] the probability to be observed in j being in j according to
3054: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3055: */
3056: double s1, lnpijopii;
3057: /*double t34;*/
3058: int i,j, nc, ii, jj;
3059:
1.234 brouard 3060: for(i=1; i<= nlstate; i++){
3061: for(j=1; j<i;j++){
3062: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3063: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3064: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3065: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3066: }
3067: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3068: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3069: }
3070: for(j=i+1; j<=nlstate+ndeath;j++){
3071: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3072: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3073: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3074: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3075: }
3076: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3077: }
3078: }
3079:
3080: for(i=1; i<= nlstate; i++){
3081: s1=0;
3082: for(j=1; j<i; j++){
3083: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3084: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3085: }
3086: for(j=i+1; j<=nlstate+ndeath; j++){
3087: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3088: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3089: }
3090: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3091: ps[i][i]=1./(s1+1.);
3092: /* Computing other pijs */
3093: for(j=1; j<i; j++)
3094: ps[i][j]= exp(ps[i][j])*ps[i][i];
3095: for(j=i+1; j<=nlstate+ndeath; j++)
3096: ps[i][j]= exp(ps[i][j])*ps[i][i];
3097: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3098: } /* end i */
3099:
3100: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3101: for(jj=1; jj<= nlstate+ndeath; jj++){
3102: ps[ii][jj]=0;
3103: ps[ii][ii]=1;
3104: }
3105: }
3106: /* Added for backcast */ /* Transposed matrix too */
3107: for(jj=1; jj<= nlstate+ndeath; jj++){
3108: s1=0.;
3109: for(ii=1; ii<= nlstate+ndeath; ii++){
3110: s1+=ps[ii][jj];
3111: }
3112: for(ii=1; ii<= nlstate; ii++){
3113: ps[ii][jj]=ps[ii][jj]/s1;
3114: }
3115: }
3116: /* Transposition */
3117: for(jj=1; jj<= nlstate+ndeath; jj++){
3118: for(ii=jj; ii<= nlstate+ndeath; ii++){
3119: s1=ps[ii][jj];
3120: ps[ii][jj]=ps[jj][ii];
3121: ps[jj][ii]=s1;
3122: }
3123: }
3124: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3125: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3126: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3127: /* } */
3128: /* printf("\n "); */
3129: /* } */
3130: /* printf("\n ");printf("%lf ",cov[2]);*/
3131: /*
3132: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3133: goto end;*/
3134: return ps;
1.217 brouard 3135: }
3136:
3137:
1.126 brouard 3138: /**************** Product of 2 matrices ******************/
3139:
1.145 brouard 3140: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3141: {
3142: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3143: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3144: /* in, b, out are matrice of pointers which should have been initialized
3145: before: only the contents of out is modified. The function returns
3146: a pointer to pointers identical to out */
1.145 brouard 3147: int i, j, k;
1.126 brouard 3148: for(i=nrl; i<= nrh; i++)
1.145 brouard 3149: for(k=ncolol; k<=ncoloh; k++){
3150: out[i][k]=0.;
3151: for(j=ncl; j<=nch; j++)
3152: out[i][k] +=in[i][j]*b[j][k];
3153: }
1.126 brouard 3154: return out;
3155: }
3156:
3157:
3158: /************* Higher Matrix Product ***************/
3159:
1.235 brouard 3160: 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 3161: {
1.218 brouard 3162: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3163: 'nhstepm*hstepm*stepm' months (i.e. until
3164: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3165: nhstepm*hstepm matrices.
3166: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3167: (typically every 2 years instead of every month which is too big
3168: for the memory).
3169: Model is determined by parameters x and covariates have to be
3170: included manually here.
3171:
3172: */
3173:
3174: int i, j, d, h, k;
1.131 brouard 3175: double **out, cov[NCOVMAX+1];
1.126 brouard 3176: double **newm;
1.187 brouard 3177: double agexact;
1.214 brouard 3178: double agebegin, ageend;
1.126 brouard 3179:
3180: /* Hstepm could be zero and should return the unit matrix */
3181: for (i=1;i<=nlstate+ndeath;i++)
3182: for (j=1;j<=nlstate+ndeath;j++){
3183: oldm[i][j]=(i==j ? 1.0 : 0.0);
3184: po[i][j][0]=(i==j ? 1.0 : 0.0);
3185: }
3186: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3187: for(h=1; h <=nhstepm; h++){
3188: for(d=1; d <=hstepm; d++){
3189: newm=savm;
3190: /* Covariates have to be included here again */
3191: cov[1]=1.;
1.214 brouard 3192: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3193: cov[2]=agexact;
3194: if(nagesqr==1)
1.227 brouard 3195: cov[3]= agexact*agexact;
1.235 brouard 3196: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3197: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3198: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3199: /* 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)); */
3200: }
3201: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3202: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3203: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3204: /* 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]); */
3205: }
3206: for (k=1; k<=cptcovage;k++){
3207: if(Dummy[Tvar[Tage[k]]]){
3208: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3209: } else{
3210: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3211: }
3212: /* 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]); */
3213: }
3214: for (k=1; k<=cptcovprod;k++){ /* */
3215: /* 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]); */
3216: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3217: }
3218: /* for (k=1; k<=cptcovn;k++) */
3219: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3220: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3221: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3222: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3223: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3224:
3225:
1.126 brouard 3226: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3227: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3228: /* right multiplication of oldm by the current matrix */
1.126 brouard 3229: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3230: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3231: /* if((int)age == 70){ */
3232: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3233: /* for(i=1; i<=nlstate+ndeath; i++) { */
3234: /* printf("%d pmmij ",i); */
3235: /* for(j=1;j<=nlstate+ndeath;j++) { */
3236: /* printf("%f ",pmmij[i][j]); */
3237: /* } */
3238: /* printf(" oldm "); */
3239: /* for(j=1;j<=nlstate+ndeath;j++) { */
3240: /* printf("%f ",oldm[i][j]); */
3241: /* } */
3242: /* printf("\n"); */
3243: /* } */
3244: /* } */
1.126 brouard 3245: savm=oldm;
3246: oldm=newm;
3247: }
3248: for(i=1; i<=nlstate+ndeath; i++)
3249: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3250: po[i][j][h]=newm[i][j];
3251: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3252: }
1.128 brouard 3253: /*printf("h=%d ",h);*/
1.126 brouard 3254: } /* end h */
1.267 brouard 3255: /* printf("\n H=%d \n",h); */
1.126 brouard 3256: return po;
3257: }
3258:
1.217 brouard 3259: /************* Higher Back Matrix Product ***************/
1.218 brouard 3260: /* 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 3261: 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 3262: {
1.266 brouard 3263: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3264: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3265: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3266: nhstepm*hstepm matrices.
3267: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3268: (typically every 2 years instead of every month which is too big
1.217 brouard 3269: for the memory).
1.218 brouard 3270: Model is determined by parameters x and covariates have to be
1.266 brouard 3271: included manually here. Then we use a call to bmij(x and cov)
3272: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3273: */
1.217 brouard 3274:
3275: int i, j, d, h, k;
1.266 brouard 3276: double **out, cov[NCOVMAX+1], **bmij();
3277: double **newm, ***newmm;
1.217 brouard 3278: double agexact;
3279: double agebegin, ageend;
1.222 brouard 3280: double **oldm, **savm;
1.217 brouard 3281:
1.266 brouard 3282: newmm=po; /* To be saved */
3283: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3284: /* Hstepm could be zero and should return the unit matrix */
3285: for (i=1;i<=nlstate+ndeath;i++)
3286: for (j=1;j<=nlstate+ndeath;j++){
3287: oldm[i][j]=(i==j ? 1.0 : 0.0);
3288: po[i][j][0]=(i==j ? 1.0 : 0.0);
3289: }
3290: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3291: for(h=1; h <=nhstepm; h++){
3292: for(d=1; d <=hstepm; d++){
3293: newm=savm;
3294: /* Covariates have to be included here again */
3295: cov[1]=1.;
1.271 brouard 3296: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3297: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3298: cov[2]=agexact;
3299: if(nagesqr==1)
1.222 brouard 3300: cov[3]= agexact*agexact;
1.266 brouard 3301: for (k=1; k<=cptcovn;k++){
3302: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3303: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3304: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3305: /* 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)); */
3306: }
1.267 brouard 3307: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3308: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3309: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3310: /* 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]); */
3311: }
3312: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3313: if(Dummy[Tvar[Tage[k]]]){
3314: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3315: } else{
3316: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3317: }
3318: /* 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]); */
3319: }
3320: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3321: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3322: }
1.217 brouard 3323: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3324: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3325:
1.218 brouard 3326: /* Careful transposed matrix */
1.266 brouard 3327: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3328: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3329: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3330: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3331: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3332: /* if((int)age == 70){ */
3333: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3334: /* for(i=1; i<=nlstate+ndeath; i++) { */
3335: /* printf("%d pmmij ",i); */
3336: /* for(j=1;j<=nlstate+ndeath;j++) { */
3337: /* printf("%f ",pmmij[i][j]); */
3338: /* } */
3339: /* printf(" oldm "); */
3340: /* for(j=1;j<=nlstate+ndeath;j++) { */
3341: /* printf("%f ",oldm[i][j]); */
3342: /* } */
3343: /* printf("\n"); */
3344: /* } */
3345: /* } */
3346: savm=oldm;
3347: oldm=newm;
3348: }
3349: for(i=1; i<=nlstate+ndeath; i++)
3350: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3351: po[i][j][h]=newm[i][j];
1.268 brouard 3352: /* if(h==nhstepm) */
3353: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3354: }
1.268 brouard 3355: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3356: } /* end h */
1.268 brouard 3357: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3358: return po;
3359: }
3360:
3361:
1.162 brouard 3362: #ifdef NLOPT
3363: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3364: double fret;
3365: double *xt;
3366: int j;
3367: myfunc_data *d2 = (myfunc_data *) pd;
3368: /* xt = (p1-1); */
3369: xt=vector(1,n);
3370: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3371:
3372: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3373: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3374: printf("Function = %.12lf ",fret);
3375: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3376: printf("\n");
3377: free_vector(xt,1,n);
3378: return fret;
3379: }
3380: #endif
1.126 brouard 3381:
3382: /*************** log-likelihood *************/
3383: double func( double *x)
3384: {
1.226 brouard 3385: int i, ii, j, k, mi, d, kk;
3386: int ioffset=0;
3387: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3388: double **out;
3389: double lli; /* Individual log likelihood */
3390: int s1, s2;
1.228 brouard 3391: 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 3392: double bbh, survp;
3393: long ipmx;
3394: double agexact;
3395: /*extern weight */
3396: /* We are differentiating ll according to initial status */
3397: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3398: /*for(i=1;i<imx;i++)
3399: printf(" %d\n",s[4][i]);
3400: */
1.162 brouard 3401:
1.226 brouard 3402: ++countcallfunc;
1.162 brouard 3403:
1.226 brouard 3404: cov[1]=1.;
1.126 brouard 3405:
1.226 brouard 3406: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3407: ioffset=0;
1.226 brouard 3408: if(mle==1){
3409: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3410: /* Computes the values of the ncovmodel covariates of the model
3411: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3412: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3413: to be observed in j being in i according to the model.
3414: */
1.243 brouard 3415: ioffset=2+nagesqr ;
1.233 brouard 3416: /* Fixed */
1.234 brouard 3417: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3418: 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)*/
3419: }
1.226 brouard 3420: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3421: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3422: has been calculated etc */
3423: /* For an individual i, wav[i] gives the number of effective waves */
3424: /* We compute the contribution to Likelihood of each effective transition
3425: mw[mi][i] is real wave of the mi th effectve wave */
3426: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3427: s2=s[mw[mi+1][i]][i];
3428: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3429: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3430: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3431: */
3432: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3433: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3434: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3435: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3436: }
3437: for (ii=1;ii<=nlstate+ndeath;ii++)
3438: for (j=1;j<=nlstate+ndeath;j++){
3439: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3440: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3441: }
3442: for(d=0; d<dh[mi][i]; d++){
3443: newm=savm;
3444: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3445: cov[2]=agexact;
3446: if(nagesqr==1)
3447: cov[3]= agexact*agexact; /* Should be changed here */
3448: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3449: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3450: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3451: else
3452: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3453: }
3454: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3455: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3456: savm=oldm;
3457: oldm=newm;
3458: } /* end mult */
3459:
3460: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3461: /* But now since version 0.9 we anticipate for bias at large stepm.
3462: * If stepm is larger than one month (smallest stepm) and if the exact delay
3463: * (in months) between two waves is not a multiple of stepm, we rounded to
3464: * the nearest (and in case of equal distance, to the lowest) interval but now
3465: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3466: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3467: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3468: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3469: * -stepm/2 to stepm/2 .
3470: * For stepm=1 the results are the same as for previous versions of Imach.
3471: * For stepm > 1 the results are less biased than in previous versions.
3472: */
1.234 brouard 3473: s1=s[mw[mi][i]][i];
3474: s2=s[mw[mi+1][i]][i];
3475: bbh=(double)bh[mi][i]/(double)stepm;
3476: /* bias bh is positive if real duration
3477: * is higher than the multiple of stepm and negative otherwise.
3478: */
3479: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3480: if( s2 > nlstate){
3481: /* i.e. if s2 is a death state and if the date of death is known
3482: then the contribution to the likelihood is the probability to
3483: die between last step unit time and current step unit time,
3484: which is also equal to probability to die before dh
3485: minus probability to die before dh-stepm .
3486: In version up to 0.92 likelihood was computed
3487: as if date of death was unknown. Death was treated as any other
3488: health state: the date of the interview describes the actual state
3489: and not the date of a change in health state. The former idea was
3490: to consider that at each interview the state was recorded
3491: (healthy, disable or death) and IMaCh was corrected; but when we
3492: introduced the exact date of death then we should have modified
3493: the contribution of an exact death to the likelihood. This new
3494: contribution is smaller and very dependent of the step unit
3495: stepm. It is no more the probability to die between last interview
3496: and month of death but the probability to survive from last
3497: interview up to one month before death multiplied by the
3498: probability to die within a month. Thanks to Chris
3499: Jackson for correcting this bug. Former versions increased
3500: mortality artificially. The bad side is that we add another loop
3501: which slows down the processing. The difference can be up to 10%
3502: lower mortality.
3503: */
3504: /* If, at the beginning of the maximization mostly, the
3505: cumulative probability or probability to be dead is
3506: constant (ie = 1) over time d, the difference is equal to
3507: 0. out[s1][3] = savm[s1][3]: probability, being at state
3508: s1 at precedent wave, to be dead a month before current
3509: wave is equal to probability, being at state s1 at
3510: precedent wave, to be dead at mont of the current
3511: wave. Then the observed probability (that this person died)
3512: is null according to current estimated parameter. In fact,
3513: it should be very low but not zero otherwise the log go to
3514: infinity.
3515: */
1.183 brouard 3516: /* #ifdef INFINITYORIGINAL */
3517: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3518: /* #else */
3519: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3520: /* lli=log(mytinydouble); */
3521: /* else */
3522: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3523: /* #endif */
1.226 brouard 3524: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3525:
1.226 brouard 3526: } else if ( s2==-1 ) { /* alive */
3527: for (j=1,survp=0. ; j<=nlstate; j++)
3528: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3529: /*survp += out[s1][j]; */
3530: lli= log(survp);
3531: }
3532: else if (s2==-4) {
3533: for (j=3,survp=0. ; j<=nlstate; j++)
3534: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3535: lli= log(survp);
3536: }
3537: else if (s2==-5) {
3538: for (j=1,survp=0. ; j<=2; j++)
3539: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3540: lli= log(survp);
3541: }
3542: else{
3543: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3544: /* 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 */
3545: }
3546: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3547: /*if(lli ==000.0)*/
3548: /*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); */
3549: ipmx +=1;
3550: sw += weight[i];
3551: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3552: /* if (lli < log(mytinydouble)){ */
3553: /* 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); */
3554: /* 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]); */
3555: /* } */
3556: } /* end of wave */
3557: } /* end of individual */
3558: } else if(mle==2){
3559: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3560: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3561: for(mi=1; mi<= wav[i]-1; mi++){
3562: for (ii=1;ii<=nlstate+ndeath;ii++)
3563: for (j=1;j<=nlstate+ndeath;j++){
3564: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3565: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3566: }
3567: for(d=0; d<=dh[mi][i]; d++){
3568: newm=savm;
3569: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3570: cov[2]=agexact;
3571: if(nagesqr==1)
3572: cov[3]= agexact*agexact;
3573: for (kk=1; kk<=cptcovage;kk++) {
3574: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3575: }
3576: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3577: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3578: savm=oldm;
3579: oldm=newm;
3580: } /* end mult */
3581:
3582: s1=s[mw[mi][i]][i];
3583: s2=s[mw[mi+1][i]][i];
3584: bbh=(double)bh[mi][i]/(double)stepm;
3585: 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 */
3586: ipmx +=1;
3587: sw += weight[i];
3588: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3589: } /* end of wave */
3590: } /* end of individual */
3591: } else if(mle==3){ /* exponential inter-extrapolation */
3592: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3593: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3594: for(mi=1; mi<= wav[i]-1; mi++){
3595: for (ii=1;ii<=nlstate+ndeath;ii++)
3596: for (j=1;j<=nlstate+ndeath;j++){
3597: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3598: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3599: }
3600: for(d=0; d<dh[mi][i]; d++){
3601: newm=savm;
3602: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3603: cov[2]=agexact;
3604: if(nagesqr==1)
3605: cov[3]= agexact*agexact;
3606: for (kk=1; kk<=cptcovage;kk++) {
3607: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3608: }
3609: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3610: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3611: savm=oldm;
3612: oldm=newm;
3613: } /* end mult */
3614:
3615: s1=s[mw[mi][i]][i];
3616: s2=s[mw[mi+1][i]][i];
3617: bbh=(double)bh[mi][i]/(double)stepm;
3618: 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 */
3619: ipmx +=1;
3620: sw += weight[i];
3621: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3622: } /* end of wave */
3623: } /* end of individual */
3624: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3625: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3626: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3627: for(mi=1; mi<= wav[i]-1; mi++){
3628: for (ii=1;ii<=nlstate+ndeath;ii++)
3629: for (j=1;j<=nlstate+ndeath;j++){
3630: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3631: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3632: }
3633: for(d=0; d<dh[mi][i]; d++){
3634: newm=savm;
3635: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3636: cov[2]=agexact;
3637: if(nagesqr==1)
3638: cov[3]= agexact*agexact;
3639: for (kk=1; kk<=cptcovage;kk++) {
3640: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3641: }
1.126 brouard 3642:
1.226 brouard 3643: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3644: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3645: savm=oldm;
3646: oldm=newm;
3647: } /* end mult */
3648:
3649: s1=s[mw[mi][i]][i];
3650: s2=s[mw[mi+1][i]][i];
3651: if( s2 > nlstate){
3652: lli=log(out[s1][s2] - savm[s1][s2]);
3653: } else if ( s2==-1 ) { /* alive */
3654: for (j=1,survp=0. ; j<=nlstate; j++)
3655: survp += out[s1][j];
3656: lli= log(survp);
3657: }else{
3658: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3659: }
3660: ipmx +=1;
3661: sw += weight[i];
3662: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3663: /* 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 3664: } /* end of wave */
3665: } /* end of individual */
3666: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3667: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3668: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3669: for(mi=1; mi<= wav[i]-1; mi++){
3670: for (ii=1;ii<=nlstate+ndeath;ii++)
3671: for (j=1;j<=nlstate+ndeath;j++){
3672: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3673: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3674: }
3675: for(d=0; d<dh[mi][i]; d++){
3676: newm=savm;
3677: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3678: cov[2]=agexact;
3679: if(nagesqr==1)
3680: cov[3]= agexact*agexact;
3681: for (kk=1; kk<=cptcovage;kk++) {
3682: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3683: }
1.126 brouard 3684:
1.226 brouard 3685: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3686: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3687: savm=oldm;
3688: oldm=newm;
3689: } /* end mult */
3690:
3691: s1=s[mw[mi][i]][i];
3692: s2=s[mw[mi+1][i]][i];
3693: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3694: ipmx +=1;
3695: sw += weight[i];
3696: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3697: /*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]);*/
3698: } /* end of wave */
3699: } /* end of individual */
3700: } /* End of if */
3701: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3702: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3703: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3704: return -l;
1.126 brouard 3705: }
3706:
3707: /*************** log-likelihood *************/
3708: double funcone( double *x)
3709: {
1.228 brouard 3710: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3711: int i, ii, j, k, mi, d, kk;
1.228 brouard 3712: int ioffset=0;
1.131 brouard 3713: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3714: double **out;
3715: double lli; /* Individual log likelihood */
3716: double llt;
3717: int s1, s2;
1.228 brouard 3718: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3719:
1.126 brouard 3720: double bbh, survp;
1.187 brouard 3721: double agexact;
1.214 brouard 3722: double agebegin, ageend;
1.126 brouard 3723: /*extern weight */
3724: /* We are differentiating ll according to initial status */
3725: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3726: /*for(i=1;i<imx;i++)
3727: printf(" %d\n",s[4][i]);
3728: */
3729: cov[1]=1.;
3730:
3731: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3732: ioffset=0;
3733: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3734: /* ioffset=2+nagesqr+cptcovage; */
3735: ioffset=2+nagesqr;
1.232 brouard 3736: /* Fixed */
1.224 brouard 3737: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3738: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3739: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3740: 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)*/
3741: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3742: /* cov[2+6]=covar[Tvar[6]][i]; */
3743: /* cov[2+6]=covar[2][i]; V2 */
3744: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3745: /* cov[2+7]=covar[Tvar[7]][i]; */
3746: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3747: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3748: /* cov[2+9]=covar[Tvar[9]][i]; */
3749: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3750: }
1.232 brouard 3751: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3752: /* 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?)*\/ */
3753: /* } */
1.231 brouard 3754: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3755: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3756: /* } */
1.225 brouard 3757:
1.233 brouard 3758:
3759: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3760: /* Wave varying (but not age varying) */
3761: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3762: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3763: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3764: }
1.232 brouard 3765: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3766: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3767: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3768: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3769: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3770: /* 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 3771: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3772: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3773: /* /\* 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]); *\/ */
3774: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3775: /* } */
1.126 brouard 3776: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3777: for (j=1;j<=nlstate+ndeath;j++){
3778: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3779: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3780: }
1.214 brouard 3781:
3782: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3783: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3784: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3785: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3786: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3787: and mw[mi+1][i]. dh depends on stepm.*/
3788: newm=savm;
1.247 brouard 3789: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3790: cov[2]=agexact;
3791: if(nagesqr==1)
3792: cov[3]= agexact*agexact;
3793: for (kk=1; kk<=cptcovage;kk++) {
3794: if(!FixedV[Tvar[Tage[kk]]])
3795: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3796: else
3797: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3798: }
3799: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3800: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3801: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3802: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3803: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3804: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3805: savm=oldm;
3806: oldm=newm;
1.126 brouard 3807: } /* end mult */
3808:
3809: s1=s[mw[mi][i]][i];
3810: s2=s[mw[mi+1][i]][i];
1.217 brouard 3811: /* if(s2==-1){ */
1.268 brouard 3812: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3813: /* /\* exit(1); *\/ */
3814: /* } */
1.126 brouard 3815: bbh=(double)bh[mi][i]/(double)stepm;
3816: /* bias is positive if real duration
3817: * is higher than the multiple of stepm and negative otherwise.
3818: */
3819: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3820: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3821: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3822: for (j=1,survp=0. ; j<=nlstate; j++)
3823: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3824: lli= log(survp);
1.126 brouard 3825: }else if (mle==1){
1.242 brouard 3826: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3827: } else if(mle==2){
1.242 brouard 3828: 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 3829: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3830: 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 3831: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3832: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3833: } else{ /* mle=0 back to 1 */
1.242 brouard 3834: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3835: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3836: } /* End of if */
3837: ipmx +=1;
3838: sw += weight[i];
3839: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3840: /*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 3841: if(globpr){
1.246 brouard 3842: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3843: %11.6f %11.6f %11.6f ", \
1.242 brouard 3844: 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 3845: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3846: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3847: llt +=ll[k]*gipmx/gsw;
3848: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3849: }
3850: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3851: }
1.232 brouard 3852: } /* end of wave */
3853: } /* end of individual */
3854: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3855: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3856: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3857: if(globpr==0){ /* First time we count the contributions and weights */
3858: gipmx=ipmx;
3859: gsw=sw;
3860: }
3861: return -l;
1.126 brouard 3862: }
3863:
3864:
3865: /*************** function likelione ***********/
3866: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3867: {
3868: /* This routine should help understanding what is done with
3869: the selection of individuals/waves and
3870: to check the exact contribution to the likelihood.
3871: Plotting could be done.
3872: */
3873: int k;
3874:
3875: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3876: strcpy(fileresilk,"ILK_");
1.202 brouard 3877: strcat(fileresilk,fileresu);
1.126 brouard 3878: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3879: printf("Problem with resultfile: %s\n", fileresilk);
3880: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3881: }
1.214 brouard 3882: 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");
3883: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3884: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3885: for(k=1; k<=nlstate; k++)
3886: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3887: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3888: }
3889:
3890: *fretone=(*funcone)(p);
3891: if(*globpri !=0){
3892: fclose(ficresilk);
1.205 brouard 3893: if (mle ==0)
3894: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3895: else if(mle >=1)
3896: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3897: 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 3898: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3899:
3900: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3901: 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 3902: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3903: }
1.207 brouard 3904: 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 3905: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3906: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3907: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3908: fflush(fichtm);
1.205 brouard 3909: }
1.126 brouard 3910: return;
3911: }
3912:
3913:
3914: /*********** Maximum Likelihood Estimation ***************/
3915:
3916: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3917: {
1.165 brouard 3918: int i,j, iter=0;
1.126 brouard 3919: double **xi;
3920: double fret;
3921: double fretone; /* Only one call to likelihood */
3922: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3923:
3924: #ifdef NLOPT
3925: int creturn;
3926: nlopt_opt opt;
3927: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3928: double *lb;
3929: double minf; /* the minimum objective value, upon return */
3930: double * p1; /* Shifted parameters from 0 instead of 1 */
3931: myfunc_data dinst, *d = &dinst;
3932: #endif
3933:
3934:
1.126 brouard 3935: xi=matrix(1,npar,1,npar);
3936: for (i=1;i<=npar;i++)
3937: for (j=1;j<=npar;j++)
3938: xi[i][j]=(i==j ? 1.0 : 0.0);
3939: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3940: strcpy(filerespow,"POW_");
1.126 brouard 3941: strcat(filerespow,fileres);
3942: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3943: printf("Problem with resultfile: %s\n", filerespow);
3944: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3945: }
3946: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3947: for (i=1;i<=nlstate;i++)
3948: for(j=1;j<=nlstate+ndeath;j++)
3949: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3950: fprintf(ficrespow,"\n");
1.162 brouard 3951: #ifdef POWELL
1.126 brouard 3952: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3953: #endif
1.126 brouard 3954:
1.162 brouard 3955: #ifdef NLOPT
3956: #ifdef NEWUOA
3957: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3958: #else
3959: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3960: #endif
3961: lb=vector(0,npar-1);
3962: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3963: nlopt_set_lower_bounds(opt, lb);
3964: nlopt_set_initial_step1(opt, 0.1);
3965:
3966: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3967: d->function = func;
3968: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3969: nlopt_set_min_objective(opt, myfunc, d);
3970: nlopt_set_xtol_rel(opt, ftol);
3971: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3972: printf("nlopt failed! %d\n",creturn);
3973: }
3974: else {
3975: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3976: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3977: iter=1; /* not equal */
3978: }
3979: nlopt_destroy(opt);
3980: #endif
1.126 brouard 3981: free_matrix(xi,1,npar,1,npar);
3982: fclose(ficrespow);
1.203 brouard 3983: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3984: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3985: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3986:
3987: }
3988:
3989: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3990: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3991: {
3992: double **a,**y,*x,pd;
1.203 brouard 3993: /* double **hess; */
1.164 brouard 3994: int i, j;
1.126 brouard 3995: int *indx;
3996:
3997: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3998: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3999: void lubksb(double **a, int npar, int *indx, double b[]) ;
4000: void ludcmp(double **a, int npar, int *indx, double *d) ;
4001: double gompertz(double p[]);
1.203 brouard 4002: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4003:
4004: printf("\nCalculation of the hessian matrix. Wait...\n");
4005: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4006: for (i=1;i<=npar;i++){
1.203 brouard 4007: printf("%d-",i);fflush(stdout);
4008: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4009:
4010: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4011:
4012: /* printf(" %f ",p[i]);
4013: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4014: }
4015:
4016: for (i=1;i<=npar;i++) {
4017: for (j=1;j<=npar;j++) {
4018: if (j>i) {
1.203 brouard 4019: printf(".%d-%d",i,j);fflush(stdout);
4020: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4021: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4022:
4023: hess[j][i]=hess[i][j];
4024: /*printf(" %lf ",hess[i][j]);*/
4025: }
4026: }
4027: }
4028: printf("\n");
4029: fprintf(ficlog,"\n");
4030:
4031: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4032: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4033:
4034: a=matrix(1,npar,1,npar);
4035: y=matrix(1,npar,1,npar);
4036: x=vector(1,npar);
4037: indx=ivector(1,npar);
4038: for (i=1;i<=npar;i++)
4039: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4040: ludcmp(a,npar,indx,&pd);
4041:
4042: for (j=1;j<=npar;j++) {
4043: for (i=1;i<=npar;i++) x[i]=0;
4044: x[j]=1;
4045: lubksb(a,npar,indx,x);
4046: for (i=1;i<=npar;i++){
4047: matcov[i][j]=x[i];
4048: }
4049: }
4050:
4051: printf("\n#Hessian matrix#\n");
4052: fprintf(ficlog,"\n#Hessian matrix#\n");
4053: for (i=1;i<=npar;i++) {
4054: for (j=1;j<=npar;j++) {
1.203 brouard 4055: printf("%.6e ",hess[i][j]);
4056: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4057: }
4058: printf("\n");
4059: fprintf(ficlog,"\n");
4060: }
4061:
1.203 brouard 4062: /* printf("\n#Covariance matrix#\n"); */
4063: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4064: /* for (i=1;i<=npar;i++) { */
4065: /* for (j=1;j<=npar;j++) { */
4066: /* printf("%.6e ",matcov[i][j]); */
4067: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4068: /* } */
4069: /* printf("\n"); */
4070: /* fprintf(ficlog,"\n"); */
4071: /* } */
4072:
1.126 brouard 4073: /* Recompute Inverse */
1.203 brouard 4074: /* for (i=1;i<=npar;i++) */
4075: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4076: /* ludcmp(a,npar,indx,&pd); */
4077:
4078: /* printf("\n#Hessian matrix recomputed#\n"); */
4079:
4080: /* for (j=1;j<=npar;j++) { */
4081: /* for (i=1;i<=npar;i++) x[i]=0; */
4082: /* x[j]=1; */
4083: /* lubksb(a,npar,indx,x); */
4084: /* for (i=1;i<=npar;i++){ */
4085: /* y[i][j]=x[i]; */
4086: /* printf("%.3e ",y[i][j]); */
4087: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4088: /* } */
4089: /* printf("\n"); */
4090: /* fprintf(ficlog,"\n"); */
4091: /* } */
4092:
4093: /* Verifying the inverse matrix */
4094: #ifdef DEBUGHESS
4095: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4096:
1.203 brouard 4097: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4098: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4099:
4100: for (j=1;j<=npar;j++) {
4101: for (i=1;i<=npar;i++){
1.203 brouard 4102: printf("%.2f ",y[i][j]);
4103: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4104: }
4105: printf("\n");
4106: fprintf(ficlog,"\n");
4107: }
1.203 brouard 4108: #endif
1.126 brouard 4109:
4110: free_matrix(a,1,npar,1,npar);
4111: free_matrix(y,1,npar,1,npar);
4112: free_vector(x,1,npar);
4113: free_ivector(indx,1,npar);
1.203 brouard 4114: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4115:
4116:
4117: }
4118:
4119: /*************** hessian matrix ****************/
4120: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4121: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4122: int i;
4123: int l=1, lmax=20;
1.203 brouard 4124: double k1,k2, res, fx;
1.132 brouard 4125: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4126: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4127: int k=0,kmax=10;
4128: double l1;
4129:
4130: fx=func(x);
4131: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4132: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4133: l1=pow(10,l);
4134: delts=delt;
4135: for(k=1 ; k <kmax; k=k+1){
4136: delt = delta*(l1*k);
4137: p2[theta]=x[theta] +delt;
1.145 brouard 4138: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4139: p2[theta]=x[theta]-delt;
4140: k2=func(p2)-fx;
4141: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4142: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4143:
1.203 brouard 4144: #ifdef DEBUGHESSII
1.126 brouard 4145: 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);
4146: 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);
4147: #endif
4148: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4149: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4150: k=kmax;
4151: }
4152: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4153: k=kmax; l=lmax*10;
1.126 brouard 4154: }
4155: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4156: delts=delt;
4157: }
1.203 brouard 4158: } /* End loop k */
1.126 brouard 4159: }
4160: delti[theta]=delts;
4161: return res;
4162:
4163: }
4164:
1.203 brouard 4165: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4166: {
4167: int i;
1.164 brouard 4168: int l=1, lmax=20;
1.126 brouard 4169: double k1,k2,k3,k4,res,fx;
1.132 brouard 4170: double p2[MAXPARM+1];
1.203 brouard 4171: int k, kmax=1;
4172: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4173:
4174: int firstime=0;
1.203 brouard 4175:
1.126 brouard 4176: fx=func(x);
1.203 brouard 4177: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4178: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4179: p2[thetai]=x[thetai]+delti[thetai]*k;
4180: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4181: k1=func(p2)-fx;
4182:
1.203 brouard 4183: p2[thetai]=x[thetai]+delti[thetai]*k;
4184: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4185: k2=func(p2)-fx;
4186:
1.203 brouard 4187: p2[thetai]=x[thetai]-delti[thetai]*k;
4188: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4189: k3=func(p2)-fx;
4190:
1.203 brouard 4191: p2[thetai]=x[thetai]-delti[thetai]*k;
4192: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4193: k4=func(p2)-fx;
1.203 brouard 4194: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4195: if(k1*k2*k3*k4 <0.){
1.208 brouard 4196: firstime=1;
1.203 brouard 4197: kmax=kmax+10;
1.208 brouard 4198: }
4199: if(kmax >=10 || firstime ==1){
1.246 brouard 4200: 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);
4201: 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 4202: 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);
4203: 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);
4204: }
4205: #ifdef DEBUGHESSIJ
4206: v1=hess[thetai][thetai];
4207: v2=hess[thetaj][thetaj];
4208: cv12=res;
4209: /* Computing eigen value of Hessian matrix */
4210: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4211: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4212: if ((lc2 <0) || (lc1 <0) ){
4213: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4214: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4215: 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);
4216: 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);
4217: }
1.126 brouard 4218: #endif
4219: }
4220: return res;
4221: }
4222:
1.203 brouard 4223: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4224: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4225: /* { */
4226: /* int i; */
4227: /* int l=1, lmax=20; */
4228: /* double k1,k2,k3,k4,res,fx; */
4229: /* double p2[MAXPARM+1]; */
4230: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4231: /* int k=0,kmax=10; */
4232: /* double l1; */
4233:
4234: /* fx=func(x); */
4235: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4236: /* l1=pow(10,l); */
4237: /* delts=delt; */
4238: /* for(k=1 ; k <kmax; k=k+1){ */
4239: /* delt = delti*(l1*k); */
4240: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4241: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4242: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4243: /* k1=func(p2)-fx; */
4244:
4245: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4246: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4247: /* k2=func(p2)-fx; */
4248:
4249: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4250: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4251: /* k3=func(p2)-fx; */
4252:
4253: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4254: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4255: /* k4=func(p2)-fx; */
4256: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4257: /* #ifdef DEBUGHESSIJ */
4258: /* 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); */
4259: /* 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); */
4260: /* #endif */
4261: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4262: /* k=kmax; */
4263: /* } */
4264: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4265: /* k=kmax; l=lmax*10; */
4266: /* } */
4267: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4268: /* delts=delt; */
4269: /* } */
4270: /* } /\* End loop k *\/ */
4271: /* } */
4272: /* delti[theta]=delts; */
4273: /* return res; */
4274: /* } */
4275:
4276:
1.126 brouard 4277: /************** Inverse of matrix **************/
4278: void ludcmp(double **a, int n, int *indx, double *d)
4279: {
4280: int i,imax,j,k;
4281: double big,dum,sum,temp;
4282: double *vv;
4283:
4284: vv=vector(1,n);
4285: *d=1.0;
4286: for (i=1;i<=n;i++) {
4287: big=0.0;
4288: for (j=1;j<=n;j++)
4289: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4290: if (big == 0.0){
4291: printf(" Singular Hessian matrix at row %d:\n",i);
4292: for (j=1;j<=n;j++) {
4293: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4294: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4295: }
4296: fflush(ficlog);
4297: fclose(ficlog);
4298: nrerror("Singular matrix in routine ludcmp");
4299: }
1.126 brouard 4300: vv[i]=1.0/big;
4301: }
4302: for (j=1;j<=n;j++) {
4303: for (i=1;i<j;i++) {
4304: sum=a[i][j];
4305: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4306: a[i][j]=sum;
4307: }
4308: big=0.0;
4309: for (i=j;i<=n;i++) {
4310: sum=a[i][j];
4311: for (k=1;k<j;k++)
4312: sum -= a[i][k]*a[k][j];
4313: a[i][j]=sum;
4314: if ( (dum=vv[i]*fabs(sum)) >= big) {
4315: big=dum;
4316: imax=i;
4317: }
4318: }
4319: if (j != imax) {
4320: for (k=1;k<=n;k++) {
4321: dum=a[imax][k];
4322: a[imax][k]=a[j][k];
4323: a[j][k]=dum;
4324: }
4325: *d = -(*d);
4326: vv[imax]=vv[j];
4327: }
4328: indx[j]=imax;
4329: if (a[j][j] == 0.0) a[j][j]=TINY;
4330: if (j != n) {
4331: dum=1.0/(a[j][j]);
4332: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4333: }
4334: }
4335: free_vector(vv,1,n); /* Doesn't work */
4336: ;
4337: }
4338:
4339: void lubksb(double **a, int n, int *indx, double b[])
4340: {
4341: int i,ii=0,ip,j;
4342: double sum;
4343:
4344: for (i=1;i<=n;i++) {
4345: ip=indx[i];
4346: sum=b[ip];
4347: b[ip]=b[i];
4348: if (ii)
4349: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4350: else if (sum) ii=i;
4351: b[i]=sum;
4352: }
4353: for (i=n;i>=1;i--) {
4354: sum=b[i];
4355: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4356: b[i]=sum/a[i][i];
4357: }
4358: }
4359:
4360: void pstamp(FILE *fichier)
4361: {
1.196 brouard 4362: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4363: }
4364:
1.253 brouard 4365:
4366:
1.126 brouard 4367: /************ Frequencies ********************/
1.251 brouard 4368: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4369: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4370: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4371: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4372:
1.265 brouard 4373: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4374: int iind=0, iage=0;
4375: int mi; /* Effective wave */
4376: int first;
4377: double ***freq; /* Frequencies */
1.268 brouard 4378: 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 */
4379: 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 4380: double *meanq, *stdq, *idq;
1.226 brouard 4381: double **meanqt;
4382: double *pp, **prop, *posprop, *pospropt;
4383: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4384: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4385: double agebegin, ageend;
4386:
4387: pp=vector(1,nlstate);
1.251 brouard 4388: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4389: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4390: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4391: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4392: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4393: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4394: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4395: meanqt=matrix(1,lastpass,1,nqtveff);
4396: strcpy(fileresp,"P_");
4397: strcat(fileresp,fileresu);
4398: /*strcat(fileresphtm,fileresu);*/
4399: if((ficresp=fopen(fileresp,"w"))==NULL) {
4400: printf("Problem with prevalence resultfile: %s\n", fileresp);
4401: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4402: exit(0);
4403: }
1.240 brouard 4404:
1.226 brouard 4405: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4406: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4407: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4408: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4409: fflush(ficlog);
4410: exit(70);
4411: }
4412: else{
4413: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4414: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4415: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4416: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4417: }
1.237 brouard 4418: 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 4419:
1.226 brouard 4420: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4421: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4422: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4423: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4424: fflush(ficlog);
4425: exit(70);
1.240 brouard 4426: } else{
1.226 brouard 4427: 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 4428: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4429: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4430: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4431: }
1.240 brouard 4432: 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);
4433:
1.253 brouard 4434: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4435: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4436: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4437: j1=0;
1.126 brouard 4438:
1.227 brouard 4439: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4440: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4441: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4442:
4443:
1.226 brouard 4444: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4445: reference=low_education V1=0,V2=0
4446: med_educ V1=1 V2=0,
4447: high_educ V1=0 V2=1
4448: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4449: */
1.249 brouard 4450: dateintsum=0;
4451: k2cpt=0;
4452:
1.253 brouard 4453: if(cptcoveff == 0 )
1.265 brouard 4454: nl=1; /* Constant and age model only */
1.253 brouard 4455: else
4456: nl=2;
1.265 brouard 4457:
4458: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4459: /* Loop on nj=1 or 2 if dummy covariates j!=0
4460: * Loop on j1(1 to 2**cptcoveff) covariate combination
4461: * freq[s1][s2][iage] =0.
4462: * Loop on iind
4463: * ++freq[s1][s2][iage] weighted
4464: * end iind
4465: * if covariate and j!0
4466: * headers Variable on one line
4467: * endif cov j!=0
4468: * header of frequency table by age
4469: * Loop on age
4470: * pp[s1]+=freq[s1][s2][iage] weighted
4471: * pos+=freq[s1][s2][iage] weighted
4472: * Loop on s1 initial state
4473: * fprintf(ficresp
4474: * end s1
4475: * end age
4476: * if j!=0 computes starting values
4477: * end compute starting values
4478: * end j1
4479: * end nl
4480: */
1.253 brouard 4481: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4482: if(nj==1)
4483: j=0; /* First pass for the constant */
1.265 brouard 4484: else{
1.253 brouard 4485: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4486: }
1.251 brouard 4487: first=1;
1.265 brouard 4488: 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 4489: posproptt=0.;
4490: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4491: scanf("%d", i);*/
4492: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4493: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4494: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4495: freq[i][s2][m]=0;
1.251 brouard 4496:
4497: for (i=1; i<=nlstate; i++) {
1.240 brouard 4498: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4499: prop[i][m]=0;
4500: posprop[i]=0;
4501: pospropt[i]=0;
4502: }
1.283 brouard 4503: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4504: idq[z1]=0.;
4505: meanq[z1]=0.;
4506: stdq[z1]=0.;
1.283 brouard 4507: }
4508: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4509: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4510: /* meanqt[m][z1]=0.; */
4511: /* } */
4512: /* } */
1.251 brouard 4513: /* dateintsum=0; */
4514: /* k2cpt=0; */
4515:
1.265 brouard 4516: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4517: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4518: bool=1;
4519: if(j !=0){
4520: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4521: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4522: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4523: /* if(Tvaraff[z1] ==-20){ */
4524: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4525: /* }else if(Tvaraff[z1] ==-10){ */
4526: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4527: /* }else */
4528: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4529: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4530: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4531: /* 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",
4532: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4533: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4534: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4535: } /* Onlyf fixed */
4536: } /* end z1 */
4537: } /* cptcovn > 0 */
4538: } /* end any */
4539: }/* end j==0 */
1.265 brouard 4540: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4541: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4542: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4543: m=mw[mi][iind];
4544: if(j!=0){
4545: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4546: for (z1=1; z1<=cptcoveff; z1++) {
4547: if( Fixed[Tmodelind[z1]]==1){
4548: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4549: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4550: value is -1, we don't select. It differs from the
4551: constant and age model which counts them. */
4552: bool=0; /* not selected */
4553: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4554: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4555: bool=0;
4556: }
4557: }
4558: }
4559: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4560: } /* end j==0 */
4561: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4562: if(bool==1){ /*Selected */
1.251 brouard 4563: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4564: and mw[mi+1][iind]. dh depends on stepm. */
4565: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4566: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4567: if(m >=firstpass && m <=lastpass){
4568: k2=anint[m][iind]+(mint[m][iind]/12.);
4569: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4570: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4571: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4572: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4573: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4574: if (m<lastpass) {
4575: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4576: /* 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]); */
4577: if(s[m][iind]==-1)
4578: 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.));
4579: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
1.284 brouard 4580: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean */
4581: idq[z1]=idq[z1]+weight[iind];
4582: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4583: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4584: }
1.251 brouard 4585: /* if((int)agev[m][iind] == 55) */
4586: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4587: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4588: 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 4589: }
1.251 brouard 4590: } /* end if between passes */
4591: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4592: dateintsum=dateintsum+k2; /* on all covariates ?*/
4593: k2cpt++;
4594: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4595: }
1.251 brouard 4596: }else{
4597: bool=1;
4598: }/* end bool 2 */
4599: } /* end m */
1.284 brouard 4600: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4601: /* idq[z1]=idq[z1]+weight[iind]; */
4602: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4603: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4604: /* } */
1.251 brouard 4605: } /* end bool */
4606: } /* end iind = 1 to imx */
4607: /* prop[s][age] is feeded for any initial and valid live state as well as
4608: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4609:
4610:
4611: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4612: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4613: pstamp(ficresp);
1.251 brouard 4614: if (cptcoveff>0 && j!=0){
1.265 brouard 4615: pstamp(ficresp);
1.251 brouard 4616: printf( "\n#********** Variable ");
4617: fprintf(ficresp, "\n#********** Variable ");
4618: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4619: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4620: fprintf(ficlog, "\n#********** Variable ");
4621: for (z1=1; z1<=cptcoveff; z1++){
4622: if(!FixedV[Tvaraff[z1]]){
4623: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4624: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4625: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4626: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4627: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4628: }else{
1.251 brouard 4629: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4630: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4631: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4632: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4633: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4634: }
4635: }
4636: printf( "**********\n#");
4637: fprintf(ficresp, "**********\n#");
4638: fprintf(ficresphtm, "**********</h3>\n");
4639: fprintf(ficresphtmfr, "**********</h3>\n");
4640: fprintf(ficlog, "**********\n");
4641: }
1.284 brouard 4642: /*
4643: Printing means of quantitative variables if any
4644: */
4645: for (z1=1; z1<= nqfveff; z1++) {
1.285 brouard 4646: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.0f individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.284 brouard 4647: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
4648: if(weightopt==1){
4649: printf(" Weighted mean and standard deviation of");
4650: fprintf(ficlog," Weighted mean and standard deviation of");
4651: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4652: }
1.285 brouard 4653: printf(" fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
4654: fprintf(ficlog," fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
4655: fprintf(ficresphtmfr," fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)<p>\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
1.284 brouard 4656: }
4657: /* for (z1=1; z1<= nqtveff; z1++) { */
4658: /* for(m=1;m<=lastpass;m++){ */
4659: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4660: /* } */
4661: /* } */
1.283 brouard 4662:
1.251 brouard 4663: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4664: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4665: fprintf(ficresp, " Age");
4666: 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 4667: for(i=1; i<=nlstate;i++) {
1.265 brouard 4668: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4669: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4670: }
1.265 brouard 4671: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4672: fprintf(ficresphtm, "\n");
4673:
4674: /* Header of frequency table by age */
4675: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4676: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4677: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4678: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4679: if(s2!=0 && m!=0)
4680: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4681: }
1.226 brouard 4682: }
1.251 brouard 4683: fprintf(ficresphtmfr, "\n");
4684:
4685: /* For each age */
4686: for(iage=iagemin; iage <= iagemax+3; iage++){
4687: fprintf(ficresphtm,"<tr>");
4688: if(iage==iagemax+1){
4689: fprintf(ficlog,"1");
4690: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4691: }else if(iage==iagemax+2){
4692: fprintf(ficlog,"0");
4693: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4694: }else if(iage==iagemax+3){
4695: fprintf(ficlog,"Total");
4696: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4697: }else{
1.240 brouard 4698: if(first==1){
1.251 brouard 4699: first=0;
4700: printf("See log file for details...\n");
4701: }
4702: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4703: fprintf(ficlog,"Age %d", iage);
4704: }
1.265 brouard 4705: for(s1=1; s1 <=nlstate ; s1++){
4706: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4707: pp[s1] += freq[s1][m][iage];
1.251 brouard 4708: }
1.265 brouard 4709: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4710: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4711: pos += freq[s1][m][iage];
4712: if(pp[s1]>=1.e-10){
1.251 brouard 4713: if(first==1){
1.265 brouard 4714: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4715: }
1.265 brouard 4716: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4717: }else{
4718: if(first==1)
1.265 brouard 4719: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4720: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4721: }
4722: }
4723:
1.265 brouard 4724: for(s1=1; s1 <=nlstate ; s1++){
4725: /* posprop[s1]=0; */
4726: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4727: pp[s1] += freq[s1][m][iage];
4728: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4729:
4730: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4731: pos += pp[s1]; /* pos is the total number of transitions until this age */
4732: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4733: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4734: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4735: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4736: }
4737:
4738: /* Writing ficresp */
4739: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4740: if( iage <= iagemax){
4741: fprintf(ficresp," %d",iage);
4742: }
4743: }else if( nj==2){
4744: if( iage <= iagemax){
4745: fprintf(ficresp," %d",iage);
4746: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4747: }
1.240 brouard 4748: }
1.265 brouard 4749: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4750: if(pos>=1.e-5){
1.251 brouard 4751: if(first==1)
1.265 brouard 4752: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4753: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4754: }else{
4755: if(first==1)
1.265 brouard 4756: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4757: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4758: }
4759: if( iage <= iagemax){
4760: if(pos>=1.e-5){
1.265 brouard 4761: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4762: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4763: }else if( nj==2){
4764: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4765: }
4766: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4767: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4768: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4769: } else{
4770: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4771: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4772: }
1.240 brouard 4773: }
1.265 brouard 4774: pospropt[s1] +=posprop[s1];
4775: } /* end loop s1 */
1.251 brouard 4776: /* pospropt=0.; */
1.265 brouard 4777: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4778: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4779: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4780: if(first==1){
1.265 brouard 4781: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4782: }
1.265 brouard 4783: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4784: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4785: }
1.265 brouard 4786: if(s1!=0 && m!=0)
4787: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4788: }
1.265 brouard 4789: } /* end loop s1 */
1.251 brouard 4790: posproptt=0.;
1.265 brouard 4791: for(s1=1; s1 <=nlstate; s1++){
4792: posproptt += pospropt[s1];
1.251 brouard 4793: }
4794: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4795: fprintf(ficresphtm,"</tr>\n");
4796: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4797: if(iage <= iagemax)
4798: fprintf(ficresp,"\n");
1.240 brouard 4799: }
1.251 brouard 4800: if(first==1)
4801: printf("Others in log...\n");
4802: fprintf(ficlog,"\n");
4803: } /* end loop age iage */
1.265 brouard 4804:
1.251 brouard 4805: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4806: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4807: if(posproptt < 1.e-5){
1.265 brouard 4808: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4809: }else{
1.265 brouard 4810: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4811: }
1.226 brouard 4812: }
1.251 brouard 4813: fprintf(ficresphtm,"</tr>\n");
4814: fprintf(ficresphtm,"</table>\n");
4815: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4816: if(posproptt < 1.e-5){
1.251 brouard 4817: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4818: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4819: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4820: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4821: invalidvarcomb[j1]=1;
1.226 brouard 4822: }else{
1.251 brouard 4823: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4824: invalidvarcomb[j1]=0;
1.226 brouard 4825: }
1.251 brouard 4826: fprintf(ficresphtmfr,"</table>\n");
4827: fprintf(ficlog,"\n");
4828: if(j!=0){
4829: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4830: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4831: for(k=1; k <=(nlstate+ndeath); k++){
4832: if (k != i) {
1.265 brouard 4833: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4834: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4835: if(j1==1){ /* All dummy covariates to zero */
4836: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4837: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4838: printf("%d%d ",i,k);
4839: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4840: 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]));
4841: 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]));
4842: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4843: }
1.253 brouard 4844: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4845: for(iage=iagemin; iage <= iagemax+3; iage++){
4846: x[iage]= (double)iage;
4847: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4848: /* 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 4849: }
1.268 brouard 4850: /* Some are not finite, but linreg will ignore these ages */
4851: no=0;
1.253 brouard 4852: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4853: pstart[s1]=b;
4854: pstart[s1-1]=a;
1.252 brouard 4855: }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 */
4856: 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]);
4857: 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 4858: 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 4859: printf("%d%d ",i,k);
4860: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4861: 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 4862: }else{ /* Other cases, like quantitative fixed or varying covariates */
4863: ;
4864: }
4865: /* printf("%12.7f )", param[i][jj][k]); */
4866: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4867: s1++;
1.251 brouard 4868: } /* end jj */
4869: } /* end k!= i */
4870: } /* end k */
1.265 brouard 4871: } /* end i, s1 */
1.251 brouard 4872: } /* end j !=0 */
4873: } /* end selected combination of covariate j1 */
4874: if(j==0){ /* We can estimate starting values from the occurences in each case */
4875: printf("#Freqsummary: Starting values for the constants:\n");
4876: fprintf(ficlog,"\n");
1.265 brouard 4877: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4878: for(k=1; k <=(nlstate+ndeath); k++){
4879: if (k != i) {
4880: printf("%d%d ",i,k);
4881: fprintf(ficlog,"%d%d ",i,k);
4882: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4883: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4884: if(jj==1){ /* Age has to be done */
1.265 brouard 4885: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4886: 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]));
4887: 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 4888: }
4889: /* printf("%12.7f )", param[i][jj][k]); */
4890: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4891: s1++;
1.250 brouard 4892: }
1.251 brouard 4893: printf("\n");
4894: fprintf(ficlog,"\n");
1.250 brouard 4895: }
4896: }
1.284 brouard 4897: } /* end of state i */
1.251 brouard 4898: printf("#Freqsummary\n");
4899: fprintf(ficlog,"\n");
1.265 brouard 4900: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4901: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4902: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4903: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4904: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4905: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4906: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4907: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4908: /* } */
4909: }
1.265 brouard 4910: } /* end loop s1 */
1.251 brouard 4911:
4912: printf("\n");
4913: fprintf(ficlog,"\n");
4914: } /* end j=0 */
1.249 brouard 4915: } /* end j */
1.252 brouard 4916:
1.253 brouard 4917: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4918: for(i=1, jk=1; i <=nlstate; i++){
4919: for(j=1; j <=nlstate+ndeath; j++){
4920: if(j!=i){
4921: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4922: printf("%1d%1d",i,j);
4923: fprintf(ficparo,"%1d%1d",i,j);
4924: for(k=1; k<=ncovmodel;k++){
4925: /* printf(" %lf",param[i][j][k]); */
4926: /* fprintf(ficparo," %lf",param[i][j][k]); */
4927: p[jk]=pstart[jk];
4928: printf(" %f ",pstart[jk]);
4929: fprintf(ficparo," %f ",pstart[jk]);
4930: jk++;
4931: }
4932: printf("\n");
4933: fprintf(ficparo,"\n");
4934: }
4935: }
4936: }
4937: } /* end mle=-2 */
1.226 brouard 4938: dateintmean=dateintsum/k2cpt;
1.240 brouard 4939:
1.226 brouard 4940: fclose(ficresp);
4941: fclose(ficresphtm);
4942: fclose(ficresphtmfr);
1.283 brouard 4943: free_vector(idq,1,nqfveff);
1.226 brouard 4944: free_vector(meanq,1,nqfveff);
1.284 brouard 4945: free_vector(stdq,1,nqfveff);
1.226 brouard 4946: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4947: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4948: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4949: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4950: free_vector(pospropt,1,nlstate);
4951: free_vector(posprop,1,nlstate);
1.251 brouard 4952: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4953: free_vector(pp,1,nlstate);
4954: /* End of freqsummary */
4955: }
1.126 brouard 4956:
1.268 brouard 4957: /* Simple linear regression */
4958: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4959:
4960: /* y=a+bx regression */
4961: double sumx = 0.0; /* sum of x */
4962: double sumx2 = 0.0; /* sum of x**2 */
4963: double sumxy = 0.0; /* sum of x * y */
4964: double sumy = 0.0; /* sum of y */
4965: double sumy2 = 0.0; /* sum of y**2 */
4966: double sume2 = 0.0; /* sum of square or residuals */
4967: double yhat;
4968:
4969: double denom=0;
4970: int i;
4971: int ne=*no;
4972:
4973: for ( i=ifi, ne=0;i<=ila;i++) {
4974: if(!isfinite(x[i]) || !isfinite(y[i])){
4975: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4976: continue;
4977: }
4978: ne=ne+1;
4979: sumx += x[i];
4980: sumx2 += x[i]*x[i];
4981: sumxy += x[i] * y[i];
4982: sumy += y[i];
4983: sumy2 += y[i]*y[i];
4984: denom = (ne * sumx2 - sumx*sumx);
4985: /* 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); */
4986: }
4987:
4988: denom = (ne * sumx2 - sumx*sumx);
4989: if (denom == 0) {
4990: // vertical, slope m is infinity
4991: *b = INFINITY;
4992: *a = 0;
4993: if (r) *r = 0;
4994: return 1;
4995: }
4996:
4997: *b = (ne * sumxy - sumx * sumy) / denom;
4998: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4999: if (r!=NULL) {
5000: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5001: sqrt((sumx2 - sumx*sumx/ne) *
5002: (sumy2 - sumy*sumy/ne));
5003: }
5004: *no=ne;
5005: for ( i=ifi, ne=0;i<=ila;i++) {
5006: if(!isfinite(x[i]) || !isfinite(y[i])){
5007: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5008: continue;
5009: }
5010: ne=ne+1;
5011: yhat = y[i] - *a -*b* x[i];
5012: sume2 += yhat * yhat ;
5013:
5014: denom = (ne * sumx2 - sumx*sumx);
5015: /* 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); */
5016: }
5017: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5018: *sa= *sb * sqrt(sumx2/ne);
5019:
5020: return 0;
5021: }
5022:
1.126 brouard 5023: /************ Prevalence ********************/
1.227 brouard 5024: 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)
5025: {
5026: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5027: in each health status at the date of interview (if between dateprev1 and dateprev2).
5028: We still use firstpass and lastpass as another selection.
5029: */
1.126 brouard 5030:
1.227 brouard 5031: int i, m, jk, j1, bool, z1,j, iv;
5032: int mi; /* Effective wave */
5033: int iage;
5034: double agebegin, ageend;
5035:
5036: double **prop;
5037: double posprop;
5038: double y2; /* in fractional years */
5039: int iagemin, iagemax;
5040: int first; /** to stop verbosity which is redirected to log file */
5041:
5042: iagemin= (int) agemin;
5043: iagemax= (int) agemax;
5044: /*pp=vector(1,nlstate);*/
1.251 brouard 5045: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5046: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5047: j1=0;
1.222 brouard 5048:
1.227 brouard 5049: /*j=cptcoveff;*/
5050: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5051:
1.227 brouard 5052: first=1;
5053: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5054: for (i=1; i<=nlstate; i++)
1.251 brouard 5055: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5056: prop[i][iage]=0.0;
5057: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5058: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5059: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5060:
5061: for (i=1; i<=imx; i++) { /* Each individual */
5062: bool=1;
5063: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5064: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5065: m=mw[mi][i];
5066: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5067: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5068: for (z1=1; z1<=cptcoveff; z1++){
5069: if( Fixed[Tmodelind[z1]]==1){
5070: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5071: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5072: bool=0;
5073: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5074: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5075: bool=0;
5076: }
5077: }
5078: if(bool==1){ /* Otherwise we skip that wave/person */
5079: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5080: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5081: if(m >=firstpass && m <=lastpass){
5082: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5083: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5084: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5085: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5086: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5087: 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);
5088: exit(1);
5089: }
5090: if (s[m][i]>0 && s[m][i]<=nlstate) {
5091: /*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]]);*/
5092: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5093: prop[s[m][i]][iagemax+3] += weight[i];
5094: } /* end valid statuses */
5095: } /* end selection of dates */
5096: } /* end selection of waves */
5097: } /* end bool */
5098: } /* end wave */
5099: } /* end individual */
5100: for(i=iagemin; i <= iagemax+3; i++){
5101: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5102: posprop += prop[jk][i];
5103: }
5104:
5105: for(jk=1; jk <=nlstate ; jk++){
5106: if( i <= iagemax){
5107: if(posprop>=1.e-5){
5108: probs[i][jk][j1]= prop[jk][i]/posprop;
5109: } else{
5110: if(first==1){
5111: first=0;
1.266 brouard 5112: 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]);
5113: fprintf(ficlog,"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]);
5114: }else{
5115: fprintf(ficlog,"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]);
1.227 brouard 5116: }
5117: }
5118: }
5119: }/* end jk */
5120: }/* end i */
1.222 brouard 5121: /*} *//* end i1 */
1.227 brouard 5122: } /* end j1 */
1.222 brouard 5123:
1.227 brouard 5124: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5125: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5126: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5127: } /* End of prevalence */
1.126 brouard 5128:
5129: /************* Waves Concatenation ***************/
5130:
5131: 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)
5132: {
5133: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5134: Death is a valid wave (if date is known).
5135: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5136: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5137: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5138: */
1.126 brouard 5139:
1.224 brouard 5140: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5141: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5142: double sum=0., jmean=0.;*/
1.224 brouard 5143: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5144: int j, k=0,jk, ju, jl;
5145: double sum=0.;
5146: first=0;
1.214 brouard 5147: firstwo=0;
1.217 brouard 5148: firsthree=0;
1.218 brouard 5149: firstfour=0;
1.164 brouard 5150: jmin=100000;
1.126 brouard 5151: jmax=-1;
5152: jmean=0.;
1.224 brouard 5153:
5154: /* Treating live states */
1.214 brouard 5155: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5156: mi=0; /* First valid wave */
1.227 brouard 5157: mli=0; /* Last valid wave */
1.126 brouard 5158: m=firstpass;
1.214 brouard 5159: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5160: 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 */
5161: mli=m-1;/* mw[++mi][i]=m-1; */
5162: }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 */
5163: mw[++mi][i]=m;
5164: mli=m;
1.224 brouard 5165: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5166: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5167: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5168: }
1.227 brouard 5169: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5170: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5171: break;
1.224 brouard 5172: #else
1.227 brouard 5173: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5174: if(firsthree == 0){
1.262 brouard 5175: 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 5176: firsthree=1;
5177: }
1.262 brouard 5178: 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);
1.227 brouard 5179: mw[++mi][i]=m;
5180: mli=m;
5181: }
5182: if(s[m][i]==-2){ /* Vital status is really unknown */
5183: nbwarn++;
5184: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5185: 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);
5186: 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);
5187: }
5188: break;
5189: }
5190: break;
1.224 brouard 5191: #endif
1.227 brouard 5192: }/* End m >= lastpass */
1.126 brouard 5193: }/* end while */
1.224 brouard 5194:
1.227 brouard 5195: /* 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 5196: /* After last pass */
1.224 brouard 5197: /* Treating death states */
1.214 brouard 5198: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5199: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5200: /* } */
1.126 brouard 5201: mi++; /* Death is another wave */
5202: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5203: /* Only death is a correct wave */
1.126 brouard 5204: mw[mi][i]=m;
1.257 brouard 5205: } /* else not in a death state */
1.224 brouard 5206: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5207: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5208: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5209: if((andc[i]+moisdc[i]/12.) <=(anint[m][i]+mint[m][i]/12.)){ /* death occured before last wave and status should have been death instead of -1 */
5210: nbwarn++;
5211: if(firstfiv==0){
5212: printf("Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d interviewed at %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 );
5213: firstfiv=1;
5214: }else{
5215: fprintf(ficlog,"Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d interviewed at %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 );
5216: }
5217: }else{ /* Death occured afer last wave potential bias */
5218: nberr++;
5219: if(firstwo==0){
1.257 brouard 5220: printf("Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%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. Please add a new fictive wave at the date of last vital status scan, with a dead status or alive but unknown state status (-1). 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], i,m );
1.227 brouard 5221: firstwo=1;
5222: }
1.257 brouard 5223: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%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. Please add a new fictive wave at the date of last vital status scan, with a dead status or alive but unknown state status (-1). See documentation\n\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
1.227 brouard 5224: }
1.257 brouard 5225: }else{ /* if date of interview is unknown */
1.227 brouard 5226: /* death is known but not confirmed by death status at any wave */
5227: if(firstfour==0){
5228: 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. 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], i,m );
5229: firstfour=1;
5230: }
5231: 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. 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], i,m );
1.214 brouard 5232: }
1.224 brouard 5233: } /* end if date of death is known */
5234: #endif
5235: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5236: /* wav[i]=mw[mi][i]; */
1.126 brouard 5237: if(mi==0){
5238: nbwarn++;
5239: if(first==0){
1.227 brouard 5240: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5241: first=1;
1.126 brouard 5242: }
5243: if(first==1){
1.227 brouard 5244: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5245: }
5246: } /* end mi==0 */
5247: } /* End individuals */
1.214 brouard 5248: /* wav and mw are no more changed */
1.223 brouard 5249:
1.214 brouard 5250:
1.126 brouard 5251: for(i=1; i<=imx; i++){
5252: for(mi=1; mi<wav[i];mi++){
5253: if (stepm <=0)
1.227 brouard 5254: dh[mi][i]=1;
1.126 brouard 5255: else{
1.260 brouard 5256: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5257: if (agedc[i] < 2*AGESUP) {
5258: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5259: if(j==0) j=1; /* Survives at least one month after exam */
5260: else if(j<0){
5261: nberr++;
5262: 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]);
5263: j=1; /* Temporary Dangerous patch */
5264: 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);
5265: 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]);
5266: 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);
5267: }
5268: k=k+1;
5269: if (j >= jmax){
5270: jmax=j;
5271: ijmax=i;
5272: }
5273: if (j <= jmin){
5274: jmin=j;
5275: ijmin=i;
5276: }
5277: sum=sum+j;
5278: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5279: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5280: }
5281: }
5282: else{
5283: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5284: /* 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 5285:
1.227 brouard 5286: k=k+1;
5287: if (j >= jmax) {
5288: jmax=j;
5289: ijmax=i;
5290: }
5291: else if (j <= jmin){
5292: jmin=j;
5293: ijmin=i;
5294: }
5295: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5296: /*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]);*/
5297: if(j<0){
5298: nberr++;
5299: 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]);
5300: 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]);
5301: }
5302: sum=sum+j;
5303: }
5304: jk= j/stepm;
5305: jl= j -jk*stepm;
5306: ju= j -(jk+1)*stepm;
5307: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5308: if(jl==0){
5309: dh[mi][i]=jk;
5310: bh[mi][i]=0;
5311: }else{ /* We want a negative bias in order to only have interpolation ie
5312: * to avoid the price of an extra matrix product in likelihood */
5313: dh[mi][i]=jk+1;
5314: bh[mi][i]=ju;
5315: }
5316: }else{
5317: if(jl <= -ju){
5318: dh[mi][i]=jk;
5319: bh[mi][i]=jl; /* bias is positive if real duration
5320: * is higher than the multiple of stepm and negative otherwise.
5321: */
5322: }
5323: else{
5324: dh[mi][i]=jk+1;
5325: bh[mi][i]=ju;
5326: }
5327: if(dh[mi][i]==0){
5328: dh[mi][i]=1; /* At least one step */
5329: bh[mi][i]=ju; /* At least one step */
5330: /* 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);*/
5331: }
5332: } /* end if mle */
1.126 brouard 5333: }
5334: } /* end wave */
5335: }
5336: jmean=sum/k;
5337: 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 5338: 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 5339: }
1.126 brouard 5340:
5341: /*********** Tricode ****************************/
1.220 brouard 5342: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5343: {
5344: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5345: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5346: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5347: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5348: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5349: */
1.130 brouard 5350:
1.242 brouard 5351: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5352: int modmaxcovj=0; /* Modality max of covariates j */
5353: int cptcode=0; /* Modality max of covariates j */
5354: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5355:
5356:
1.242 brouard 5357: /* cptcoveff=0; */
5358: /* *cptcov=0; */
1.126 brouard 5359:
1.242 brouard 5360: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5361: for (k=1; k <= maxncov; k++)
5362: for(j=1; j<=2; j++)
5363: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5364:
1.242 brouard 5365: /* Loop on covariates without age and products and no quantitative variable */
5366: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5367: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5368: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5369: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5370: switch(Fixed[k]) {
5371: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5372: 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*/
5373: ij=(int)(covar[Tvar[k]][i]);
5374: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5375: * If product of Vn*Vm, still boolean *:
5376: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5377: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5378: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5379: modality of the nth covariate of individual i. */
5380: if (ij > modmaxcovj)
5381: modmaxcovj=ij;
5382: else if (ij < modmincovj)
5383: modmincovj=ij;
5384: if ((ij < -1) && (ij > NCOVMAX)){
5385: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5386: exit(1);
5387: }else
5388: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5389: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5390: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5391: /* getting the maximum value of the modality of the covariate
5392: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5393: female ies 1, then modmaxcovj=1.
5394: */
5395: } /* end for loop on individuals i */
5396: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5397: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5398: cptcode=modmaxcovj;
5399: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5400: /*for (i=0; i<=cptcode; i++) {*/
5401: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5402: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5403: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5404: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5405: if( j != -1){
5406: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5407: covariate for which somebody answered excluding
5408: undefined. Usually 2: 0 and 1. */
5409: }
5410: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5411: covariate for which somebody answered including
5412: undefined. Usually 3: -1, 0 and 1. */
5413: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5414: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5415: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5416:
1.242 brouard 5417: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5418: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5419: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5420: /* modmincovj=3; modmaxcovj = 7; */
5421: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5422: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5423: /* defining two dummy variables: variables V1_1 and V1_2.*/
5424: /* nbcode[Tvar[j]][ij]=k; */
5425: /* nbcode[Tvar[j]][1]=0; */
5426: /* nbcode[Tvar[j]][2]=1; */
5427: /* nbcode[Tvar[j]][3]=2; */
5428: /* To be continued (not working yet). */
5429: ij=0; /* ij is similar to i but can jump over null modalities */
5430: 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*/
5431: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5432: break;
5433: }
5434: ij++;
5435: 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*/
5436: cptcode = ij; /* New max modality for covar j */
5437: } /* end of loop on modality i=-1 to 1 or more */
5438: break;
5439: case 1: /* Testing on varying covariate, could be simple and
5440: * should look at waves or product of fixed *
5441: * varying. No time to test -1, assuming 0 and 1 only */
5442: ij=0;
5443: for(i=0; i<=1;i++){
5444: nbcode[Tvar[k]][++ij]=i;
5445: }
5446: break;
5447: default:
5448: break;
5449: } /* end switch */
5450: } /* end dummy test */
5451:
5452: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5453: /* /\*recode from 0 *\/ */
5454: /* k is a modality. If we have model=V1+V1*sex */
5455: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5456: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5457: /* } */
5458: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5459: /* if (ij > ncodemax[j]) { */
5460: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5461: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5462: /* break; */
5463: /* } */
5464: /* } /\* end of loop on modality k *\/ */
5465: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5466:
5467: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5468: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5469: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5470: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5471: 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 */
5472: 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 */
5473: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5474: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5475:
5476: ij=0;
5477: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5478: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5479: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5480: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5481: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5482: /* If product not in single variable we don't print results */
5483: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5484: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5485: 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*/
5486: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5487: 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 */
5488: if(Fixed[k]!=0)
5489: anyvaryingduminmodel=1;
5490: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5491: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5492: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5493: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5494: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5495: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5496: }
5497: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5498: /* ij--; */
5499: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5500: *cptcov=ij; /*Number of total real effective covariates: effective
5501: * because they can be excluded from the model and real
5502: * if in the model but excluded because missing values, but how to get k from ij?*/
5503: for(j=ij+1; j<= cptcovt; j++){
5504: Tvaraff[j]=0;
5505: Tmodelind[j]=0;
5506: }
5507: for(j=ntveff+1; j<= cptcovt; j++){
5508: TmodelInvind[j]=0;
5509: }
5510: /* To be sorted */
5511: ;
5512: }
1.126 brouard 5513:
1.145 brouard 5514:
1.126 brouard 5515: /*********** Health Expectancies ****************/
5516:
1.235 brouard 5517: 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 5518:
5519: {
5520: /* Health expectancies, no variances */
1.164 brouard 5521: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5522: int nhstepma, nstepma; /* Decreasing with age */
5523: double age, agelim, hf;
5524: double ***p3mat;
5525: double eip;
5526:
1.238 brouard 5527: /* pstamp(ficreseij); */
1.126 brouard 5528: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5529: fprintf(ficreseij,"# Age");
5530: for(i=1; i<=nlstate;i++){
5531: for(j=1; j<=nlstate;j++){
5532: fprintf(ficreseij," e%1d%1d ",i,j);
5533: }
5534: fprintf(ficreseij," e%1d. ",i);
5535: }
5536: fprintf(ficreseij,"\n");
5537:
5538:
5539: if(estepm < stepm){
5540: printf ("Problem %d lower than %d\n",estepm, stepm);
5541: }
5542: else hstepm=estepm;
5543: /* We compute the life expectancy from trapezoids spaced every estepm months
5544: * This is mainly to measure the difference between two models: for example
5545: * if stepm=24 months pijx are given only every 2 years and by summing them
5546: * we are calculating an estimate of the Life Expectancy assuming a linear
5547: * progression in between and thus overestimating or underestimating according
5548: * to the curvature of the survival function. If, for the same date, we
5549: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5550: * to compare the new estimate of Life expectancy with the same linear
5551: * hypothesis. A more precise result, taking into account a more precise
5552: * curvature will be obtained if estepm is as small as stepm. */
5553:
5554: /* For example we decided to compute the life expectancy with the smallest unit */
5555: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5556: nhstepm is the number of hstepm from age to agelim
5557: nstepm is the number of stepm from age to agelin.
1.270 brouard 5558: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5559: and note for a fixed period like estepm months */
5560: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5561: survival function given by stepm (the optimization length). Unfortunately it
5562: means that if the survival funtion is printed only each two years of age and if
5563: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5564: results. So we changed our mind and took the option of the best precision.
5565: */
5566: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5567:
5568: agelim=AGESUP;
5569: /* If stepm=6 months */
5570: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5571: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5572:
5573: /* nhstepm age range expressed in number of stepm */
5574: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5575: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5576: /* if (stepm >= YEARM) hstepm=1;*/
5577: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5578: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5579:
5580: for (age=bage; age<=fage; age ++){
5581: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5582: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5583: /* if (stepm >= YEARM) hstepm=1;*/
5584: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5585:
5586: /* If stepm=6 months */
5587: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5588: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5589:
1.235 brouard 5590: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5591:
5592: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5593:
5594: printf("%d|",(int)age);fflush(stdout);
5595: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5596:
5597: /* Computing expectancies */
5598: for(i=1; i<=nlstate;i++)
5599: for(j=1; j<=nlstate;j++)
5600: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5601: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5602:
5603: /* 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]);*/
5604:
5605: }
5606:
5607: fprintf(ficreseij,"%3.0f",age );
5608: for(i=1; i<=nlstate;i++){
5609: eip=0;
5610: for(j=1; j<=nlstate;j++){
5611: eip +=eij[i][j][(int)age];
5612: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5613: }
5614: fprintf(ficreseij,"%9.4f", eip );
5615: }
5616: fprintf(ficreseij,"\n");
5617:
5618: }
5619: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5620: printf("\n");
5621: fprintf(ficlog,"\n");
5622:
5623: }
5624:
1.235 brouard 5625: 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 5626:
5627: {
5628: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5629: to initial status i, ei. .
1.126 brouard 5630: */
5631: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5632: int nhstepma, nstepma; /* Decreasing with age */
5633: double age, agelim, hf;
5634: double ***p3matp, ***p3matm, ***varhe;
5635: double **dnewm,**doldm;
5636: double *xp, *xm;
5637: double **gp, **gm;
5638: double ***gradg, ***trgradg;
5639: int theta;
5640:
5641: double eip, vip;
5642:
5643: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5644: xp=vector(1,npar);
5645: xm=vector(1,npar);
5646: dnewm=matrix(1,nlstate*nlstate,1,npar);
5647: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5648:
5649: pstamp(ficresstdeij);
5650: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5651: fprintf(ficresstdeij,"# Age");
5652: for(i=1; i<=nlstate;i++){
5653: for(j=1; j<=nlstate;j++)
5654: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5655: fprintf(ficresstdeij," e%1d. ",i);
5656: }
5657: fprintf(ficresstdeij,"\n");
5658:
5659: pstamp(ficrescveij);
5660: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5661: fprintf(ficrescveij,"# Age");
5662: for(i=1; i<=nlstate;i++)
5663: for(j=1; j<=nlstate;j++){
5664: cptj= (j-1)*nlstate+i;
5665: for(i2=1; i2<=nlstate;i2++)
5666: for(j2=1; j2<=nlstate;j2++){
5667: cptj2= (j2-1)*nlstate+i2;
5668: if(cptj2 <= cptj)
5669: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5670: }
5671: }
5672: fprintf(ficrescveij,"\n");
5673:
5674: if(estepm < stepm){
5675: printf ("Problem %d lower than %d\n",estepm, stepm);
5676: }
5677: else hstepm=estepm;
5678: /* We compute the life expectancy from trapezoids spaced every estepm months
5679: * This is mainly to measure the difference between two models: for example
5680: * if stepm=24 months pijx are given only every 2 years and by summing them
5681: * we are calculating an estimate of the Life Expectancy assuming a linear
5682: * progression in between and thus overestimating or underestimating according
5683: * to the curvature of the survival function. If, for the same date, we
5684: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5685: * to compare the new estimate of Life expectancy with the same linear
5686: * hypothesis. A more precise result, taking into account a more precise
5687: * curvature will be obtained if estepm is as small as stepm. */
5688:
5689: /* For example we decided to compute the life expectancy with the smallest unit */
5690: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5691: nhstepm is the number of hstepm from age to agelim
5692: nstepm is the number of stepm from age to agelin.
5693: Look at hpijx to understand the reason of that which relies in memory size
5694: and note for a fixed period like estepm months */
5695: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5696: survival function given by stepm (the optimization length). Unfortunately it
5697: means that if the survival funtion is printed only each two years of age and if
5698: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5699: results. So we changed our mind and took the option of the best precision.
5700: */
5701: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5702:
5703: /* If stepm=6 months */
5704: /* nhstepm age range expressed in number of stepm */
5705: agelim=AGESUP;
5706: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5707: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5708: /* if (stepm >= YEARM) hstepm=1;*/
5709: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5710:
5711: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5712: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5713: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5714: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5715: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5716: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5717:
5718: for (age=bage; age<=fage; age ++){
5719: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5720: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5721: /* if (stepm >= YEARM) hstepm=1;*/
5722: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5723:
1.126 brouard 5724: /* If stepm=6 months */
5725: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5726: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5727:
5728: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5729:
1.126 brouard 5730: /* Computing Variances of health expectancies */
5731: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5732: decrease memory allocation */
5733: for(theta=1; theta <=npar; theta++){
5734: for(i=1; i<=npar; i++){
1.222 brouard 5735: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5736: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5737: }
1.235 brouard 5738: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5739: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5740:
1.126 brouard 5741: for(j=1; j<= nlstate; j++){
1.222 brouard 5742: for(i=1; i<=nlstate; i++){
5743: for(h=0; h<=nhstepm-1; h++){
5744: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5745: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5746: }
5747: }
1.126 brouard 5748: }
1.218 brouard 5749:
1.126 brouard 5750: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5751: for(h=0; h<=nhstepm-1; h++){
5752: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5753: }
1.126 brouard 5754: }/* End theta */
5755:
5756:
5757: for(h=0; h<=nhstepm-1; h++)
5758: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5759: for(theta=1; theta <=npar; theta++)
5760: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5761:
1.218 brouard 5762:
1.222 brouard 5763: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5764: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5765: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5766:
1.222 brouard 5767: printf("%d|",(int)age);fflush(stdout);
5768: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5769: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5770: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5771: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5772: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5773: for(ij=1;ij<=nlstate*nlstate;ij++)
5774: for(ji=1;ji<=nlstate*nlstate;ji++)
5775: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5776: }
5777: }
1.218 brouard 5778:
1.126 brouard 5779: /* Computing expectancies */
1.235 brouard 5780: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5781: for(i=1; i<=nlstate;i++)
5782: for(j=1; j<=nlstate;j++)
1.222 brouard 5783: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5784: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5785:
1.222 brouard 5786: /* 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 5787:
1.222 brouard 5788: }
1.269 brouard 5789:
5790: /* Standard deviation of expectancies ij */
1.126 brouard 5791: fprintf(ficresstdeij,"%3.0f",age );
5792: for(i=1; i<=nlstate;i++){
5793: eip=0.;
5794: vip=0.;
5795: for(j=1; j<=nlstate;j++){
1.222 brouard 5796: eip += eij[i][j][(int)age];
5797: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5798: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5799: 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 5800: }
5801: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5802: }
5803: fprintf(ficresstdeij,"\n");
1.218 brouard 5804:
1.269 brouard 5805: /* Variance of expectancies ij */
1.126 brouard 5806: fprintf(ficrescveij,"%3.0f",age );
5807: for(i=1; i<=nlstate;i++)
5808: for(j=1; j<=nlstate;j++){
1.222 brouard 5809: cptj= (j-1)*nlstate+i;
5810: for(i2=1; i2<=nlstate;i2++)
5811: for(j2=1; j2<=nlstate;j2++){
5812: cptj2= (j2-1)*nlstate+i2;
5813: if(cptj2 <= cptj)
5814: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5815: }
1.126 brouard 5816: }
5817: fprintf(ficrescveij,"\n");
1.218 brouard 5818:
1.126 brouard 5819: }
5820: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5821: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5822: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5823: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5824: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5825: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5826: printf("\n");
5827: fprintf(ficlog,"\n");
1.218 brouard 5828:
1.126 brouard 5829: free_vector(xm,1,npar);
5830: free_vector(xp,1,npar);
5831: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5832: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5833: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5834: }
1.218 brouard 5835:
1.126 brouard 5836: /************ Variance ******************/
1.235 brouard 5837: 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 5838: {
1.279 brouard 5839: /** Variance of health expectancies
5840: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5841: * double **newm;
5842: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5843: */
1.218 brouard 5844:
5845: /* int movingaverage(); */
5846: double **dnewm,**doldm;
5847: double **dnewmp,**doldmp;
5848: int i, j, nhstepm, hstepm, h, nstepm ;
5849: int k;
5850: double *xp;
1.279 brouard 5851: double **gp, **gm; /**< for var eij */
5852: double ***gradg, ***trgradg; /**< for var eij */
5853: double **gradgp, **trgradgp; /**< for var p point j */
5854: double *gpp, *gmp; /**< for var p point j */
5855: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5856: double ***p3mat;
5857: double age,agelim, hf;
5858: /* double ***mobaverage; */
5859: int theta;
5860: char digit[4];
5861: char digitp[25];
5862:
5863: char fileresprobmorprev[FILENAMELENGTH];
5864:
5865: if(popbased==1){
5866: if(mobilav!=0)
5867: strcpy(digitp,"-POPULBASED-MOBILAV_");
5868: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5869: }
5870: else
5871: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5872:
1.218 brouard 5873: /* if (mobilav!=0) { */
5874: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5875: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5876: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5877: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5878: /* } */
5879: /* } */
5880:
5881: strcpy(fileresprobmorprev,"PRMORPREV-");
5882: sprintf(digit,"%-d",ij);
5883: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5884: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5885: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5886: strcat(fileresprobmorprev,fileresu);
5887: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5888: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5889: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5890: }
5891: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5892: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5893: pstamp(ficresprobmorprev);
5894: 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 5895: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5896: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5897: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5898: }
5899: for(j=1;j<=cptcoveff;j++)
5900: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5901: fprintf(ficresprobmorprev,"\n");
5902:
1.218 brouard 5903: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5904: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5905: fprintf(ficresprobmorprev," p.%-d SE",j);
5906: for(i=1; i<=nlstate;i++)
5907: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5908: }
5909: fprintf(ficresprobmorprev,"\n");
5910:
5911: fprintf(ficgp,"\n# Routine varevsij");
5912: fprintf(ficgp,"\nunset title \n");
5913: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5914: 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");
5915: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 5916:
1.218 brouard 5917: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5918: pstamp(ficresvij);
5919: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5920: if(popbased==1)
5921: 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);
5922: else
5923: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5924: fprintf(ficresvij,"# Age");
5925: for(i=1; i<=nlstate;i++)
5926: for(j=1; j<=nlstate;j++)
5927: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5928: fprintf(ficresvij,"\n");
5929:
5930: xp=vector(1,npar);
5931: dnewm=matrix(1,nlstate,1,npar);
5932: doldm=matrix(1,nlstate,1,nlstate);
5933: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5934: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5935:
5936: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5937: gpp=vector(nlstate+1,nlstate+ndeath);
5938: gmp=vector(nlstate+1,nlstate+ndeath);
5939: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5940:
1.218 brouard 5941: if(estepm < stepm){
5942: printf ("Problem %d lower than %d\n",estepm, stepm);
5943: }
5944: else hstepm=estepm;
5945: /* For example we decided to compute the life expectancy with the smallest unit */
5946: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5947: nhstepm is the number of hstepm from age to agelim
5948: nstepm is the number of stepm from age to agelim.
5949: Look at function hpijx to understand why because of memory size limitations,
5950: we decided (b) to get a life expectancy respecting the most precise curvature of the
5951: survival function given by stepm (the optimization length). Unfortunately it
5952: means that if the survival funtion is printed every two years of age and if
5953: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5954: results. So we changed our mind and took the option of the best precision.
5955: */
5956: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5957: agelim = AGESUP;
5958: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5959: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5960: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5961: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5962: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5963: gp=matrix(0,nhstepm,1,nlstate);
5964: gm=matrix(0,nhstepm,1,nlstate);
5965:
5966:
5967: for(theta=1; theta <=npar; theta++){
5968: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5969: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5970: }
1.279 brouard 5971: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
5972: * returns into prlim .
5973: */
1.242 brouard 5974: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 5975:
5976: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 5977: if (popbased==1) {
5978: if(mobilav ==0){
5979: for(i=1; i<=nlstate;i++)
5980: prlim[i][i]=probs[(int)age][i][ij];
5981: }else{ /* mobilav */
5982: for(i=1; i<=nlstate;i++)
5983: prlim[i][i]=mobaverage[(int)age][i][ij];
5984: }
5985: }
1.279 brouard 5986: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}_x\f$ at horizon h.
5987: */
5988: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=0 to nhstepm */
5989: /**< And for each alive state j, sums over i \f$ w^i_x {}{h}_p^{ij}_x\f$, which are the probability
5990: * at horizon h in state j including mortality.
5991: */
1.218 brouard 5992: for(j=1; j<= nlstate; j++){
5993: for(h=0; h<=nhstepm; h++){
5994: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5995: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5996: }
5997: }
1.279 brouard 5998: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 5999: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6000: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6001: */
6002: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6003: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6004: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6005: }
6006:
6007: /* Again with minus shift */
1.218 brouard 6008:
6009: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6010: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6011:
1.242 brouard 6012: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6013:
6014: if (popbased==1) {
6015: if(mobilav ==0){
6016: for(i=1; i<=nlstate;i++)
6017: prlim[i][i]=probs[(int)age][i][ij];
6018: }else{ /* mobilav */
6019: for(i=1; i<=nlstate;i++)
6020: prlim[i][i]=mobaverage[(int)age][i][ij];
6021: }
6022: }
6023:
1.235 brouard 6024: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6025:
6026: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6027: for(h=0; h<=nhstepm; h++){
6028: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6029: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6030: }
6031: }
6032: /* This for computing probability of death (h=1 means
6033: computed over hstepm matrices product = hstepm*stepm months)
6034: as a weighted average of prlim.
6035: */
6036: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6037: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6038: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6039: }
1.279 brouard 6040: /* end shifting computations */
6041:
6042: /**< Computing gradient matrix at horizon h
6043: */
1.218 brouard 6044: for(j=1; j<= nlstate; j++) /* vareij */
6045: for(h=0; h<=nhstepm; h++){
6046: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6047: }
1.279 brouard 6048: /**< Gradient of overall mortality p.3 (or p.j)
6049: */
6050: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6051: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6052: }
6053:
6054: } /* End theta */
1.279 brouard 6055:
6056: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6057: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6058:
6059: for(h=0; h<=nhstepm; h++) /* veij */
6060: for(j=1; j<=nlstate;j++)
6061: for(theta=1; theta <=npar; theta++)
6062: trgradg[h][j][theta]=gradg[h][theta][j];
6063:
6064: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6065: for(theta=1; theta <=npar; theta++)
6066: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6067: /**< as well as its transposed matrix
6068: */
1.218 brouard 6069:
6070: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6071: for(i=1;i<=nlstate;i++)
6072: for(j=1;j<=nlstate;j++)
6073: vareij[i][j][(int)age] =0.;
1.279 brouard 6074:
6075: /* Computing trgradg by matcov by gradg at age and summing over h
6076: * and k (nhstepm) formula 15 of article
6077: * Lievre-Brouard-Heathcote
6078: */
6079:
1.218 brouard 6080: for(h=0;h<=nhstepm;h++){
6081: for(k=0;k<=nhstepm;k++){
6082: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6083: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6084: for(i=1;i<=nlstate;i++)
6085: for(j=1;j<=nlstate;j++)
6086: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6087: }
6088: }
6089:
1.279 brouard 6090: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6091: * p.j overall mortality formula 49 but computed directly because
6092: * we compute the grad (wix pijx) instead of grad (pijx),even if
6093: * wix is independent of theta.
6094: */
1.218 brouard 6095: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6096: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6097: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6098: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6099: varppt[j][i]=doldmp[j][i];
6100: /* end ppptj */
6101: /* x centered again */
6102:
1.242 brouard 6103: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6104:
6105: if (popbased==1) {
6106: if(mobilav ==0){
6107: for(i=1; i<=nlstate;i++)
6108: prlim[i][i]=probs[(int)age][i][ij];
6109: }else{ /* mobilav */
6110: for(i=1; i<=nlstate;i++)
6111: prlim[i][i]=mobaverage[(int)age][i][ij];
6112: }
6113: }
6114:
6115: /* This for computing probability of death (h=1 means
6116: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6117: as a weighted average of prlim.
6118: */
1.235 brouard 6119: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6120: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6121: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6122: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6123: }
6124: /* end probability of death */
6125:
6126: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6127: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6128: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6129: for(i=1; i<=nlstate;i++){
6130: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6131: }
6132: }
6133: fprintf(ficresprobmorprev,"\n");
6134:
6135: fprintf(ficresvij,"%.0f ",age );
6136: for(i=1; i<=nlstate;i++)
6137: for(j=1; j<=nlstate;j++){
6138: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6139: }
6140: fprintf(ficresvij,"\n");
6141: free_matrix(gp,0,nhstepm,1,nlstate);
6142: free_matrix(gm,0,nhstepm,1,nlstate);
6143: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6144: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6145: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6146: } /* End age */
6147: free_vector(gpp,nlstate+1,nlstate+ndeath);
6148: free_vector(gmp,nlstate+1,nlstate+ndeath);
6149: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6150: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6151: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6152: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6153: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6154: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6155: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6156: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6157: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6158: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6159: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6160: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6161: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6162: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6163: 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);
6164: /* 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 6165: */
1.218 brouard 6166: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6167: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6168:
1.218 brouard 6169: free_vector(xp,1,npar);
6170: free_matrix(doldm,1,nlstate,1,nlstate);
6171: free_matrix(dnewm,1,nlstate,1,npar);
6172: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6173: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6174: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6175: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6176: fclose(ficresprobmorprev);
6177: fflush(ficgp);
6178: fflush(fichtm);
6179: } /* end varevsij */
1.126 brouard 6180:
6181: /************ Variance of prevlim ******************/
1.269 brouard 6182: 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 6183: {
1.205 brouard 6184: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6185: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6186:
1.268 brouard 6187: double **dnewmpar,**doldm;
1.126 brouard 6188: int i, j, nhstepm, hstepm;
6189: double *xp;
6190: double *gp, *gm;
6191: double **gradg, **trgradg;
1.208 brouard 6192: double **mgm, **mgp;
1.126 brouard 6193: double age,agelim;
6194: int theta;
6195:
6196: pstamp(ficresvpl);
6197: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6198: fprintf(ficresvpl,"# Age ");
6199: if(nresult >=1)
6200: fprintf(ficresvpl," Result# ");
1.126 brouard 6201: for(i=1; i<=nlstate;i++)
6202: fprintf(ficresvpl," %1d-%1d",i,i);
6203: fprintf(ficresvpl,"\n");
6204:
6205: xp=vector(1,npar);
1.268 brouard 6206: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6207: doldm=matrix(1,nlstate,1,nlstate);
6208:
6209: hstepm=1*YEARM; /* Every year of age */
6210: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6211: agelim = AGESUP;
6212: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6213: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6214: if (stepm >= YEARM) hstepm=1;
6215: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6216: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6217: mgp=matrix(1,npar,1,nlstate);
6218: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6219: gp=vector(1,nlstate);
6220: gm=vector(1,nlstate);
6221:
6222: for(theta=1; theta <=npar; theta++){
6223: for(i=1; i<=npar; i++){ /* Computes gradient */
6224: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6225: }
1.209 brouard 6226: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6227: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6228: else
1.235 brouard 6229: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6230: for(i=1;i<=nlstate;i++){
1.126 brouard 6231: gp[i] = prlim[i][i];
1.208 brouard 6232: mgp[theta][i] = prlim[i][i];
6233: }
1.126 brouard 6234: for(i=1; i<=npar; i++) /* Computes gradient */
6235: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6236: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6237: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6238: else
1.235 brouard 6239: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6240: for(i=1;i<=nlstate;i++){
1.126 brouard 6241: gm[i] = prlim[i][i];
1.208 brouard 6242: mgm[theta][i] = prlim[i][i];
6243: }
1.126 brouard 6244: for(i=1;i<=nlstate;i++)
6245: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6246: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6247: } /* End theta */
6248:
6249: trgradg =matrix(1,nlstate,1,npar);
6250:
6251: for(j=1; j<=nlstate;j++)
6252: for(theta=1; theta <=npar; theta++)
6253: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6254: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6255: /* printf("\nmgm mgp %d ",(int)age); */
6256: /* for(j=1; j<=nlstate;j++){ */
6257: /* printf(" %d ",j); */
6258: /* for(theta=1; theta <=npar; theta++) */
6259: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6260: /* printf("\n "); */
6261: /* } */
6262: /* } */
6263: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6264: /* printf("\n gradg %d ",(int)age); */
6265: /* for(j=1; j<=nlstate;j++){ */
6266: /* printf("%d ",j); */
6267: /* for(theta=1; theta <=npar; theta++) */
6268: /* printf("%d %lf ",theta,gradg[theta][j]); */
6269: /* printf("\n "); */
6270: /* } */
6271: /* } */
1.126 brouard 6272:
6273: for(i=1;i<=nlstate;i++)
6274: varpl[i][(int)age] =0.;
1.209 brouard 6275: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6276: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6277: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6278: }else{
1.268 brouard 6279: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6280: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6281: }
1.126 brouard 6282: for(i=1;i<=nlstate;i++)
6283: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6284:
6285: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6286: if(nresult >=1)
6287: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6288: for(i=1; i<=nlstate;i++)
6289: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6290: fprintf(ficresvpl,"\n");
6291: free_vector(gp,1,nlstate);
6292: free_vector(gm,1,nlstate);
1.208 brouard 6293: free_matrix(mgm,1,npar,1,nlstate);
6294: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6295: free_matrix(gradg,1,npar,1,nlstate);
6296: free_matrix(trgradg,1,nlstate,1,npar);
6297: } /* End age */
6298:
6299: free_vector(xp,1,npar);
6300: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6301: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6302:
6303: }
6304:
6305:
6306: /************ Variance of backprevalence limit ******************/
1.269 brouard 6307: 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 6308: {
6309: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6310: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6311:
6312: double **dnewmpar,**doldm;
6313: int i, j, nhstepm, hstepm;
6314: double *xp;
6315: double *gp, *gm;
6316: double **gradg, **trgradg;
6317: double **mgm, **mgp;
6318: double age,agelim;
6319: int theta;
6320:
6321: pstamp(ficresvbl);
6322: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6323: fprintf(ficresvbl,"# Age ");
6324: if(nresult >=1)
6325: fprintf(ficresvbl," Result# ");
6326: for(i=1; i<=nlstate;i++)
6327: fprintf(ficresvbl," %1d-%1d",i,i);
6328: fprintf(ficresvbl,"\n");
6329:
6330: xp=vector(1,npar);
6331: dnewmpar=matrix(1,nlstate,1,npar);
6332: doldm=matrix(1,nlstate,1,nlstate);
6333:
6334: hstepm=1*YEARM; /* Every year of age */
6335: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6336: agelim = AGEINF;
6337: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6338: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6339: if (stepm >= YEARM) hstepm=1;
6340: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6341: gradg=matrix(1,npar,1,nlstate);
6342: mgp=matrix(1,npar,1,nlstate);
6343: mgm=matrix(1,npar,1,nlstate);
6344: gp=vector(1,nlstate);
6345: gm=vector(1,nlstate);
6346:
6347: for(theta=1; theta <=npar; theta++){
6348: for(i=1; i<=npar; i++){ /* Computes gradient */
6349: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6350: }
6351: if(mobilavproj > 0 )
6352: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6353: else
6354: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6355: for(i=1;i<=nlstate;i++){
6356: gp[i] = bprlim[i][i];
6357: mgp[theta][i] = bprlim[i][i];
6358: }
6359: for(i=1; i<=npar; i++) /* Computes gradient */
6360: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6361: if(mobilavproj > 0 )
6362: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6363: else
6364: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6365: for(i=1;i<=nlstate;i++){
6366: gm[i] = bprlim[i][i];
6367: mgm[theta][i] = bprlim[i][i];
6368: }
6369: for(i=1;i<=nlstate;i++)
6370: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6371: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6372: } /* End theta */
6373:
6374: trgradg =matrix(1,nlstate,1,npar);
6375:
6376: for(j=1; j<=nlstate;j++)
6377: for(theta=1; theta <=npar; theta++)
6378: trgradg[j][theta]=gradg[theta][j];
6379: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6380: /* printf("\nmgm mgp %d ",(int)age); */
6381: /* for(j=1; j<=nlstate;j++){ */
6382: /* printf(" %d ",j); */
6383: /* for(theta=1; theta <=npar; theta++) */
6384: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6385: /* printf("\n "); */
6386: /* } */
6387: /* } */
6388: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6389: /* printf("\n gradg %d ",(int)age); */
6390: /* for(j=1; j<=nlstate;j++){ */
6391: /* printf("%d ",j); */
6392: /* for(theta=1; theta <=npar; theta++) */
6393: /* printf("%d %lf ",theta,gradg[theta][j]); */
6394: /* printf("\n "); */
6395: /* } */
6396: /* } */
6397:
6398: for(i=1;i<=nlstate;i++)
6399: varbpl[i][(int)age] =0.;
6400: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6401: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6402: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6403: }else{
6404: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6405: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6406: }
6407: for(i=1;i<=nlstate;i++)
6408: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6409:
6410: fprintf(ficresvbl,"%.0f ",age );
6411: if(nresult >=1)
6412: fprintf(ficresvbl,"%d ",nres );
6413: for(i=1; i<=nlstate;i++)
6414: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6415: fprintf(ficresvbl,"\n");
6416: free_vector(gp,1,nlstate);
6417: free_vector(gm,1,nlstate);
6418: free_matrix(mgm,1,npar,1,nlstate);
6419: free_matrix(mgp,1,npar,1,nlstate);
6420: free_matrix(gradg,1,npar,1,nlstate);
6421: free_matrix(trgradg,1,nlstate,1,npar);
6422: } /* End age */
6423:
6424: free_vector(xp,1,npar);
6425: free_matrix(doldm,1,nlstate,1,npar);
6426: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6427:
6428: }
6429:
6430: /************ Variance of one-step probabilities ******************/
6431: 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 6432: {
6433: int i, j=0, k1, l1, tj;
6434: int k2, l2, j1, z1;
6435: int k=0, l;
6436: int first=1, first1, first2;
6437: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6438: double **dnewm,**doldm;
6439: double *xp;
6440: double *gp, *gm;
6441: double **gradg, **trgradg;
6442: double **mu;
6443: double age, cov[NCOVMAX+1];
6444: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6445: int theta;
6446: char fileresprob[FILENAMELENGTH];
6447: char fileresprobcov[FILENAMELENGTH];
6448: char fileresprobcor[FILENAMELENGTH];
6449: double ***varpij;
6450:
6451: strcpy(fileresprob,"PROB_");
6452: strcat(fileresprob,fileres);
6453: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6454: printf("Problem with resultfile: %s\n", fileresprob);
6455: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6456: }
6457: strcpy(fileresprobcov,"PROBCOV_");
6458: strcat(fileresprobcov,fileresu);
6459: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6460: printf("Problem with resultfile: %s\n", fileresprobcov);
6461: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6462: }
6463: strcpy(fileresprobcor,"PROBCOR_");
6464: strcat(fileresprobcor,fileresu);
6465: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6466: printf("Problem with resultfile: %s\n", fileresprobcor);
6467: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6468: }
6469: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6470: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6471: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6472: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6473: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6474: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6475: pstamp(ficresprob);
6476: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6477: fprintf(ficresprob,"# Age");
6478: pstamp(ficresprobcov);
6479: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6480: fprintf(ficresprobcov,"# Age");
6481: pstamp(ficresprobcor);
6482: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6483: fprintf(ficresprobcor,"# Age");
1.126 brouard 6484:
6485:
1.222 brouard 6486: for(i=1; i<=nlstate;i++)
6487: for(j=1; j<=(nlstate+ndeath);j++){
6488: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6489: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6490: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6491: }
6492: /* fprintf(ficresprob,"\n");
6493: fprintf(ficresprobcov,"\n");
6494: fprintf(ficresprobcor,"\n");
6495: */
6496: xp=vector(1,npar);
6497: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6498: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6499: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6500: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6501: first=1;
6502: fprintf(ficgp,"\n# Routine varprob");
6503: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6504: fprintf(fichtm,"\n");
6505:
1.266 brouard 6506: 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. %s</li>\n",optionfilehtmcov,optionfilehtmcov);
1.222 brouard 6507: 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);
6508: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6509: and drawn. It helps understanding how is the covariance between two incidences.\
6510: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6511: 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 6512: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6513: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6514: standard deviations wide on each axis. <br>\
6515: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6516: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6517: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6518:
1.222 brouard 6519: cov[1]=1;
6520: /* tj=cptcoveff; */
1.225 brouard 6521: tj = (int) pow(2,cptcoveff);
1.222 brouard 6522: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6523: j1=0;
1.224 brouard 6524: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6525: if (cptcovn>0) {
6526: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6527: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6528: fprintf(ficresprob, "**********\n#\n");
6529: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6530: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6531: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6532:
1.222 brouard 6533: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6534: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6535: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6536:
6537:
1.222 brouard 6538: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6539: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6540: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6541:
1.222 brouard 6542: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6543: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6544: fprintf(ficresprobcor, "**********\n#");
6545: if(invalidvarcomb[j1]){
6546: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6547: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6548: continue;
6549: }
6550: }
6551: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6552: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6553: gp=vector(1,(nlstate)*(nlstate+ndeath));
6554: gm=vector(1,(nlstate)*(nlstate+ndeath));
6555: for (age=bage; age<=fage; age ++){
6556: cov[2]=age;
6557: if(nagesqr==1)
6558: cov[3]= age*age;
6559: for (k=1; k<=cptcovn;k++) {
6560: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6561: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6562: * 1 1 1 1 1
6563: * 2 2 1 1 1
6564: * 3 1 2 1 1
6565: */
6566: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6567: }
6568: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6569: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6570: for (k=1; k<=cptcovprod;k++)
6571: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6572:
6573:
1.222 brouard 6574: for(theta=1; theta <=npar; theta++){
6575: for(i=1; i<=npar; i++)
6576: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6577:
1.222 brouard 6578: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6579:
1.222 brouard 6580: k=0;
6581: for(i=1; i<= (nlstate); i++){
6582: for(j=1; j<=(nlstate+ndeath);j++){
6583: k=k+1;
6584: gp[k]=pmmij[i][j];
6585: }
6586: }
1.220 brouard 6587:
1.222 brouard 6588: for(i=1; i<=npar; i++)
6589: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6590:
1.222 brouard 6591: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6592: k=0;
6593: for(i=1; i<=(nlstate); i++){
6594: for(j=1; j<=(nlstate+ndeath);j++){
6595: k=k+1;
6596: gm[k]=pmmij[i][j];
6597: }
6598: }
1.220 brouard 6599:
1.222 brouard 6600: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6601: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6602: }
1.126 brouard 6603:
1.222 brouard 6604: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6605: for(theta=1; theta <=npar; theta++)
6606: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6607:
1.222 brouard 6608: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6609: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6610:
1.222 brouard 6611: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6612:
1.222 brouard 6613: k=0;
6614: for(i=1; i<=(nlstate); i++){
6615: for(j=1; j<=(nlstate+ndeath);j++){
6616: k=k+1;
6617: mu[k][(int) age]=pmmij[i][j];
6618: }
6619: }
6620: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6621: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6622: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6623:
1.222 brouard 6624: /*printf("\n%d ",(int)age);
6625: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6626: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6627: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6628: }*/
1.220 brouard 6629:
1.222 brouard 6630: fprintf(ficresprob,"\n%d ",(int)age);
6631: fprintf(ficresprobcov,"\n%d ",(int)age);
6632: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6633:
1.222 brouard 6634: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6635: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6636: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6637: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6638: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6639: }
6640: i=0;
6641: for (k=1; k<=(nlstate);k++){
6642: for (l=1; l<=(nlstate+ndeath);l++){
6643: i++;
6644: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6645: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6646: for (j=1; j<=i;j++){
6647: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6648: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6649: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6650: }
6651: }
6652: }/* end of loop for state */
6653: } /* end of loop for age */
6654: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6655: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6656: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6657: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6658:
6659: /* Confidence intervalle of pij */
6660: /*
6661: fprintf(ficgp,"\nunset parametric;unset label");
6662: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6663: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6664: 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);
6665: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6666: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6667: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6668: */
6669:
6670: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6671: first1=1;first2=2;
6672: for (k2=1; k2<=(nlstate);k2++){
6673: for (l2=1; l2<=(nlstate+ndeath);l2++){
6674: if(l2==k2) continue;
6675: j=(k2-1)*(nlstate+ndeath)+l2;
6676: for (k1=1; k1<=(nlstate);k1++){
6677: for (l1=1; l1<=(nlstate+ndeath);l1++){
6678: if(l1==k1) continue;
6679: i=(k1-1)*(nlstate+ndeath)+l1;
6680: if(i<=j) continue;
6681: for (age=bage; age<=fage; age ++){
6682: if ((int)age %5==0){
6683: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6684: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6685: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6686: mu1=mu[i][(int) age]/stepm*YEARM ;
6687: mu2=mu[j][(int) age]/stepm*YEARM;
6688: c12=cv12/sqrt(v1*v2);
6689: /* Computing eigen value of matrix of covariance */
6690: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6691: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6692: if ((lc2 <0) || (lc1 <0) ){
6693: if(first2==1){
6694: first1=0;
6695: 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);
6696: }
6697: 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);
6698: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6699: /* lc2=fabs(lc2); */
6700: }
1.220 brouard 6701:
1.222 brouard 6702: /* Eigen vectors */
1.280 brouard 6703: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6704: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6705: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6706: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6707: }else
6708: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6709: /*v21=sqrt(1.-v11*v11); *//* error */
6710: v21=(lc1-v1)/cv12*v11;
6711: v12=-v21;
6712: v22=v11;
6713: tnalp=v21/v11;
6714: if(first1==1){
6715: first1=0;
6716: 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);
6717: }
6718: 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);
6719: /*printf(fignu*/
6720: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6721: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6722: if(first==1){
6723: first=0;
6724: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6725: fprintf(ficgp,"\nset parametric;unset label");
6726: 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);
6727: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6728: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6729: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6730: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6731: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6732: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6733: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6734: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6735: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6736: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6737: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6738: 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 6739: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6740: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6741: }else{
6742: first=0;
6743: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6744: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6745: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6746: 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 6747: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6748: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6749: }/* if first */
6750: } /* age mod 5 */
6751: } /* end loop age */
6752: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6753: first=1;
6754: } /*l12 */
6755: } /* k12 */
6756: } /*l1 */
6757: }/* k1 */
6758: } /* loop on combination of covariates j1 */
6759: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6760: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6761: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6762: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6763: free_vector(xp,1,npar);
6764: fclose(ficresprob);
6765: fclose(ficresprobcov);
6766: fclose(ficresprobcor);
6767: fflush(ficgp);
6768: fflush(fichtmcov);
6769: }
1.126 brouard 6770:
6771:
6772: /******************* Printing html file ***********/
1.201 brouard 6773: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6774: int lastpass, int stepm, int weightopt, char model[],\
6775: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6776: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6777: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6778: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6779: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6780:
6781: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6782: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6783: </ul>");
1.237 brouard 6784: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6785: </ul>", model);
1.214 brouard 6786: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6787: 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",
6788: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6789: 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 6790: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6791: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6792: fprintf(fichtm,"\
6793: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6794: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6795: fprintf(fichtm,"\
1.217 brouard 6796: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6797: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6798: fprintf(fichtm,"\
1.126 brouard 6799: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6800: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6801: fprintf(fichtm,"\
1.217 brouard 6802: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6803: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6804: fprintf(fichtm,"\
1.211 brouard 6805: - (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 6806: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6807: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6808: if(prevfcast==1){
6809: fprintf(fichtm,"\
6810: - Prevalence projections by age and states: \
1.201 brouard 6811: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6812: }
1.126 brouard 6813:
6814:
1.225 brouard 6815: m=pow(2,cptcoveff);
1.222 brouard 6816: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6817:
1.264 brouard 6818: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6819:
6820: jj1=0;
6821:
6822: fprintf(fichtm," \n<ul>");
6823: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6824: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6825: if(m != 1 && TKresult[nres]!= k1)
6826: continue;
6827: jj1++;
6828: if (cptcovn > 0) {
6829: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6830: for (cpt=1; cpt<=cptcoveff;cpt++){
6831: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6832: }
6833: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6834: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6835: }
6836: fprintf(fichtm,"\">");
6837:
6838: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6839: fprintf(fichtm,"************ Results for covariates");
6840: for (cpt=1; cpt<=cptcoveff;cpt++){
6841: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6842: }
6843: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6844: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6845: }
6846: if(invalidvarcomb[k1]){
6847: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6848: continue;
6849: }
6850: fprintf(fichtm,"</a></li>");
6851: } /* cptcovn >0 */
6852: }
6853: fprintf(fichtm," \n</ul>");
6854:
1.222 brouard 6855: jj1=0;
1.237 brouard 6856:
6857: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6858: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6859: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6860: continue;
1.220 brouard 6861:
1.222 brouard 6862: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6863: jj1++;
6864: if (cptcovn > 0) {
1.264 brouard 6865: fprintf(fichtm,"\n<p><a name=\"rescov");
6866: for (cpt=1; cpt<=cptcoveff;cpt++){
6867: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6868: }
6869: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6870: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6871: }
6872: fprintf(fichtm,"\"</a>");
6873:
1.222 brouard 6874: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6875: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6876: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6877: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6878: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6879: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6880: }
1.237 brouard 6881: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6882: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6883: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6884: }
6885:
1.230 brouard 6886: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6887: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6888: if(invalidvarcomb[k1]){
6889: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6890: printf("\nCombination (%d) ignored because no cases \n",k1);
6891: continue;
6892: }
6893: }
6894: /* aij, bij */
1.259 brouard 6895: 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 6896: <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 6897: /* Pij */
1.241 brouard 6898: 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> \
6899: <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 6900: /* Quasi-incidences */
6901: 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 6902: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6903: 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 6904: 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> \
6905: <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 6906: /* Survival functions (period) in state j */
6907: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6908: fprintf(fichtm,"<br>\n- Survival functions in state %d. Or probability to survive 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> \
6909: <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 6910: }
6911: /* State specific survival functions (period) */
6912: for(cpt=1; cpt<=nlstate;cpt++){
6913: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6914: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.283 brouard 6915: <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 6916: }
6917: /* Period (stable) prevalence in each health state */
6918: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6919: 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> \
6920: <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 6921: }
6922: if(backcast==1){
6923: /* Period (stable) back prevalence in each health state */
6924: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6925: 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 6926: <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 6927: }
1.217 brouard 6928: }
1.222 brouard 6929: if(prevfcast==1){
6930: /* Projection of prevalence up to period (stable) prevalence in each health state */
6931: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6932: 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) 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><br> \
6933: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateproj1, dateproj2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6934: }
6935: }
1.268 brouard 6936: if(backcast==1){
6937: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6938: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6939: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6940: 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 \
6941: 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) \
6942: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6943: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateback1, dateback2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 6944: }
6945: }
1.220 brouard 6946:
1.222 brouard 6947: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6948: 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> <br> \
6949: <img src=\"%s_%d-%d-%d.svg\">",cpt,nlstate,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.222 brouard 6950: }
6951: /* } /\* end i1 *\/ */
6952: }/* End k1 */
6953: fprintf(fichtm,"</ul>");
1.126 brouard 6954:
1.222 brouard 6955: fprintf(fichtm,"\
1.126 brouard 6956: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6957: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6958: - 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 6959: But because parameters are usually highly correlated (a higher incidence of disability \
6960: and a higher incidence of recovery can give very close observed transition) it might \
6961: be very useful to look not only at linear confidence intervals estimated from the \
6962: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6963: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6964: covariance matrix of the one-step probabilities. \
6965: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6966:
1.222 brouard 6967: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6968: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6969: fprintf(fichtm,"\
1.126 brouard 6970: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6971: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6972:
1.222 brouard 6973: fprintf(fichtm,"\
1.126 brouard 6974: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6975: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6976: fprintf(fichtm,"\
1.126 brouard 6977: - 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): \
6978: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6979: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6980: fprintf(fichtm,"\
1.126 brouard 6981: - (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): \
6982: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6983: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6984: fprintf(fichtm,"\
1.128 brouard 6985: - Variances and covariances of health expectancies by age. Status (i) based health expectancies (in state j), e<sup>ij</sup> are weighted by the 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 6986: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6987: fprintf(fichtm,"\
1.128 brouard 6988: - 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 6989: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6990: fprintf(fichtm,"\
1.126 brouard 6991: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6992: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6993:
6994: /* if(popforecast==1) fprintf(fichtm,"\n */
6995: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6996: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6997: /* <br>",fileres,fileres,fileres,fileres); */
6998: /* else */
6999: /* 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 7000: fflush(fichtm);
7001: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7002:
1.225 brouard 7003: m=pow(2,cptcoveff);
1.222 brouard 7004: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7005:
1.222 brouard 7006: jj1=0;
1.237 brouard 7007:
1.241 brouard 7008: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7009: for(k1=1; k1<=m;k1++){
1.253 brouard 7010: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7011: continue;
1.222 brouard 7012: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7013: jj1++;
1.126 brouard 7014: if (cptcovn > 0) {
7015: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7016: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7017: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7018: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7019: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7020: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7021: }
7022:
1.126 brouard 7023: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7024:
1.222 brouard 7025: if(invalidvarcomb[k1]){
7026: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7027: continue;
7028: }
1.126 brouard 7029: }
7030: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7031: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 7032: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
1.258 brouard 7033: <img src=\"%s_%d-%d-%d.svg\">",mobilav,cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
1.126 brouard 7034: }
7035: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 7036: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
7037: true period expectancies (those weighted with period prevalences are also\
7038: drawn in addition to the population based expectancies computed using\
1.241 brouard 7039: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
7040: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7041: /* } /\* end i1 *\/ */
7042: }/* End k1 */
1.241 brouard 7043: }/* End nres */
1.222 brouard 7044: fprintf(fichtm,"</ul>");
7045: fflush(fichtm);
1.126 brouard 7046: }
7047:
7048: /******************* Gnuplot file **************/
1.270 brouard 7049: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double bage, double fage , int prevfcast, int backcast, char pathc[], double p[], int offyear, int offbyear){
1.126 brouard 7050:
7051: char dirfileres[132],optfileres[132];
1.264 brouard 7052: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7053: 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 7054: int lv=0, vlv=0, kl=0;
1.130 brouard 7055: int ng=0;
1.201 brouard 7056: int vpopbased;
1.223 brouard 7057: int ioffset; /* variable offset for columns */
1.270 brouard 7058: int iyearc=1; /* variable column for year of projection */
7059: int iagec=1; /* variable column for age of projection */
1.235 brouard 7060: int nres=0; /* Index of resultline */
1.266 brouard 7061: int istart=1; /* For starting graphs in projections */
1.219 brouard 7062:
1.126 brouard 7063: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7064: /* printf("Problem with file %s",optionfilegnuplot); */
7065: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7066: /* } */
7067:
7068: /*#ifdef windows */
7069: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7070: /*#endif */
1.225 brouard 7071: m=pow(2,cptcoveff);
1.126 brouard 7072:
1.274 brouard 7073: /* diagram of the model */
7074: fprintf(ficgp,"\n#Diagram of the model \n");
7075: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7076: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7077: 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);
7078:
7079: 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);
7080: fprintf(ficgp,"\n#show arrow\nunset label\n");
7081: 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);
7082: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7083: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7084: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7085: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7086:
1.202 brouard 7087: /* Contribution to likelihood */
7088: /* Plot the probability implied in the likelihood */
1.223 brouard 7089: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7090: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7091: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7092: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7093: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7094: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7095: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7096: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7097: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7098: 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));
7099: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7100: 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));
7101: for (i=1; i<= nlstate ; i ++) {
7102: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7103: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7104: 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);
7105: for (j=2; j<= nlstate+ndeath ; j ++) {
7106: 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);
7107: }
7108: fprintf(ficgp,";\nset out; unset ylabel;\n");
7109: }
7110: /* 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 */
7111: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7112: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7113: fprintf(ficgp,"\nset out;unset log\n");
7114: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7115:
1.126 brouard 7116: strcpy(dirfileres,optionfilefiname);
7117: strcpy(optfileres,"vpl");
1.223 brouard 7118: /* 1eme*/
1.238 brouard 7119: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7120: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7121: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7122: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7123: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7124: continue;
7125: /* We are interested in selected combination by the resultline */
1.246 brouard 7126: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7127: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7128: strcpy(gplotlabel,"(");
1.238 brouard 7129: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7130: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7131: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7132: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7133: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7134: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7135: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7136: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7137: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7138: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7139: }
7140: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7141: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7142: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7143: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7144: }
7145: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7146: /* printf("\n#\n"); */
1.238 brouard 7147: fprintf(ficgp,"\n#\n");
7148: if(invalidvarcomb[k1]){
1.260 brouard 7149: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7150: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7151: continue;
7152: }
1.235 brouard 7153:
1.241 brouard 7154: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7155: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7156: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7157: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7158: 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);
7159: /* 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); */
7160: /* k1-1 error should be nres-1*/
1.238 brouard 7161: for (i=1; i<= nlstate ; i ++) {
7162: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7163: else fprintf(ficgp," %%*lf (%%*lf)");
7164: }
1.260 brouard 7165: fprintf(ficgp,"\" t\"Period (stable) 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 7166: for (i=1; i<= nlstate ; i ++) {
7167: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7168: else fprintf(ficgp," %%*lf (%%*lf)");
7169: }
1.260 brouard 7170: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7171: for (i=1; i<= nlstate ; i ++) {
7172: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7173: else fprintf(ficgp," %%*lf (%%*lf)");
7174: }
1.265 brouard 7175: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7176:
7177: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7178: if(cptcoveff ==0){
1.271 brouard 7179: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7180: }else{
7181: kl=0;
7182: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7183: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7184: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7185: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7186: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7187: vlv= nbcode[Tvaraff[k]][lv];
7188: kl++;
7189: /* 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 *\/ */
7190: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7191: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7192: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7193: if(k==cptcoveff){
7194: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7195: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7196: }else{
7197: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7198: kl++;
7199: }
7200: } /* end covariate */
7201: } /* end if no covariate */
7202:
1.238 brouard 7203: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7204: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7205: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7206: if(cptcoveff ==0){
1.245 brouard 7207: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7208: }else{
7209: kl=0;
7210: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7211: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7212: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7213: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7214: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7215: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7216: kl++;
1.238 brouard 7217: /* 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 *\/ */
7218: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7219: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7220: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7221: if(k==cptcoveff){
1.245 brouard 7222: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7223: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7224: }else{
7225: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7226: kl++;
7227: }
7228: } /* end covariate */
7229: } /* end if no covariate */
1.268 brouard 7230: if(backcast == 1){
7231: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7232: /* k1-1 error should be nres-1*/
7233: for (i=1; i<= nlstate ; i ++) {
7234: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7235: else fprintf(ficgp," %%*lf (%%*lf)");
7236: }
1.271 brouard 7237: 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 7238: for (i=1; i<= nlstate ; i ++) {
7239: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7240: else fprintf(ficgp," %%*lf (%%*lf)");
7241: }
1.276 brouard 7242: 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 7243: for (i=1; i<= nlstate ; i ++) {
7244: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7245: else fprintf(ficgp," %%*lf (%%*lf)");
7246: }
1.274 brouard 7247: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7248: } /* end if backprojcast */
1.238 brouard 7249: } /* end if backcast */
1.276 brouard 7250: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7251: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7252: } /* nres */
1.201 brouard 7253: } /* k1 */
7254: } /* cpt */
1.235 brouard 7255:
7256:
1.126 brouard 7257: /*2 eme*/
1.238 brouard 7258: for (k1=1; k1<= m ; k1 ++){
7259: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7260: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7261: continue;
7262: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7263: strcpy(gplotlabel,"(");
1.238 brouard 7264: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7265: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7266: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7267: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7268: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7269: vlv= nbcode[Tvaraff[k]][lv];
7270: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7271: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7272: }
1.237 brouard 7273: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7274: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7275: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7276: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7277: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7278: }
1.264 brouard 7279: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7280: fprintf(ficgp,"\n#\n");
1.223 brouard 7281: if(invalidvarcomb[k1]){
7282: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7283: continue;
7284: }
1.219 brouard 7285:
1.241 brouard 7286: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7287: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7288: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7289: if(vpopbased==0){
1.238 brouard 7290: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7291: }else
1.238 brouard 7292: fprintf(ficgp,"\nreplot ");
7293: for (i=1; i<= nlstate+1 ; i ++) {
7294: k=2*i;
1.261 brouard 7295: 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 7296: for (j=1; j<= nlstate+1 ; j ++) {
7297: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7298: else fprintf(ficgp," %%*lf (%%*lf)");
7299: }
7300: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7301: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7302: 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 7303: for (j=1; j<= nlstate+1 ; j ++) {
7304: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7305: else fprintf(ficgp," %%*lf (%%*lf)");
7306: }
7307: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7308: 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 7309: for (j=1; j<= nlstate+1 ; j ++) {
7310: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7311: else fprintf(ficgp," %%*lf (%%*lf)");
7312: }
7313: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7314: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7315: } /* state */
7316: } /* vpopbased */
1.264 brouard 7317: 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 7318: } /* end nres */
7319: } /* k1 end 2 eme*/
7320:
7321:
7322: /*3eme*/
7323: for (k1=1; k1<= m ; k1 ++){
7324: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7325: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7326: continue;
7327:
7328: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7329: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7330: strcpy(gplotlabel,"(");
1.238 brouard 7331: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7332: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7333: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7334: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7335: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7336: vlv= nbcode[Tvaraff[k]][lv];
7337: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7338: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7339: }
7340: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7341: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7342: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7343: }
1.264 brouard 7344: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7345: fprintf(ficgp,"\n#\n");
7346: if(invalidvarcomb[k1]){
7347: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7348: continue;
7349: }
7350:
7351: /* k=2+nlstate*(2*cpt-2); */
7352: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7353: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7354: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7355: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7356: 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 7357: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7358: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7359: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7360: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7361: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7362: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7363:
1.238 brouard 7364: */
7365: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7366: 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 7367: /* 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 7368:
1.238 brouard 7369: }
1.261 brouard 7370: 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 7371: }
1.264 brouard 7372: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7373: } /* end nres */
7374: } /* end kl 3eme */
1.126 brouard 7375:
1.223 brouard 7376: /* 4eme */
1.201 brouard 7377: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7378: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7379: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7380: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7381: continue;
1.238 brouard 7382: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7383: strcpy(gplotlabel,"(");
1.238 brouard 7384: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7385: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7386: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7387: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7388: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7389: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7390: vlv= nbcode[Tvaraff[k]][lv];
7391: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7392: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7393: }
7394: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7395: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7396: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7397: }
1.264 brouard 7398: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7399: fprintf(ficgp,"\n#\n");
7400: if(invalidvarcomb[k1]){
7401: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7402: continue;
1.223 brouard 7403: }
1.238 brouard 7404:
1.241 brouard 7405: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7406: 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 7407: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7408: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7409: k=3;
7410: for (i=1; i<= nlstate ; i ++){
7411: if(i==1){
7412: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7413: }else{
7414: fprintf(ficgp,", '' ");
7415: }
7416: l=(nlstate+ndeath)*(i-1)+1;
7417: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7418: for (j=2; j<= nlstate+ndeath ; j ++)
7419: fprintf(ficgp,"+$%d",k+l+j-1);
7420: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7421: } /* nlstate */
1.264 brouard 7422: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7423: } /* end cpt state*/
7424: } /* end nres */
7425: } /* end covariate k1 */
7426:
1.220 brouard 7427: /* 5eme */
1.201 brouard 7428: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7429: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7430: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7431: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7432: continue;
1.238 brouard 7433: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7434: strcpy(gplotlabel,"(");
1.238 brouard 7435: 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);
7436: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7437: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7438: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7439: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7440: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7441: vlv= nbcode[Tvaraff[k]][lv];
7442: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7443: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7444: }
7445: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7446: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7447: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7448: }
1.264 brouard 7449: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7450: fprintf(ficgp,"\n#\n");
7451: if(invalidvarcomb[k1]){
7452: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7453: continue;
7454: }
1.227 brouard 7455:
1.241 brouard 7456: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7457: 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 7458: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7459: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7460: k=3;
7461: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7462: if(j==1)
7463: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7464: else
7465: fprintf(ficgp,", '' ");
7466: l=(nlstate+ndeath)*(cpt-1) +j;
7467: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7468: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7469: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7470: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7471: } /* nlstate */
7472: fprintf(ficgp,", '' ");
7473: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7474: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7475: l=(nlstate+ndeath)*(cpt-1) +j;
7476: if(j < nlstate)
7477: fprintf(ficgp,"$%d +",k+l);
7478: else
7479: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7480: }
1.264 brouard 7481: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7482: } /* end cpt state*/
7483: } /* end covariate */
7484: } /* end nres */
1.227 brouard 7485:
1.220 brouard 7486: /* 6eme */
1.202 brouard 7487: /* CV preval stable (period) for each covariate */
1.237 brouard 7488: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7489: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7490: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7491: continue;
1.255 brouard 7492: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7493: strcpy(gplotlabel,"(");
1.211 brouard 7494: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7495: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7496: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7497: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7498: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7499: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7500: vlv= nbcode[Tvaraff[k]][lv];
7501: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7502: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7503: }
1.237 brouard 7504: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7505: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7506: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7507: }
1.264 brouard 7508: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7509: fprintf(ficgp,"\n#\n");
1.223 brouard 7510: if(invalidvarcomb[k1]){
1.227 brouard 7511: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7512: continue;
1.223 brouard 7513: }
1.227 brouard 7514:
1.241 brouard 7515: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7516: 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 7517: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7518: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7519: k=3; /* Offset */
1.255 brouard 7520: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7521: if(i==1)
7522: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7523: else
7524: fprintf(ficgp,", '' ");
1.255 brouard 7525: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7526: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7527: for (j=2; j<= nlstate ; j ++)
7528: fprintf(ficgp,"+$%d",k+l+j-1);
7529: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7530: } /* nlstate */
1.264 brouard 7531: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7532: } /* end cpt state*/
7533: } /* end covariate */
1.227 brouard 7534:
7535:
1.220 brouard 7536: /* 7eme */
1.218 brouard 7537: if(backcast == 1){
1.217 brouard 7538: /* CV back preval stable (period) for each covariate */
1.237 brouard 7539: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7540: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7541: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7542: continue;
1.268 brouard 7543: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7544: strcpy(gplotlabel,"(");
7545: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7546: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7547: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7548: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7549: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7550: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7551: vlv= nbcode[Tvaraff[k]][lv];
7552: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7553: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7554: }
1.237 brouard 7555: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7556: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7557: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7558: }
1.264 brouard 7559: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7560: fprintf(ficgp,"\n#\n");
7561: if(invalidvarcomb[k1]){
7562: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7563: continue;
7564: }
7565:
1.241 brouard 7566: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7567: 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 7568: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7569: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7570: k=3; /* Offset */
1.268 brouard 7571: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7572: if(i==1)
7573: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7574: else
7575: fprintf(ficgp,", '' ");
7576: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7577: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7578: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7579: /* 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 7580: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7581: /* for (j=2; j<= nlstate ; j ++) */
7582: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7583: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7584: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7585: } /* nlstate */
1.264 brouard 7586: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7587: } /* end cpt state*/
7588: } /* end covariate */
7589: } /* End if backcast */
7590:
1.223 brouard 7591: /* 8eme */
1.218 brouard 7592: if(prevfcast==1){
7593: /* Projection from cross-sectional to stable (period) for each covariate */
7594:
1.237 brouard 7595: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7596: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7597: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7598: continue;
1.211 brouard 7599: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7600: strcpy(gplotlabel,"(");
1.227 brouard 7601: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7602: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7603: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7604: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7605: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7606: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7607: vlv= nbcode[Tvaraff[k]][lv];
7608: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7609: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7610: }
1.237 brouard 7611: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7612: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7613: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7614: }
1.264 brouard 7615: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7616: fprintf(ficgp,"\n#\n");
7617: if(invalidvarcomb[k1]){
7618: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7619: continue;
7620: }
7621:
7622: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7623: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7624: 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 7625: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7626: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7627:
7628: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7629: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7630: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7631: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7632: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7633: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7634: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7635: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7636: if(i==istart){
1.227 brouard 7637: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7638: }else{
7639: fprintf(ficgp,",\\\n '' ");
7640: }
7641: if(cptcoveff ==0){ /* No covariate */
7642: ioffset=2; /* Age is in 2 */
7643: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7644: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7645: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7646: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7647: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7648: if(i==nlstate+1){
1.270 brouard 7649: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7650: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7651: fprintf(ficgp,",\\\n '' ");
7652: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7653: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7654: offyear, \
1.268 brouard 7655: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7656: }else
1.227 brouard 7657: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7658: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7659: }else{ /* more than 2 covariates */
1.270 brouard 7660: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7661: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7662: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7663: iyearc=ioffset-1;
7664: iagec=ioffset;
1.227 brouard 7665: fprintf(ficgp," u %d:(",ioffset);
7666: kl=0;
7667: strcpy(gplotcondition,"(");
7668: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7669: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7670: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7671: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7672: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7673: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7674: kl++;
7675: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7676: kl++;
7677: if(k <cptcoveff && cptcoveff>1)
7678: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7679: }
7680: strcpy(gplotcondition+strlen(gplotcondition),")");
7681: /* 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 *\/ */
7682: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7683: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7684: /* '' 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*/
7685: if(i==nlstate+1){
1.270 brouard 7686: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7687: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7688: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7689: fprintf(ficgp," u %d:(",iagec);
7690: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7691: iyearc, iagec, offyear, \
7692: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7693: /* '' 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 7694: }else{
7695: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7696: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7697: }
7698: } /* end if covariate */
7699: } /* nlstate */
1.264 brouard 7700: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7701: } /* end cpt state*/
7702: } /* end covariate */
7703: } /* End if prevfcast */
1.227 brouard 7704:
1.268 brouard 7705: if(backcast==1){
7706: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7707:
7708: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7709: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7710: if(m != 1 && TKresult[nres]!= k1)
7711: continue;
7712: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7713: strcpy(gplotlabel,"(");
7714: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7715: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7716: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7717: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7718: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7719: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7720: vlv= nbcode[Tvaraff[k]][lv];
7721: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7722: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7723: }
7724: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7725: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7726: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7727: }
7728: strcpy(gplotlabel+strlen(gplotlabel),")");
7729: fprintf(ficgp,"\n#\n");
7730: if(invalidvarcomb[k1]){
7731: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7732: continue;
7733: }
7734:
7735: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7736: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7737: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7738: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7739: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7740:
7741: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7742: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7743: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7744: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7745: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7746: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7747: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7748: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7749: if(i==istart){
7750: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7751: }else{
7752: fprintf(ficgp,",\\\n '' ");
7753: }
7754: if(cptcoveff ==0){ /* No covariate */
7755: ioffset=2; /* Age is in 2 */
7756: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7757: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7758: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7759: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7760: fprintf(ficgp," u %d:(", ioffset);
7761: if(i==nlstate+1){
1.270 brouard 7762: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7763: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7764: fprintf(ficgp,",\\\n '' ");
7765: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7766: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7767: offbyear, \
7768: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7769: }else
7770: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7771: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7772: }else{ /* more than 2 covariates */
1.270 brouard 7773: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7774: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7775: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7776: iyearc=ioffset-1;
7777: iagec=ioffset;
1.268 brouard 7778: fprintf(ficgp," u %d:(",ioffset);
7779: kl=0;
7780: strcpy(gplotcondition,"(");
7781: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7782: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7783: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7784: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7785: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7786: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7787: kl++;
7788: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7789: kl++;
7790: if(k <cptcoveff && cptcoveff>1)
7791: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7792: }
7793: strcpy(gplotcondition+strlen(gplotcondition),")");
7794: /* 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 *\/ */
7795: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7796: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7797: /* '' 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*/
7798: if(i==nlstate+1){
1.270 brouard 7799: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7800: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7801: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7802: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7803: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7804: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7805: iyearc,iagec,offbyear, \
7806: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7807: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7808: }else{
7809: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7810: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7811: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7812: }
7813: } /* end if covariate */
7814: } /* nlstate */
7815: fprintf(ficgp,"\nset out; unset label;\n");
7816: } /* end cpt state*/
7817: } /* end covariate */
7818: } /* End if backcast */
7819:
1.227 brouard 7820:
1.238 brouard 7821: /* 9eme writing MLE parameters */
7822: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7823: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7824: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7825: for(k=1; k <=(nlstate+ndeath); k++){
7826: if (k != i) {
1.227 brouard 7827: fprintf(ficgp,"# current state %d\n",k);
7828: for(j=1; j <=ncovmodel; j++){
7829: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7830: jk++;
7831: }
7832: fprintf(ficgp,"\n");
1.126 brouard 7833: }
7834: }
1.223 brouard 7835: }
1.187 brouard 7836: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7837:
1.145 brouard 7838: /*goto avoid;*/
1.238 brouard 7839: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7840: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7841: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7842: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7843: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7844: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7845: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7846: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7847: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7848: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7849: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7850: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7851: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7852: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7853: fprintf(ficgp,"#\n");
1.223 brouard 7854: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7855: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7856: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7857: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7858: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7859: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7860: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7861: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7862: continue;
1.264 brouard 7863: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7864: strcpy(gplotlabel,"(");
1.276 brouard 7865: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7866: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7867: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7868: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7869: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7870: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7871: vlv= nbcode[Tvaraff[k]][lv];
7872: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7873: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7874: }
1.237 brouard 7875: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7876: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7877: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7878: }
1.264 brouard 7879: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7880: fprintf(ficgp,"\n#\n");
1.264 brouard 7881: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7882: fprintf(ficgp,"\nset key outside ");
7883: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7884: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7885: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7886: if (ng==1){
7887: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7888: fprintf(ficgp,"\nunset log y");
7889: }else if (ng==2){
7890: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7891: fprintf(ficgp,"\nset log y");
7892: }else if (ng==3){
7893: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7894: fprintf(ficgp,"\nset log y");
7895: }else
7896: fprintf(ficgp,"\nunset title ");
7897: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7898: i=1;
7899: for(k2=1; k2<=nlstate; k2++) {
7900: k3=i;
7901: for(k=1; k<=(nlstate+ndeath); k++) {
7902: if (k != k2){
7903: switch( ng) {
7904: case 1:
7905: if(nagesqr==0)
7906: fprintf(ficgp," p%d+p%d*x",i,i+1);
7907: else /* nagesqr =1 */
7908: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7909: break;
7910: case 2: /* ng=2 */
7911: if(nagesqr==0)
7912: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7913: else /* nagesqr =1 */
7914: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7915: break;
7916: case 3:
7917: if(nagesqr==0)
7918: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7919: else /* nagesqr =1 */
7920: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7921: break;
7922: }
7923: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7924: ijp=1; /* product no age */
7925: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7926: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7927: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7928: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7929: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7930: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7931: if(DummyV[j]==0){
7932: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7933: }else{ /* quantitative */
7934: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7935: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7936: }
7937: ij++;
1.237 brouard 7938: }
1.268 brouard 7939: }
7940: }else if(cptcovprod >0){
7941: if(j==Tprod[ijp]) { /* */
7942: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7943: if(ijp <=cptcovprod) { /* Product */
7944: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7945: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7946: /* 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)]); */
7947: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7948: }else{ /* Vn is dummy and Vm is quanti */
7949: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7950: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7951: }
7952: }else{ /* Vn*Vm Vn is quanti */
7953: if(DummyV[Tvard[ijp][2]]==0){
7954: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7955: }else{ /* Both quanti */
7956: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7957: }
1.237 brouard 7958: }
1.268 brouard 7959: ijp++;
1.237 brouard 7960: }
1.268 brouard 7961: } /* end Tprod */
1.237 brouard 7962: } else{ /* simple covariate */
1.264 brouard 7963: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7964: if(Dummy[j]==0){
7965: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7966: }else{ /* quantitative */
7967: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7968: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7969: }
1.237 brouard 7970: } /* end simple */
7971: } /* end j */
1.223 brouard 7972: }else{
7973: i=i-ncovmodel;
7974: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7975: fprintf(ficgp," (1.");
7976: }
1.227 brouard 7977:
1.223 brouard 7978: if(ng != 1){
7979: fprintf(ficgp,")/(1");
1.227 brouard 7980:
1.264 brouard 7981: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7982: if(nagesqr==0)
1.264 brouard 7983: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7984: else /* nagesqr =1 */
1.264 brouard 7985: 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 7986:
1.223 brouard 7987: ij=1;
7988: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7989: if(cptcovage >0){
7990: if((j-2)==Tage[ij]) { /* Bug valgrind */
7991: if(ij <=cptcovage) { /* Bug valgrind */
7992: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7993: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7994: ij++;
7995: }
7996: }
7997: }else
7998: 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 7999: }
8000: fprintf(ficgp,")");
8001: }
8002: fprintf(ficgp,")");
8003: if(ng ==2)
1.276 brouard 8004: 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 8005: else /* ng= 3 */
1.276 brouard 8006: 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 8007: }else{ /* end ng <> 1 */
8008: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8009: 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 8010: }
8011: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8012: fprintf(ficgp,",");
8013: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8014: fprintf(ficgp,",");
8015: i=i+ncovmodel;
8016: } /* end k */
8017: } /* end k2 */
1.276 brouard 8018: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8019: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8020: } /* end k1 */
1.223 brouard 8021: } /* end ng */
8022: /* avoid: */
8023: fflush(ficgp);
1.126 brouard 8024: } /* end gnuplot */
8025:
8026:
8027: /*************** Moving average **************/
1.219 brouard 8028: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8029: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8030:
1.222 brouard 8031: int i, cpt, cptcod;
8032: int modcovmax =1;
8033: int mobilavrange, mob;
8034: int iage=0;
8035:
1.266 brouard 8036: double sum=0., sumr=0.;
1.222 brouard 8037: double age;
1.266 brouard 8038: double *sumnewp, *sumnewm, *sumnewmr;
8039: double *agemingood, *agemaxgood;
8040: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8041:
8042:
1.278 brouard 8043: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8044: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8045:
8046: sumnewp = vector(1,ncovcombmax);
8047: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8048: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8049: agemingood = vector(1,ncovcombmax);
1.266 brouard 8050: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8051: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8052: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8053:
8054: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8055: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8056: sumnewp[cptcod]=0.;
1.266 brouard 8057: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8058: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8059: }
8060: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8061:
1.266 brouard 8062: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8063: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8064: else mobilavrange=mobilav;
8065: for (age=bage; age<=fage; age++)
8066: for (i=1; i<=nlstate;i++)
8067: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8068: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8069: /* We keep the original values on the extreme ages bage, fage and for
8070: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8071: we use a 5 terms etc. until the borders are no more concerned.
8072: */
8073: for (mob=3;mob <=mobilavrange;mob=mob+2){
8074: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8075: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8076: sumnewm[cptcod]=0.;
8077: for (i=1; i<=nlstate;i++){
1.222 brouard 8078: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8079: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8080: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8081: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8082: }
8083: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8084: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8085: } /* end i */
8086: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8087: } /* end cptcod */
1.222 brouard 8088: }/* end age */
8089: }/* end mob */
1.266 brouard 8090: }else{
8091: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8092: return -1;
1.266 brouard 8093: }
8094:
8095: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8096: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8097: if(invalidvarcomb[cptcod]){
8098: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8099: continue;
8100: }
1.219 brouard 8101:
1.266 brouard 8102: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8103: sumnewm[cptcod]=0.;
8104: sumnewmr[cptcod]=0.;
8105: for (i=1; i<=nlstate;i++){
8106: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8107: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8108: }
8109: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8110: agemingoodr[cptcod]=age;
8111: }
8112: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8113: agemingood[cptcod]=age;
8114: }
8115: } /* age */
8116: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8117: sumnewm[cptcod]=0.;
1.266 brouard 8118: sumnewmr[cptcod]=0.;
1.222 brouard 8119: for (i=1; i<=nlstate;i++){
8120: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8121: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8122: }
8123: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8124: agemaxgoodr[cptcod]=age;
1.222 brouard 8125: }
8126: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8127: agemaxgood[cptcod]=age;
8128: }
8129: } /* age */
8130: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8131: /* but they will change */
8132: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8133: sumnewm[cptcod]=0.;
8134: sumnewmr[cptcod]=0.;
8135: for (i=1; i<=nlstate;i++){
8136: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8137: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8138: }
8139: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8140: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8141: agemaxgoodr[cptcod]=age; /* age min */
8142: for (i=1; i<=nlstate;i++)
8143: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8144: }else{ /* bad we change the value with the values of good ages */
8145: for (i=1; i<=nlstate;i++){
8146: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8147: } /* i */
8148: } /* end bad */
8149: }else{
8150: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8151: agemaxgood[cptcod]=age;
8152: }else{ /* bad we change the value with the values of good ages */
8153: for (i=1; i<=nlstate;i++){
8154: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8155: } /* i */
8156: } /* end bad */
8157: }/* end else */
8158: sum=0.;sumr=0.;
8159: for (i=1; i<=nlstate;i++){
8160: sum+=mobaverage[(int)age][i][cptcod];
8161: sumr+=probs[(int)age][i][cptcod];
8162: }
8163: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8164: 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\n",cptcod,sumr, (int)age, (int)bage);
1.266 brouard 8165: } /* end bad */
8166: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8167: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8168: 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\n",cptcod,sumr, (int)age, (int)bage);
1.222 brouard 8169: } /* end bad */
8170: }/* age */
1.266 brouard 8171:
8172: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8173: sumnewm[cptcod]=0.;
1.266 brouard 8174: sumnewmr[cptcod]=0.;
1.222 brouard 8175: for (i=1; i<=nlstate;i++){
8176: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8177: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8178: }
8179: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8180: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8181: agemingoodr[cptcod]=age;
8182: for (i=1; i<=nlstate;i++)
8183: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8184: }else{ /* bad we change the value with the values of good ages */
8185: for (i=1; i<=nlstate;i++){
8186: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8187: } /* i */
8188: } /* end bad */
8189: }else{
8190: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8191: agemingood[cptcod]=age;
8192: }else{ /* bad */
8193: for (i=1; i<=nlstate;i++){
8194: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8195: } /* i */
8196: } /* end bad */
8197: }/* end else */
8198: sum=0.;sumr=0.;
8199: for (i=1; i<=nlstate;i++){
8200: sum+=mobaverage[(int)age][i][cptcod];
8201: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8202: }
1.266 brouard 8203: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8204: 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 8205: } /* end bad */
8206: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8207: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8208: 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 8209: } /* end bad */
8210: }/* age */
1.266 brouard 8211:
1.222 brouard 8212:
8213: for (age=bage; age<=fage; age++){
1.235 brouard 8214: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8215: sumnewp[cptcod]=0.;
8216: sumnewm[cptcod]=0.;
8217: for (i=1; i<=nlstate;i++){
8218: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8219: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8220: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8221: }
8222: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8223: }
8224: /* printf("\n"); */
8225: /* } */
1.266 brouard 8226:
1.222 brouard 8227: /* brutal averaging */
1.266 brouard 8228: /* for (i=1; i<=nlstate;i++){ */
8229: /* for (age=1; age<=bage; age++){ */
8230: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8231: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8232: /* } */
8233: /* for (age=fage; age<=AGESUP; age++){ */
8234: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8235: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8236: /* } */
8237: /* } /\* end i status *\/ */
8238: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8239: /* for (age=1; age<=AGESUP; age++){ */
8240: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8241: /* mobaverage[(int)age][i][cptcod]=0.; */
8242: /* } */
8243: /* } */
1.222 brouard 8244: }/* end cptcod */
1.266 brouard 8245: free_vector(agemaxgoodr,1, ncovcombmax);
8246: free_vector(agemaxgood,1, ncovcombmax);
8247: free_vector(agemingood,1, ncovcombmax);
8248: free_vector(agemingoodr,1, ncovcombmax);
8249: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8250: free_vector(sumnewm,1, ncovcombmax);
8251: free_vector(sumnewp,1, ncovcombmax);
8252: return 0;
8253: }/* End movingaverage */
1.218 brouard 8254:
1.126 brouard 8255:
8256: /************** Forecasting ******************/
1.269 brouard 8257: void prevforecast(char fileres[], double anproj1, double mproj1, double jproj1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double anproj2, double p[], int cptcoveff){
1.126 brouard 8258: /* proj1, year, month, day of starting projection
8259: agemin, agemax range of age
8260: dateprev1 dateprev2 range of dates during which prevalence is computed
8261: anproj2 year of en of projection (same day and month as proj1).
8262: */
1.267 brouard 8263: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8264: double agec; /* generic age */
8265: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8266: double *popeffectif,*popcount;
8267: double ***p3mat;
1.218 brouard 8268: /* double ***mobaverage; */
1.126 brouard 8269: char fileresf[FILENAMELENGTH];
8270:
8271: agelim=AGESUP;
1.211 brouard 8272: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8273: in each health status at the date of interview (if between dateprev1 and dateprev2).
8274: We still use firstpass and lastpass as another selection.
8275: */
1.214 brouard 8276: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8277: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8278:
1.201 brouard 8279: strcpy(fileresf,"F_");
8280: strcat(fileresf,fileresu);
1.126 brouard 8281: if((ficresf=fopen(fileresf,"w"))==NULL) {
8282: printf("Problem with forecast resultfile: %s\n", fileresf);
8283: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8284: }
1.235 brouard 8285: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8286: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8287:
1.225 brouard 8288: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8289:
8290:
8291: stepsize=(int) (stepm+YEARM-1)/YEARM;
8292: if (stepm<=12) stepsize=1;
8293: if(estepm < stepm){
8294: printf ("Problem %d lower than %d\n",estepm, stepm);
8295: }
1.270 brouard 8296: else{
8297: hstepm=estepm;
8298: }
8299: if(estepm > stepm){ /* Yes every two year */
8300: stepsize=2;
8301: }
1.126 brouard 8302:
8303: hstepm=hstepm/stepm;
8304: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8305: fractional in yp1 */
8306: anprojmean=yp;
8307: yp2=modf((yp1*12),&yp);
8308: mprojmean=yp;
8309: yp1=modf((yp2*30.5),&yp);
8310: jprojmean=yp;
8311: if(jprojmean==0) jprojmean=1;
8312: if(mprojmean==0) jprojmean=1;
8313:
1.227 brouard 8314: i1=pow(2,cptcoveff);
1.126 brouard 8315: if (cptcovn < 1){i1=1;}
8316:
8317: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8318:
8319: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8320:
1.126 brouard 8321: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8322: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8323: for(k=1; k<=i1;k++){
1.253 brouard 8324: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8325: continue;
1.227 brouard 8326: if(invalidvarcomb[k]){
8327: printf("\nCombination (%d) projection ignored because no cases \n",k);
8328: continue;
8329: }
8330: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8331: for(j=1;j<=cptcoveff;j++) {
8332: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8333: }
1.235 brouard 8334: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8335: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8336: }
1.227 brouard 8337: fprintf(ficresf," yearproj age");
8338: for(j=1; j<=nlstate+ndeath;j++){
8339: for(i=1; i<=nlstate;i++)
8340: fprintf(ficresf," p%d%d",i,j);
8341: fprintf(ficresf," wp.%d",j);
8342: }
8343: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8344: fprintf(ficresf,"\n");
8345: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8346: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8347: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8348: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8349: nhstepm = nhstepm/hstepm;
8350: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8351: oldm=oldms;savm=savms;
1.268 brouard 8352: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8353: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8354: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8355: for (h=0; h<=nhstepm; h++){
8356: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8357: break;
8358: }
8359: }
8360: fprintf(ficresf,"\n");
8361: for(j=1;j<=cptcoveff;j++)
8362: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8363: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8364:
8365: for(j=1; j<=nlstate+ndeath;j++) {
8366: ppij=0.;
8367: for(i=1; i<=nlstate;i++) {
1.278 brouard 8368: if (mobilav>=1)
8369: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8370: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8371: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8372: }
1.268 brouard 8373: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8374: } /* end i */
8375: fprintf(ficresf," %.3f", ppij);
8376: }/* end j */
1.227 brouard 8377: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8378: } /* end agec */
1.266 brouard 8379: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8380: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8381: } /* end yearp */
8382: } /* end k */
1.219 brouard 8383:
1.126 brouard 8384: fclose(ficresf);
1.215 brouard 8385: printf("End of Computing forecasting \n");
8386: fprintf(ficlog,"End of Computing forecasting\n");
8387:
1.126 brouard 8388: }
8389:
1.269 brouard 8390: /************** Back Forecasting ******************/
8391: 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){
1.267 brouard 8392: /* back1, year, month, day of starting backection
8393: agemin, agemax range of age
8394: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8395: anback2 year of end of backprojection (same day and month as back1).
8396: prevacurrent and prev are prevalences.
1.267 brouard 8397: */
8398: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8399: double agec; /* generic age */
1.268 brouard 8400: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8401: double *popeffectif,*popcount;
8402: double ***p3mat;
8403: /* double ***mobaverage; */
8404: char fileresfb[FILENAMELENGTH];
8405:
1.268 brouard 8406: agelim=AGEINF;
1.267 brouard 8407: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8408: in each health status at the date of interview (if between dateprev1 and dateprev2).
8409: We still use firstpass and lastpass as another selection.
8410: */
8411: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8412: /* firstpass, lastpass, stepm, weightopt, model); */
8413:
8414: /*Do we need to compute prevalence again?*/
8415:
8416: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8417:
8418: strcpy(fileresfb,"FB_");
8419: strcat(fileresfb,fileresu);
8420: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8421: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8422: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8423: }
8424: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8425: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8426:
8427: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8428:
8429:
8430: stepsize=(int) (stepm+YEARM-1)/YEARM;
8431: if (stepm<=12) stepsize=1;
8432: if(estepm < stepm){
8433: printf ("Problem %d lower than %d\n",estepm, stepm);
8434: }
1.270 brouard 8435: else{
8436: hstepm=estepm;
8437: }
8438: if(estepm >= stepm){ /* Yes every two year */
8439: stepsize=2;
8440: }
1.267 brouard 8441:
8442: hstepm=hstepm/stepm;
8443: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8444: fractional in yp1 */
8445: anprojmean=yp;
8446: yp2=modf((yp1*12),&yp);
8447: mprojmean=yp;
8448: yp1=modf((yp2*30.5),&yp);
8449: jprojmean=yp;
8450: if(jprojmean==0) jprojmean=1;
8451: if(mprojmean==0) jprojmean=1;
8452:
8453: i1=pow(2,cptcoveff);
8454: if (cptcovn < 1){i1=1;}
8455:
8456: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8457: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8458:
8459: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8460:
8461: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8462: for(k=1; k<=i1;k++){
8463: if(i1 != 1 && TKresult[nres]!= k)
8464: continue;
8465: if(invalidvarcomb[k]){
8466: printf("\nCombination (%d) projection ignored because no cases \n",k);
8467: continue;
8468: }
1.268 brouard 8469: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8470: for(j=1;j<=cptcoveff;j++) {
8471: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8472: }
8473: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8474: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8475: }
8476: fprintf(ficresfb," yearbproj age");
8477: for(j=1; j<=nlstate+ndeath;j++){
8478: for(i=1; i<=nlstate;i++)
1.268 brouard 8479: fprintf(ficresfb," b%d%d",i,j);
8480: fprintf(ficresfb," b.%d",j);
1.267 brouard 8481: }
8482: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8483: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8484: fprintf(ficresfb,"\n");
8485: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8486: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8487: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8488: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8489: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8490: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8491: nhstepm = nhstepm/hstepm;
8492: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8493: oldm=oldms;savm=savms;
1.268 brouard 8494: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8495: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8496: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8497: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8498: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8499: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8500: for (h=0; h<=nhstepm; h++){
1.268 brouard 8501: if (h*hstepm/YEARM*stepm ==-yearp) {
8502: break;
8503: }
8504: }
8505: fprintf(ficresfb,"\n");
8506: for(j=1;j<=cptcoveff;j++)
8507: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8508: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8509: for(i=1; i<=nlstate+ndeath;i++) {
8510: ppij=0.;ppi=0.;
8511: for(j=1; j<=nlstate;j++) {
8512: /* if (mobilav==1) */
1.269 brouard 8513: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8514: ppi=ppi+prevacurrent[(int)agec][j][k];
8515: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8516: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8517: /* else { */
8518: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8519: /* } */
1.268 brouard 8520: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8521: } /* end j */
8522: if(ppi <0.99){
8523: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8524: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8525: }
8526: fprintf(ficresfb," %.3f", ppij);
8527: }/* end j */
1.267 brouard 8528: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8529: } /* end agec */
8530: } /* end yearp */
8531: } /* end k */
1.217 brouard 8532:
1.267 brouard 8533: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8534:
1.267 brouard 8535: fclose(ficresfb);
8536: printf("End of Computing Back forecasting \n");
8537: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8538:
1.267 brouard 8539: }
1.217 brouard 8540:
1.269 brouard 8541: /* Variance of prevalence limit: varprlim */
8542: 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){
8543: /*------- Variance of period (stable) prevalence------*/
8544:
8545: char fileresvpl[FILENAMELENGTH];
8546: FILE *ficresvpl;
8547: double **oldm, **savm;
8548: double **varpl; /* Variances of prevalence limits by age */
8549: int i1, k, nres, j ;
8550:
8551: strcpy(fileresvpl,"VPL_");
8552: strcat(fileresvpl,fileresu);
8553: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8554: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8555: exit(0);
8556: }
8557: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8558: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8559:
8560: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8561: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8562:
8563: i1=pow(2,cptcoveff);
8564: if (cptcovn < 1){i1=1;}
8565:
8566: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8567: for(k=1; k<=i1;k++){
8568: if(i1 != 1 && TKresult[nres]!= k)
8569: continue;
8570: fprintf(ficresvpl,"\n#****** ");
8571: printf("\n#****** ");
8572: fprintf(ficlog,"\n#****** ");
8573: for(j=1;j<=cptcoveff;j++) {
8574: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8575: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8576: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8577: }
8578: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8579: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8580: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8581: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8582: }
8583: fprintf(ficresvpl,"******\n");
8584: printf("******\n");
8585: fprintf(ficlog,"******\n");
8586:
8587: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8588: oldm=oldms;savm=savms;
8589: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8590: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8591: /*}*/
8592: }
8593:
8594: fclose(ficresvpl);
8595: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8596: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8597:
8598: }
8599: /* Variance of back prevalence: varbprlim */
8600: 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){
8601: /*------- Variance of back (stable) prevalence------*/
8602:
8603: char fileresvbl[FILENAMELENGTH];
8604: FILE *ficresvbl;
8605:
8606: double **oldm, **savm;
8607: double **varbpl; /* Variances of back prevalence limits by age */
8608: int i1, k, nres, j ;
8609:
8610: strcpy(fileresvbl,"VBL_");
8611: strcat(fileresvbl,fileresu);
8612: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8613: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8614: exit(0);
8615: }
8616: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8617: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8618:
8619:
8620: i1=pow(2,cptcoveff);
8621: if (cptcovn < 1){i1=1;}
8622:
8623: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8624: for(k=1; k<=i1;k++){
8625: if(i1 != 1 && TKresult[nres]!= k)
8626: continue;
8627: fprintf(ficresvbl,"\n#****** ");
8628: printf("\n#****** ");
8629: fprintf(ficlog,"\n#****** ");
8630: for(j=1;j<=cptcoveff;j++) {
8631: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8632: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8633: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8634: }
8635: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8636: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8637: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8638: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8639: }
8640: fprintf(ficresvbl,"******\n");
8641: printf("******\n");
8642: fprintf(ficlog,"******\n");
8643:
8644: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8645: oldm=oldms;savm=savms;
8646:
8647: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8648: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8649: /*}*/
8650: }
8651:
8652: fclose(ficresvbl);
8653: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8654: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8655:
8656: } /* End of varbprlim */
8657:
1.126 brouard 8658: /************** Forecasting *****not tested NB*************/
1.227 brouard 8659: /* 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 8660:
1.227 brouard 8661: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8662: /* int *popage; */
8663: /* double calagedatem, agelim, kk1, kk2; */
8664: /* double *popeffectif,*popcount; */
8665: /* double ***p3mat,***tabpop,***tabpopprev; */
8666: /* /\* double ***mobaverage; *\/ */
8667: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8668:
1.227 brouard 8669: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8670: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8671: /* agelim=AGESUP; */
8672: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8673:
1.227 brouard 8674: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8675:
8676:
1.227 brouard 8677: /* strcpy(filerespop,"POP_"); */
8678: /* strcat(filerespop,fileresu); */
8679: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8680: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8681: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8682: /* } */
8683: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8684: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8685:
1.227 brouard 8686: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8687:
1.227 brouard 8688: /* /\* if (mobilav!=0) { *\/ */
8689: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8690: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8691: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8692: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8693: /* /\* } *\/ */
8694: /* /\* } *\/ */
1.126 brouard 8695:
1.227 brouard 8696: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8697: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8698:
1.227 brouard 8699: /* agelim=AGESUP; */
1.126 brouard 8700:
1.227 brouard 8701: /* hstepm=1; */
8702: /* hstepm=hstepm/stepm; */
1.218 brouard 8703:
1.227 brouard 8704: /* if (popforecast==1) { */
8705: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8706: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8707: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8708: /* } */
8709: /* popage=ivector(0,AGESUP); */
8710: /* popeffectif=vector(0,AGESUP); */
8711: /* popcount=vector(0,AGESUP); */
1.126 brouard 8712:
1.227 brouard 8713: /* i=1; */
8714: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8715:
1.227 brouard 8716: /* imx=i; */
8717: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8718: /* } */
1.218 brouard 8719:
1.227 brouard 8720: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8721: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8722: /* k=k+1; */
8723: /* fprintf(ficrespop,"\n#******"); */
8724: /* for(j=1;j<=cptcoveff;j++) { */
8725: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8726: /* } */
8727: /* fprintf(ficrespop,"******\n"); */
8728: /* fprintf(ficrespop,"# Age"); */
8729: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8730: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8731:
1.227 brouard 8732: /* for (cpt=0; cpt<=0;cpt++) { */
8733: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8734:
1.227 brouard 8735: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8736: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8737: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8738:
1.227 brouard 8739: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8740: /* oldm=oldms;savm=savms; */
8741: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8742:
1.227 brouard 8743: /* for (h=0; h<=nhstepm; h++){ */
8744: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8745: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8746: /* } */
8747: /* for(j=1; j<=nlstate+ndeath;j++) { */
8748: /* kk1=0.;kk2=0; */
8749: /* for(i=1; i<=nlstate;i++) { */
8750: /* if (mobilav==1) */
8751: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8752: /* else { */
8753: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8754: /* } */
8755: /* } */
8756: /* if (h==(int)(calagedatem+12*cpt)){ */
8757: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8758: /* /\*fprintf(ficrespop," %.3f", kk1); */
8759: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8760: /* } */
8761: /* } */
8762: /* for(i=1; i<=nlstate;i++){ */
8763: /* kk1=0.; */
8764: /* for(j=1; j<=nlstate;j++){ */
8765: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8766: /* } */
8767: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8768: /* } */
1.218 brouard 8769:
1.227 brouard 8770: /* if (h==(int)(calagedatem+12*cpt)) */
8771: /* for(j=1; j<=nlstate;j++) */
8772: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8773: /* } */
8774: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8775: /* } */
8776: /* } */
1.218 brouard 8777:
1.227 brouard 8778: /* /\******\/ */
1.218 brouard 8779:
1.227 brouard 8780: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8781: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8782: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8783: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8784: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8785:
1.227 brouard 8786: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8787: /* oldm=oldms;savm=savms; */
8788: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8789: /* for (h=0; h<=nhstepm; h++){ */
8790: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8791: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8792: /* } */
8793: /* for(j=1; j<=nlstate+ndeath;j++) { */
8794: /* kk1=0.;kk2=0; */
8795: /* for(i=1; i<=nlstate;i++) { */
8796: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8797: /* } */
8798: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8799: /* } */
8800: /* } */
8801: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8802: /* } */
8803: /* } */
8804: /* } */
8805: /* } */
1.218 brouard 8806:
1.227 brouard 8807: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8808:
1.227 brouard 8809: /* if (popforecast==1) { */
8810: /* free_ivector(popage,0,AGESUP); */
8811: /* free_vector(popeffectif,0,AGESUP); */
8812: /* free_vector(popcount,0,AGESUP); */
8813: /* } */
8814: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8815: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8816: /* fclose(ficrespop); */
8817: /* } /\* End of popforecast *\/ */
1.218 brouard 8818:
1.126 brouard 8819: int fileappend(FILE *fichier, char *optionfich)
8820: {
8821: if((fichier=fopen(optionfich,"a"))==NULL) {
8822: printf("Problem with file: %s\n", optionfich);
8823: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8824: return (0);
8825: }
8826: fflush(fichier);
8827: return (1);
8828: }
8829:
8830:
8831: /**************** function prwizard **********************/
8832: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8833: {
8834:
8835: /* Wizard to print covariance matrix template */
8836:
1.164 brouard 8837: char ca[32], cb[32];
8838: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8839: int numlinepar;
8840:
8841: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8842: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8843: for(i=1; i <=nlstate; i++){
8844: jj=0;
8845: for(j=1; j <=nlstate+ndeath; j++){
8846: if(j==i) continue;
8847: jj++;
8848: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8849: printf("%1d%1d",i,j);
8850: fprintf(ficparo,"%1d%1d",i,j);
8851: for(k=1; k<=ncovmodel;k++){
8852: /* printf(" %lf",param[i][j][k]); */
8853: /* fprintf(ficparo," %lf",param[i][j][k]); */
8854: printf(" 0.");
8855: fprintf(ficparo," 0.");
8856: }
8857: printf("\n");
8858: fprintf(ficparo,"\n");
8859: }
8860: }
8861: printf("# Scales (for hessian or gradient estimation)\n");
8862: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8863: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8864: for(i=1; i <=nlstate; i++){
8865: jj=0;
8866: for(j=1; j <=nlstate+ndeath; j++){
8867: if(j==i) continue;
8868: jj++;
8869: fprintf(ficparo,"%1d%1d",i,j);
8870: printf("%1d%1d",i,j);
8871: fflush(stdout);
8872: for(k=1; k<=ncovmodel;k++){
8873: /* printf(" %le",delti3[i][j][k]); */
8874: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8875: printf(" 0.");
8876: fprintf(ficparo," 0.");
8877: }
8878: numlinepar++;
8879: printf("\n");
8880: fprintf(ficparo,"\n");
8881: }
8882: }
8883: printf("# Covariance matrix\n");
8884: /* # 121 Var(a12)\n\ */
8885: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8886: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8887: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8888: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8889: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8890: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8891: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8892: fflush(stdout);
8893: fprintf(ficparo,"# Covariance matrix\n");
8894: /* # 121 Var(a12)\n\ */
8895: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8896: /* # ...\n\ */
8897: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8898:
8899: for(itimes=1;itimes<=2;itimes++){
8900: jj=0;
8901: for(i=1; i <=nlstate; i++){
8902: for(j=1; j <=nlstate+ndeath; j++){
8903: if(j==i) continue;
8904: for(k=1; k<=ncovmodel;k++){
8905: jj++;
8906: ca[0]= k+'a'-1;ca[1]='\0';
8907: if(itimes==1){
8908: printf("#%1d%1d%d",i,j,k);
8909: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8910: }else{
8911: printf("%1d%1d%d",i,j,k);
8912: fprintf(ficparo,"%1d%1d%d",i,j,k);
8913: /* printf(" %.5le",matcov[i][j]); */
8914: }
8915: ll=0;
8916: for(li=1;li <=nlstate; li++){
8917: for(lj=1;lj <=nlstate+ndeath; lj++){
8918: if(lj==li) continue;
8919: for(lk=1;lk<=ncovmodel;lk++){
8920: ll++;
8921: if(ll<=jj){
8922: cb[0]= lk +'a'-1;cb[1]='\0';
8923: if(ll<jj){
8924: if(itimes==1){
8925: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8926: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8927: }else{
8928: printf(" 0.");
8929: fprintf(ficparo," 0.");
8930: }
8931: }else{
8932: if(itimes==1){
8933: printf(" Var(%s%1d%1d)",ca,i,j);
8934: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8935: }else{
8936: printf(" 0.");
8937: fprintf(ficparo," 0.");
8938: }
8939: }
8940: }
8941: } /* end lk */
8942: } /* end lj */
8943: } /* end li */
8944: printf("\n");
8945: fprintf(ficparo,"\n");
8946: numlinepar++;
8947: } /* end k*/
8948: } /*end j */
8949: } /* end i */
8950: } /* end itimes */
8951:
8952: } /* end of prwizard */
8953: /******************* Gompertz Likelihood ******************************/
8954: double gompertz(double x[])
8955: {
8956: double A,B,L=0.0,sump=0.,num=0.;
8957: int i,n=0; /* n is the size of the sample */
8958:
1.220 brouard 8959: for (i=1;i<=imx ; i++) {
1.126 brouard 8960: sump=sump+weight[i];
8961: /* sump=sump+1;*/
8962: num=num+1;
8963: }
8964:
8965:
8966: /* for (i=0; i<=imx; i++)
8967: 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]);*/
8968:
8969: for (i=1;i<=imx ; i++)
8970: {
8971: if (cens[i] == 1 && wav[i]>1)
8972: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8973:
8974: if (cens[i] == 0 && wav[i]>1)
8975: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8976: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8977:
8978: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8979: if (wav[i] > 1 ) { /* ??? */
8980: L=L+A*weight[i];
8981: /* 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]);*/
8982: }
8983: }
8984:
8985: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8986:
8987: return -2*L*num/sump;
8988: }
8989:
1.136 brouard 8990: #ifdef GSL
8991: /******************* Gompertz_f Likelihood ******************************/
8992: double gompertz_f(const gsl_vector *v, void *params)
8993: {
8994: double A,B,LL=0.0,sump=0.,num=0.;
8995: double *x= (double *) v->data;
8996: int i,n=0; /* n is the size of the sample */
8997:
8998: for (i=0;i<=imx-1 ; i++) {
8999: sump=sump+weight[i];
9000: /* sump=sump+1;*/
9001: num=num+1;
9002: }
9003:
9004:
9005: /* for (i=0; i<=imx; i++)
9006: 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]);*/
9007: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9008: for (i=1;i<=imx ; i++)
9009: {
9010: if (cens[i] == 1 && wav[i]>1)
9011: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9012:
9013: if (cens[i] == 0 && wav[i]>1)
9014: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9015: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9016:
9017: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9018: if (wav[i] > 1 ) { /* ??? */
9019: LL=LL+A*weight[i];
9020: /* 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]);*/
9021: }
9022: }
9023:
9024: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9025: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9026:
9027: return -2*LL*num/sump;
9028: }
9029: #endif
9030:
1.126 brouard 9031: /******************* Printing html file ***********/
1.201 brouard 9032: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9033: int lastpass, int stepm, int weightopt, char model[],\
9034: int imx, double p[],double **matcov,double agemortsup){
9035: int i,k;
9036:
9037: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9038: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9039: for (i=1;i<=2;i++)
9040: 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 9041: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9042: fprintf(fichtm,"</ul>");
9043:
9044: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9045:
9046: 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>");
9047:
9048: for (k=agegomp;k<(agemortsup-2);k++)
9049: 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]);
9050:
9051:
9052: fflush(fichtm);
9053: }
9054:
9055: /******************* Gnuplot file **************/
1.201 brouard 9056: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9057:
9058: char dirfileres[132],optfileres[132];
1.164 brouard 9059:
1.126 brouard 9060: int ng;
9061:
9062:
9063: /*#ifdef windows */
9064: fprintf(ficgp,"cd \"%s\" \n",pathc);
9065: /*#endif */
9066:
9067:
9068: strcpy(dirfileres,optionfilefiname);
9069: strcpy(optfileres,"vpl");
1.199 brouard 9070: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9071: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9072: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9073: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9074: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9075:
9076: }
9077:
1.136 brouard 9078: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9079: {
1.126 brouard 9080:
1.136 brouard 9081: /*-------- data file ----------*/
9082: FILE *fic;
9083: char dummy[]=" ";
1.240 brouard 9084: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9085: int lstra;
1.136 brouard 9086: int linei, month, year,iout;
9087: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9088: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9089: char *stratrunc;
1.223 brouard 9090:
1.240 brouard 9091: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9092: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9093:
1.240 brouard 9094: for(v=1; v <=ncovcol;v++){
9095: DummyV[v]=0;
9096: FixedV[v]=0;
9097: }
9098: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9099: DummyV[v]=1;
9100: FixedV[v]=0;
9101: }
9102: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9103: DummyV[v]=0;
9104: FixedV[v]=1;
9105: }
9106: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9107: DummyV[v]=1;
9108: FixedV[v]=1;
9109: }
9110: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9111: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9112: 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]);
9113: }
1.126 brouard 9114:
1.136 brouard 9115: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9116: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9117: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9118: }
1.126 brouard 9119:
1.136 brouard 9120: i=1;
9121: linei=0;
9122: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9123: linei=linei+1;
9124: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9125: if(line[j] == '\t')
9126: line[j] = ' ';
9127: }
9128: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9129: ;
9130: };
9131: line[j+1]=0; /* Trims blanks at end of line */
9132: if(line[0]=='#'){
9133: fprintf(ficlog,"Comment line\n%s\n",line);
9134: printf("Comment line\n%s\n",line);
9135: continue;
9136: }
9137: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9138: strcpy(line, linetmp);
1.223 brouard 9139:
9140: /* Loops on waves */
9141: for (j=maxwav;j>=1;j--){
9142: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9143: cutv(stra, strb, line, ' ');
9144: if(strb[0]=='.') { /* Missing value */
9145: lval=-1;
9146: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9147: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9148: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9149: 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);
9150: 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);
9151: return 1;
9152: }
9153: }else{
9154: errno=0;
9155: /* what_kind_of_number(strb); */
9156: dval=strtod(strb,&endptr);
9157: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9158: /* if(strb != endptr && *endptr == '\0') */
9159: /* dval=dlval; */
9160: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9161: if( strb[0]=='\0' || (*endptr != '\0')){
9162: 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);
9163: 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);
9164: return 1;
9165: }
9166: cotqvar[j][iv][i]=dval;
9167: cotvar[j][ntv+iv][i]=dval;
9168: }
9169: strcpy(line,stra);
1.223 brouard 9170: }/* end loop ntqv */
1.225 brouard 9171:
1.223 brouard 9172: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9173: cutv(stra, strb, line, ' ');
9174: if(strb[0]=='.') { /* Missing value */
9175: lval=-1;
9176: }else{
9177: errno=0;
9178: lval=strtol(strb,&endptr,10);
9179: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9180: if( strb[0]=='\0' || (*endptr != '\0')){
9181: 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);
9182: 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);
9183: return 1;
9184: }
9185: }
9186: if(lval <-1 || lval >1){
9187: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9188: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9189: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9190: For example, for multinomial values like 1, 2 and 3,\n \
9191: build V1=0 V2=0 for the reference value (1),\n \
9192: V1=1 V2=0 for (2) \n \
1.223 brouard 9193: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9194: output of IMaCh is often meaningless.\n \
1.223 brouard 9195: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9196: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9197: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9198: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9199: For example, for multinomial values like 1, 2 and 3,\n \
9200: build V1=0 V2=0 for the reference value (1),\n \
9201: V1=1 V2=0 for (2) \n \
1.223 brouard 9202: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9203: output of IMaCh is often meaningless.\n \
1.223 brouard 9204: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9205: return 1;
9206: }
9207: cotvar[j][iv][i]=(double)(lval);
9208: strcpy(line,stra);
1.223 brouard 9209: }/* end loop ntv */
1.225 brouard 9210:
1.223 brouard 9211: /* Statuses at wave */
1.137 brouard 9212: cutv(stra, strb, line, ' ');
1.223 brouard 9213: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9214: lval=-1;
1.136 brouard 9215: }else{
1.238 brouard 9216: errno=0;
9217: lval=strtol(strb,&endptr,10);
9218: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9219: if( strb[0]=='\0' || (*endptr != '\0')){
9220: 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);
9221: 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);
9222: return 1;
9223: }
1.136 brouard 9224: }
1.225 brouard 9225:
1.136 brouard 9226: s[j][i]=lval;
1.225 brouard 9227:
1.223 brouard 9228: /* Date of Interview */
1.136 brouard 9229: strcpy(line,stra);
9230: cutv(stra, strb,line,' ');
1.169 brouard 9231: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9232: }
1.169 brouard 9233: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9234: month=99;
9235: year=9999;
1.136 brouard 9236: }else{
1.225 brouard 9237: 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);
9238: 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);
9239: return 1;
1.136 brouard 9240: }
9241: anint[j][i]= (double) year;
9242: mint[j][i]= (double)month;
9243: strcpy(line,stra);
1.223 brouard 9244: } /* End loop on waves */
1.225 brouard 9245:
1.223 brouard 9246: /* Date of death */
1.136 brouard 9247: cutv(stra, strb,line,' ');
1.169 brouard 9248: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9249: }
1.169 brouard 9250: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9251: month=99;
9252: year=9999;
9253: }else{
1.141 brouard 9254: 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 9255: 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);
9256: return 1;
1.136 brouard 9257: }
9258: andc[i]=(double) year;
9259: moisdc[i]=(double) month;
9260: strcpy(line,stra);
9261:
1.223 brouard 9262: /* Date of birth */
1.136 brouard 9263: cutv(stra, strb,line,' ');
1.169 brouard 9264: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9265: }
1.169 brouard 9266: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9267: month=99;
9268: year=9999;
9269: }else{
1.141 brouard 9270: 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);
9271: 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 9272: return 1;
1.136 brouard 9273: }
9274: if (year==9999) {
1.141 brouard 9275: 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);
9276: 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 9277: return 1;
9278:
1.136 brouard 9279: }
9280: annais[i]=(double)(year);
9281: moisnais[i]=(double)(month);
9282: strcpy(line,stra);
1.225 brouard 9283:
1.223 brouard 9284: /* Sample weight */
1.136 brouard 9285: cutv(stra, strb,line,' ');
9286: errno=0;
9287: dval=strtod(strb,&endptr);
9288: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9289: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9290: 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 9291: fflush(ficlog);
9292: return 1;
9293: }
9294: weight[i]=dval;
9295: strcpy(line,stra);
1.225 brouard 9296:
1.223 brouard 9297: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9298: cutv(stra, strb, line, ' ');
9299: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9300: lval=-1;
1.223 brouard 9301: }else{
1.225 brouard 9302: errno=0;
9303: /* what_kind_of_number(strb); */
9304: dval=strtod(strb,&endptr);
9305: /* if(strb != endptr && *endptr == '\0') */
9306: /* dval=dlval; */
9307: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9308: if( strb[0]=='\0' || (*endptr != '\0')){
9309: 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);
9310: 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);
9311: return 1;
9312: }
9313: coqvar[iv][i]=dval;
1.226 brouard 9314: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9315: }
9316: strcpy(line,stra);
9317: }/* end loop nqv */
1.136 brouard 9318:
1.223 brouard 9319: /* Covariate values */
1.136 brouard 9320: for (j=ncovcol;j>=1;j--){
9321: cutv(stra, strb,line,' ');
1.223 brouard 9322: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9323: lval=-1;
1.136 brouard 9324: }else{
1.225 brouard 9325: errno=0;
9326: lval=strtol(strb,&endptr,10);
9327: if( strb[0]=='\0' || (*endptr != '\0')){
9328: 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);
9329: 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);
9330: return 1;
9331: }
1.136 brouard 9332: }
9333: if(lval <-1 || lval >1){
1.225 brouard 9334: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9335: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9336: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9337: For example, for multinomial values like 1, 2 and 3,\n \
9338: build V1=0 V2=0 for the reference value (1),\n \
9339: V1=1 V2=0 for (2) \n \
1.136 brouard 9340: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9341: output of IMaCh is often meaningless.\n \
1.136 brouard 9342: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9343: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9344: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9345: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9346: For example, for multinomial values like 1, 2 and 3,\n \
9347: build V1=0 V2=0 for the reference value (1),\n \
9348: V1=1 V2=0 for (2) \n \
1.136 brouard 9349: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9350: output of IMaCh is often meaningless.\n \
1.136 brouard 9351: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9352: return 1;
1.136 brouard 9353: }
9354: covar[j][i]=(double)(lval);
9355: strcpy(line,stra);
9356: }
9357: lstra=strlen(stra);
1.225 brouard 9358:
1.136 brouard 9359: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9360: stratrunc = &(stra[lstra-9]);
9361: num[i]=atol(stratrunc);
9362: }
9363: else
9364: num[i]=atol(stra);
9365: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9366: 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;}*/
9367:
9368: i=i+1;
9369: } /* End loop reading data */
1.225 brouard 9370:
1.136 brouard 9371: *imax=i-1; /* Number of individuals */
9372: fclose(fic);
1.225 brouard 9373:
1.136 brouard 9374: return (0);
1.164 brouard 9375: /* endread: */
1.225 brouard 9376: printf("Exiting readdata: ");
9377: fclose(fic);
9378: return (1);
1.223 brouard 9379: }
1.126 brouard 9380:
1.234 brouard 9381: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9382: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9383: while (*p2 == ' ')
1.234 brouard 9384: p2++;
9385: /* while ((*p1++ = *p2++) !=0) */
9386: /* ; */
9387: /* do */
9388: /* while (*p2 == ' ') */
9389: /* p2++; */
9390: /* while (*p1++ == *p2++); */
9391: *stri=p2;
1.145 brouard 9392: }
9393:
1.235 brouard 9394: int decoderesult ( char resultline[], int nres)
1.230 brouard 9395: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9396: {
1.235 brouard 9397: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9398: char resultsav[MAXLINE];
1.234 brouard 9399: int resultmodel[MAXLINE];
9400: int modelresult[MAXLINE];
1.230 brouard 9401: char stra[80], strb[80], strc[80], strd[80],stre[80];
9402:
1.234 brouard 9403: removefirstspace(&resultline);
1.233 brouard 9404: printf("decoderesult:%s\n",resultline);
1.230 brouard 9405:
9406: if (strstr(resultline,"v") !=0){
9407: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9408: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9409: return 1;
9410: }
9411: trimbb(resultsav, resultline);
9412: if (strlen(resultsav) >1){
9413: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9414: }
1.253 brouard 9415: if(j == 0){ /* Resultline but no = */
9416: TKresult[nres]=0; /* Combination for the nresult and the model */
9417: return (0);
9418: }
9419:
1.234 brouard 9420: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9421: printf("ERROR: the number of variable in the resultline, %d, differs from the number of variable used in the model line, %d.\n",j, cptcovs);
9422: fprintf(ficlog,"ERROR: the number of variable in the resultline, %d, differs from the number of variable used in the model line, %d.\n",j, cptcovs);
9423: }
9424: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9425: if(nbocc(resultsav,'=') >1){
9426: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9427: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9428: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9429: }else
9430: cutl(strc,strd,resultsav,'=');
1.230 brouard 9431: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9432:
1.230 brouard 9433: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9434: Tvarsel[k]=atoi(strc);
9435: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9436: /* cptcovsel++; */
9437: if (nbocc(stra,'=') >0)
9438: strcpy(resultsav,stra); /* and analyzes it */
9439: }
1.235 brouard 9440: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9441: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9442: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9443: match=0;
1.236 brouard 9444: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9445: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9446: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9447: match=1;
9448: break;
9449: }
9450: }
9451: if(match == 0){
9452: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9453: }
9454: }
9455: }
1.235 brouard 9456: /* Checking for missing or useless values in comparison of current model needs */
9457: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9458: match=0;
1.235 brouard 9459: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9460: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9461: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9462: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9463: ++match;
9464: }
9465: }
9466: }
9467: if(match == 0){
9468: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9469: }else if(match > 1){
9470: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9471: }
9472: }
1.235 brouard 9473:
1.234 brouard 9474: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9475: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9476: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9477: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9478: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9479: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9480: /* 1 0 0 0 */
9481: /* 2 1 0 0 */
9482: /* 3 0 1 0 */
9483: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9484: /* 5 0 0 1 */
9485: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9486: /* 7 0 1 1 */
9487: /* 8 1 1 1 */
1.237 brouard 9488: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9489: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9490: /* V5*age V5 known which value for nres? */
9491: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9492: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9493: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9494: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9495: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9496: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9497: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9498: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9499: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9500: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9501: k4++;;
9502: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9503: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9504: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9505: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9506: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9507: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9508: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9509: k4q++;;
9510: }
9511: }
1.234 brouard 9512:
1.235 brouard 9513: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9514: return (0);
9515: }
1.235 brouard 9516:
1.230 brouard 9517: int decodemodel( char model[], int lastobs)
9518: /**< This routine decodes the model and returns:
1.224 brouard 9519: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9520: * - nagesqr = 1 if age*age in the model, otherwise 0.
9521: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9522: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9523: * - cptcovage number of covariates with age*products =2
9524: * - cptcovs number of simple covariates
9525: * - 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
9526: * which is a new column after the 9 (ncovcol) variables.
9527: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9528: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9529: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9530: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9531: */
1.136 brouard 9532: {
1.238 brouard 9533: int i, j, k, ks, v;
1.227 brouard 9534: int j1, k1, k2, k3, k4;
1.136 brouard 9535: char modelsav[80];
1.145 brouard 9536: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9537: char *strpt;
1.136 brouard 9538:
1.145 brouard 9539: /*removespace(model);*/
1.136 brouard 9540: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9541: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9542: if (strstr(model,"AGE") !=0){
1.192 brouard 9543: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9544: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9545: return 1;
9546: }
1.141 brouard 9547: if (strstr(model,"v") !=0){
9548: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9549: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9550: return 1;
9551: }
1.187 brouard 9552: strcpy(modelsav,model);
9553: if ((strpt=strstr(model,"age*age")) !=0){
9554: printf(" strpt=%s, model=%s\n",strpt, model);
9555: if(strpt != model){
1.234 brouard 9556: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9557: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9558: corresponding column of parameters.\n",model);
1.234 brouard 9559: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9560: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9561: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9562: return 1;
1.225 brouard 9563: }
1.187 brouard 9564: nagesqr=1;
9565: if (strstr(model,"+age*age") !=0)
1.234 brouard 9566: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9567: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9568: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9569: else
1.234 brouard 9570: substrchaine(modelsav, model, "age*age");
1.187 brouard 9571: }else
9572: nagesqr=0;
9573: if (strlen(modelsav) >1){
9574: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9575: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9576: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9577: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9578: * cst, age and age*age
9579: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9580: /* including age products which are counted in cptcovage.
9581: * but the covariates which are products must be treated
9582: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9583: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9584: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9585:
9586:
1.187 brouard 9587: /* Design
9588: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9589: * < ncovcol=8 >
9590: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9591: * k= 1 2 3 4 5 6 7 8
9592: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9593: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9594: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9595: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9596: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9597: * Tage[++cptcovage]=k
9598: * if products, new covar are created after ncovcol with k1
9599: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9600: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9601: * 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
9602: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9603: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9604: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9605: * < ncovcol=8 >
9606: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9607: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9608: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9609: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9610: * p Tprod[1]@2={ 6, 5}
9611: *p Tvard[1][1]@4= {7, 8, 5, 6}
9612: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9613: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9614: *How to reorganize?
9615: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9616: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9617: * {2, 1, 4, 8, 5, 6, 3, 7}
9618: * Struct []
9619: */
1.225 brouard 9620:
1.187 brouard 9621: /* This loop fills the array Tvar from the string 'model'.*/
9622: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9623: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9624: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9625: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9626: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9627: /* k=1 Tvar[1]=2 (from V2) */
9628: /* k=5 Tvar[5] */
9629: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9630: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9631: /* } */
1.198 brouard 9632: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9633: /*
9634: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9635: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9636: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9637: }
1.187 brouard 9638: cptcovage=0;
9639: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9640: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9641: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9642: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9643: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9644: /*scanf("%d",i);*/
9645: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9646: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9647: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9648: /* covar is not filled and then is empty */
9649: cptcovprod--;
9650: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9651: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9652: Typevar[k]=1; /* 1 for age product */
9653: cptcovage++; /* Sums the number of covariates which include age as a product */
9654: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9655: /*printf("stre=%s ", stre);*/
9656: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9657: cptcovprod--;
9658: cutl(stre,strb,strc,'V');
9659: Tvar[k]=atoi(stre);
9660: Typevar[k]=1; /* 1 for age product */
9661: cptcovage++;
9662: Tage[cptcovage]=k;
9663: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9664: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9665: cptcovn++;
9666: cptcovprodnoage++;k1++;
9667: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9668: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9669: because this model-covariate is a construction we invent a new column
9670: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9671: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9672: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9673: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9674: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9675: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9676: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9677: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9678: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9679: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9680: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9681: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9682: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9683: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9684: for (i=1; i<=lastobs;i++){
9685: /* Computes the new covariate which is a product of
9686: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9687: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9688: }
9689: } /* End age is not in the model */
9690: } /* End if model includes a product */
9691: else { /* no more sum */
9692: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9693: /* scanf("%d",i);*/
9694: cutl(strd,strc,strb,'V');
9695: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9696: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9697: Tvar[k]=atoi(strd);
9698: Typevar[k]=0; /* 0 for simple covariates */
9699: }
9700: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9701: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9702: scanf("%d",i);*/
1.187 brouard 9703: } /* end of loop + on total covariates */
9704: } /* end if strlen(modelsave == 0) age*age might exist */
9705: } /* end if strlen(model == 0) */
1.136 brouard 9706:
9707: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9708: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9709:
1.136 brouard 9710: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9711: printf("cptcovprod=%d ", cptcovprod);
9712: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9713: scanf("%d ",i);*/
9714:
9715:
1.230 brouard 9716: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9717: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9718: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9719: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9720: k = 1 2 3 4 5 6 7 8 9
9721: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9722: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9723: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9724: Dummy[k] 1 0 0 0 3 1 1 2 3
9725: Tmodelind[combination of covar]=k;
1.225 brouard 9726: */
9727: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9728: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9729: /* 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 9730: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9731: printf("Model=%s\n\
9732: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9733: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9734: 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);
9735: fprintf(ficlog,"Model=%s\n\
9736: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9737: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9738: 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 9739: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9740: 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 */
9741: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9742: Fixed[k]= 0;
9743: Dummy[k]= 0;
1.225 brouard 9744: ncoveff++;
1.232 brouard 9745: ncovf++;
1.234 brouard 9746: nsd++;
9747: modell[k].maintype= FTYPE;
9748: TvarsD[nsd]=Tvar[k];
9749: TvarsDind[nsd]=k;
9750: TvarF[ncovf]=Tvar[k];
9751: TvarFind[ncovf]=k;
9752: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9753: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9754: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9755: Fixed[k]= 0;
9756: Dummy[k]= 0;
9757: ncoveff++;
9758: ncovf++;
9759: modell[k].maintype= FTYPE;
9760: TvarF[ncovf]=Tvar[k];
9761: TvarFind[ncovf]=k;
1.230 brouard 9762: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9763: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9764: }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 9765: Fixed[k]= 0;
9766: Dummy[k]= 1;
1.230 brouard 9767: nqfveff++;
1.234 brouard 9768: modell[k].maintype= FTYPE;
9769: modell[k].subtype= FQ;
9770: nsq++;
9771: TvarsQ[nsq]=Tvar[k];
9772: TvarsQind[nsq]=k;
1.232 brouard 9773: ncovf++;
1.234 brouard 9774: TvarF[ncovf]=Tvar[k];
9775: TvarFind[ncovf]=k;
1.231 brouard 9776: 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 9777: 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 9778: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9779: Fixed[k]= 1;
9780: Dummy[k]= 0;
1.225 brouard 9781: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9782: modell[k].maintype= VTYPE;
9783: modell[k].subtype= VD;
9784: nsd++;
9785: TvarsD[nsd]=Tvar[k];
9786: TvarsDind[nsd]=k;
9787: ncovv++; /* Only simple time varying variables */
9788: TvarV[ncovv]=Tvar[k];
1.242 brouard 9789: 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 9790: 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 */
9791: 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 9792: 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);
9793: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9794: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9795: Fixed[k]= 1;
9796: Dummy[k]= 1;
9797: nqtveff++;
9798: modell[k].maintype= VTYPE;
9799: modell[k].subtype= VQ;
9800: ncovv++; /* Only simple time varying variables */
9801: nsq++;
9802: TvarsQ[nsq]=Tvar[k];
9803: TvarsQind[nsq]=k;
9804: TvarV[ncovv]=Tvar[k];
1.242 brouard 9805: 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 9806: 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 */
9807: 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 9808: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9809: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9810: 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 9811: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9812: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9813: ncova++;
9814: TvarA[ncova]=Tvar[k];
9815: TvarAind[ncova]=k;
1.231 brouard 9816: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9817: Fixed[k]= 2;
9818: Dummy[k]= 2;
9819: modell[k].maintype= ATYPE;
9820: modell[k].subtype= APFD;
9821: /* ncoveff++; */
1.227 brouard 9822: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9823: Fixed[k]= 2;
9824: Dummy[k]= 3;
9825: modell[k].maintype= ATYPE;
9826: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9827: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9828: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9829: Fixed[k]= 3;
9830: Dummy[k]= 2;
9831: modell[k].maintype= ATYPE;
9832: modell[k].subtype= APVD; /* Product age * varying dummy */
9833: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9834: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9835: Fixed[k]= 3;
9836: Dummy[k]= 3;
9837: modell[k].maintype= ATYPE;
9838: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9839: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9840: }
9841: }else if (Typevar[k] == 2) { /* product without age */
9842: k1=Tposprod[k];
9843: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9844: if(Tvard[k1][2] <=ncovcol){
9845: Fixed[k]= 1;
9846: Dummy[k]= 0;
9847: modell[k].maintype= FTYPE;
9848: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9849: ncovf++; /* Fixed variables without age */
9850: TvarF[ncovf]=Tvar[k];
9851: TvarFind[ncovf]=k;
9852: }else if(Tvard[k1][2] <=ncovcol+nqv){
9853: Fixed[k]= 0; /* or 2 ?*/
9854: Dummy[k]= 1;
9855: modell[k].maintype= FTYPE;
9856: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9857: ncovf++; /* Varying variables without age */
9858: TvarF[ncovf]=Tvar[k];
9859: TvarFind[ncovf]=k;
9860: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9861: Fixed[k]= 1;
9862: Dummy[k]= 0;
9863: modell[k].maintype= VTYPE;
9864: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9865: ncovv++; /* Varying variables without age */
9866: TvarV[ncovv]=Tvar[k];
9867: TvarVind[ncovv]=k;
9868: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9869: Fixed[k]= 1;
9870: Dummy[k]= 1;
9871: modell[k].maintype= VTYPE;
9872: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9873: ncovv++; /* Varying variables without age */
9874: TvarV[ncovv]=Tvar[k];
9875: TvarVind[ncovv]=k;
9876: }
1.227 brouard 9877: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9878: if(Tvard[k1][2] <=ncovcol){
9879: Fixed[k]= 0; /* or 2 ?*/
9880: Dummy[k]= 1;
9881: modell[k].maintype= FTYPE;
9882: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9883: ncovf++; /* Fixed variables without age */
9884: TvarF[ncovf]=Tvar[k];
9885: TvarFind[ncovf]=k;
9886: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9887: Fixed[k]= 1;
9888: Dummy[k]= 1;
9889: modell[k].maintype= VTYPE;
9890: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9891: ncovv++; /* Varying variables without age */
9892: TvarV[ncovv]=Tvar[k];
9893: TvarVind[ncovv]=k;
9894: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9895: Fixed[k]= 1;
9896: Dummy[k]= 1;
9897: modell[k].maintype= VTYPE;
9898: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9899: ncovv++; /* Varying variables without age */
9900: TvarV[ncovv]=Tvar[k];
9901: TvarVind[ncovv]=k;
9902: ncovv++; /* Varying variables without age */
9903: TvarV[ncovv]=Tvar[k];
9904: TvarVind[ncovv]=k;
9905: }
1.227 brouard 9906: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9907: if(Tvard[k1][2] <=ncovcol){
9908: Fixed[k]= 1;
9909: Dummy[k]= 1;
9910: modell[k].maintype= VTYPE;
9911: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9912: ncovv++; /* Varying variables without age */
9913: TvarV[ncovv]=Tvar[k];
9914: TvarVind[ncovv]=k;
9915: }else if(Tvard[k1][2] <=ncovcol+nqv){
9916: Fixed[k]= 1;
9917: Dummy[k]= 1;
9918: modell[k].maintype= VTYPE;
9919: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9920: ncovv++; /* Varying variables without age */
9921: TvarV[ncovv]=Tvar[k];
9922: TvarVind[ncovv]=k;
9923: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9924: Fixed[k]= 1;
9925: Dummy[k]= 0;
9926: modell[k].maintype= VTYPE;
9927: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9928: ncovv++; /* Varying variables without age */
9929: TvarV[ncovv]=Tvar[k];
9930: TvarVind[ncovv]=k;
9931: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9932: Fixed[k]= 1;
9933: Dummy[k]= 1;
9934: modell[k].maintype= VTYPE;
9935: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9936: ncovv++; /* Varying variables without age */
9937: TvarV[ncovv]=Tvar[k];
9938: TvarVind[ncovv]=k;
9939: }
1.227 brouard 9940: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9941: if(Tvard[k1][2] <=ncovcol){
9942: Fixed[k]= 1;
9943: Dummy[k]= 1;
9944: modell[k].maintype= VTYPE;
9945: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9946: ncovv++; /* Varying variables without age */
9947: TvarV[ncovv]=Tvar[k];
9948: TvarVind[ncovv]=k;
9949: }else if(Tvard[k1][2] <=ncovcol+nqv){
9950: Fixed[k]= 1;
9951: Dummy[k]= 1;
9952: modell[k].maintype= VTYPE;
9953: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9954: ncovv++; /* Varying variables without age */
9955: TvarV[ncovv]=Tvar[k];
9956: TvarVind[ncovv]=k;
9957: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9958: Fixed[k]= 1;
9959: Dummy[k]= 1;
9960: modell[k].maintype= VTYPE;
9961: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9962: ncovv++; /* Varying variables without age */
9963: TvarV[ncovv]=Tvar[k];
9964: TvarVind[ncovv]=k;
9965: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9966: Fixed[k]= 1;
9967: Dummy[k]= 1;
9968: modell[k].maintype= VTYPE;
9969: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9970: ncovv++; /* Varying variables without age */
9971: TvarV[ncovv]=Tvar[k];
9972: TvarVind[ncovv]=k;
9973: }
1.227 brouard 9974: }else{
1.240 brouard 9975: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9976: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9977: } /*end k1*/
1.225 brouard 9978: }else{
1.226 brouard 9979: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9980: 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 9981: }
1.227 brouard 9982: 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 9983: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9984: 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]);
9985: }
9986: /* Searching for doublons in the model */
9987: for(k1=1; k1<= cptcovt;k1++){
9988: for(k2=1; k2 <k1;k2++){
1.285 brouard 9989: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
9990: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 9991: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9992: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 9993: 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]);
9994: 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 9995: return(1);
9996: }
9997: }else if (Typevar[k1] ==2){
9998: k3=Tposprod[k1];
9999: k4=Tposprod[k2];
10000: 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])) ){
10001: 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]]);
10002: 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);
10003: return(1);
10004: }
10005: }
1.227 brouard 10006: }
10007: }
1.225 brouard 10008: }
10009: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10010: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10011: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10012: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10013: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10014: /*endread:*/
1.225 brouard 10015: printf("Exiting decodemodel: ");
10016: return (1);
1.136 brouard 10017: }
10018:
1.169 brouard 10019: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10020: {/* Check ages at death */
1.136 brouard 10021: int i, m;
1.218 brouard 10022: int firstone=0;
10023:
1.136 brouard 10024: for (i=1; i<=imx; i++) {
10025: for(m=2; (m<= maxwav); m++) {
10026: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10027: anint[m][i]=9999;
1.216 brouard 10028: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10029: s[m][i]=-1;
1.136 brouard 10030: }
10031: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10032: *nberr = *nberr + 1;
1.218 brouard 10033: if(firstone == 0){
10034: firstone=1;
1.260 brouard 10035: 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 10036: }
1.262 brouard 10037: 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 10038: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10039: }
10040: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10041: (*nberr)++;
1.259 brouard 10042: 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 10043: 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 10044: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10045: }
10046: }
10047: }
10048:
10049: for (i=1; i<=imx; i++) {
10050: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10051: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10052: 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 10053: if (s[m][i] >= nlstate+1) {
1.169 brouard 10054: if(agedc[i]>0){
10055: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10056: agev[m][i]=agedc[i];
1.214 brouard 10057: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10058: }else {
1.136 brouard 10059: if ((int)andc[i]!=9999){
10060: nbwarn++;
10061: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10062: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10063: agev[m][i]=-1;
10064: }
10065: }
1.169 brouard 10066: } /* agedc > 0 */
1.214 brouard 10067: } /* end if */
1.136 brouard 10068: else if(s[m][i] !=9){ /* Standard case, age in fractional
10069: years but with the precision of a month */
10070: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10071: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10072: agev[m][i]=1;
10073: else if(agev[m][i] < *agemin){
10074: *agemin=agev[m][i];
10075: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10076: }
10077: else if(agev[m][i] >*agemax){
10078: *agemax=agev[m][i];
1.156 brouard 10079: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10080: }
10081: /*agev[m][i]=anint[m][i]-annais[i];*/
10082: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10083: } /* en if 9*/
1.136 brouard 10084: else { /* =9 */
1.214 brouard 10085: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10086: agev[m][i]=1;
10087: s[m][i]=-1;
10088: }
10089: }
1.214 brouard 10090: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10091: agev[m][i]=1;
1.214 brouard 10092: else{
10093: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10094: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10095: agev[m][i]=0;
10096: }
10097: } /* End for lastpass */
10098: }
1.136 brouard 10099:
10100: for (i=1; i<=imx; i++) {
10101: for(m=firstpass; (m<=lastpass); m++){
10102: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10103: (*nberr)++;
1.136 brouard 10104: 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);
10105: 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);
10106: return 1;
10107: }
10108: }
10109: }
10110:
10111: /*for (i=1; i<=imx; i++){
10112: for (m=firstpass; (m<lastpass); m++){
10113: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10114: }
10115:
10116: }*/
10117:
10118:
1.139 brouard 10119: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10120: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10121:
10122: return (0);
1.164 brouard 10123: /* endread:*/
1.136 brouard 10124: printf("Exiting calandcheckages: ");
10125: return (1);
10126: }
10127:
1.172 brouard 10128: #if defined(_MSC_VER)
10129: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10130: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10131: //#include "stdafx.h"
10132: //#include <stdio.h>
10133: //#include <tchar.h>
10134: //#include <windows.h>
10135: //#include <iostream>
10136: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10137:
10138: LPFN_ISWOW64PROCESS fnIsWow64Process;
10139:
10140: BOOL IsWow64()
10141: {
10142: BOOL bIsWow64 = FALSE;
10143:
10144: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10145: // (HANDLE, PBOOL);
10146:
10147: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10148:
10149: HMODULE module = GetModuleHandle(_T("kernel32"));
10150: const char funcName[] = "IsWow64Process";
10151: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10152: GetProcAddress(module, funcName);
10153:
10154: if (NULL != fnIsWow64Process)
10155: {
10156: if (!fnIsWow64Process(GetCurrentProcess(),
10157: &bIsWow64))
10158: //throw std::exception("Unknown error");
10159: printf("Unknown error\n");
10160: }
10161: return bIsWow64 != FALSE;
10162: }
10163: #endif
1.177 brouard 10164:
1.191 brouard 10165: void syscompilerinfo(int logged)
1.167 brouard 10166: {
10167: /* #include "syscompilerinfo.h"*/
1.185 brouard 10168: /* command line Intel compiler 32bit windows, XP compatible:*/
10169: /* /GS /W3 /Gy
10170: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10171: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10172: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10173: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10174: */
10175: /* 64 bits */
1.185 brouard 10176: /*
10177: /GS /W3 /Gy
10178: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10179: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10180: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10181: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10182: /* Optimization are useless and O3 is slower than O2 */
10183: /*
10184: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10185: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10186: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10187: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10188: */
1.186 brouard 10189: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10190: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10191: /PDB:"visual studio
10192: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10193: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10194: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10195: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10196: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10197: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10198: uiAccess='false'"
10199: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10200: /NOLOGO /TLBID:1
10201: */
1.177 brouard 10202: #if defined __INTEL_COMPILER
1.178 brouard 10203: #if defined(__GNUC__)
10204: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10205: #endif
1.177 brouard 10206: #elif defined(__GNUC__)
1.179 brouard 10207: #ifndef __APPLE__
1.174 brouard 10208: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10209: #endif
1.177 brouard 10210: struct utsname sysInfo;
1.178 brouard 10211: int cross = CROSS;
10212: if (cross){
10213: printf("Cross-");
1.191 brouard 10214: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10215: }
1.174 brouard 10216: #endif
10217:
1.171 brouard 10218: #include <stdint.h>
1.178 brouard 10219:
1.191 brouard 10220: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10221: #if defined(__clang__)
1.191 brouard 10222: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10223: #endif
10224: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10225: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10226: #endif
10227: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10228: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10229: #endif
10230: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10231: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10232: #endif
10233: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10234: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10235: #endif
10236: #if defined(_MSC_VER)
1.191 brouard 10237: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10238: #endif
10239: #if defined(__PGI)
1.191 brouard 10240: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10241: #endif
10242: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10243: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10244: #endif
1.191 brouard 10245: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10246:
1.167 brouard 10247: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10248: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10249: // Windows (x64 and x86)
1.191 brouard 10250: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10251: #elif __unix__ // all unices, not all compilers
10252: // Unix
1.191 brouard 10253: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10254: #elif __linux__
10255: // linux
1.191 brouard 10256: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10257: #elif __APPLE__
1.174 brouard 10258: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10259: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10260: #endif
10261:
10262: /* __MINGW32__ */
10263: /* __CYGWIN__ */
10264: /* __MINGW64__ */
10265: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10266: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10267: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10268: /* _WIN64 // Defined for applications for Win64. */
10269: /* _M_X64 // Defined for compilations that target x64 processors. */
10270: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10271:
1.167 brouard 10272: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10273: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10274: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10275: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10276: #else
1.191 brouard 10277: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10278: #endif
10279:
1.169 brouard 10280: #if defined(__GNUC__)
10281: # if defined(__GNUC_PATCHLEVEL__)
10282: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10283: + __GNUC_MINOR__ * 100 \
10284: + __GNUC_PATCHLEVEL__)
10285: # else
10286: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10287: + __GNUC_MINOR__ * 100)
10288: # endif
1.174 brouard 10289: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10290: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10291:
10292: if (uname(&sysInfo) != -1) {
10293: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10294: 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 10295: }
10296: else
10297: perror("uname() error");
1.179 brouard 10298: //#ifndef __INTEL_COMPILER
10299: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10300: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10301: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10302: #endif
1.169 brouard 10303: #endif
1.172 brouard 10304:
1.286 ! brouard 10305: // void main ()
1.172 brouard 10306: // {
1.169 brouard 10307: #if defined(_MSC_VER)
1.174 brouard 10308: if (IsWow64()){
1.191 brouard 10309: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10310: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10311: }
10312: else{
1.191 brouard 10313: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10314: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10315: }
1.172 brouard 10316: // printf("\nPress Enter to continue...");
10317: // getchar();
10318: // }
10319:
1.169 brouard 10320: #endif
10321:
1.167 brouard 10322:
1.219 brouard 10323: }
1.136 brouard 10324:
1.219 brouard 10325: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10326: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10327: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10328: /* double ftolpl = 1.e-10; */
1.180 brouard 10329: double age, agebase, agelim;
1.203 brouard 10330: double tot;
1.180 brouard 10331:
1.202 brouard 10332: strcpy(filerespl,"PL_");
10333: strcat(filerespl,fileresu);
10334: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10335: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10336: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10337: }
1.227 brouard 10338: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10339: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10340: pstamp(ficrespl);
1.203 brouard 10341: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10342: fprintf(ficrespl,"#Age ");
10343: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10344: fprintf(ficrespl,"\n");
1.180 brouard 10345:
1.219 brouard 10346: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10347:
1.219 brouard 10348: agebase=ageminpar;
10349: agelim=agemaxpar;
1.180 brouard 10350:
1.227 brouard 10351: /* i1=pow(2,ncoveff); */
1.234 brouard 10352: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10353: if (cptcovn < 1){i1=1;}
1.180 brouard 10354:
1.238 brouard 10355: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10356: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10357: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10358: continue;
1.235 brouard 10359:
1.238 brouard 10360: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10361: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10362: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10363: /* k=k+1; */
10364: /* to clean */
10365: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10366: fprintf(ficrespl,"#******");
10367: printf("#******");
10368: fprintf(ficlog,"#******");
10369: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10370: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10371: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10372: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10373: }
10374: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10375: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10376: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10377: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10378: }
10379: fprintf(ficrespl,"******\n");
10380: printf("******\n");
10381: fprintf(ficlog,"******\n");
10382: if(invalidvarcomb[k]){
10383: printf("\nCombination (%d) ignored because no case \n",k);
10384: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10385: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10386: continue;
10387: }
1.219 brouard 10388:
1.238 brouard 10389: fprintf(ficrespl,"#Age ");
10390: for(j=1;j<=cptcoveff;j++) {
10391: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10392: }
10393: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10394: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10395:
1.238 brouard 10396: for (age=agebase; age<=agelim; age++){
10397: /* for (age=agebase; age<=agebase; age++){ */
10398: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10399: fprintf(ficrespl,"%.0f ",age );
10400: for(j=1;j<=cptcoveff;j++)
10401: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10402: tot=0.;
10403: for(i=1; i<=nlstate;i++){
10404: tot += prlim[i][i];
10405: fprintf(ficrespl," %.5f", prlim[i][i]);
10406: }
10407: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10408: } /* Age */
10409: /* was end of cptcod */
10410: } /* cptcov */
10411: } /* nres */
1.219 brouard 10412: return 0;
1.180 brouard 10413: }
10414:
1.218 brouard 10415: 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){
10416: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10417:
10418: /* Computes the back prevalence limit for any combination of covariate values
10419: * at any age between ageminpar and agemaxpar
10420: */
1.235 brouard 10421: int i, j, k, i1, nres=0 ;
1.217 brouard 10422: /* double ftolpl = 1.e-10; */
10423: double age, agebase, agelim;
10424: double tot;
1.218 brouard 10425: /* double ***mobaverage; */
10426: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10427:
10428: strcpy(fileresplb,"PLB_");
10429: strcat(fileresplb,fileresu);
10430: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10431: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10432: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10433: }
10434: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10435: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10436: pstamp(ficresplb);
10437: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10438: fprintf(ficresplb,"#Age ");
10439: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10440: fprintf(ficresplb,"\n");
10441:
1.218 brouard 10442:
10443: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10444:
10445: agebase=ageminpar;
10446: agelim=agemaxpar;
10447:
10448:
1.227 brouard 10449: i1=pow(2,cptcoveff);
1.218 brouard 10450: if (cptcovn < 1){i1=1;}
1.227 brouard 10451:
1.238 brouard 10452: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10453: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10454: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10455: continue;
10456: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10457: fprintf(ficresplb,"#******");
10458: printf("#******");
10459: fprintf(ficlog,"#******");
10460: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10461: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10462: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10463: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10464: }
10465: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10466: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10467: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10468: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10469: }
10470: fprintf(ficresplb,"******\n");
10471: printf("******\n");
10472: fprintf(ficlog,"******\n");
10473: if(invalidvarcomb[k]){
10474: printf("\nCombination (%d) ignored because no cases \n",k);
10475: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10476: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10477: continue;
10478: }
1.218 brouard 10479:
1.238 brouard 10480: fprintf(ficresplb,"#Age ");
10481: for(j=1;j<=cptcoveff;j++) {
10482: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10483: }
10484: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10485: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10486:
10487:
1.238 brouard 10488: for (age=agebase; age<=agelim; age++){
10489: /* for (age=agebase; age<=agebase; age++){ */
10490: if(mobilavproj > 0){
10491: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10492: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10493: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10494: }else if (mobilavproj == 0){
10495: 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);
10496: 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);
10497: exit(1);
10498: }else{
10499: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10500: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10501: /* printf("TOTOT\n"); */
10502: /* exit(1); */
1.238 brouard 10503: }
10504: fprintf(ficresplb,"%.0f ",age );
10505: for(j=1;j<=cptcoveff;j++)
10506: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10507: tot=0.;
10508: for(i=1; i<=nlstate;i++){
10509: tot += bprlim[i][i];
10510: fprintf(ficresplb," %.5f", bprlim[i][i]);
10511: }
10512: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10513: } /* Age */
10514: /* was end of cptcod */
1.255 brouard 10515: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10516: } /* end of any combination */
10517: } /* end of nres */
1.218 brouard 10518: /* hBijx(p, bage, fage); */
10519: /* fclose(ficrespijb); */
10520:
10521: return 0;
1.217 brouard 10522: }
1.218 brouard 10523:
1.180 brouard 10524: int hPijx(double *p, int bage, int fage){
10525: /*------------- h Pij x at various ages ------------*/
10526:
10527: int stepsize;
10528: int agelim;
10529: int hstepm;
10530: int nhstepm;
1.235 brouard 10531: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10532:
10533: double agedeb;
10534: double ***p3mat;
10535:
1.201 brouard 10536: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10537: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10538: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10539: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10540: }
10541: printf("Computing pij: result on file '%s' \n", filerespij);
10542: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10543:
10544: stepsize=(int) (stepm+YEARM-1)/YEARM;
10545: /*if (stepm<=24) stepsize=2;*/
10546:
10547: agelim=AGESUP;
10548: hstepm=stepsize*YEARM; /* Every year of age */
10549: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10550:
1.180 brouard 10551: /* hstepm=1; aff par mois*/
10552: pstamp(ficrespij);
10553: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10554: i1= pow(2,cptcoveff);
1.218 brouard 10555: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10556: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10557: /* k=k+1; */
1.235 brouard 10558: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10559: for(k=1; k<=i1;k++){
1.253 brouard 10560: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10561: continue;
1.183 brouard 10562: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10563: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10564: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10565: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10566: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10567: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10568: }
1.183 brouard 10569: fprintf(ficrespij,"******\n");
10570:
10571: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10572: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10573: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10574:
10575: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10576:
1.183 brouard 10577: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10578: oldm=oldms;savm=savms;
1.235 brouard 10579: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10580: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10581: for(i=1; i<=nlstate;i++)
10582: for(j=1; j<=nlstate+ndeath;j++)
10583: fprintf(ficrespij," %1d-%1d",i,j);
10584: fprintf(ficrespij,"\n");
10585: for (h=0; h<=nhstepm; h++){
10586: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10587: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10588: for(i=1; i<=nlstate;i++)
10589: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10590: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10591: fprintf(ficrespij,"\n");
10592: }
1.183 brouard 10593: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10594: fprintf(ficrespij,"\n");
10595: }
1.180 brouard 10596: /*}*/
10597: }
1.218 brouard 10598: return 0;
1.180 brouard 10599: }
1.218 brouard 10600:
10601: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10602: /*------------- h Bij x at various ages ------------*/
10603:
10604: int stepsize;
1.218 brouard 10605: /* int agelim; */
10606: int ageminl;
1.217 brouard 10607: int hstepm;
10608: int nhstepm;
1.238 brouard 10609: int h, i, i1, j, k, nres;
1.218 brouard 10610:
1.217 brouard 10611: double agedeb;
10612: double ***p3mat;
1.218 brouard 10613:
10614: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10615: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10616: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10617: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10618: }
10619: printf("Computing pij back: result on file '%s' \n", filerespijb);
10620: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10621:
10622: stepsize=(int) (stepm+YEARM-1)/YEARM;
10623: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10624:
1.218 brouard 10625: /* agelim=AGESUP; */
10626: ageminl=30;
10627: hstepm=stepsize*YEARM; /* Every year of age */
10628: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10629:
10630: /* hstepm=1; aff par mois*/
10631: pstamp(ficrespijb);
1.255 brouard 10632: 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 10633: i1= pow(2,cptcoveff);
1.218 brouard 10634: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10635: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10636: /* k=k+1; */
1.238 brouard 10637: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10638: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10639: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10640: continue;
10641: fprintf(ficrespijb,"\n#****** ");
10642: for(j=1;j<=cptcoveff;j++)
10643: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10644: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10645: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10646: }
10647: fprintf(ficrespijb,"******\n");
1.264 brouard 10648: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10649: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10650: continue;
10651: }
10652:
10653: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10654: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10655: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10656: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10657: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10658:
10659: /* nhstepm=nhstepm*YEARM; aff par mois*/
10660:
1.266 brouard 10661: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10662: /* and memory limitations if stepm is small */
10663:
1.238 brouard 10664: /* oldm=oldms;savm=savms; */
10665: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10666: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10667: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10668: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10669: for(i=1; i<=nlstate;i++)
10670: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10671: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10672: fprintf(ficrespijb,"\n");
1.238 brouard 10673: for (h=0; h<=nhstepm; h++){
10674: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10675: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10676: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10677: for(i=1; i<=nlstate;i++)
10678: for(j=1; j<=nlstate+ndeath;j++)
10679: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10680: fprintf(ficrespijb,"\n");
10681: }
10682: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10683: fprintf(ficrespijb,"\n");
10684: } /* end age deb */
10685: } /* end combination */
10686: } /* end nres */
1.218 brouard 10687: return 0;
10688: } /* hBijx */
1.217 brouard 10689:
1.180 brouard 10690:
1.136 brouard 10691: /***********************************************/
10692: /**************** Main Program *****************/
10693: /***********************************************/
10694:
10695: int main(int argc, char *argv[])
10696: {
10697: #ifdef GSL
10698: const gsl_multimin_fminimizer_type *T;
10699: size_t iteri = 0, it;
10700: int rval = GSL_CONTINUE;
10701: int status = GSL_SUCCESS;
10702: double ssval;
10703: #endif
10704: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10705: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10706: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10707: int jj, ll, li, lj, lk;
1.136 brouard 10708: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10709: int num_filled;
1.136 brouard 10710: int itimes;
10711: int NDIM=2;
10712: int vpopbased=0;
1.235 brouard 10713: int nres=0;
1.258 brouard 10714: int endishere=0;
1.277 brouard 10715: int noffset=0;
1.274 brouard 10716: int ncurrv=0; /* Temporary variable */
10717:
1.164 brouard 10718: char ca[32], cb[32];
1.136 brouard 10719: /* FILE *fichtm; *//* Html File */
10720: /* FILE *ficgp;*/ /*Gnuplot File */
10721: struct stat info;
1.191 brouard 10722: double agedeb=0.;
1.194 brouard 10723:
10724: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10725: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10726:
1.165 brouard 10727: double fret;
1.191 brouard 10728: double dum=0.; /* Dummy variable */
1.136 brouard 10729: double ***p3mat;
1.218 brouard 10730: /* double ***mobaverage; */
1.164 brouard 10731:
10732: char line[MAXLINE];
1.197 brouard 10733: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10734:
1.234 brouard 10735: char modeltemp[MAXLINE];
1.230 brouard 10736: char resultline[MAXLINE];
10737:
1.136 brouard 10738: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10739: char *tok, *val; /* pathtot */
1.136 brouard 10740: int firstobs=1, lastobs=10;
1.195 brouard 10741: int c, h , cpt, c2;
1.191 brouard 10742: int jl=0;
10743: int i1, j1, jk, stepsize=0;
1.194 brouard 10744: int count=0;
10745:
1.164 brouard 10746: int *tab;
1.136 brouard 10747: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10748: int backcast=0;
1.136 brouard 10749: int mobilav=0,popforecast=0;
1.191 brouard 10750: int hstepm=0, nhstepm=0;
1.136 brouard 10751: int agemortsup;
10752: float sumlpop=0.;
10753: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10754: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10755:
1.191 brouard 10756: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10757: double ftolpl=FTOL;
10758: double **prlim;
1.217 brouard 10759: double **bprlim;
1.136 brouard 10760: double ***param; /* Matrix of parameters */
1.251 brouard 10761: double ***paramstart; /* Matrix of starting parameter values */
10762: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10763: double **matcov; /* Matrix of covariance */
1.203 brouard 10764: double **hess; /* Hessian matrix */
1.136 brouard 10765: double ***delti3; /* Scale */
10766: double *delti; /* Scale */
10767: double ***eij, ***vareij;
10768: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10769:
1.136 brouard 10770: double *epj, vepp;
1.164 brouard 10771:
1.273 brouard 10772: double dateprev1, dateprev2;
10773: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10774: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10775:
1.136 brouard 10776: double **ximort;
1.145 brouard 10777: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10778: int *dcwave;
10779:
1.164 brouard 10780: char z[1]="c";
1.136 brouard 10781:
10782: /*char *strt;*/
10783: char strtend[80];
1.126 brouard 10784:
1.164 brouard 10785:
1.126 brouard 10786: /* setlocale (LC_ALL, ""); */
10787: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10788: /* textdomain (PACKAGE); */
10789: /* setlocale (LC_CTYPE, ""); */
10790: /* setlocale (LC_MESSAGES, ""); */
10791:
10792: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10793: rstart_time = time(NULL);
10794: /* (void) gettimeofday(&start_time,&tzp);*/
10795: start_time = *localtime(&rstart_time);
1.126 brouard 10796: curr_time=start_time;
1.157 brouard 10797: /*tml = *localtime(&start_time.tm_sec);*/
10798: /* strcpy(strstart,asctime(&tml)); */
10799: strcpy(strstart,asctime(&start_time));
1.126 brouard 10800:
10801: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10802: /* tp.tm_sec = tp.tm_sec +86400; */
10803: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10804: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10805: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10806: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10807: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10808: /* strt=asctime(&tmg); */
10809: /* printf("Time(after) =%s",strstart); */
10810: /* (void) time (&time_value);
10811: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10812: * tm = *localtime(&time_value);
10813: * strstart=asctime(&tm);
10814: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10815: */
10816:
10817: nberr=0; /* Number of errors and warnings */
10818: nbwarn=0;
1.184 brouard 10819: #ifdef WIN32
10820: _getcwd(pathcd, size);
10821: #else
1.126 brouard 10822: getcwd(pathcd, size);
1.184 brouard 10823: #endif
1.191 brouard 10824: syscompilerinfo(0);
1.196 brouard 10825: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10826: if(argc <=1){
10827: printf("\nEnter the parameter file name: ");
1.205 brouard 10828: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10829: printf("ERROR Empty parameter file name\n");
10830: goto end;
10831: }
1.126 brouard 10832: i=strlen(pathr);
10833: if(pathr[i-1]=='\n')
10834: pathr[i-1]='\0';
1.156 brouard 10835: i=strlen(pathr);
1.205 brouard 10836: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10837: pathr[i-1]='\0';
1.205 brouard 10838: }
10839: i=strlen(pathr);
10840: if( i==0 ){
10841: printf("ERROR Empty parameter file name\n");
10842: goto end;
10843: }
10844: for (tok = pathr; tok != NULL; ){
1.126 brouard 10845: printf("Pathr |%s|\n",pathr);
10846: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10847: printf("val= |%s| pathr=%s\n",val,pathr);
10848: strcpy (pathtot, val);
10849: if(pathr[0] == '\0') break; /* Dirty */
10850: }
10851: }
1.281 brouard 10852: else if (argc<=2){
10853: strcpy(pathtot,argv[1]);
10854: }
1.126 brouard 10855: else{
10856: strcpy(pathtot,argv[1]);
1.281 brouard 10857: strcpy(z,argv[2]);
10858: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 10859: }
10860: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10861: /*cygwin_split_path(pathtot,path,optionfile);
10862: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10863: /* cutv(path,optionfile,pathtot,'\\');*/
10864:
10865: /* Split argv[0], imach program to get pathimach */
10866: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10867: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10868: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10869: /* strcpy(pathimach,argv[0]); */
10870: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10871: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10872: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10873: #ifdef WIN32
10874: _chdir(path); /* Can be a relative path */
10875: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10876: #else
1.126 brouard 10877: chdir(path); /* Can be a relative path */
1.184 brouard 10878: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10879: #endif
10880: printf("Current directory %s!\n",pathcd);
1.126 brouard 10881: strcpy(command,"mkdir ");
10882: strcat(command,optionfilefiname);
10883: if((outcmd=system(command)) != 0){
1.169 brouard 10884: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10885: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10886: /* fclose(ficlog); */
10887: /* exit(1); */
10888: }
10889: /* if((imk=mkdir(optionfilefiname))<0){ */
10890: /* perror("mkdir"); */
10891: /* } */
10892:
10893: /*-------- arguments in the command line --------*/
10894:
1.186 brouard 10895: /* Main Log file */
1.126 brouard 10896: strcat(filelog, optionfilefiname);
10897: strcat(filelog,".log"); /* */
10898: if((ficlog=fopen(filelog,"w"))==NULL) {
10899: printf("Problem with logfile %s\n",filelog);
10900: goto end;
10901: }
10902: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10903: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10904: fprintf(ficlog,"\nEnter the parameter file name: \n");
10905: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10906: path=%s \n\
10907: optionfile=%s\n\
10908: optionfilext=%s\n\
1.156 brouard 10909: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10910:
1.197 brouard 10911: syscompilerinfo(1);
1.167 brouard 10912:
1.126 brouard 10913: printf("Local time (at start):%s",strstart);
10914: fprintf(ficlog,"Local time (at start): %s",strstart);
10915: fflush(ficlog);
10916: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10917: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10918:
10919: /* */
10920: strcpy(fileres,"r");
10921: strcat(fileres, optionfilefiname);
1.201 brouard 10922: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10923: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10924: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10925:
1.186 brouard 10926: /* Main ---------arguments file --------*/
1.126 brouard 10927:
10928: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10929: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10930: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10931: fflush(ficlog);
1.149 brouard 10932: /* goto end; */
10933: exit(70);
1.126 brouard 10934: }
10935:
10936: strcpy(filereso,"o");
1.201 brouard 10937: strcat(filereso,fileresu);
1.126 brouard 10938: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10939: printf("Problem with Output resultfile: %s\n", filereso);
10940: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10941: fflush(ficlog);
10942: goto end;
10943: }
1.278 brouard 10944: /*-------- Rewriting parameter file ----------*/
10945: strcpy(rfileres,"r"); /* "Rparameterfile */
10946: strcat(rfileres,optionfilefiname); /* Parameter file first name */
10947: strcat(rfileres,"."); /* */
10948: strcat(rfileres,optionfilext); /* Other files have txt extension */
10949: if((ficres =fopen(rfileres,"w"))==NULL) {
10950: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10951: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
10952: fflush(ficlog);
10953: goto end;
10954: }
10955: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 10956:
1.278 brouard 10957:
1.126 brouard 10958: /* Reads comments: lines beginning with '#' */
10959: numlinepar=0;
1.277 brouard 10960: /* Is it a BOM UTF-8 Windows file? */
10961: /* First parameter line */
1.197 brouard 10962: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 10963: noffset=0;
10964: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10965: {
10966: noffset=noffset+3;
10967: printf("# File is an UTF8 Bom.\n"); // 0xBF
10968: }
10969: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
10970: {
10971: noffset=noffset+2;
10972: printf("# File is an UTF16BE BOM file\n");
10973: }
10974: else if( line[0] == 0 && line[1] == 0)
10975: {
10976: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10977: noffset=noffset+4;
10978: printf("# File is an UTF16BE BOM file\n");
10979: }
10980: } else{
10981: ;/*printf(" Not a BOM file\n");*/
10982: }
10983:
1.197 brouard 10984: /* If line starts with a # it is a comment */
1.277 brouard 10985: if (line[noffset] == '#') {
1.197 brouard 10986: numlinepar++;
10987: fputs(line,stdout);
10988: fputs(line,ficparo);
1.278 brouard 10989: fputs(line,ficres);
1.197 brouard 10990: fputs(line,ficlog);
10991: continue;
10992: }else
10993: break;
10994: }
10995: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10996: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10997: if (num_filled != 5) {
10998: printf("Should be 5 parameters\n");
1.283 brouard 10999: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11000: }
1.126 brouard 11001: numlinepar++;
1.197 brouard 11002: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11003: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11004: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11005: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11006: }
11007: /* Second parameter line */
11008: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11009: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11010: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11011: if (line[0] == '#') {
11012: numlinepar++;
1.283 brouard 11013: printf("%s",line);
11014: fprintf(ficres,"%s",line);
11015: fprintf(ficparo,"%s",line);
11016: fprintf(ficlog,"%s",line);
1.197 brouard 11017: continue;
11018: }else
11019: break;
11020: }
1.223 brouard 11021: 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", \
11022: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11023: if (num_filled != 11) {
11024: 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 11025: printf("but line=%s\n",line);
1.283 brouard 11026: 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");
11027: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11028: }
1.286 ! brouard 11029: if( lastpass > maxwav){
! 11030: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
! 11031: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
! 11032: fflush(ficlog);
! 11033: goto end;
! 11034: }
! 11035: 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 11036: 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 11037: 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 11038: 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 11039: }
1.203 brouard 11040: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11041: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11042: /* Third parameter line */
11043: while(fgets(line, MAXLINE, ficpar)) {
11044: /* If line starts with a # it is a comment */
11045: if (line[0] == '#') {
11046: numlinepar++;
1.283 brouard 11047: printf("%s",line);
11048: fprintf(ficres,"%s",line);
11049: fprintf(ficparo,"%s",line);
11050: fprintf(ficlog,"%s",line);
1.197 brouard 11051: continue;
11052: }else
11053: break;
11054: }
1.201 brouard 11055: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11056: if (num_filled != 1){
11057: printf("ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
11058: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
1.197 brouard 11059: model[0]='\0';
11060: goto end;
11061: }
11062: else{
11063: if (model[0]=='+'){
11064: for(i=1; i<=strlen(model);i++)
11065: modeltemp[i-1]=model[i];
1.201 brouard 11066: strcpy(model,modeltemp);
1.197 brouard 11067: }
11068: }
1.199 brouard 11069: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11070: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11071: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11072: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11073: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11074: }
11075: /* 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); */
11076: /* numlinepar=numlinepar+3; /\* In general *\/ */
11077: /* 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 11078: /* 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); */
11079: /* 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 11080: fflush(ficlog);
1.190 brouard 11081: /* if(model[0]=='#'|| model[0]== '\0'){ */
11082: if(model[0]=='#'){
1.279 brouard 11083: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11084: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11085: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11086: if(mle != -1){
1.279 brouard 11087: 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 11088: exit(1);
11089: }
11090: }
1.126 brouard 11091: while((c=getc(ficpar))=='#' && c!= EOF){
11092: ungetc(c,ficpar);
11093: fgets(line, MAXLINE, ficpar);
11094: numlinepar++;
1.195 brouard 11095: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11096: z[0]=line[1];
11097: }
11098: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11099: fputs(line, stdout);
11100: //puts(line);
1.126 brouard 11101: fputs(line,ficparo);
11102: fputs(line,ficlog);
11103: }
11104: ungetc(c,ficpar);
11105:
11106:
1.145 brouard 11107: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 11108: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
11109: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
11110: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11111: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11112: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11113: v1+v2*age+v2*v3 makes cptcovn = 3
11114: */
11115: if (strlen(model)>1)
1.187 brouard 11116: 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 11117: else
1.187 brouard 11118: ncovmodel=2; /* Constant and age */
1.133 brouard 11119: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11120: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11121: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11122: 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);
11123: 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);
11124: fflush(stdout);
11125: fclose (ficlog);
11126: goto end;
11127: }
1.126 brouard 11128: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11129: delti=delti3[1][1];
11130: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11131: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11132: /* We could also provide initial parameters values giving by simple logistic regression
11133: * only one way, that is without matrix product. We will have nlstate maximizations */
11134: /* for(i=1;i<nlstate;i++){ */
11135: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11136: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11137: /* } */
1.126 brouard 11138: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11139: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11140: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11141: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11142: fclose (ficparo);
11143: fclose (ficlog);
11144: goto end;
11145: exit(0);
1.220 brouard 11146: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11147: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11148: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11149: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11150: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11151: matcov=matrix(1,npar,1,npar);
1.203 brouard 11152: hess=matrix(1,npar,1,npar);
1.220 brouard 11153: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11154: /* Read guessed parameters */
1.126 brouard 11155: /* Reads comments: lines beginning with '#' */
11156: while((c=getc(ficpar))=='#' && c!= EOF){
11157: ungetc(c,ficpar);
11158: fgets(line, MAXLINE, ficpar);
11159: numlinepar++;
1.141 brouard 11160: fputs(line,stdout);
1.126 brouard 11161: fputs(line,ficparo);
11162: fputs(line,ficlog);
11163: }
11164: ungetc(c,ficpar);
11165:
11166: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11167: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11168: for(i=1; i <=nlstate; i++){
1.234 brouard 11169: j=0;
1.126 brouard 11170: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11171: if(jj==i) continue;
11172: j++;
11173: fscanf(ficpar,"%1d%1d",&i1,&j1);
11174: if ((i1 != i) || (j1 != jj)){
11175: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11176: It might be a problem of design; if ncovcol and the model are correct\n \
11177: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11178: exit(1);
11179: }
11180: fprintf(ficparo,"%1d%1d",i1,j1);
11181: if(mle==1)
11182: printf("%1d%1d",i,jj);
11183: fprintf(ficlog,"%1d%1d",i,jj);
11184: for(k=1; k<=ncovmodel;k++){
11185: fscanf(ficpar," %lf",¶m[i][j][k]);
11186: if(mle==1){
11187: printf(" %lf",param[i][j][k]);
11188: fprintf(ficlog," %lf",param[i][j][k]);
11189: }
11190: else
11191: fprintf(ficlog," %lf",param[i][j][k]);
11192: fprintf(ficparo," %lf",param[i][j][k]);
11193: }
11194: fscanf(ficpar,"\n");
11195: numlinepar++;
11196: if(mle==1)
11197: printf("\n");
11198: fprintf(ficlog,"\n");
11199: fprintf(ficparo,"\n");
1.126 brouard 11200: }
11201: }
11202: fflush(ficlog);
1.234 brouard 11203:
1.251 brouard 11204: /* Reads parameters values */
1.126 brouard 11205: p=param[1][1];
1.251 brouard 11206: pstart=paramstart[1][1];
1.126 brouard 11207:
11208: /* Reads comments: lines beginning with '#' */
11209: while((c=getc(ficpar))=='#' && c!= EOF){
11210: ungetc(c,ficpar);
11211: fgets(line, MAXLINE, ficpar);
11212: numlinepar++;
1.141 brouard 11213: fputs(line,stdout);
1.126 brouard 11214: fputs(line,ficparo);
11215: fputs(line,ficlog);
11216: }
11217: ungetc(c,ficpar);
11218:
11219: for(i=1; i <=nlstate; i++){
11220: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11221: fscanf(ficpar,"%1d%1d",&i1,&j1);
11222: if ( (i1-i) * (j1-j) != 0){
11223: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11224: exit(1);
11225: }
11226: printf("%1d%1d",i,j);
11227: fprintf(ficparo,"%1d%1d",i1,j1);
11228: fprintf(ficlog,"%1d%1d",i1,j1);
11229: for(k=1; k<=ncovmodel;k++){
11230: fscanf(ficpar,"%le",&delti3[i][j][k]);
11231: printf(" %le",delti3[i][j][k]);
11232: fprintf(ficparo," %le",delti3[i][j][k]);
11233: fprintf(ficlog," %le",delti3[i][j][k]);
11234: }
11235: fscanf(ficpar,"\n");
11236: numlinepar++;
11237: printf("\n");
11238: fprintf(ficparo,"\n");
11239: fprintf(ficlog,"\n");
1.126 brouard 11240: }
11241: }
11242: fflush(ficlog);
1.234 brouard 11243:
1.145 brouard 11244: /* Reads covariance matrix */
1.126 brouard 11245: delti=delti3[1][1];
1.220 brouard 11246:
11247:
1.126 brouard 11248: /* 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 11249:
1.126 brouard 11250: /* Reads comments: lines beginning with '#' */
11251: while((c=getc(ficpar))=='#' && c!= EOF){
11252: ungetc(c,ficpar);
11253: fgets(line, MAXLINE, ficpar);
11254: numlinepar++;
1.141 brouard 11255: fputs(line,stdout);
1.126 brouard 11256: fputs(line,ficparo);
11257: fputs(line,ficlog);
11258: }
11259: ungetc(c,ficpar);
1.220 brouard 11260:
1.126 brouard 11261: matcov=matrix(1,npar,1,npar);
1.203 brouard 11262: hess=matrix(1,npar,1,npar);
1.131 brouard 11263: for(i=1; i <=npar; i++)
11264: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11265:
1.194 brouard 11266: /* Scans npar lines */
1.126 brouard 11267: for(i=1; i <=npar; i++){
1.226 brouard 11268: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11269: if(count != 3){
1.226 brouard 11270: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11271: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11272: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11273: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11274: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11275: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11276: exit(1);
1.220 brouard 11277: }else{
1.226 brouard 11278: if(mle==1)
11279: printf("%1d%1d%d",i1,j1,jk);
11280: }
11281: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11282: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11283: for(j=1; j <=i; j++){
1.226 brouard 11284: fscanf(ficpar," %le",&matcov[i][j]);
11285: if(mle==1){
11286: printf(" %.5le",matcov[i][j]);
11287: }
11288: fprintf(ficlog," %.5le",matcov[i][j]);
11289: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11290: }
11291: fscanf(ficpar,"\n");
11292: numlinepar++;
11293: if(mle==1)
1.220 brouard 11294: printf("\n");
1.126 brouard 11295: fprintf(ficlog,"\n");
11296: fprintf(ficparo,"\n");
11297: }
1.194 brouard 11298: /* End of read covariance matrix npar lines */
1.126 brouard 11299: for(i=1; i <=npar; i++)
11300: for(j=i+1;j<=npar;j++)
1.226 brouard 11301: matcov[i][j]=matcov[j][i];
1.126 brouard 11302:
11303: if(mle==1)
11304: printf("\n");
11305: fprintf(ficlog,"\n");
11306:
11307: fflush(ficlog);
11308:
11309: } /* End of mle != -3 */
1.218 brouard 11310:
1.186 brouard 11311: /* Main data
11312: */
1.126 brouard 11313: n= lastobs;
11314: num=lvector(1,n);
11315: moisnais=vector(1,n);
11316: annais=vector(1,n);
11317: moisdc=vector(1,n);
11318: andc=vector(1,n);
1.220 brouard 11319: weight=vector(1,n);
1.126 brouard 11320: agedc=vector(1,n);
11321: cod=ivector(1,n);
1.220 brouard 11322: for(i=1;i<=n;i++){
1.234 brouard 11323: num[i]=0;
11324: moisnais[i]=0;
11325: annais[i]=0;
11326: moisdc[i]=0;
11327: andc[i]=0;
11328: agedc[i]=0;
11329: cod[i]=0;
11330: weight[i]=1.0; /* Equal weights, 1 by default */
11331: }
1.126 brouard 11332: mint=matrix(1,maxwav,1,n);
11333: anint=matrix(1,maxwav,1,n);
1.131 brouard 11334: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11335: tab=ivector(1,NCOVMAX);
1.144 brouard 11336: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11337: 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 11338:
1.136 brouard 11339: /* Reads data from file datafile */
11340: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11341: goto end;
11342:
11343: /* Calculation of the number of parameters from char model */
1.234 brouard 11344: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11345: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11346: k=3 V4 Tvar[k=3]= 4 (from V4)
11347: k=2 V1 Tvar[k=2]= 1 (from V1)
11348: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11349: */
11350:
11351: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11352: TvarsDind=ivector(1,NCOVMAX); /* */
11353: TvarsD=ivector(1,NCOVMAX); /* */
11354: TvarsQind=ivector(1,NCOVMAX); /* */
11355: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11356: TvarF=ivector(1,NCOVMAX); /* */
11357: TvarFind=ivector(1,NCOVMAX); /* */
11358: TvarV=ivector(1,NCOVMAX); /* */
11359: TvarVind=ivector(1,NCOVMAX); /* */
11360: TvarA=ivector(1,NCOVMAX); /* */
11361: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11362: TvarFD=ivector(1,NCOVMAX); /* */
11363: TvarFDind=ivector(1,NCOVMAX); /* */
11364: TvarFQ=ivector(1,NCOVMAX); /* */
11365: TvarFQind=ivector(1,NCOVMAX); /* */
11366: TvarVD=ivector(1,NCOVMAX); /* */
11367: TvarVDind=ivector(1,NCOVMAX); /* */
11368: TvarVQ=ivector(1,NCOVMAX); /* */
11369: TvarVQind=ivector(1,NCOVMAX); /* */
11370:
1.230 brouard 11371: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11372: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11373: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11374: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11375: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11376: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11377: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11378: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11379: */
11380: /* For model-covariate k tells which data-covariate to use but
11381: because this model-covariate is a construction we invent a new column
11382: ncovcol + k1
11383: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11384: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11385: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11386: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11387: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11388: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11389: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11390: */
1.145 brouard 11391: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11392: 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 11393: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11394: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11395: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11396: 4 covariates (3 plus signs)
11397: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11398: */
1.230 brouard 11399: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11400: * individual dummy, fixed or varying:
11401: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11402: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11403: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11404: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11405: * Tmodelind[1]@9={9,0,3,2,}*/
11406: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11407: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11408: * individual quantitative, fixed or varying:
11409: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11410: * 3, 1, 0, 0, 0, 0, 0, 0},
11411: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11412: /* Main decodemodel */
11413:
1.187 brouard 11414:
1.223 brouard 11415: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11416: goto end;
11417:
1.137 brouard 11418: if((double)(lastobs-imx)/(double)imx > 1.10){
11419: nbwarn++;
11420: 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);
11421: 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);
11422: }
1.136 brouard 11423: /* if(mle==1){*/
1.137 brouard 11424: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11425: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11426: }
11427:
11428: /*-calculation of age at interview from date of interview and age at death -*/
11429: agev=matrix(1,maxwav,1,imx);
11430:
11431: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11432: goto end;
11433:
1.126 brouard 11434:
1.136 brouard 11435: agegomp=(int)agemin;
11436: free_vector(moisnais,1,n);
11437: free_vector(annais,1,n);
1.126 brouard 11438: /* free_matrix(mint,1,maxwav,1,n);
11439: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11440: /* free_vector(moisdc,1,n); */
11441: /* free_vector(andc,1,n); */
1.145 brouard 11442: /* */
11443:
1.126 brouard 11444: wav=ivector(1,imx);
1.214 brouard 11445: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11446: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11447: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11448: 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.*/
11449: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11450: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11451:
11452: /* Concatenates waves */
1.214 brouard 11453: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11454: Death is a valid wave (if date is known).
11455: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11456: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11457: and mw[mi+1][i]. dh depends on stepm.
11458: */
11459:
1.126 brouard 11460: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11461: /* Concatenates waves */
1.145 brouard 11462:
1.215 brouard 11463: free_vector(moisdc,1,n);
11464: free_vector(andc,1,n);
11465:
1.126 brouard 11466: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11467: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11468: ncodemax[1]=1;
1.145 brouard 11469: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11470: cptcoveff=0;
1.220 brouard 11471: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11472: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11473: }
11474:
11475: ncovcombmax=pow(2,cptcoveff);
11476: invalidvarcomb=ivector(1, ncovcombmax);
11477: for(i=1;i<ncovcombmax;i++)
11478: invalidvarcomb[i]=0;
11479:
1.211 brouard 11480: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11481: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11482: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11483:
1.200 brouard 11484: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11485: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11486: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11487: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11488: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11489: * (currently 0 or 1) in the data.
11490: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11491: * corresponding modality (h,j).
11492: */
11493:
1.145 brouard 11494: h=0;
11495: /*if (cptcovn > 0) */
1.126 brouard 11496: m=pow(2,cptcoveff);
11497:
1.144 brouard 11498: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11499: * For k=4 covariates, h goes from 1 to m=2**k
11500: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11501: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11502: * h\k 1 2 3 4
1.143 brouard 11503: *______________________________
11504: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11505: * 2 2 1 1 1
11506: * 3 i=2 1 2 1 1
11507: * 4 2 2 1 1
11508: * 5 i=3 1 i=2 1 2 1
11509: * 6 2 1 2 1
11510: * 7 i=4 1 2 2 1
11511: * 8 2 2 2 1
1.197 brouard 11512: * 9 i=5 1 i=3 1 i=2 1 2
11513: * 10 2 1 1 2
11514: * 11 i=6 1 2 1 2
11515: * 12 2 2 1 2
11516: * 13 i=7 1 i=4 1 2 2
11517: * 14 2 1 2 2
11518: * 15 i=8 1 2 2 2
11519: * 16 2 2 2 2
1.143 brouard 11520: */
1.212 brouard 11521: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11522: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11523: * and the value of each covariate?
11524: * V1=1, V2=1, V3=2, V4=1 ?
11525: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11526: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11527: * In order to get the real value in the data, we use nbcode
11528: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11529: * We are keeping this crazy system in order to be able (in the future?)
11530: * to have more than 2 values (0 or 1) for a covariate.
11531: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11532: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11533: * bbbbbbbb
11534: * 76543210
11535: * h-1 00000101 (6-1=5)
1.219 brouard 11536: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11537: * &
11538: * 1 00000001 (1)
1.219 brouard 11539: * 00000000 = 1 & ((h-1) >> (k-1))
11540: * +1= 00000001 =1
1.211 brouard 11541: *
11542: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11543: * h' 1101 =2^3+2^2+0x2^1+2^0
11544: * >>k' 11
11545: * & 00000001
11546: * = 00000001
11547: * +1 = 00000010=2 = codtabm(14,3)
11548: * Reverse h=6 and m=16?
11549: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11550: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11551: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11552: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11553: * V3=decodtabm(14,3,2**4)=2
11554: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11555: *(h-1) >> (j-1) 0011 =13 >> 2
11556: * &1 000000001
11557: * = 000000001
11558: * +1= 000000010 =2
11559: * 2211
11560: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11561: * V3=2
1.220 brouard 11562: * codtabm and decodtabm are identical
1.211 brouard 11563: */
11564:
1.145 brouard 11565:
11566: free_ivector(Ndum,-1,NCOVMAX);
11567:
11568:
1.126 brouard 11569:
1.186 brouard 11570: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11571: strcpy(optionfilegnuplot,optionfilefiname);
11572: if(mle==-3)
1.201 brouard 11573: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11574: strcat(optionfilegnuplot,".gp");
11575:
11576: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11577: printf("Problem with file %s",optionfilegnuplot);
11578: }
11579: else{
1.204 brouard 11580: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11581: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11582: //fprintf(ficgp,"set missing 'NaNq'\n");
11583: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11584: }
11585: /* fclose(ficgp);*/
1.186 brouard 11586:
11587:
11588: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11589:
11590: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11591: if(mle==-3)
1.201 brouard 11592: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11593: strcat(optionfilehtm,".htm");
11594: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11595: printf("Problem with %s \n",optionfilehtm);
11596: exit(0);
1.126 brouard 11597: }
11598:
11599: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11600: strcat(optionfilehtmcov,"-cov.htm");
11601: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11602: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11603: }
11604: else{
11605: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11606: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11607: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11608: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11609: }
11610:
1.213 brouard 11611: fprintf(fichtm,"<html><head>\n<head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<title>IMaCh %s</title></head>\n <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n<font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>-EUROREVES-Institut de longévité-2013-2016-Japan Society for the Promotion of Sciences 日本学術振興会 (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \
1.204 brouard 11612: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11613: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11614: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11615: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11616: \n\
11617: <hr size=\"2\" color=\"#EC5E5E\">\
11618: <ul><li><h4>Parameter files</h4>\n\
11619: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11620: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11621: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11622: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11623: - Date and time at start: %s</ul>\n",\
11624: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11625: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11626: fileres,fileres,\
11627: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11628: fflush(fichtm);
11629:
11630: strcpy(pathr,path);
11631: strcat(pathr,optionfilefiname);
1.184 brouard 11632: #ifdef WIN32
11633: _chdir(optionfilefiname); /* Move to directory named optionfile */
11634: #else
1.126 brouard 11635: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11636: #endif
11637:
1.126 brouard 11638:
1.220 brouard 11639: /* Calculates basic frequencies. Computes observed prevalence at single age
11640: and for any valid combination of covariates
1.126 brouard 11641: and prints on file fileres'p'. */
1.251 brouard 11642: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11643: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11644:
11645: fprintf(fichtm,"\n");
1.286 ! brouard 11646: 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 11647: ftol, stepm);
11648: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11649: ncurrv=1;
11650: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11651: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11652: ncurrv=i;
11653: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
11654: fprintf(fichtm,"\n<li> Number of time varying (wave varying) covariates: ntv=%d ", ntv);
11655: ncurrv=i;
11656: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
11657: fprintf(fichtm,"\n<li>Number of quantitative time varying covariates: nqtv=%d ", nqtv);
11658: ncurrv=i;
11659: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11660: 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", \
11661: nlstate, ndeath, maxwav, mle, weightopt);
11662:
11663: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11664: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11665:
11666:
11667: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11668: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11669: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11670: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11671: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11672: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11673: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11674: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11675: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11676:
1.126 brouard 11677: /* For Powell, parameters are in a vector p[] starting at p[1]
11678: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11679: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11680:
11681: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11682: /* For mortality only */
1.126 brouard 11683: if (mle==-3){
1.136 brouard 11684: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11685: for(i=1;i<=NDIM;i++)
11686: for(j=1;j<=NDIM;j++)
11687: ximort[i][j]=0.;
1.186 brouard 11688: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11689: cens=ivector(1,n);
11690: ageexmed=vector(1,n);
11691: agecens=vector(1,n);
11692: dcwave=ivector(1,n);
1.223 brouard 11693:
1.126 brouard 11694: for (i=1; i<=imx; i++){
11695: dcwave[i]=-1;
11696: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11697: if (s[m][i]>nlstate) {
11698: dcwave[i]=m;
11699: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11700: break;
11701: }
1.126 brouard 11702: }
1.226 brouard 11703:
1.126 brouard 11704: for (i=1; i<=imx; i++) {
11705: if (wav[i]>0){
1.226 brouard 11706: ageexmed[i]=agev[mw[1][i]][i];
11707: j=wav[i];
11708: agecens[i]=1.;
11709:
11710: if (ageexmed[i]> 1 && wav[i] > 0){
11711: agecens[i]=agev[mw[j][i]][i];
11712: cens[i]= 1;
11713: }else if (ageexmed[i]< 1)
11714: cens[i]= -1;
11715: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11716: cens[i]=0 ;
1.126 brouard 11717: }
11718: else cens[i]=-1;
11719: }
11720:
11721: for (i=1;i<=NDIM;i++) {
11722: for (j=1;j<=NDIM;j++)
1.226 brouard 11723: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11724: }
11725:
1.145 brouard 11726: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11727: /*printf("%lf %lf", p[1], p[2]);*/
11728:
11729:
1.136 brouard 11730: #ifdef GSL
11731: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11732: #else
1.126 brouard 11733: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11734: #endif
1.201 brouard 11735: strcpy(filerespow,"POW-MORT_");
11736: strcat(filerespow,fileresu);
1.126 brouard 11737: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11738: printf("Problem with resultfile: %s\n", filerespow);
11739: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11740: }
1.136 brouard 11741: #ifdef GSL
11742: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11743: #else
1.126 brouard 11744: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11745: #endif
1.126 brouard 11746: /* for (i=1;i<=nlstate;i++)
11747: for(j=1;j<=nlstate+ndeath;j++)
11748: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11749: */
11750: fprintf(ficrespow,"\n");
1.136 brouard 11751: #ifdef GSL
11752: /* gsl starts here */
11753: T = gsl_multimin_fminimizer_nmsimplex;
11754: gsl_multimin_fminimizer *sfm = NULL;
11755: gsl_vector *ss, *x;
11756: gsl_multimin_function minex_func;
11757:
11758: /* Initial vertex size vector */
11759: ss = gsl_vector_alloc (NDIM);
11760:
11761: if (ss == NULL){
11762: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11763: }
11764: /* Set all step sizes to 1 */
11765: gsl_vector_set_all (ss, 0.001);
11766:
11767: /* Starting point */
1.126 brouard 11768:
1.136 brouard 11769: x = gsl_vector_alloc (NDIM);
11770:
11771: if (x == NULL){
11772: gsl_vector_free(ss);
11773: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11774: }
11775:
11776: /* Initialize method and iterate */
11777: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11778: /* gsl_vector_set(x, 0, 0.0268); */
11779: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11780: gsl_vector_set(x, 0, p[1]);
11781: gsl_vector_set(x, 1, p[2]);
11782:
11783: minex_func.f = &gompertz_f;
11784: minex_func.n = NDIM;
11785: minex_func.params = (void *)&p; /* ??? */
11786:
11787: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11788: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11789:
11790: printf("Iterations beginning .....\n\n");
11791: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11792:
11793: iteri=0;
11794: while (rval == GSL_CONTINUE){
11795: iteri++;
11796: status = gsl_multimin_fminimizer_iterate(sfm);
11797:
11798: if (status) printf("error: %s\n", gsl_strerror (status));
11799: fflush(0);
11800:
11801: if (status)
11802: break;
11803:
11804: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11805: ssval = gsl_multimin_fminimizer_size (sfm);
11806:
11807: if (rval == GSL_SUCCESS)
11808: printf ("converged to a local maximum at\n");
11809:
11810: printf("%5d ", iteri);
11811: for (it = 0; it < NDIM; it++){
11812: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11813: }
11814: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11815: }
11816:
11817: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11818:
11819: gsl_vector_free(x); /* initial values */
11820: gsl_vector_free(ss); /* inital step size */
11821: for (it=0; it<NDIM; it++){
11822: p[it+1]=gsl_vector_get(sfm->x,it);
11823: fprintf(ficrespow," %.12lf", p[it]);
11824: }
11825: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11826: #endif
11827: #ifdef POWELL
11828: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11829: #endif
1.126 brouard 11830: fclose(ficrespow);
11831:
1.203 brouard 11832: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11833:
11834: for(i=1; i <=NDIM; i++)
11835: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11836: matcov[i][j]=matcov[j][i];
1.126 brouard 11837:
11838: printf("\nCovariance matrix\n ");
1.203 brouard 11839: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11840: for(i=1; i <=NDIM; i++) {
11841: for(j=1;j<=NDIM;j++){
1.220 brouard 11842: printf("%f ",matcov[i][j]);
11843: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11844: }
1.203 brouard 11845: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11846: }
11847:
11848: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11849: for (i=1;i<=NDIM;i++) {
1.126 brouard 11850: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11851: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11852: }
1.126 brouard 11853: lsurv=vector(1,AGESUP);
11854: lpop=vector(1,AGESUP);
11855: tpop=vector(1,AGESUP);
11856: lsurv[agegomp]=100000;
11857:
11858: for (k=agegomp;k<=AGESUP;k++) {
11859: agemortsup=k;
11860: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11861: }
11862:
11863: for (k=agegomp;k<agemortsup;k++)
11864: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11865:
11866: for (k=agegomp;k<agemortsup;k++){
11867: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11868: sumlpop=sumlpop+lpop[k];
11869: }
11870:
11871: tpop[agegomp]=sumlpop;
11872: for (k=agegomp;k<(agemortsup-3);k++){
11873: /* tpop[k+1]=2;*/
11874: tpop[k+1]=tpop[k]-lpop[k];
11875: }
11876:
11877:
11878: printf("\nAge lx qx dx Lx Tx e(x)\n");
11879: for (k=agegomp;k<(agemortsup-2);k++)
11880: 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]);
11881:
11882:
11883: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11884: ageminpar=50;
11885: agemaxpar=100;
1.194 brouard 11886: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11887: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11888: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11889: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11890: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11891: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11892: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11893: }else{
11894: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11895: 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 11896: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11897: }
1.201 brouard 11898: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11899: stepm, weightopt,\
11900: model,imx,p,matcov,agemortsup);
11901:
11902: free_vector(lsurv,1,AGESUP);
11903: free_vector(lpop,1,AGESUP);
11904: free_vector(tpop,1,AGESUP);
1.220 brouard 11905: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11906: free_ivector(cens,1,n);
11907: free_vector(agecens,1,n);
11908: free_ivector(dcwave,1,n);
1.220 brouard 11909: #ifdef GSL
1.136 brouard 11910: #endif
1.186 brouard 11911: } /* Endof if mle==-3 mortality only */
1.205 brouard 11912: /* Standard */
11913: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11914: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11915: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11916: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11917: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11918: for (k=1; k<=npar;k++)
11919: printf(" %d %8.5f",k,p[k]);
11920: printf("\n");
1.205 brouard 11921: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11922: /* mlikeli uses func not funcone */
1.247 brouard 11923: /* for(i=1;i<nlstate;i++){ */
11924: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11925: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11926: /* } */
1.205 brouard 11927: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11928: }
11929: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11930: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11931: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11932: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11933: }
11934: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11935: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11936: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11937: for (k=1; k<=npar;k++)
11938: printf(" %d %8.5f",k,p[k]);
11939: printf("\n");
11940:
11941: /*--------- results files --------------*/
1.283 brouard 11942: /* 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 11943:
11944:
11945: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11946: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11947: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11948: for(i=1,jk=1; i <=nlstate; i++){
11949: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11950: if (k != i) {
11951: printf("%d%d ",i,k);
11952: fprintf(ficlog,"%d%d ",i,k);
11953: fprintf(ficres,"%1d%1d ",i,k);
11954: for(j=1; j <=ncovmodel; j++){
11955: printf("%12.7f ",p[jk]);
11956: fprintf(ficlog,"%12.7f ",p[jk]);
11957: fprintf(ficres,"%12.7f ",p[jk]);
11958: jk++;
11959: }
11960: printf("\n");
11961: fprintf(ficlog,"\n");
11962: fprintf(ficres,"\n");
11963: }
1.126 brouard 11964: }
11965: }
1.203 brouard 11966: if(mle != 0){
11967: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11968: ftolhess=ftol; /* Usually correct */
1.203 brouard 11969: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11970: 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");
11971: 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");
11972: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11973: for(k=1; k <=(nlstate+ndeath); k++){
11974: if (k != i) {
11975: printf("%d%d ",i,k);
11976: fprintf(ficlog,"%d%d ",i,k);
11977: for(j=1; j <=ncovmodel; j++){
11978: 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]));
11979: 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]));
11980: jk++;
11981: }
11982: printf("\n");
11983: fprintf(ficlog,"\n");
11984: }
11985: }
1.193 brouard 11986: }
1.203 brouard 11987: } /* end of hesscov and Wald tests */
1.225 brouard 11988:
1.203 brouard 11989: /* */
1.126 brouard 11990: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11991: printf("# Scales (for hessian or gradient estimation)\n");
11992: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11993: for(i=1,jk=1; i <=nlstate; i++){
11994: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11995: if (j!=i) {
11996: fprintf(ficres,"%1d%1d",i,j);
11997: printf("%1d%1d",i,j);
11998: fprintf(ficlog,"%1d%1d",i,j);
11999: for(k=1; k<=ncovmodel;k++){
12000: printf(" %.5e",delti[jk]);
12001: fprintf(ficlog," %.5e",delti[jk]);
12002: fprintf(ficres," %.5e",delti[jk]);
12003: jk++;
12004: }
12005: printf("\n");
12006: fprintf(ficlog,"\n");
12007: fprintf(ficres,"\n");
12008: }
1.126 brouard 12009: }
12010: }
12011:
12012: 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 12013: if(mle >= 1) /* To big for the screen */
1.126 brouard 12014: 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");
12015: 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");
12016: /* # 121 Var(a12)\n\ */
12017: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12018: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12019: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12020: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12021: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12022: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12023: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12024:
12025:
12026: /* Just to have a covariance matrix which will be more understandable
12027: even is we still don't want to manage dictionary of variables
12028: */
12029: for(itimes=1;itimes<=2;itimes++){
12030: jj=0;
12031: for(i=1; i <=nlstate; i++){
1.225 brouard 12032: for(j=1; j <=nlstate+ndeath; j++){
12033: if(j==i) continue;
12034: for(k=1; k<=ncovmodel;k++){
12035: jj++;
12036: ca[0]= k+'a'-1;ca[1]='\0';
12037: if(itimes==1){
12038: if(mle>=1)
12039: printf("#%1d%1d%d",i,j,k);
12040: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12041: fprintf(ficres,"#%1d%1d%d",i,j,k);
12042: }else{
12043: if(mle>=1)
12044: printf("%1d%1d%d",i,j,k);
12045: fprintf(ficlog,"%1d%1d%d",i,j,k);
12046: fprintf(ficres,"%1d%1d%d",i,j,k);
12047: }
12048: ll=0;
12049: for(li=1;li <=nlstate; li++){
12050: for(lj=1;lj <=nlstate+ndeath; lj++){
12051: if(lj==li) continue;
12052: for(lk=1;lk<=ncovmodel;lk++){
12053: ll++;
12054: if(ll<=jj){
12055: cb[0]= lk +'a'-1;cb[1]='\0';
12056: if(ll<jj){
12057: if(itimes==1){
12058: if(mle>=1)
12059: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12060: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12061: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12062: }else{
12063: if(mle>=1)
12064: printf(" %.5e",matcov[jj][ll]);
12065: fprintf(ficlog," %.5e",matcov[jj][ll]);
12066: fprintf(ficres," %.5e",matcov[jj][ll]);
12067: }
12068: }else{
12069: if(itimes==1){
12070: if(mle>=1)
12071: printf(" Var(%s%1d%1d)",ca,i,j);
12072: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12073: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12074: }else{
12075: if(mle>=1)
12076: printf(" %.7e",matcov[jj][ll]);
12077: fprintf(ficlog," %.7e",matcov[jj][ll]);
12078: fprintf(ficres," %.7e",matcov[jj][ll]);
12079: }
12080: }
12081: }
12082: } /* end lk */
12083: } /* end lj */
12084: } /* end li */
12085: if(mle>=1)
12086: printf("\n");
12087: fprintf(ficlog,"\n");
12088: fprintf(ficres,"\n");
12089: numlinepar++;
12090: } /* end k*/
12091: } /*end j */
1.126 brouard 12092: } /* end i */
12093: } /* end itimes */
12094:
12095: fflush(ficlog);
12096: fflush(ficres);
1.225 brouard 12097: while(fgets(line, MAXLINE, ficpar)) {
12098: /* If line starts with a # it is a comment */
12099: if (line[0] == '#') {
12100: numlinepar++;
12101: fputs(line,stdout);
12102: fputs(line,ficparo);
12103: fputs(line,ficlog);
12104: continue;
12105: }else
12106: break;
12107: }
12108:
1.209 brouard 12109: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12110: /* ungetc(c,ficpar); */
12111: /* fgets(line, MAXLINE, ficpar); */
12112: /* fputs(line,stdout); */
12113: /* fputs(line,ficparo); */
12114: /* } */
12115: /* ungetc(c,ficpar); */
1.126 brouard 12116:
12117: estepm=0;
1.209 brouard 12118: 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 12119:
12120: if (num_filled != 6) {
12121: 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);
12122: 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);
12123: goto end;
12124: }
12125: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12126: }
12127: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12128: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12129:
1.209 brouard 12130: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12131: if (estepm==0 || estepm < stepm) estepm=stepm;
12132: if (fage <= 2) {
12133: bage = ageminpar;
12134: fage = agemaxpar;
12135: }
12136:
12137: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12138: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12139: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12140:
1.186 brouard 12141: /* Other stuffs, more or less useful */
1.254 brouard 12142: while(fgets(line, MAXLINE, ficpar)) {
12143: /* If line starts with a # it is a comment */
12144: if (line[0] == '#') {
12145: numlinepar++;
12146: fputs(line,stdout);
12147: fputs(line,ficparo);
12148: fputs(line,ficlog);
12149: continue;
12150: }else
12151: break;
12152: }
12153:
12154: 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){
12155:
12156: if (num_filled != 7) {
12157: 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);
12158: 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);
12159: goto end;
12160: }
12161: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12162: 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);
12163: 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);
12164: 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 12165: }
1.254 brouard 12166:
12167: while(fgets(line, MAXLINE, ficpar)) {
12168: /* If line starts with a # it is a comment */
12169: if (line[0] == '#') {
12170: numlinepar++;
12171: fputs(line,stdout);
12172: fputs(line,ficparo);
12173: fputs(line,ficlog);
12174: continue;
12175: }else
12176: break;
1.126 brouard 12177: }
12178:
12179:
12180: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12181: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12182:
1.254 brouard 12183: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12184: if (num_filled != 1) {
12185: 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);
12186: 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);
12187: goto end;
12188: }
12189: printf("pop_based=%d\n",popbased);
12190: fprintf(ficlog,"pop_based=%d\n",popbased);
12191: fprintf(ficparo,"pop_based=%d\n",popbased);
12192: fprintf(ficres,"pop_based=%d\n",popbased);
12193: }
12194:
1.258 brouard 12195: /* Results */
12196: nresult=0;
12197: do{
12198: if(!fgets(line, MAXLINE, ficpar)){
12199: endishere=1;
12200: parameterline=14;
12201: }else if (line[0] == '#') {
12202: /* If line starts with a # it is a comment */
1.254 brouard 12203: numlinepar++;
12204: fputs(line,stdout);
12205: fputs(line,ficparo);
12206: fputs(line,ficlog);
12207: continue;
1.258 brouard 12208: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12209: parameterline=11;
12210: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12211: parameterline=12;
12212: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12213: parameterline=13;
12214: else{
12215: parameterline=14;
1.254 brouard 12216: }
1.258 brouard 12217: switch (parameterline){
12218: case 11:
12219: 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){
12220: if (num_filled != 8) {
12221: 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\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
12222: 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 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
12223: goto end;
12224: }
12225: 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);
12226: 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);
12227: 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);
12228: 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);
12229: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12230: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12231: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12232:
1.258 brouard 12233: }
1.254 brouard 12234: break;
1.258 brouard 12235: case 12:
12236: /*fscanf(ficpar,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj);*/
12237: if((num_filled=sscanf(line,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj)) !=EOF){
12238: if (num_filled != 8) {
1.262 brouard 12239: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
12240: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
1.258 brouard 12241: goto end;
12242: }
12243: printf("backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12244: fprintf(ficparo,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12245: fprintf(ficlog,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12246: fprintf(ficres,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12247: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12248: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12249: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12250: }
1.230 brouard 12251: break;
1.258 brouard 12252: case 13:
12253: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12254: if (num_filled == 0){
12255: resultline[0]='\0';
12256: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12257: fprintf(ficlog,"Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12258: break;
12259: } else if (num_filled != 1){
12260: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12261: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12262: }
12263: nresult++; /* Sum of resultlines */
12264: printf("Result %d: result=%s\n",nresult, resultline);
12265: if(nresult > MAXRESULTLINES){
12266: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12267: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12268: goto end;
12269: }
12270: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12271: fprintf(ficparo,"result: %s\n",resultline);
12272: fprintf(ficres,"result: %s\n",resultline);
12273: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12274: break;
1.258 brouard 12275: case 14:
1.259 brouard 12276: if(ncovmodel >2 && nresult==0 ){
12277: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12278: goto end;
12279: }
1.259 brouard 12280: break;
1.258 brouard 12281: default:
12282: nresult=1;
12283: decoderesult(".",nresult ); /* No covariate */
12284: }
12285: } /* End switch parameterline */
12286: }while(endishere==0); /* End do */
1.126 brouard 12287:
1.230 brouard 12288: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12289: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12290:
12291: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12292: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12293: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12294: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12295: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12296: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12297: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12298: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12299: }else{
1.270 brouard 12300: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12301: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12302: }
12303: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12304: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12305: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12306:
1.225 brouard 12307: /*------------ free_vector -------------*/
12308: /* chdir(path); */
1.220 brouard 12309:
1.215 brouard 12310: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12311: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12312: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12313: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12314: free_lvector(num,1,n);
12315: free_vector(agedc,1,n);
12316: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12317: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12318: fclose(ficparo);
12319: fclose(ficres);
1.220 brouard 12320:
12321:
1.186 brouard 12322: /* Other results (useful)*/
1.220 brouard 12323:
12324:
1.126 brouard 12325: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12326: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12327: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12328: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12329: fclose(ficrespl);
12330:
12331: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12332: /*#include "hpijx.h"*/
12333: hPijx(p, bage, fage);
1.145 brouard 12334: fclose(ficrespij);
1.227 brouard 12335:
1.220 brouard 12336: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12337: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12338: k=1;
1.126 brouard 12339: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12340:
1.269 brouard 12341: /* Prevalence for each covariate combination in probs[age][status][cov] */
12342: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12343: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12344: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12345: for(k=1;k<=ncovcombmax;k++)
12346: probs[i][j][k]=0.;
1.269 brouard 12347: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12348: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12349: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12350: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12351: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12352: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12353: for(k=1;k<=ncovcombmax;k++)
12354: mobaverages[i][j][k]=0.;
1.219 brouard 12355: mobaverage=mobaverages;
12356: if (mobilav!=0) {
1.235 brouard 12357: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12358: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12359: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12360: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12361: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12362: }
1.269 brouard 12363: } else if (mobilavproj !=0) {
1.235 brouard 12364: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12365: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12366: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12367: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12368: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12369: }
1.269 brouard 12370: }else{
12371: printf("Internal error moving average\n");
12372: fflush(stdout);
12373: exit(1);
1.219 brouard 12374: }
12375: }/* end if moving average */
1.227 brouard 12376:
1.126 brouard 12377: /*---------- Forecasting ------------------*/
12378: if(prevfcast==1){
12379: /* if(stepm ==1){*/
1.269 brouard 12380: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12381: }
1.269 brouard 12382:
12383: /* Backcasting */
1.217 brouard 12384: if(backcast==1){
1.219 brouard 12385: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12386: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12387: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12388:
12389: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12390:
12391: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12392:
1.219 brouard 12393: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12394: fclose(ficresplb);
12395:
1.222 brouard 12396: hBijx(p, bage, fage, mobaverage);
12397: fclose(ficrespijb);
1.219 brouard 12398:
1.269 brouard 12399: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12400: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12401: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12402:
12403:
1.269 brouard 12404: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12405: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12406: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12407: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12408: } /* end Backcasting */
1.268 brouard 12409:
1.186 brouard 12410:
12411: /* ------ Other prevalence ratios------------ */
1.126 brouard 12412:
1.215 brouard 12413: free_ivector(wav,1,imx);
12414: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12415: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12416: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12417:
12418:
1.127 brouard 12419: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12420:
1.201 brouard 12421: strcpy(filerese,"E_");
12422: strcat(filerese,fileresu);
1.126 brouard 12423: if((ficreseij=fopen(filerese,"w"))==NULL) {
12424: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12425: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12426: }
1.208 brouard 12427: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12428: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12429:
12430: pstamp(ficreseij);
1.219 brouard 12431:
1.235 brouard 12432: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12433: if (cptcovn < 1){i1=1;}
12434:
12435: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12436: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12437: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12438: continue;
1.219 brouard 12439: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12440: printf("\n#****** ");
1.225 brouard 12441: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12442: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12443: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12444: }
12445: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12446: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12447: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12448: }
12449: fprintf(ficreseij,"******\n");
1.235 brouard 12450: printf("******\n");
1.219 brouard 12451:
12452: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12453: oldm=oldms;savm=savms;
1.235 brouard 12454: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12455:
1.219 brouard 12456: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12457: }
12458: fclose(ficreseij);
1.208 brouard 12459: printf("done evsij\n");fflush(stdout);
12460: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12461:
1.218 brouard 12462:
1.227 brouard 12463: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12464:
1.201 brouard 12465: strcpy(filerest,"T_");
12466: strcat(filerest,fileresu);
1.127 brouard 12467: if((ficrest=fopen(filerest,"w"))==NULL) {
12468: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12469: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12470: }
1.208 brouard 12471: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12472: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12473: strcpy(fileresstde,"STDE_");
12474: strcat(fileresstde,fileresu);
1.126 brouard 12475: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12476: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12477: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12478: }
1.227 brouard 12479: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12480: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12481:
1.201 brouard 12482: strcpy(filerescve,"CVE_");
12483: strcat(filerescve,fileresu);
1.126 brouard 12484: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12485: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12486: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12487: }
1.227 brouard 12488: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12489: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12490:
1.201 brouard 12491: strcpy(fileresv,"V_");
12492: strcat(fileresv,fileresu);
1.126 brouard 12493: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12494: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12495: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12496: }
1.227 brouard 12497: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12498: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12499:
1.235 brouard 12500: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12501: if (cptcovn < 1){i1=1;}
12502:
12503: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12504: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12505: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12506: continue;
1.242 brouard 12507: printf("\n#****** Result for:");
12508: fprintf(ficrest,"\n#****** Result for:");
12509: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12510: for(j=1;j<=cptcoveff;j++){
12511: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12512: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12513: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12514: }
1.235 brouard 12515: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12516: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12517: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12518: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12519: }
1.208 brouard 12520: fprintf(ficrest,"******\n");
1.227 brouard 12521: fprintf(ficlog,"******\n");
12522: printf("******\n");
1.208 brouard 12523:
12524: fprintf(ficresstdeij,"\n#****** ");
12525: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12526: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12527: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12528: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12529: }
1.235 brouard 12530: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12531: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12532: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12533: }
1.208 brouard 12534: fprintf(ficresstdeij,"******\n");
12535: fprintf(ficrescveij,"******\n");
12536:
12537: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12538: /* pstamp(ficresvij); */
1.225 brouard 12539: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12540: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12541: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12542: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12543: }
1.208 brouard 12544: fprintf(ficresvij,"******\n");
12545:
12546: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12547: oldm=oldms;savm=savms;
1.235 brouard 12548: printf(" cvevsij ");
12549: fprintf(ficlog, " cvevsij ");
12550: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12551: printf(" end cvevsij \n ");
12552: fprintf(ficlog, " end cvevsij \n ");
12553:
12554: /*
12555: */
12556: /* goto endfree; */
12557:
12558: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12559: pstamp(ficrest);
12560:
1.269 brouard 12561: epj=vector(1,nlstate+1);
1.208 brouard 12562: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12563: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12564: cptcod= 0; /* To be deleted */
12565: printf("varevsij vpopbased=%d \n",vpopbased);
12566: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12567: 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 12568: 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 ");
12569: if(vpopbased==1)
12570: 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);
12571: else
12572: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12573: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12574: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12575: fprintf(ficrest,"\n");
12576: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12577: printf("Computing age specific period (stable) prevalences in each health state \n");
12578: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12579: for(age=bage; age <=fage ;age++){
1.235 brouard 12580: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12581: if (vpopbased==1) {
12582: if(mobilav ==0){
12583: for(i=1; i<=nlstate;i++)
12584: prlim[i][i]=probs[(int)age][i][k];
12585: }else{ /* mobilav */
12586: for(i=1; i<=nlstate;i++)
12587: prlim[i][i]=mobaverage[(int)age][i][k];
12588: }
12589: }
1.219 brouard 12590:
1.227 brouard 12591: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12592: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12593: /* printf(" age %4.0f ",age); */
12594: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12595: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12596: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12597: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12598: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12599: }
12600: epj[nlstate+1] +=epj[j];
12601: }
12602: /* printf(" age %4.0f \n",age); */
1.219 brouard 12603:
1.227 brouard 12604: for(i=1, vepp=0.;i <=nlstate;i++)
12605: for(j=1;j <=nlstate;j++)
12606: vepp += vareij[i][j][(int)age];
12607: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12608: for(j=1;j <=nlstate;j++){
12609: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12610: }
12611: fprintf(ficrest,"\n");
12612: }
1.208 brouard 12613: } /* End vpopbased */
1.269 brouard 12614: free_vector(epj,1,nlstate+1);
1.208 brouard 12615: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12616: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12617: printf("done selection\n");fflush(stdout);
12618: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12619:
1.235 brouard 12620: } /* End k selection */
1.227 brouard 12621:
12622: printf("done State-specific expectancies\n");fflush(stdout);
12623: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12624:
1.269 brouard 12625: /* variance-covariance of period prevalence*/
12626: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12627:
1.227 brouard 12628:
12629: free_vector(weight,1,n);
12630: free_imatrix(Tvard,1,NCOVMAX,1,2);
12631: free_imatrix(s,1,maxwav+1,1,n);
12632: free_matrix(anint,1,maxwav,1,n);
12633: free_matrix(mint,1,maxwav,1,n);
12634: free_ivector(cod,1,n);
12635: free_ivector(tab,1,NCOVMAX);
12636: fclose(ficresstdeij);
12637: fclose(ficrescveij);
12638: fclose(ficresvij);
12639: fclose(ficrest);
12640: fclose(ficpar);
12641:
12642:
1.126 brouard 12643: /*---------- End : free ----------------*/
1.219 brouard 12644: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12645: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12646: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12647: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12648: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12649: } /* mle==-3 arrives here for freeing */
1.227 brouard 12650: /* endfree:*/
12651: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12652: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12653: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12654: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12655: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12656: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12657: free_matrix(covar,0,NCOVMAX,1,n);
12658: free_matrix(matcov,1,npar,1,npar);
12659: free_matrix(hess,1,npar,1,npar);
12660: /*free_vector(delti,1,npar);*/
12661: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12662: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12663: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12664: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12665:
12666: free_ivector(ncodemax,1,NCOVMAX);
12667: free_ivector(ncodemaxwundef,1,NCOVMAX);
12668: free_ivector(Dummy,-1,NCOVMAX);
12669: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12670: free_ivector(DummyV,1,NCOVMAX);
12671: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12672: free_ivector(Typevar,-1,NCOVMAX);
12673: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12674: free_ivector(TvarsQ,1,NCOVMAX);
12675: free_ivector(TvarsQind,1,NCOVMAX);
12676: free_ivector(TvarsD,1,NCOVMAX);
12677: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12678: free_ivector(TvarFD,1,NCOVMAX);
12679: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12680: free_ivector(TvarF,1,NCOVMAX);
12681: free_ivector(TvarFind,1,NCOVMAX);
12682: free_ivector(TvarV,1,NCOVMAX);
12683: free_ivector(TvarVind,1,NCOVMAX);
12684: free_ivector(TvarA,1,NCOVMAX);
12685: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12686: free_ivector(TvarFQ,1,NCOVMAX);
12687: free_ivector(TvarFQind,1,NCOVMAX);
12688: free_ivector(TvarVD,1,NCOVMAX);
12689: free_ivector(TvarVDind,1,NCOVMAX);
12690: free_ivector(TvarVQ,1,NCOVMAX);
12691: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12692: free_ivector(Tvarsel,1,NCOVMAX);
12693: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12694: free_ivector(Tposprod,1,NCOVMAX);
12695: free_ivector(Tprod,1,NCOVMAX);
12696: free_ivector(Tvaraff,1,NCOVMAX);
12697: free_ivector(invalidvarcomb,1,ncovcombmax);
12698: free_ivector(Tage,1,NCOVMAX);
12699: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12700: free_ivector(TmodelInvind,1,NCOVMAX);
12701: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12702:
12703: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12704: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12705: fflush(fichtm);
12706: fflush(ficgp);
12707:
1.227 brouard 12708:
1.126 brouard 12709: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12710: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12711: 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 12712: }else{
12713: printf("End of Imach\n");
12714: fprintf(ficlog,"End of Imach\n");
12715: }
12716: printf("See log file on %s\n",filelog);
12717: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12718: /*(void) gettimeofday(&end_time,&tzp);*/
12719: rend_time = time(NULL);
12720: end_time = *localtime(&rend_time);
12721: /* tml = *localtime(&end_time.tm_sec); */
12722: strcpy(strtend,asctime(&end_time));
1.126 brouard 12723: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12724: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12725: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12726:
1.157 brouard 12727: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12728: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12729: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12730: /* printf("Total time was %d uSec.\n", total_usecs);*/
12731: /* if(fileappend(fichtm,optionfilehtm)){ */
12732: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12733: fclose(fichtm);
12734: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12735: fclose(fichtmcov);
12736: fclose(ficgp);
12737: fclose(ficlog);
12738: /*------ End -----------*/
1.227 brouard 12739:
1.281 brouard 12740:
12741: /* Executes gnuplot */
1.227 brouard 12742:
12743: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12744: #ifdef WIN32
1.227 brouard 12745: if (_chdir(pathcd) != 0)
12746: printf("Can't move to directory %s!\n",path);
12747: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12748: #else
1.227 brouard 12749: if(chdir(pathcd) != 0)
12750: printf("Can't move to directory %s!\n", path);
12751: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12752: #endif
1.126 brouard 12753: printf("Current directory %s!\n",pathcd);
12754: /*strcat(plotcmd,CHARSEPARATOR);*/
12755: sprintf(plotcmd,"gnuplot");
1.157 brouard 12756: #ifdef _WIN32
1.126 brouard 12757: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12758: #endif
12759: if(!stat(plotcmd,&info)){
1.158 brouard 12760: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12761: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12762: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12763: }else
12764: strcpy(pplotcmd,plotcmd);
1.157 brouard 12765: #ifdef __unix
1.126 brouard 12766: strcpy(plotcmd,GNUPLOTPROGRAM);
12767: if(!stat(plotcmd,&info)){
1.158 brouard 12768: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12769: }else
12770: strcpy(pplotcmd,plotcmd);
12771: #endif
12772: }else
12773: strcpy(pplotcmd,plotcmd);
12774:
12775: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12776: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12777:
1.126 brouard 12778: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12779: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12780: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12781: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12782: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12783: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12784: }
1.158 brouard 12785: printf(" Successful, please wait...");
1.126 brouard 12786: while (z[0] != 'q') {
12787: /* chdir(path); */
1.154 brouard 12788: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12789: scanf("%s",z);
12790: /* if (z[0] == 'c') system("./imach"); */
12791: if (z[0] == 'e') {
1.158 brouard 12792: #ifdef __APPLE__
1.152 brouard 12793: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12794: #elif __linux
12795: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12796: #else
1.152 brouard 12797: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12798: #endif
12799: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12800: system(pplotcmd);
1.126 brouard 12801: }
12802: else if (z[0] == 'g') system(plotcmd);
12803: else if (z[0] == 'q') exit(0);
12804: }
1.227 brouard 12805: end:
1.126 brouard 12806: while (z[0] != 'q') {
1.195 brouard 12807: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12808: scanf("%s",z);
12809: }
1.283 brouard 12810: printf("End\n");
1.282 brouard 12811: exit(0);
1.126 brouard 12812: }
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