Annotation of imach/src/imach.c, revision 1.293
1.293 ! brouard 1: /* $Id: imach.c,v 1.292 2019/05/09 14:17:20 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.293 ! brouard 4: Revision 1.292 2019/05/09 14:17:20 brouard
! 5: Summary: Some updates
! 6:
1.292 brouard 7: Revision 1.291 2019/05/09 13:44:18 brouard
8: Summary: Before ncovmax
9:
1.291 brouard 10: Revision 1.290 2019/05/09 13:39:37 brouard
11: Summary: 0.99r18 unlimited number of individuals
12:
13: The number n which was limited to 20,000 cases is now unlimited, from firstobs to lastobs. If the number is too for the virtual memory, probably an error will occur.
14:
1.290 brouard 15: Revision 1.289 2018/12/13 09:16:26 brouard
16: Summary: Bug for young ages (<-30) will be in r17
17:
1.289 brouard 18: Revision 1.288 2018/05/02 20:58:27 brouard
19: Summary: Some bugs fixed
20:
1.288 brouard 21: Revision 1.287 2018/05/01 17:57:25 brouard
22: Summary: Bug fixed by providing frequencies only for non missing covariates
23:
1.287 brouard 24: Revision 1.286 2018/04/27 14:27:04 brouard
25: Summary: some minor bugs
26:
1.286 brouard 27: Revision 1.285 2018/04/21 21:02:16 brouard
28: Summary: Some bugs fixed, valgrind tested
29:
1.285 brouard 30: Revision 1.284 2018/04/20 05:22:13 brouard
31: Summary: Computing mean and stdeviation of fixed quantitative variables
32:
1.284 brouard 33: Revision 1.283 2018/04/19 14:49:16 brouard
34: Summary: Some minor bugs fixed
35:
1.283 brouard 36: Revision 1.282 2018/02/27 22:50:02 brouard
37: *** empty log message ***
38:
1.282 brouard 39: Revision 1.281 2018/02/27 19:25:23 brouard
40: Summary: Adding second argument for quitting
41:
1.281 brouard 42: Revision 1.280 2018/02/21 07:58:13 brouard
43: Summary: 0.99r15
44:
45: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
46:
1.280 brouard 47: Revision 1.279 2017/07/20 13:35:01 brouard
48: Summary: temporary working
49:
1.279 brouard 50: Revision 1.278 2017/07/19 14:09:02 brouard
51: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
52:
1.278 brouard 53: Revision 1.277 2017/07/17 08:53:49 brouard
54: Summary: BOM files can be read now
55:
1.277 brouard 56: Revision 1.276 2017/06/30 15:48:31 brouard
57: Summary: Graphs improvements
58:
1.276 brouard 59: Revision 1.275 2017/06/30 13:39:33 brouard
60: Summary: Saito's color
61:
1.275 brouard 62: Revision 1.274 2017/06/29 09:47:08 brouard
63: Summary: Version 0.99r14
64:
1.274 brouard 65: Revision 1.273 2017/06/27 11:06:02 brouard
66: Summary: More documentation on projections
67:
1.273 brouard 68: Revision 1.272 2017/06/27 10:22:40 brouard
69: Summary: Color of backprojection changed from 6 to 5(yellow)
70:
1.272 brouard 71: Revision 1.271 2017/06/27 10:17:50 brouard
72: Summary: Some bug with rint
73:
1.271 brouard 74: Revision 1.270 2017/05/24 05:45:29 brouard
75: *** empty log message ***
76:
1.270 brouard 77: Revision 1.269 2017/05/23 08:39:25 brouard
78: Summary: Code into subroutine, cleanings
79:
1.269 brouard 80: Revision 1.268 2017/05/18 20:09:32 brouard
81: Summary: backprojection and confidence intervals of backprevalence
82:
1.268 brouard 83: Revision 1.267 2017/05/13 10:25:05 brouard
84: Summary: temporary save for backprojection
85:
1.267 brouard 86: Revision 1.266 2017/05/13 07:26:12 brouard
87: Summary: Version 0.99r13 (improvements and bugs fixed)
88:
1.266 brouard 89: Revision 1.265 2017/04/26 16:22:11 brouard
90: Summary: imach 0.99r13 Some bugs fixed
91:
1.265 brouard 92: Revision 1.264 2017/04/26 06:01:29 brouard
93: Summary: Labels in graphs
94:
1.264 brouard 95: Revision 1.263 2017/04/24 15:23:15 brouard
96: Summary: to save
97:
1.263 brouard 98: Revision 1.262 2017/04/18 16:48:12 brouard
99: *** empty log message ***
100:
1.262 brouard 101: Revision 1.261 2017/04/05 10:14:09 brouard
102: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
103:
1.261 brouard 104: Revision 1.260 2017/04/04 17:46:59 brouard
105: Summary: Gnuplot indexations fixed (humm)
106:
1.260 brouard 107: Revision 1.259 2017/04/04 13:01:16 brouard
108: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
109:
1.259 brouard 110: Revision 1.258 2017/04/03 10:17:47 brouard
111: Summary: Version 0.99r12
112:
113: Some cleanings, conformed with updated documentation.
114:
1.258 brouard 115: Revision 1.257 2017/03/29 16:53:30 brouard
116: Summary: Temp
117:
1.257 brouard 118: Revision 1.256 2017/03/27 05:50:23 brouard
119: Summary: Temporary
120:
1.256 brouard 121: Revision 1.255 2017/03/08 16:02:28 brouard
122: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
123:
1.255 brouard 124: Revision 1.254 2017/03/08 07:13:00 brouard
125: Summary: Fixing data parameter line
126:
1.254 brouard 127: Revision 1.253 2016/12/15 11:59:41 brouard
128: Summary: 0.99 in progress
129:
1.253 brouard 130: Revision 1.252 2016/09/15 21:15:37 brouard
131: *** empty log message ***
132:
1.252 brouard 133: Revision 1.251 2016/09/15 15:01:13 brouard
134: Summary: not working
135:
1.251 brouard 136: Revision 1.250 2016/09/08 16:07:27 brouard
137: Summary: continue
138:
1.250 brouard 139: Revision 1.249 2016/09/07 17:14:18 brouard
140: Summary: Starting values from frequencies
141:
1.249 brouard 142: Revision 1.248 2016/09/07 14:10:18 brouard
143: *** empty log message ***
144:
1.248 brouard 145: Revision 1.247 2016/09/02 11:11:21 brouard
146: *** empty log message ***
147:
1.247 brouard 148: Revision 1.246 2016/09/02 08:49:22 brouard
149: *** empty log message ***
150:
1.246 brouard 151: Revision 1.245 2016/09/02 07:25:01 brouard
152: *** empty log message ***
153:
1.245 brouard 154: Revision 1.244 2016/09/02 07:17:34 brouard
155: *** empty log message ***
156:
1.244 brouard 157: Revision 1.243 2016/09/02 06:45:35 brouard
158: *** empty log message ***
159:
1.243 brouard 160: Revision 1.242 2016/08/30 15:01:20 brouard
161: Summary: Fixing a lots
162:
1.242 brouard 163: Revision 1.241 2016/08/29 17:17:25 brouard
164: Summary: gnuplot problem in Back projection to fix
165:
1.241 brouard 166: Revision 1.240 2016/08/29 07:53:18 brouard
167: Summary: Better
168:
1.240 brouard 169: Revision 1.239 2016/08/26 15:51:03 brouard
170: Summary: Improvement in Powell output in order to copy and paste
171:
172: Author:
173:
1.239 brouard 174: Revision 1.238 2016/08/26 14:23:35 brouard
175: Summary: Starting tests of 0.99
176:
1.238 brouard 177: Revision 1.237 2016/08/26 09:20:19 brouard
178: Summary: to valgrind
179:
1.237 brouard 180: Revision 1.236 2016/08/25 10:50:18 brouard
181: *** empty log message ***
182:
1.236 brouard 183: Revision 1.235 2016/08/25 06:59:23 brouard
184: *** empty log message ***
185:
1.235 brouard 186: Revision 1.234 2016/08/23 16:51:20 brouard
187: *** empty log message ***
188:
1.234 brouard 189: Revision 1.233 2016/08/23 07:40:50 brouard
190: Summary: not working
191:
1.233 brouard 192: Revision 1.232 2016/08/22 14:20:21 brouard
193: Summary: not working
194:
1.232 brouard 195: Revision 1.231 2016/08/22 07:17:15 brouard
196: Summary: not working
197:
1.231 brouard 198: Revision 1.230 2016/08/22 06:55:53 brouard
199: Summary: Not working
200:
1.230 brouard 201: Revision 1.229 2016/07/23 09:45:53 brouard
202: Summary: Completing for func too
203:
1.229 brouard 204: Revision 1.228 2016/07/22 17:45:30 brouard
205: Summary: Fixing some arrays, still debugging
206:
1.227 brouard 207: Revision 1.226 2016/07/12 18:42:34 brouard
208: Summary: temp
209:
1.226 brouard 210: Revision 1.225 2016/07/12 08:40:03 brouard
211: Summary: saving but not running
212:
1.225 brouard 213: Revision 1.224 2016/07/01 13:16:01 brouard
214: Summary: Fixes
215:
1.224 brouard 216: Revision 1.223 2016/02/19 09:23:35 brouard
217: Summary: temporary
218:
1.223 brouard 219: Revision 1.222 2016/02/17 08:14:50 brouard
220: Summary: Probably last 0.98 stable version 0.98r6
221:
1.222 brouard 222: Revision 1.221 2016/02/15 23:35:36 brouard
223: Summary: minor bug
224:
1.220 brouard 225: Revision 1.219 2016/02/15 00:48:12 brouard
226: *** empty log message ***
227:
1.219 brouard 228: Revision 1.218 2016/02/12 11:29:23 brouard
229: Summary: 0.99 Back projections
230:
1.218 brouard 231: Revision 1.217 2015/12/23 17:18:31 brouard
232: Summary: Experimental backcast
233:
1.217 brouard 234: Revision 1.216 2015/12/18 17:32:11 brouard
235: Summary: 0.98r4 Warning and status=-2
236:
237: Version 0.98r4 is now:
238: - displaying an error when status is -1, date of interview unknown and date of death known;
239: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
240: Older changes concerning s=-2, dating from 2005 have been supersed.
241:
1.216 brouard 242: Revision 1.215 2015/12/16 08:52:24 brouard
243: Summary: 0.98r4 working
244:
1.215 brouard 245: Revision 1.214 2015/12/16 06:57:54 brouard
246: Summary: temporary not working
247:
1.214 brouard 248: Revision 1.213 2015/12/11 18:22:17 brouard
249: Summary: 0.98r4
250:
1.213 brouard 251: Revision 1.212 2015/11/21 12:47:24 brouard
252: Summary: minor typo
253:
1.212 brouard 254: Revision 1.211 2015/11/21 12:41:11 brouard
255: Summary: 0.98r3 with some graph of projected cross-sectional
256:
257: Author: Nicolas Brouard
258:
1.211 brouard 259: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 260: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 261: Summary: Adding ftolpl parameter
262: Author: N Brouard
263:
264: We had difficulties to get smoothed confidence intervals. It was due
265: to the period prevalence which wasn't computed accurately. The inner
266: parameter ftolpl is now an outer parameter of the .imach parameter
267: file after estepm. If ftolpl is small 1.e-4 and estepm too,
268: computation are long.
269:
1.209 brouard 270: Revision 1.208 2015/11/17 14:31:57 brouard
271: Summary: temporary
272:
1.208 brouard 273: Revision 1.207 2015/10/27 17:36:57 brouard
274: *** empty log message ***
275:
1.207 brouard 276: Revision 1.206 2015/10/24 07:14:11 brouard
277: *** empty log message ***
278:
1.206 brouard 279: Revision 1.205 2015/10/23 15:50:53 brouard
280: Summary: 0.98r3 some clarification for graphs on likelihood contributions
281:
1.205 brouard 282: Revision 1.204 2015/10/01 16:20:26 brouard
283: Summary: Some new graphs of contribution to likelihood
284:
1.204 brouard 285: Revision 1.203 2015/09/30 17:45:14 brouard
286: Summary: looking at better estimation of the hessian
287:
288: Also a better criteria for convergence to the period prevalence And
289: therefore adding the number of years needed to converge. (The
290: prevalence in any alive state shold sum to one
291:
1.203 brouard 292: Revision 1.202 2015/09/22 19:45:16 brouard
293: Summary: Adding some overall graph on contribution to likelihood. Might change
294:
1.202 brouard 295: Revision 1.201 2015/09/15 17:34:58 brouard
296: Summary: 0.98r0
297:
298: - Some new graphs like suvival functions
299: - Some bugs fixed like model=1+age+V2.
300:
1.201 brouard 301: Revision 1.200 2015/09/09 16:53:55 brouard
302: Summary: Big bug thanks to Flavia
303:
304: Even model=1+age+V2. did not work anymore
305:
1.200 brouard 306: Revision 1.199 2015/09/07 14:09:23 brouard
307: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
308:
1.199 brouard 309: Revision 1.198 2015/09/03 07:14:39 brouard
310: Summary: 0.98q5 Flavia
311:
1.198 brouard 312: Revision 1.197 2015/09/01 18:24:39 brouard
313: *** empty log message ***
314:
1.197 brouard 315: Revision 1.196 2015/08/18 23:17:52 brouard
316: Summary: 0.98q5
317:
1.196 brouard 318: Revision 1.195 2015/08/18 16:28:39 brouard
319: Summary: Adding a hack for testing purpose
320:
321: After reading the title, ftol and model lines, if the comment line has
322: a q, starting with #q, the answer at the end of the run is quit. It
323: permits to run test files in batch with ctest. The former workaround was
324: $ echo q | imach foo.imach
325:
1.195 brouard 326: Revision 1.194 2015/08/18 13:32:00 brouard
327: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
328:
1.194 brouard 329: Revision 1.193 2015/08/04 07:17:42 brouard
330: Summary: 0.98q4
331:
1.193 brouard 332: Revision 1.192 2015/07/16 16:49:02 brouard
333: Summary: Fixing some outputs
334:
1.192 brouard 335: Revision 1.191 2015/07/14 10:00:33 brouard
336: Summary: Some fixes
337:
1.191 brouard 338: Revision 1.190 2015/05/05 08:51:13 brouard
339: Summary: Adding digits in output parameters (7 digits instead of 6)
340:
341: Fix 1+age+.
342:
1.190 brouard 343: Revision 1.189 2015/04/30 14:45:16 brouard
344: Summary: 0.98q2
345:
1.189 brouard 346: Revision 1.188 2015/04/30 08:27:53 brouard
347: *** empty log message ***
348:
1.188 brouard 349: Revision 1.187 2015/04/29 09:11:15 brouard
350: *** empty log message ***
351:
1.187 brouard 352: Revision 1.186 2015/04/23 12:01:52 brouard
353: Summary: V1*age is working now, version 0.98q1
354:
355: Some codes had been disabled in order to simplify and Vn*age was
356: working in the optimization phase, ie, giving correct MLE parameters,
357: but, as usual, outputs were not correct and program core dumped.
358:
1.186 brouard 359: Revision 1.185 2015/03/11 13:26:42 brouard
360: Summary: Inclusion of compile and links command line for Intel Compiler
361:
1.185 brouard 362: Revision 1.184 2015/03/11 11:52:39 brouard
363: Summary: Back from Windows 8. Intel Compiler
364:
1.184 brouard 365: Revision 1.183 2015/03/10 20:34:32 brouard
366: Summary: 0.98q0, trying with directest, mnbrak fixed
367:
368: We use directest instead of original Powell test; probably no
369: incidence on the results, but better justifications;
370: We fixed Numerical Recipes mnbrak routine which was wrong and gave
371: wrong results.
372:
1.183 brouard 373: Revision 1.182 2015/02/12 08:19:57 brouard
374: Summary: Trying to keep directest which seems simpler and more general
375: Author: Nicolas Brouard
376:
1.182 brouard 377: Revision 1.181 2015/02/11 23:22:24 brouard
378: Summary: Comments on Powell added
379:
380: Author:
381:
1.181 brouard 382: Revision 1.180 2015/02/11 17:33:45 brouard
383: Summary: Finishing move from main to function (hpijx and prevalence_limit)
384:
1.180 brouard 385: Revision 1.179 2015/01/04 09:57:06 brouard
386: Summary: back to OS/X
387:
1.179 brouard 388: Revision 1.178 2015/01/04 09:35:48 brouard
389: *** empty log message ***
390:
1.178 brouard 391: Revision 1.177 2015/01/03 18:40:56 brouard
392: Summary: Still testing ilc32 on OSX
393:
1.177 brouard 394: Revision 1.176 2015/01/03 16:45:04 brouard
395: *** empty log message ***
396:
1.176 brouard 397: Revision 1.175 2015/01/03 16:33:42 brouard
398: *** empty log message ***
399:
1.175 brouard 400: Revision 1.174 2015/01/03 16:15:49 brouard
401: Summary: Still in cross-compilation
402:
1.174 brouard 403: Revision 1.173 2015/01/03 12:06:26 brouard
404: Summary: trying to detect cross-compilation
405:
1.173 brouard 406: Revision 1.172 2014/12/27 12:07:47 brouard
407: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
408:
1.172 brouard 409: Revision 1.171 2014/12/23 13:26:59 brouard
410: Summary: Back from Visual C
411:
412: Still problem with utsname.h on Windows
413:
1.171 brouard 414: Revision 1.170 2014/12/23 11:17:12 brouard
415: Summary: Cleaning some \%% back to %%
416:
417: The escape was mandatory for a specific compiler (which one?), but too many warnings.
418:
1.170 brouard 419: Revision 1.169 2014/12/22 23:08:31 brouard
420: Summary: 0.98p
421:
422: Outputs some informations on compiler used, OS etc. Testing on different platforms.
423:
1.169 brouard 424: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 425: Summary: update
1.169 brouard 426:
1.168 brouard 427: Revision 1.167 2014/12/22 13:50:56 brouard
428: Summary: Testing uname and compiler version and if compiled 32 or 64
429:
430: Testing on Linux 64
431:
1.167 brouard 432: Revision 1.166 2014/12/22 11:40:47 brouard
433: *** empty log message ***
434:
1.166 brouard 435: Revision 1.165 2014/12/16 11:20:36 brouard
436: Summary: After compiling on Visual C
437:
438: * imach.c (Module): Merging 1.61 to 1.162
439:
1.165 brouard 440: Revision 1.164 2014/12/16 10:52:11 brouard
441: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
442:
443: * imach.c (Module): Merging 1.61 to 1.162
444:
1.164 brouard 445: Revision 1.163 2014/12/16 10:30:11 brouard
446: * imach.c (Module): Merging 1.61 to 1.162
447:
1.163 brouard 448: Revision 1.162 2014/09/25 11:43:39 brouard
449: Summary: temporary backup 0.99!
450:
1.162 brouard 451: Revision 1.1 2014/09/16 11:06:58 brouard
452: Summary: With some code (wrong) for nlopt
453:
454: Author:
455:
456: Revision 1.161 2014/09/15 20:41:41 brouard
457: Summary: Problem with macro SQR on Intel compiler
458:
1.161 brouard 459: Revision 1.160 2014/09/02 09:24:05 brouard
460: *** empty log message ***
461:
1.160 brouard 462: Revision 1.159 2014/09/01 10:34:10 brouard
463: Summary: WIN32
464: Author: Brouard
465:
1.159 brouard 466: Revision 1.158 2014/08/27 17:11:51 brouard
467: *** empty log message ***
468:
1.158 brouard 469: Revision 1.157 2014/08/27 16:26:55 brouard
470: Summary: Preparing windows Visual studio version
471: Author: Brouard
472:
473: In order to compile on Visual studio, time.h is now correct and time_t
474: and tm struct should be used. difftime should be used but sometimes I
475: just make the differences in raw time format (time(&now).
476: Trying to suppress #ifdef LINUX
477: Add xdg-open for __linux in order to open default browser.
478:
1.157 brouard 479: Revision 1.156 2014/08/25 20:10:10 brouard
480: *** empty log message ***
481:
1.156 brouard 482: Revision 1.155 2014/08/25 18:32:34 brouard
483: Summary: New compile, minor changes
484: Author: Brouard
485:
1.155 brouard 486: Revision 1.154 2014/06/20 17:32:08 brouard
487: Summary: Outputs now all graphs of convergence to period prevalence
488:
1.154 brouard 489: Revision 1.153 2014/06/20 16:45:46 brouard
490: Summary: If 3 live state, convergence to period prevalence on same graph
491: Author: Brouard
492:
1.153 brouard 493: Revision 1.152 2014/06/18 17:54:09 brouard
494: Summary: open browser, use gnuplot on same dir than imach if not found in the path
495:
1.152 brouard 496: Revision 1.151 2014/06/18 16:43:30 brouard
497: *** empty log message ***
498:
1.151 brouard 499: Revision 1.150 2014/06/18 16:42:35 brouard
500: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
501: Author: brouard
502:
1.150 brouard 503: Revision 1.149 2014/06/18 15:51:14 brouard
504: Summary: Some fixes in parameter files errors
505: Author: Nicolas Brouard
506:
1.149 brouard 507: Revision 1.148 2014/06/17 17:38:48 brouard
508: Summary: Nothing new
509: Author: Brouard
510:
511: Just a new packaging for OS/X version 0.98nS
512:
1.148 brouard 513: Revision 1.147 2014/06/16 10:33:11 brouard
514: *** empty log message ***
515:
1.147 brouard 516: Revision 1.146 2014/06/16 10:20:28 brouard
517: Summary: Merge
518: Author: Brouard
519:
520: Merge, before building revised version.
521:
1.146 brouard 522: Revision 1.145 2014/06/10 21:23:15 brouard
523: Summary: Debugging with valgrind
524: Author: Nicolas Brouard
525:
526: Lot of changes in order to output the results with some covariates
527: After the Edimburgh REVES conference 2014, it seems mandatory to
528: improve the code.
529: No more memory valgrind error but a lot has to be done in order to
530: continue the work of splitting the code into subroutines.
531: Also, decodemodel has been improved. Tricode is still not
532: optimal. nbcode should be improved. Documentation has been added in
533: the source code.
534:
1.144 brouard 535: Revision 1.143 2014/01/26 09:45:38 brouard
536: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
537:
538: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
539: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
540:
1.143 brouard 541: Revision 1.142 2014/01/26 03:57:36 brouard
542: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
543:
544: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
545:
1.142 brouard 546: Revision 1.141 2014/01/26 02:42:01 brouard
547: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
548:
1.141 brouard 549: Revision 1.140 2011/09/02 10:37:54 brouard
550: Summary: times.h is ok with mingw32 now.
551:
1.140 brouard 552: Revision 1.139 2010/06/14 07:50:17 brouard
553: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
554: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
555:
1.139 brouard 556: Revision 1.138 2010/04/30 18:19:40 brouard
557: *** empty log message ***
558:
1.138 brouard 559: Revision 1.137 2010/04/29 18:11:38 brouard
560: (Module): Checking covariates for more complex models
561: than V1+V2. A lot of change to be done. Unstable.
562:
1.137 brouard 563: Revision 1.136 2010/04/26 20:30:53 brouard
564: (Module): merging some libgsl code. Fixing computation
565: of likelione (using inter/intrapolation if mle = 0) in order to
566: get same likelihood as if mle=1.
567: Some cleaning of code and comments added.
568:
1.136 brouard 569: Revision 1.135 2009/10/29 15:33:14 brouard
570: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
571:
1.135 brouard 572: Revision 1.134 2009/10/29 13:18:53 brouard
573: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
574:
1.134 brouard 575: Revision 1.133 2009/07/06 10:21:25 brouard
576: just nforces
577:
1.133 brouard 578: Revision 1.132 2009/07/06 08:22:05 brouard
579: Many tings
580:
1.132 brouard 581: Revision 1.131 2009/06/20 16:22:47 brouard
582: Some dimensions resccaled
583:
1.131 brouard 584: Revision 1.130 2009/05/26 06:44:34 brouard
585: (Module): Max Covariate is now set to 20 instead of 8. A
586: lot of cleaning with variables initialized to 0. Trying to make
587: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
588:
1.130 brouard 589: Revision 1.129 2007/08/31 13:49:27 lievre
590: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
591:
1.129 lievre 592: Revision 1.128 2006/06/30 13:02:05 brouard
593: (Module): Clarifications on computing e.j
594:
1.128 brouard 595: Revision 1.127 2006/04/28 18:11:50 brouard
596: (Module): Yes the sum of survivors was wrong since
597: imach-114 because nhstepm was no more computed in the age
598: loop. Now we define nhstepma in the age loop.
599: (Module): In order to speed up (in case of numerous covariates) we
600: compute health expectancies (without variances) in a first step
601: and then all the health expectancies with variances or standard
602: deviation (needs data from the Hessian matrices) which slows the
603: computation.
604: In the future we should be able to stop the program is only health
605: expectancies and graph are needed without standard deviations.
606:
1.127 brouard 607: Revision 1.126 2006/04/28 17:23:28 brouard
608: (Module): Yes the sum of survivors was wrong since
609: imach-114 because nhstepm was no more computed in the age
610: loop. Now we define nhstepma in the age loop.
611: Version 0.98h
612:
1.126 brouard 613: Revision 1.125 2006/04/04 15:20:31 lievre
614: Errors in calculation of health expectancies. Age was not initialized.
615: Forecasting file added.
616:
617: Revision 1.124 2006/03/22 17:13:53 lievre
618: Parameters are printed with %lf instead of %f (more numbers after the comma).
619: The log-likelihood is printed in the log file
620:
621: Revision 1.123 2006/03/20 10:52:43 brouard
622: * imach.c (Module): <title> changed, corresponds to .htm file
623: name. <head> headers where missing.
624:
625: * imach.c (Module): Weights can have a decimal point as for
626: English (a comma might work with a correct LC_NUMERIC environment,
627: otherwise the weight is truncated).
628: Modification of warning when the covariates values are not 0 or
629: 1.
630: Version 0.98g
631:
632: Revision 1.122 2006/03/20 09:45:41 brouard
633: (Module): Weights can have a decimal point as for
634: English (a comma might work with a correct LC_NUMERIC environment,
635: otherwise the weight is truncated).
636: Modification of warning when the covariates values are not 0 or
637: 1.
638: Version 0.98g
639:
640: Revision 1.121 2006/03/16 17:45:01 lievre
641: * imach.c (Module): Comments concerning covariates added
642:
643: * imach.c (Module): refinements in the computation of lli if
644: status=-2 in order to have more reliable computation if stepm is
645: not 1 month. Version 0.98f
646:
647: Revision 1.120 2006/03/16 15:10:38 lievre
648: (Module): refinements in the computation of lli if
649: status=-2 in order to have more reliable computation if stepm is
650: not 1 month. Version 0.98f
651:
652: Revision 1.119 2006/03/15 17:42:26 brouard
653: (Module): Bug if status = -2, the loglikelihood was
654: computed as likelihood omitting the logarithm. Version O.98e
655:
656: Revision 1.118 2006/03/14 18:20:07 brouard
657: (Module): varevsij Comments added explaining the second
658: table of variances if popbased=1 .
659: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
660: (Module): Function pstamp added
661: (Module): Version 0.98d
662:
663: Revision 1.117 2006/03/14 17:16:22 brouard
664: (Module): varevsij Comments added explaining the second
665: table of variances if popbased=1 .
666: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
667: (Module): Function pstamp added
668: (Module): Version 0.98d
669:
670: Revision 1.116 2006/03/06 10:29:27 brouard
671: (Module): Variance-covariance wrong links and
672: varian-covariance of ej. is needed (Saito).
673:
674: Revision 1.115 2006/02/27 12:17:45 brouard
675: (Module): One freematrix added in mlikeli! 0.98c
676:
677: Revision 1.114 2006/02/26 12:57:58 brouard
678: (Module): Some improvements in processing parameter
679: filename with strsep.
680:
681: Revision 1.113 2006/02/24 14:20:24 brouard
682: (Module): Memory leaks checks with valgrind and:
683: datafile was not closed, some imatrix were not freed and on matrix
684: allocation too.
685:
686: Revision 1.112 2006/01/30 09:55:26 brouard
687: (Module): Back to gnuplot.exe instead of wgnuplot.exe
688:
689: Revision 1.111 2006/01/25 20:38:18 brouard
690: (Module): Lots of cleaning and bugs added (Gompertz)
691: (Module): Comments can be added in data file. Missing date values
692: can be a simple dot '.'.
693:
694: Revision 1.110 2006/01/25 00:51:50 brouard
695: (Module): Lots of cleaning and bugs added (Gompertz)
696:
697: Revision 1.109 2006/01/24 19:37:15 brouard
698: (Module): Comments (lines starting with a #) are allowed in data.
699:
700: Revision 1.108 2006/01/19 18:05:42 lievre
701: Gnuplot problem appeared...
702: To be fixed
703:
704: Revision 1.107 2006/01/19 16:20:37 brouard
705: Test existence of gnuplot in imach path
706:
707: Revision 1.106 2006/01/19 13:24:36 brouard
708: Some cleaning and links added in html output
709:
710: Revision 1.105 2006/01/05 20:23:19 lievre
711: *** empty log message ***
712:
713: Revision 1.104 2005/09/30 16:11:43 lievre
714: (Module): sump fixed, loop imx fixed, and simplifications.
715: (Module): If the status is missing at the last wave but we know
716: that the person is alive, then we can code his/her status as -2
717: (instead of missing=-1 in earlier versions) and his/her
718: contributions to the likelihood is 1 - Prob of dying from last
719: health status (= 1-p13= p11+p12 in the easiest case of somebody in
720: the healthy state at last known wave). Version is 0.98
721:
722: Revision 1.103 2005/09/30 15:54:49 lievre
723: (Module): sump fixed, loop imx fixed, and simplifications.
724:
725: Revision 1.102 2004/09/15 17:31:30 brouard
726: Add the possibility to read data file including tab characters.
727:
728: Revision 1.101 2004/09/15 10:38:38 brouard
729: Fix on curr_time
730:
731: Revision 1.100 2004/07/12 18:29:06 brouard
732: Add version for Mac OS X. Just define UNIX in Makefile
733:
734: Revision 1.99 2004/06/05 08:57:40 brouard
735: *** empty log message ***
736:
737: Revision 1.98 2004/05/16 15:05:56 brouard
738: New version 0.97 . First attempt to estimate force of mortality
739: directly from the data i.e. without the need of knowing the health
740: state at each age, but using a Gompertz model: log u =a + b*age .
741: This is the basic analysis of mortality and should be done before any
742: other analysis, in order to test if the mortality estimated from the
743: cross-longitudinal survey is different from the mortality estimated
744: from other sources like vital statistic data.
745:
746: The same imach parameter file can be used but the option for mle should be -3.
747:
1.133 brouard 748: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 749: former routines in order to include the new code within the former code.
750:
751: The output is very simple: only an estimate of the intercept and of
752: the slope with 95% confident intervals.
753:
754: Current limitations:
755: A) Even if you enter covariates, i.e. with the
756: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
757: B) There is no computation of Life Expectancy nor Life Table.
758:
759: Revision 1.97 2004/02/20 13:25:42 lievre
760: Version 0.96d. Population forecasting command line is (temporarily)
761: suppressed.
762:
763: Revision 1.96 2003/07/15 15:38:55 brouard
764: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
765: rewritten within the same printf. Workaround: many printfs.
766:
767: Revision 1.95 2003/07/08 07:54:34 brouard
768: * imach.c (Repository):
769: (Repository): Using imachwizard code to output a more meaningful covariance
770: matrix (cov(a12,c31) instead of numbers.
771:
772: Revision 1.94 2003/06/27 13:00:02 brouard
773: Just cleaning
774:
775: Revision 1.93 2003/06/25 16:33:55 brouard
776: (Module): On windows (cygwin) function asctime_r doesn't
777: exist so I changed back to asctime which exists.
778: (Module): Version 0.96b
779:
780: Revision 1.92 2003/06/25 16:30:45 brouard
781: (Module): On windows (cygwin) function asctime_r doesn't
782: exist so I changed back to asctime which exists.
783:
784: Revision 1.91 2003/06/25 15:30:29 brouard
785: * imach.c (Repository): Duplicated warning errors corrected.
786: (Repository): Elapsed time after each iteration is now output. It
787: helps to forecast when convergence will be reached. Elapsed time
788: is stamped in powell. We created a new html file for the graphs
789: concerning matrix of covariance. It has extension -cov.htm.
790:
791: Revision 1.90 2003/06/24 12:34:15 brouard
792: (Module): Some bugs corrected for windows. Also, when
793: mle=-1 a template is output in file "or"mypar.txt with the design
794: of the covariance matrix to be input.
795:
796: Revision 1.89 2003/06/24 12:30:52 brouard
797: (Module): Some bugs corrected for windows. Also, when
798: mle=-1 a template is output in file "or"mypar.txt with the design
799: of the covariance matrix to be input.
800:
801: Revision 1.88 2003/06/23 17:54:56 brouard
802: * 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.
803:
804: Revision 1.87 2003/06/18 12:26:01 brouard
805: Version 0.96
806:
807: Revision 1.86 2003/06/17 20:04:08 brouard
808: (Module): Change position of html and gnuplot routines and added
809: routine fileappend.
810:
811: Revision 1.85 2003/06/17 13:12:43 brouard
812: * imach.c (Repository): Check when date of death was earlier that
813: current date of interview. It may happen when the death was just
814: prior to the death. In this case, dh was negative and likelihood
815: was wrong (infinity). We still send an "Error" but patch by
816: assuming that the date of death was just one stepm after the
817: interview.
818: (Repository): Because some people have very long ID (first column)
819: we changed int to long in num[] and we added a new lvector for
820: memory allocation. But we also truncated to 8 characters (left
821: truncation)
822: (Repository): No more line truncation errors.
823:
824: Revision 1.84 2003/06/13 21:44:43 brouard
825: * imach.c (Repository): Replace "freqsummary" at a correct
826: place. It differs from routine "prevalence" which may be called
827: many times. Probs is memory consuming and must be used with
828: parcimony.
829: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
830:
831: Revision 1.83 2003/06/10 13:39:11 lievre
832: *** empty log message ***
833:
834: Revision 1.82 2003/06/05 15:57:20 brouard
835: Add log in imach.c and fullversion number is now printed.
836:
837: */
838: /*
839: Interpolated Markov Chain
840:
841: Short summary of the programme:
842:
1.227 brouard 843: This program computes Healthy Life Expectancies or State-specific
844: (if states aren't health statuses) Expectancies from
845: cross-longitudinal data. Cross-longitudinal data consist in:
846:
847: -1- a first survey ("cross") where individuals from different ages
848: are interviewed on their health status or degree of disability (in
849: the case of a health survey which is our main interest)
850:
851: -2- at least a second wave of interviews ("longitudinal") which
852: measure each change (if any) in individual health status. Health
853: expectancies are computed from the time spent in each health state
854: according to a model. More health states you consider, more time is
855: necessary to reach the Maximum Likelihood of the parameters involved
856: in the model. The simplest model is the multinomial logistic model
857: where pij is the probability to be observed in state j at the second
858: wave conditional to be observed in state i at the first
859: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
860: etc , where 'age' is age and 'sex' is a covariate. If you want to
861: have a more complex model than "constant and age", you should modify
862: the program where the markup *Covariates have to be included here
863: again* invites you to do it. More covariates you add, slower the
1.126 brouard 864: convergence.
865:
866: The advantage of this computer programme, compared to a simple
867: multinomial logistic model, is clear when the delay between waves is not
868: identical for each individual. Also, if a individual missed an
869: intermediate interview, the information is lost, but taken into
870: account using an interpolation or extrapolation.
871:
872: hPijx is the probability to be observed in state i at age x+h
873: conditional to the observed state i at age x. The delay 'h' can be
874: split into an exact number (nh*stepm) of unobserved intermediate
875: states. This elementary transition (by month, quarter,
876: semester or year) is modelled as a multinomial logistic. The hPx
877: matrix is simply the matrix product of nh*stepm elementary matrices
878: and the contribution of each individual to the likelihood is simply
879: hPijx.
880:
881: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 882: of the life expectancies. It also computes the period (stable) prevalence.
883:
884: Back prevalence and projections:
1.227 brouard 885:
886: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
887: double agemaxpar, double ftolpl, int *ncvyearp, double
888: dateprev1,double dateprev2, int firstpass, int lastpass, int
889: mobilavproj)
890:
891: Computes the back prevalence limit for any combination of
892: covariate values k at any age between ageminpar and agemaxpar and
893: returns it in **bprlim. In the loops,
894:
895: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
896: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
897:
898: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 899: Computes for any combination of covariates k and any age between bage and fage
900: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
901: oldm=oldms;savm=savms;
1.227 brouard 902:
1.267 brouard 903: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 904: Computes the transition matrix starting at age 'age' over
905: 'nhstepm*hstepm*stepm' months (i.e. until
906: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 907: nhstepm*hstepm matrices.
908:
909: Returns p3mat[i][j][h] after calling
910: p3mat[i][j][h]=matprod2(newm,
911: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
912: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
913: oldm);
1.226 brouard 914:
915: Important routines
916:
917: - func (or funcone), computes logit (pij) distinguishing
918: o fixed variables (single or product dummies or quantitative);
919: o varying variables by:
920: (1) wave (single, product dummies, quantitative),
921: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
922: % fixed dummy (treated) or quantitative (not done because time-consuming);
923: % varying dummy (not done) or quantitative (not done);
924: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
925: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
926: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
927: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
928: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 929:
1.226 brouard 930:
931:
1.133 brouard 932: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
933: Institut national d'études démographiques, Paris.
1.126 brouard 934: This software have been partly granted by Euro-REVES, a concerted action
935: from the European Union.
936: It is copyrighted identically to a GNU software product, ie programme and
937: software can be distributed freely for non commercial use. Latest version
938: can be accessed at http://euroreves.ined.fr/imach .
939:
940: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
941: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
942:
943: **********************************************************************/
944: /*
945: main
946: read parameterfile
947: read datafile
948: concatwav
949: freqsummary
950: if (mle >= 1)
951: mlikeli
952: print results files
953: if mle==1
954: computes hessian
955: read end of parameter file: agemin, agemax, bage, fage, estepm
956: begin-prev-date,...
957: open gnuplot file
958: open html file
1.145 brouard 959: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
960: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
961: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
962: freexexit2 possible for memory heap.
963:
964: h Pij x | pij_nom ficrestpij
965: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
966: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
967: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
968:
969: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
970: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
971: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
972: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
973: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
974:
1.126 brouard 975: forecasting if prevfcast==1 prevforecast call prevalence()
976: health expectancies
977: Variance-covariance of DFLE
978: prevalence()
979: movingaverage()
980: varevsij()
981: if popbased==1 varevsij(,popbased)
982: total life expectancies
983: Variance of period (stable) prevalence
984: end
985: */
986:
1.187 brouard 987: /* #define DEBUG */
988: /* #define DEBUGBRENT */
1.203 brouard 989: /* #define DEBUGLINMIN */
990: /* #define DEBUGHESS */
991: #define DEBUGHESSIJ
1.224 brouard 992: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 993: #define POWELL /* Instead of NLOPT */
1.224 brouard 994: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 995: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
996: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 997:
998: #include <math.h>
999: #include <stdio.h>
1000: #include <stdlib.h>
1001: #include <string.h>
1.226 brouard 1002: #include <ctype.h>
1.159 brouard 1003:
1004: #ifdef _WIN32
1005: #include <io.h>
1.172 brouard 1006: #include <windows.h>
1007: #include <tchar.h>
1.159 brouard 1008: #else
1.126 brouard 1009: #include <unistd.h>
1.159 brouard 1010: #endif
1.126 brouard 1011:
1012: #include <limits.h>
1013: #include <sys/types.h>
1.171 brouard 1014:
1015: #if defined(__GNUC__)
1016: #include <sys/utsname.h> /* Doesn't work on Windows */
1017: #endif
1018:
1.126 brouard 1019: #include <sys/stat.h>
1020: #include <errno.h>
1.159 brouard 1021: /* extern int errno; */
1.126 brouard 1022:
1.157 brouard 1023: /* #ifdef LINUX */
1024: /* #include <time.h> */
1025: /* #include "timeval.h" */
1026: /* #else */
1027: /* #include <sys/time.h> */
1028: /* #endif */
1029:
1.126 brouard 1030: #include <time.h>
1031:
1.136 brouard 1032: #ifdef GSL
1033: #include <gsl/gsl_errno.h>
1034: #include <gsl/gsl_multimin.h>
1035: #endif
1036:
1.167 brouard 1037:
1.162 brouard 1038: #ifdef NLOPT
1039: #include <nlopt.h>
1040: typedef struct {
1041: double (* function)(double [] );
1042: } myfunc_data ;
1043: #endif
1044:
1.126 brouard 1045: /* #include <libintl.h> */
1046: /* #define _(String) gettext (String) */
1047:
1.251 brouard 1048: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1049:
1050: #define GNUPLOTPROGRAM "gnuplot"
1051: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1052: #define FILENAMELENGTH 132
1053:
1054: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1055: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1056:
1.144 brouard 1057: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1058: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1059:
1060: #define NINTERVMAX 8
1.144 brouard 1061: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1062: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1063: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1064: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1065: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1066: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1067: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1068: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1069: /* #define AGESUP 130 */
1.288 brouard 1070: /* #define AGESUP 150 */
1071: #define AGESUP 200
1.268 brouard 1072: #define AGEINF 0
1.218 brouard 1073: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1074: #define AGEBASE 40
1.194 brouard 1075: #define AGEOVERFLOW 1.e20
1.164 brouard 1076: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1077: #ifdef _WIN32
1078: #define DIRSEPARATOR '\\'
1079: #define CHARSEPARATOR "\\"
1080: #define ODIRSEPARATOR '/'
1081: #else
1.126 brouard 1082: #define DIRSEPARATOR '/'
1083: #define CHARSEPARATOR "/"
1084: #define ODIRSEPARATOR '\\'
1085: #endif
1086:
1.293 ! brouard 1087: /* $Id: imach.c,v 1.292 2019/05/09 14:17:20 brouard Exp $ */
1.126 brouard 1088: /* $State: Exp $ */
1.196 brouard 1089: #include "version.h"
1090: char version[]=__IMACH_VERSION__;
1.283 brouard 1091: 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";
1.293 ! brouard 1092: char fullversion[]="$Revision: 1.292 $ $Date: 2019/05/09 14:17:20 $";
1.126 brouard 1093: char strstart[80];
1094: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1095: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1096: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1097: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1098: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1099: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1100: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1101: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1102: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1103: int cptcovprodnoage=0; /**< Number of covariate products without age */
1104: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1105: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1106: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1107: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1108: int nsd=0; /**< Total number of single dummy variables (output) */
1109: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1110: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1111: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1112: int ntveff=0; /**< ntveff number of effective time varying variables */
1113: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1114: int cptcov=0; /* Working variable */
1.290 brouard 1115: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1116: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1117: int npar=NPARMAX;
1118: int nlstate=2; /* Number of live states */
1119: int ndeath=1; /* Number of dead states */
1.130 brouard 1120: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1121: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1122: int popbased=0;
1123:
1124: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1125: int maxwav=0; /* Maxim number of waves */
1126: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1127: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1128: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1129: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1130: int mle=1, weightopt=0;
1.126 brouard 1131: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1132: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1133: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1134: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1135: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1136: int selected(int kvar); /* Is covariate kvar selected for printing results */
1137:
1.130 brouard 1138: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1139: double **matprod2(); /* test */
1.126 brouard 1140: double **oldm, **newm, **savm; /* Working pointers to matrices */
1141: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1142: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1143:
1.136 brouard 1144: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1145: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1146: FILE *ficlog, *ficrespow;
1.130 brouard 1147: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1148: double fretone; /* Only one call to likelihood */
1.130 brouard 1149: long ipmx=0; /* Number of contributions */
1.126 brouard 1150: double sw; /* Sum of weights */
1151: char filerespow[FILENAMELENGTH];
1152: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1153: FILE *ficresilk;
1154: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1155: FILE *ficresprobmorprev;
1156: FILE *fichtm, *fichtmcov; /* Html File */
1157: FILE *ficreseij;
1158: char filerese[FILENAMELENGTH];
1159: FILE *ficresstdeij;
1160: char fileresstde[FILENAMELENGTH];
1161: FILE *ficrescveij;
1162: char filerescve[FILENAMELENGTH];
1163: FILE *ficresvij;
1164: char fileresv[FILENAMELENGTH];
1.269 brouard 1165:
1.126 brouard 1166: char title[MAXLINE];
1.234 brouard 1167: char model[MAXLINE]; /**< The model line */
1.217 brouard 1168: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1169: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1170: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1171: char command[FILENAMELENGTH];
1172: int outcmd=0;
1173:
1.217 brouard 1174: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1175: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1176: char filelog[FILENAMELENGTH]; /* Log file */
1177: char filerest[FILENAMELENGTH];
1178: char fileregp[FILENAMELENGTH];
1179: char popfile[FILENAMELENGTH];
1180:
1181: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1182:
1.157 brouard 1183: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1184: /* struct timezone tzp; */
1185: /* extern int gettimeofday(); */
1186: struct tm tml, *gmtime(), *localtime();
1187:
1188: extern time_t time();
1189:
1190: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1191: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1192: struct tm tm;
1193:
1.126 brouard 1194: char strcurr[80], strfor[80];
1195:
1196: char *endptr;
1197: long lval;
1198: double dval;
1199:
1200: #define NR_END 1
1201: #define FREE_ARG char*
1202: #define FTOL 1.0e-10
1203:
1204: #define NRANSI
1.240 brouard 1205: #define ITMAX 200
1206: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1207:
1208: #define TOL 2.0e-4
1209:
1210: #define CGOLD 0.3819660
1211: #define ZEPS 1.0e-10
1212: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1213:
1214: #define GOLD 1.618034
1215: #define GLIMIT 100.0
1216: #define TINY 1.0e-20
1217:
1218: static double maxarg1,maxarg2;
1219: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1220: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1221:
1222: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1223: #define rint(a) floor(a+0.5)
1.166 brouard 1224: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1225: #define mytinydouble 1.0e-16
1.166 brouard 1226: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1227: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1228: /* static double dsqrarg; */
1229: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1230: static double sqrarg;
1231: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1232: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1233: int agegomp= AGEGOMP;
1234:
1235: int imx;
1236: int stepm=1;
1237: /* Stepm, step in month: minimum step interpolation*/
1238:
1239: int estepm;
1240: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1241:
1242: int m,nb;
1243: long *num;
1.197 brouard 1244: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1245: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1246: covariate for which somebody answered excluding
1247: undefined. Usually 2: 0 and 1. */
1248: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1249: covariate for which somebody answered including
1250: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1251: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1252: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1253: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1254: double *ageexmed,*agecens;
1255: double dateintmean=0;
1256:
1257: double *weight;
1258: int **s; /* Status */
1.141 brouard 1259: double *agedc;
1.145 brouard 1260: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1261: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1262: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1263: double **coqvar; /* Fixed quantitative covariate nqv */
1264: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1265: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1266: double idx;
1267: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1268: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1269: /*k 1 2 3 4 5 6 7 8 9 */
1270: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1271: /* Tndvar[k] 1 2 3 4 5 */
1272: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1273: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1274: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1275: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1276: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1277: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1278: /* Tprod[i]=k 4 7 */
1279: /* Tage[i]=k 5 8 */
1280: /* */
1281: /* Type */
1282: /* V 1 2 3 4 5 */
1283: /* F F V V V */
1284: /* D Q D D Q */
1285: /* */
1286: int *TvarsD;
1287: int *TvarsDind;
1288: int *TvarsQ;
1289: int *TvarsQind;
1290:
1.235 brouard 1291: #define MAXRESULTLINES 10
1292: int nresult=0;
1.258 brouard 1293: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1294: int TKresult[MAXRESULTLINES];
1.237 brouard 1295: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1296: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1297: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1298: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1299: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1300: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1301:
1.234 brouard 1302: /* 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 1303: 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 */
1304: 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 */
1305: 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 */
1306: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1307: 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 */
1308: 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 1309: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1310: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1311: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1312: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1313: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1314: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1315: 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 */
1316: 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 */
1317:
1.230 brouard 1318: int *Tvarsel; /**< Selected covariates for output */
1319: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1320: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1321: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1322: 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 1323: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1324: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1325: int *Tage;
1.227 brouard 1326: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1327: 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 1328: 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*/
1329: 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 1330: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1331: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1332: int **Tvard;
1333: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1334: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1335: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1336: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1337: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1338: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1339: double *lsurv, *lpop, *tpop;
1340:
1.231 brouard 1341: #define FD 1; /* Fixed dummy covariate */
1342: #define FQ 2; /* Fixed quantitative covariate */
1343: #define FP 3; /* Fixed product covariate */
1344: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1345: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1346: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1347: #define VD 10; /* Varying dummy covariate */
1348: #define VQ 11; /* Varying quantitative covariate */
1349: #define VP 12; /* Varying product covariate */
1350: #define VPDD 13; /* Varying product dummy*dummy covariate */
1351: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1352: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1353: #define APFD 16; /* Age product * fixed dummy covariate */
1354: #define APFQ 17; /* Age product * fixed quantitative covariate */
1355: #define APVD 18; /* Age product * varying dummy covariate */
1356: #define APVQ 19; /* Age product * varying quantitative covariate */
1357:
1358: #define FTYPE 1; /* Fixed covariate */
1359: #define VTYPE 2; /* Varying covariate (loop in wave) */
1360: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1361:
1362: struct kmodel{
1363: int maintype; /* main type */
1364: int subtype; /* subtype */
1365: };
1366: struct kmodel modell[NCOVMAX];
1367:
1.143 brouard 1368: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1369: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1370:
1371: /**************** split *************************/
1372: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1373: {
1374: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1375: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1376: */
1377: char *ss; /* pointer */
1.186 brouard 1378: int l1=0, l2=0; /* length counters */
1.126 brouard 1379:
1380: l1 = strlen(path ); /* length of path */
1381: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1382: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1383: if ( ss == NULL ) { /* no directory, so determine current directory */
1384: strcpy( name, path ); /* we got the fullname name because no directory */
1385: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1386: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1387: /* get current working directory */
1388: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1389: #ifdef WIN32
1390: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1391: #else
1392: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1393: #endif
1.126 brouard 1394: return( GLOCK_ERROR_GETCWD );
1395: }
1396: /* got dirc from getcwd*/
1397: printf(" DIRC = %s \n",dirc);
1.205 brouard 1398: } else { /* strip directory from path */
1.126 brouard 1399: ss++; /* after this, the filename */
1400: l2 = strlen( ss ); /* length of filename */
1401: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1402: strcpy( name, ss ); /* save file name */
1403: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1404: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1405: printf(" DIRC2 = %s \n",dirc);
1406: }
1407: /* We add a separator at the end of dirc if not exists */
1408: l1 = strlen( dirc ); /* length of directory */
1409: if( dirc[l1-1] != DIRSEPARATOR ){
1410: dirc[l1] = DIRSEPARATOR;
1411: dirc[l1+1] = 0;
1412: printf(" DIRC3 = %s \n",dirc);
1413: }
1414: ss = strrchr( name, '.' ); /* find last / */
1415: if (ss >0){
1416: ss++;
1417: strcpy(ext,ss); /* save extension */
1418: l1= strlen( name);
1419: l2= strlen(ss)+1;
1420: strncpy( finame, name, l1-l2);
1421: finame[l1-l2]= 0;
1422: }
1423:
1424: return( 0 ); /* we're done */
1425: }
1426:
1427:
1428: /******************************************/
1429:
1430: void replace_back_to_slash(char *s, char*t)
1431: {
1432: int i;
1433: int lg=0;
1434: i=0;
1435: lg=strlen(t);
1436: for(i=0; i<= lg; i++) {
1437: (s[i] = t[i]);
1438: if (t[i]== '\\') s[i]='/';
1439: }
1440: }
1441:
1.132 brouard 1442: char *trimbb(char *out, char *in)
1.137 brouard 1443: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1444: char *s;
1445: s=out;
1446: while (*in != '\0'){
1.137 brouard 1447: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1448: in++;
1449: }
1450: *out++ = *in++;
1451: }
1452: *out='\0';
1453: return s;
1454: }
1455:
1.187 brouard 1456: /* char *substrchaine(char *out, char *in, char *chain) */
1457: /* { */
1458: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1459: /* char *s, *t; */
1460: /* t=in;s=out; */
1461: /* while ((*in != *chain) && (*in != '\0')){ */
1462: /* *out++ = *in++; */
1463: /* } */
1464:
1465: /* /\* *in matches *chain *\/ */
1466: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1467: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1468: /* } */
1469: /* in--; chain--; */
1470: /* while ( (*in != '\0')){ */
1471: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1472: /* *out++ = *in++; */
1473: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1474: /* } */
1475: /* *out='\0'; */
1476: /* out=s; */
1477: /* return out; */
1478: /* } */
1479: char *substrchaine(char *out, char *in, char *chain)
1480: {
1481: /* Substract chain 'chain' from 'in', return and output 'out' */
1482: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1483:
1484: char *strloc;
1485:
1486: strcpy (out, in);
1487: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1488: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1489: if(strloc != NULL){
1490: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1491: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1492: /* strcpy (strloc, strloc +strlen(chain));*/
1493: }
1494: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1495: return out;
1496: }
1497:
1498:
1.145 brouard 1499: char *cutl(char *blocc, char *alocc, char *in, char occ)
1500: {
1.187 brouard 1501: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1502: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1503: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1504: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1505: */
1.160 brouard 1506: char *s, *t;
1.145 brouard 1507: t=in;s=in;
1508: while ((*in != occ) && (*in != '\0')){
1509: *alocc++ = *in++;
1510: }
1511: if( *in == occ){
1512: *(alocc)='\0';
1513: s=++in;
1514: }
1515:
1516: if (s == t) {/* occ not found */
1517: *(alocc-(in-s))='\0';
1518: in=s;
1519: }
1520: while ( *in != '\0'){
1521: *blocc++ = *in++;
1522: }
1523:
1524: *blocc='\0';
1525: return t;
1526: }
1.137 brouard 1527: char *cutv(char *blocc, char *alocc, char *in, char occ)
1528: {
1.187 brouard 1529: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1530: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1531: gives blocc="abcdef2ghi" and alocc="j".
1532: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1533: */
1534: char *s, *t;
1535: t=in;s=in;
1536: while (*in != '\0'){
1537: while( *in == occ){
1538: *blocc++ = *in++;
1539: s=in;
1540: }
1541: *blocc++ = *in++;
1542: }
1543: if (s == t) /* occ not found */
1544: *(blocc-(in-s))='\0';
1545: else
1546: *(blocc-(in-s)-1)='\0';
1547: in=s;
1548: while ( *in != '\0'){
1549: *alocc++ = *in++;
1550: }
1551:
1552: *alocc='\0';
1553: return s;
1554: }
1555:
1.126 brouard 1556: int nbocc(char *s, char occ)
1557: {
1558: int i,j=0;
1559: int lg=20;
1560: i=0;
1561: lg=strlen(s);
1562: for(i=0; i<= lg; i++) {
1.234 brouard 1563: if (s[i] == occ ) j++;
1.126 brouard 1564: }
1565: return j;
1566: }
1567:
1.137 brouard 1568: /* void cutv(char *u,char *v, char*t, char occ) */
1569: /* { */
1570: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1571: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1572: /* gives u="abcdef2ghi" and v="j" *\/ */
1573: /* int i,lg,j,p=0; */
1574: /* i=0; */
1575: /* lg=strlen(t); */
1576: /* for(j=0; j<=lg-1; j++) { */
1577: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1578: /* } */
1.126 brouard 1579:
1.137 brouard 1580: /* for(j=0; j<p; j++) { */
1581: /* (u[j] = t[j]); */
1582: /* } */
1583: /* u[p]='\0'; */
1.126 brouard 1584:
1.137 brouard 1585: /* for(j=0; j<= lg; j++) { */
1586: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1587: /* } */
1588: /* } */
1.126 brouard 1589:
1.160 brouard 1590: #ifdef _WIN32
1591: char * strsep(char **pp, const char *delim)
1592: {
1593: char *p, *q;
1594:
1595: if ((p = *pp) == NULL)
1596: return 0;
1597: if ((q = strpbrk (p, delim)) != NULL)
1598: {
1599: *pp = q + 1;
1600: *q = '\0';
1601: }
1602: else
1603: *pp = 0;
1604: return p;
1605: }
1606: #endif
1607:
1.126 brouard 1608: /********************** nrerror ********************/
1609:
1610: void nrerror(char error_text[])
1611: {
1612: fprintf(stderr,"ERREUR ...\n");
1613: fprintf(stderr,"%s\n",error_text);
1614: exit(EXIT_FAILURE);
1615: }
1616: /*********************** vector *******************/
1617: double *vector(int nl, int nh)
1618: {
1619: double *v;
1620: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1621: if (!v) nrerror("allocation failure in vector");
1622: return v-nl+NR_END;
1623: }
1624:
1625: /************************ free vector ******************/
1626: void free_vector(double*v, int nl, int nh)
1627: {
1628: free((FREE_ARG)(v+nl-NR_END));
1629: }
1630:
1631: /************************ivector *******************************/
1632: int *ivector(long nl,long nh)
1633: {
1634: int *v;
1635: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1636: if (!v) nrerror("allocation failure in ivector");
1637: return v-nl+NR_END;
1638: }
1639:
1640: /******************free ivector **************************/
1641: void free_ivector(int *v, long nl, long nh)
1642: {
1643: free((FREE_ARG)(v+nl-NR_END));
1644: }
1645:
1646: /************************lvector *******************************/
1647: long *lvector(long nl,long nh)
1648: {
1649: long *v;
1650: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1651: if (!v) nrerror("allocation failure in ivector");
1652: return v-nl+NR_END;
1653: }
1654:
1655: /******************free lvector **************************/
1656: void free_lvector(long *v, long nl, long nh)
1657: {
1658: free((FREE_ARG)(v+nl-NR_END));
1659: }
1660:
1661: /******************* imatrix *******************************/
1662: int **imatrix(long nrl, long nrh, long ncl, long nch)
1663: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1664: {
1665: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1666: int **m;
1667:
1668: /* allocate pointers to rows */
1669: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1670: if (!m) nrerror("allocation failure 1 in matrix()");
1671: m += NR_END;
1672: m -= nrl;
1673:
1674:
1675: /* allocate rows and set pointers to them */
1676: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1677: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1678: m[nrl] += NR_END;
1679: m[nrl] -= ncl;
1680:
1681: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1682:
1683: /* return pointer to array of pointers to rows */
1684: return m;
1685: }
1686:
1687: /****************** free_imatrix *************************/
1688: void free_imatrix(m,nrl,nrh,ncl,nch)
1689: int **m;
1690: long nch,ncl,nrh,nrl;
1691: /* free an int matrix allocated by imatrix() */
1692: {
1693: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1694: free((FREE_ARG) (m+nrl-NR_END));
1695: }
1696:
1697: /******************* matrix *******************************/
1698: double **matrix(long nrl, long nrh, long ncl, long nch)
1699: {
1700: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1701: double **m;
1702:
1703: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1704: if (!m) nrerror("allocation failure 1 in matrix()");
1705: m += NR_END;
1706: m -= nrl;
1707:
1708: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1709: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1710: m[nrl] += NR_END;
1711: m[nrl] -= ncl;
1712:
1713: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1714: return m;
1.145 brouard 1715: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1716: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1717: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1718: */
1719: }
1720:
1721: /*************************free matrix ************************/
1722: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1723: {
1724: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1725: free((FREE_ARG)(m+nrl-NR_END));
1726: }
1727:
1728: /******************* ma3x *******************************/
1729: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1730: {
1731: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1732: double ***m;
1733:
1734: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1735: if (!m) nrerror("allocation failure 1 in matrix()");
1736: m += NR_END;
1737: m -= nrl;
1738:
1739: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1740: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1741: m[nrl] += NR_END;
1742: m[nrl] -= ncl;
1743:
1744: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1745:
1746: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1747: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1748: m[nrl][ncl] += NR_END;
1749: m[nrl][ncl] -= nll;
1750: for (j=ncl+1; j<=nch; j++)
1751: m[nrl][j]=m[nrl][j-1]+nlay;
1752:
1753: for (i=nrl+1; i<=nrh; i++) {
1754: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1755: for (j=ncl+1; j<=nch; j++)
1756: m[i][j]=m[i][j-1]+nlay;
1757: }
1758: return m;
1759: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1760: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1761: */
1762: }
1763:
1764: /*************************free ma3x ************************/
1765: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1766: {
1767: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1768: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1769: free((FREE_ARG)(m+nrl-NR_END));
1770: }
1771:
1772: /*************** function subdirf ***********/
1773: char *subdirf(char fileres[])
1774: {
1775: /* Caution optionfilefiname is hidden */
1776: strcpy(tmpout,optionfilefiname);
1777: strcat(tmpout,"/"); /* Add to the right */
1778: strcat(tmpout,fileres);
1779: return tmpout;
1780: }
1781:
1782: /*************** function subdirf2 ***********/
1783: char *subdirf2(char fileres[], char *preop)
1784: {
1785:
1786: /* Caution optionfilefiname is hidden */
1787: strcpy(tmpout,optionfilefiname);
1788: strcat(tmpout,"/");
1789: strcat(tmpout,preop);
1790: strcat(tmpout,fileres);
1791: return tmpout;
1792: }
1793:
1794: /*************** function subdirf3 ***********/
1795: char *subdirf3(char fileres[], char *preop, char *preop2)
1796: {
1797:
1798: /* Caution optionfilefiname is hidden */
1799: strcpy(tmpout,optionfilefiname);
1800: strcat(tmpout,"/");
1801: strcat(tmpout,preop);
1802: strcat(tmpout,preop2);
1803: strcat(tmpout,fileres);
1804: return tmpout;
1805: }
1.213 brouard 1806:
1807: /*************** function subdirfext ***********/
1808: char *subdirfext(char fileres[], char *preop, char *postop)
1809: {
1810:
1811: strcpy(tmpout,preop);
1812: strcat(tmpout,fileres);
1813: strcat(tmpout,postop);
1814: return tmpout;
1815: }
1.126 brouard 1816:
1.213 brouard 1817: /*************** function subdirfext3 ***********/
1818: char *subdirfext3(char fileres[], char *preop, char *postop)
1819: {
1820:
1821: /* Caution optionfilefiname is hidden */
1822: strcpy(tmpout,optionfilefiname);
1823: strcat(tmpout,"/");
1824: strcat(tmpout,preop);
1825: strcat(tmpout,fileres);
1826: strcat(tmpout,postop);
1827: return tmpout;
1828: }
1829:
1.162 brouard 1830: char *asc_diff_time(long time_sec, char ascdiff[])
1831: {
1832: long sec_left, days, hours, minutes;
1833: days = (time_sec) / (60*60*24);
1834: sec_left = (time_sec) % (60*60*24);
1835: hours = (sec_left) / (60*60) ;
1836: sec_left = (sec_left) %(60*60);
1837: minutes = (sec_left) /60;
1838: sec_left = (sec_left) % (60);
1839: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1840: return ascdiff;
1841: }
1842:
1.126 brouard 1843: /***************** f1dim *************************/
1844: extern int ncom;
1845: extern double *pcom,*xicom;
1846: extern double (*nrfunc)(double []);
1847:
1848: double f1dim(double x)
1849: {
1850: int j;
1851: double f;
1852: double *xt;
1853:
1854: xt=vector(1,ncom);
1855: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1856: f=(*nrfunc)(xt);
1857: free_vector(xt,1,ncom);
1858: return f;
1859: }
1860:
1861: /*****************brent *************************/
1862: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1863: {
1864: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1865: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1866: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1867: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1868: * returned function value.
1869: */
1.126 brouard 1870: int iter;
1871: double a,b,d,etemp;
1.159 brouard 1872: double fu=0,fv,fw,fx;
1.164 brouard 1873: double ftemp=0.;
1.126 brouard 1874: double p,q,r,tol1,tol2,u,v,w,x,xm;
1875: double e=0.0;
1876:
1877: a=(ax < cx ? ax : cx);
1878: b=(ax > cx ? ax : cx);
1879: x=w=v=bx;
1880: fw=fv=fx=(*f)(x);
1881: for (iter=1;iter<=ITMAX;iter++) {
1882: xm=0.5*(a+b);
1883: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1884: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1885: printf(".");fflush(stdout);
1886: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1887: #ifdef DEBUGBRENT
1.126 brouard 1888: 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);
1889: 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);
1890: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1891: #endif
1892: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1893: *xmin=x;
1894: return fx;
1895: }
1896: ftemp=fu;
1897: if (fabs(e) > tol1) {
1898: r=(x-w)*(fx-fv);
1899: q=(x-v)*(fx-fw);
1900: p=(x-v)*q-(x-w)*r;
1901: q=2.0*(q-r);
1902: if (q > 0.0) p = -p;
1903: q=fabs(q);
1904: etemp=e;
1905: e=d;
1906: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1907: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1908: else {
1.224 brouard 1909: d=p/q;
1910: u=x+d;
1911: if (u-a < tol2 || b-u < tol2)
1912: d=SIGN(tol1,xm-x);
1.126 brouard 1913: }
1914: } else {
1915: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1916: }
1917: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1918: fu=(*f)(u);
1919: if (fu <= fx) {
1920: if (u >= x) a=x; else b=x;
1921: SHFT(v,w,x,u)
1.183 brouard 1922: SHFT(fv,fw,fx,fu)
1923: } else {
1924: if (u < x) a=u; else b=u;
1925: if (fu <= fw || w == x) {
1.224 brouard 1926: v=w;
1927: w=u;
1928: fv=fw;
1929: fw=fu;
1.183 brouard 1930: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1931: v=u;
1932: fv=fu;
1.183 brouard 1933: }
1934: }
1.126 brouard 1935: }
1936: nrerror("Too many iterations in brent");
1937: *xmin=x;
1938: return fx;
1939: }
1940:
1941: /****************** mnbrak ***********************/
1942:
1943: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1944: double (*func)(double))
1.183 brouard 1945: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1946: the downhill direction (defined by the function as evaluated at the initial points) and returns
1947: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1948: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1949: */
1.126 brouard 1950: double ulim,u,r,q, dum;
1951: double fu;
1.187 brouard 1952:
1953: double scale=10.;
1954: int iterscale=0;
1955:
1956: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1957: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1958:
1959:
1960: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1961: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1962: /* *bx = *ax - (*ax - *bx)/scale; */
1963: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1964: /* } */
1965:
1.126 brouard 1966: if (*fb > *fa) {
1967: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1968: SHFT(dum,*fb,*fa,dum)
1969: }
1.126 brouard 1970: *cx=(*bx)+GOLD*(*bx-*ax);
1971: *fc=(*func)(*cx);
1.183 brouard 1972: #ifdef DEBUG
1.224 brouard 1973: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1974: 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 1975: #endif
1.224 brouard 1976: 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 1977: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1978: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1979: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1980: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1981: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1982: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1983: fu=(*func)(u);
1.163 brouard 1984: #ifdef DEBUG
1985: /* f(x)=A(x-u)**2+f(u) */
1986: double A, fparabu;
1987: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1988: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1989: 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);
1990: 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 1991: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1992: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1993: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1994: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1995: #endif
1.184 brouard 1996: #ifdef MNBRAKORIGINAL
1.183 brouard 1997: #else
1.191 brouard 1998: /* if (fu > *fc) { */
1999: /* #ifdef DEBUG */
2000: /* printf("mnbrak4 fu > fc \n"); */
2001: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2002: /* #endif */
2003: /* /\* 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 *\\/ *\/ */
2004: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2005: /* dum=u; /\* Shifting c and u *\/ */
2006: /* u = *cx; */
2007: /* *cx = dum; */
2008: /* dum = fu; */
2009: /* fu = *fc; */
2010: /* *fc =dum; */
2011: /* } else { /\* end *\/ */
2012: /* #ifdef DEBUG */
2013: /* printf("mnbrak3 fu < fc \n"); */
2014: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2015: /* #endif */
2016: /* dum=u; /\* Shifting c and u *\/ */
2017: /* u = *cx; */
2018: /* *cx = dum; */
2019: /* dum = fu; */
2020: /* fu = *fc; */
2021: /* *fc =dum; */
2022: /* } */
1.224 brouard 2023: #ifdef DEBUGMNBRAK
2024: double A, fparabu;
2025: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2026: fparabu= *fa - A*(*ax-u)*(*ax-u);
2027: 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);
2028: 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 2029: #endif
1.191 brouard 2030: dum=u; /* Shifting c and u */
2031: u = *cx;
2032: *cx = dum;
2033: dum = fu;
2034: fu = *fc;
2035: *fc =dum;
1.183 brouard 2036: #endif
1.162 brouard 2037: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2038: #ifdef DEBUG
1.224 brouard 2039: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2040: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2041: #endif
1.126 brouard 2042: fu=(*func)(u);
2043: if (fu < *fc) {
1.183 brouard 2044: #ifdef DEBUG
1.224 brouard 2045: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2046: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2047: #endif
2048: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2049: SHFT(*fb,*fc,fu,(*func)(u))
2050: #ifdef DEBUG
2051: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2052: #endif
2053: }
1.162 brouard 2054: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2055: #ifdef DEBUG
1.224 brouard 2056: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2057: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2058: #endif
1.126 brouard 2059: u=ulim;
2060: fu=(*func)(u);
1.183 brouard 2061: } else { /* u could be left to b (if r > q parabola has a maximum) */
2062: #ifdef DEBUG
1.224 brouard 2063: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2064: 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 2065: #endif
1.126 brouard 2066: u=(*cx)+GOLD*(*cx-*bx);
2067: fu=(*func)(u);
1.224 brouard 2068: #ifdef DEBUG
2069: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2070: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2071: #endif
1.183 brouard 2072: } /* end tests */
1.126 brouard 2073: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2074: SHFT(*fa,*fb,*fc,fu)
2075: #ifdef DEBUG
1.224 brouard 2076: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2077: 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 2078: #endif
2079: } /* 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 2080: }
2081:
2082: /*************** linmin ************************/
1.162 brouard 2083: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2084: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2085: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2086: the value of func at the returned location p . This is actually all accomplished by calling the
2087: routines mnbrak and brent .*/
1.126 brouard 2088: int ncom;
2089: double *pcom,*xicom;
2090: double (*nrfunc)(double []);
2091:
1.224 brouard 2092: #ifdef LINMINORIGINAL
1.126 brouard 2093: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2094: #else
2095: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2096: #endif
1.126 brouard 2097: {
2098: double brent(double ax, double bx, double cx,
2099: double (*f)(double), double tol, double *xmin);
2100: double f1dim(double x);
2101: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2102: double *fc, double (*func)(double));
2103: int j;
2104: double xx,xmin,bx,ax;
2105: double fx,fb,fa;
1.187 brouard 2106:
1.203 brouard 2107: #ifdef LINMINORIGINAL
2108: #else
2109: double scale=10., axs, xxs; /* Scale added for infinity */
2110: #endif
2111:
1.126 brouard 2112: ncom=n;
2113: pcom=vector(1,n);
2114: xicom=vector(1,n);
2115: nrfunc=func;
2116: for (j=1;j<=n;j++) {
2117: pcom[j]=p[j];
1.202 brouard 2118: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2119: }
1.187 brouard 2120:
1.203 brouard 2121: #ifdef LINMINORIGINAL
2122: xx=1.;
2123: #else
2124: axs=0.0;
2125: xxs=1.;
2126: do{
2127: xx= xxs;
2128: #endif
1.187 brouard 2129: ax=0.;
2130: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2131: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2132: /* 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)) */
2133: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2134: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2135: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2136: /* 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 2137: #ifdef LINMINORIGINAL
2138: #else
2139: if (fx != fx){
1.224 brouard 2140: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2141: printf("|");
2142: fprintf(ficlog,"|");
1.203 brouard 2143: #ifdef DEBUGLINMIN
1.224 brouard 2144: 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 2145: #endif
2146: }
1.224 brouard 2147: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2148: #endif
2149:
1.191 brouard 2150: #ifdef DEBUGLINMIN
2151: 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 2152: 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 2153: #endif
1.224 brouard 2154: #ifdef LINMINORIGINAL
2155: #else
2156: if(fb == fx){ /* Flat function in the direction */
2157: xmin=xx;
2158: *flat=1;
2159: }else{
2160: *flat=0;
2161: #endif
2162: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2163: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2164: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2165: /* fmin = f(p[j] + xmin * xi[j]) */
2166: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2167: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2168: #ifdef DEBUG
1.224 brouard 2169: 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);
2170: 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);
2171: #endif
2172: #ifdef LINMINORIGINAL
2173: #else
2174: }
1.126 brouard 2175: #endif
1.191 brouard 2176: #ifdef DEBUGLINMIN
2177: printf("linmin end ");
1.202 brouard 2178: fprintf(ficlog,"linmin end ");
1.191 brouard 2179: #endif
1.126 brouard 2180: for (j=1;j<=n;j++) {
1.203 brouard 2181: #ifdef LINMINORIGINAL
2182: xi[j] *= xmin;
2183: #else
2184: #ifdef DEBUGLINMIN
2185: if(xxs <1.0)
2186: printf(" before xi[%d]=%12.8f", j,xi[j]);
2187: #endif
2188: 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) */
2189: #ifdef DEBUGLINMIN
2190: if(xxs <1.0)
2191: 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 );
2192: #endif
2193: #endif
1.187 brouard 2194: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2195: }
1.191 brouard 2196: #ifdef DEBUGLINMIN
1.203 brouard 2197: printf("\n");
1.191 brouard 2198: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2199: 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 2200: for (j=1;j<=n;j++) {
1.202 brouard 2201: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2202: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2203: if(j % ncovmodel == 0){
1.191 brouard 2204: printf("\n");
1.202 brouard 2205: fprintf(ficlog,"\n");
2206: }
1.191 brouard 2207: }
1.203 brouard 2208: #else
1.191 brouard 2209: #endif
1.126 brouard 2210: free_vector(xicom,1,n);
2211: free_vector(pcom,1,n);
2212: }
2213:
2214:
2215: /*************** powell ************************/
1.162 brouard 2216: /*
2217: Minimization of a function func of n variables. Input consists of an initial starting point
2218: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2219: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2220: such that failure to decrease by more than this amount on one iteration signals doneness. On
2221: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2222: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2223: */
1.224 brouard 2224: #ifdef LINMINORIGINAL
2225: #else
2226: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2227: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2228: #endif
1.126 brouard 2229: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2230: double (*func)(double []))
2231: {
1.224 brouard 2232: #ifdef LINMINORIGINAL
2233: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2234: double (*func)(double []));
1.224 brouard 2235: #else
1.241 brouard 2236: void linmin(double p[], double xi[], int n, double *fret,
2237: double (*func)(double []),int *flat);
1.224 brouard 2238: #endif
1.239 brouard 2239: int i,ibig,j,jk,k;
1.126 brouard 2240: double del,t,*pt,*ptt,*xit;
1.181 brouard 2241: double directest;
1.126 brouard 2242: double fp,fptt;
2243: double *xits;
2244: int niterf, itmp;
1.224 brouard 2245: #ifdef LINMINORIGINAL
2246: #else
2247:
2248: flatdir=ivector(1,n);
2249: for (j=1;j<=n;j++) flatdir[j]=0;
2250: #endif
1.126 brouard 2251:
2252: pt=vector(1,n);
2253: ptt=vector(1,n);
2254: xit=vector(1,n);
2255: xits=vector(1,n);
2256: *fret=(*func)(p);
2257: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2258: rcurr_time = time(NULL);
1.126 brouard 2259: for (*iter=1;;++(*iter)) {
1.187 brouard 2260: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2261: ibig=0;
2262: del=0.0;
1.157 brouard 2263: rlast_time=rcurr_time;
2264: /* (void) gettimeofday(&curr_time,&tzp); */
2265: rcurr_time = time(NULL);
2266: curr_time = *localtime(&rcurr_time);
2267: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2268: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2269: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2270: for (i=1;i<=n;i++) {
1.126 brouard 2271: fprintf(ficrespow," %.12lf", p[i]);
2272: }
1.239 brouard 2273: fprintf(ficrespow,"\n");fflush(ficrespow);
2274: printf("\n#model= 1 + age ");
2275: fprintf(ficlog,"\n#model= 1 + age ");
2276: if(nagesqr==1){
1.241 brouard 2277: printf(" + age*age ");
2278: fprintf(ficlog," + age*age ");
1.239 brouard 2279: }
2280: for(j=1;j <=ncovmodel-2;j++){
2281: if(Typevar[j]==0) {
2282: printf(" + V%d ",Tvar[j]);
2283: fprintf(ficlog," + V%d ",Tvar[j]);
2284: }else if(Typevar[j]==1) {
2285: printf(" + V%d*age ",Tvar[j]);
2286: fprintf(ficlog," + V%d*age ",Tvar[j]);
2287: }else if(Typevar[j]==2) {
2288: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2289: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2290: }
2291: }
1.126 brouard 2292: printf("\n");
1.239 brouard 2293: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2294: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2295: fprintf(ficlog,"\n");
1.239 brouard 2296: for(i=1,jk=1; i <=nlstate; i++){
2297: for(k=1; k <=(nlstate+ndeath); k++){
2298: if (k != i) {
2299: printf("%d%d ",i,k);
2300: fprintf(ficlog,"%d%d ",i,k);
2301: for(j=1; j <=ncovmodel; j++){
2302: printf("%12.7f ",p[jk]);
2303: fprintf(ficlog,"%12.7f ",p[jk]);
2304: jk++;
2305: }
2306: printf("\n");
2307: fprintf(ficlog,"\n");
2308: }
2309: }
2310: }
1.241 brouard 2311: if(*iter <=3 && *iter >1){
1.157 brouard 2312: tml = *localtime(&rcurr_time);
2313: strcpy(strcurr,asctime(&tml));
2314: rforecast_time=rcurr_time;
1.126 brouard 2315: itmp = strlen(strcurr);
2316: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2317: strcurr[itmp-1]='\0';
1.162 brouard 2318: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2319: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2320: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2321: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2322: forecast_time = *localtime(&rforecast_time);
2323: strcpy(strfor,asctime(&forecast_time));
2324: itmp = strlen(strfor);
2325: if(strfor[itmp-1]=='\n')
2326: strfor[itmp-1]='\0';
2327: 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);
2328: 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 2329: }
2330: }
1.187 brouard 2331: for (i=1;i<=n;i++) { /* For each direction i */
2332: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2333: fptt=(*fret);
2334: #ifdef DEBUG
1.203 brouard 2335: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2336: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2337: #endif
1.203 brouard 2338: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2339: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2340: #ifdef LINMINORIGINAL
1.188 brouard 2341: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2342: #else
2343: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2344: flatdir[i]=flat; /* Function is vanishing in that direction i */
2345: #endif
2346: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2347: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2348: /* because that direction will be replaced unless the gain del is small */
2349: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2350: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2351: /* with the new direction. */
2352: del=fabs(fptt-(*fret));
2353: ibig=i;
1.126 brouard 2354: }
2355: #ifdef DEBUG
2356: printf("%d %.12e",i,(*fret));
2357: fprintf(ficlog,"%d %.12e",i,(*fret));
2358: for (j=1;j<=n;j++) {
1.224 brouard 2359: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2360: printf(" x(%d)=%.12e",j,xit[j]);
2361: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2362: }
2363: for(j=1;j<=n;j++) {
1.225 brouard 2364: printf(" p(%d)=%.12e",j,p[j]);
2365: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2366: }
2367: printf("\n");
2368: fprintf(ficlog,"\n");
2369: #endif
1.187 brouard 2370: } /* end loop on each direction i */
2371: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2372: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2373: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2374: for(j=1;j<=n;j++) {
1.225 brouard 2375: if(flatdir[j] >0){
2376: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2377: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2378: }
2379: /* printf("\n"); */
2380: /* fprintf(ficlog,"\n"); */
2381: }
1.243 brouard 2382: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2383: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2384: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2385: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2386: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2387: /* decreased of more than 3.84 */
2388: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2389: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2390: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2391:
1.188 brouard 2392: /* Starting the program with initial values given by a former maximization will simply change */
2393: /* the scales of the directions and the directions, because the are reset to canonical directions */
2394: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2395: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2396: #ifdef DEBUG
2397: int k[2],l;
2398: k[0]=1;
2399: k[1]=-1;
2400: printf("Max: %.12e",(*func)(p));
2401: fprintf(ficlog,"Max: %.12e",(*func)(p));
2402: for (j=1;j<=n;j++) {
2403: printf(" %.12e",p[j]);
2404: fprintf(ficlog," %.12e",p[j]);
2405: }
2406: printf("\n");
2407: fprintf(ficlog,"\n");
2408: for(l=0;l<=1;l++) {
2409: for (j=1;j<=n;j++) {
2410: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2411: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2412: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2413: }
2414: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2415: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2416: }
2417: #endif
2418:
1.224 brouard 2419: #ifdef LINMINORIGINAL
2420: #else
2421: free_ivector(flatdir,1,n);
2422: #endif
1.126 brouard 2423: free_vector(xit,1,n);
2424: free_vector(xits,1,n);
2425: free_vector(ptt,1,n);
2426: free_vector(pt,1,n);
2427: return;
1.192 brouard 2428: } /* enough precision */
1.240 brouard 2429: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2430: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2431: ptt[j]=2.0*p[j]-pt[j];
2432: xit[j]=p[j]-pt[j];
2433: pt[j]=p[j];
2434: }
1.181 brouard 2435: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2436: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2437: if (*iter <=4) {
1.225 brouard 2438: #else
2439: #endif
1.224 brouard 2440: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2441: #else
1.161 brouard 2442: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2443: #endif
1.162 brouard 2444: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2445: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2446: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2447: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2448: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2449: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2450: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2451: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2452: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2453: /* Even if f3 <f1, directest can be negative and t >0 */
2454: /* mu² and del² are equal when f3=f1 */
2455: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2456: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2457: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2458: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2459: #ifdef NRCORIGINAL
2460: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2461: #else
2462: 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 2463: t= t- del*SQR(fp-fptt);
1.183 brouard 2464: #endif
1.202 brouard 2465: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2466: #ifdef DEBUG
1.181 brouard 2467: 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);
2468: 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 2469: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2470: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2471: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2472: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2473: 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);
2474: 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);
2475: #endif
1.183 brouard 2476: #ifdef POWELLORIGINAL
2477: if (t < 0.0) { /* Then we use it for new direction */
2478: #else
1.182 brouard 2479: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2480: 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 2481: 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 2482: 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 2483: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2484: }
1.181 brouard 2485: if (directest < 0.0) { /* Then we use it for new direction */
2486: #endif
1.191 brouard 2487: #ifdef DEBUGLINMIN
1.234 brouard 2488: printf("Before linmin in direction P%d-P0\n",n);
2489: for (j=1;j<=n;j++) {
2490: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2491: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2492: if(j % ncovmodel == 0){
2493: printf("\n");
2494: fprintf(ficlog,"\n");
2495: }
2496: }
1.224 brouard 2497: #endif
2498: #ifdef LINMINORIGINAL
1.234 brouard 2499: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2500: #else
1.234 brouard 2501: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2502: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2503: #endif
1.234 brouard 2504:
1.191 brouard 2505: #ifdef DEBUGLINMIN
1.234 brouard 2506: for (j=1;j<=n;j++) {
2507: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2508: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2509: if(j % ncovmodel == 0){
2510: printf("\n");
2511: fprintf(ficlog,"\n");
2512: }
2513: }
1.224 brouard 2514: #endif
1.234 brouard 2515: for (j=1;j<=n;j++) {
2516: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2517: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2518: }
1.224 brouard 2519: #ifdef LINMINORIGINAL
2520: #else
1.234 brouard 2521: for (j=1, flatd=0;j<=n;j++) {
2522: if(flatdir[j]>0)
2523: flatd++;
2524: }
2525: if(flatd >0){
1.255 brouard 2526: printf("%d flat directions: ",flatd);
2527: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2528: for (j=1;j<=n;j++) {
2529: if(flatdir[j]>0){
2530: printf("%d ",j);
2531: fprintf(ficlog,"%d ",j);
2532: }
2533: }
2534: printf("\n");
2535: fprintf(ficlog,"\n");
2536: }
1.191 brouard 2537: #endif
1.234 brouard 2538: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2539: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2540:
1.126 brouard 2541: #ifdef DEBUG
1.234 brouard 2542: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2543: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2544: for(j=1;j<=n;j++){
2545: printf(" %lf",xit[j]);
2546: fprintf(ficlog," %lf",xit[j]);
2547: }
2548: printf("\n");
2549: fprintf(ficlog,"\n");
1.126 brouard 2550: #endif
1.192 brouard 2551: } /* end of t or directest negative */
1.224 brouard 2552: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2553: #else
1.234 brouard 2554: } /* end if (fptt < fp) */
1.192 brouard 2555: #endif
1.225 brouard 2556: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2557: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2558: #else
1.224 brouard 2559: #endif
1.234 brouard 2560: } /* loop iteration */
1.126 brouard 2561: }
1.234 brouard 2562:
1.126 brouard 2563: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2564:
1.235 brouard 2565: 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 2566: {
1.279 brouard 2567: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2568: * (and selected quantitative values in nres)
2569: * by left multiplying the unit
2570: * matrix by transitions matrix until convergence is reached with precision ftolpl
2571: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2572: * Wx is row vector: population in state 1, population in state 2, population dead
2573: * or prevalence in state 1, prevalence in state 2, 0
2574: * newm is the matrix after multiplications, its rows are identical at a factor.
2575: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2576: * Output is prlim.
2577: * Initial matrix pimij
2578: */
1.206 brouard 2579: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2580: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2581: /* 0, 0 , 1} */
2582: /*
2583: * and after some iteration: */
2584: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2585: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2586: /* 0, 0 , 1} */
2587: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2588: /* {0.51571254859325999, 0.4842874514067399, */
2589: /* 0.51326036147820708, 0.48673963852179264} */
2590: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2591:
1.126 brouard 2592: int i, ii,j,k;
1.209 brouard 2593: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2594: /* double **matprod2(); */ /* test */
1.218 brouard 2595: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2596: double **newm;
1.209 brouard 2597: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2598: int ncvloop=0;
1.288 brouard 2599: int first=0;
1.169 brouard 2600:
1.209 brouard 2601: min=vector(1,nlstate);
2602: max=vector(1,nlstate);
2603: meandiff=vector(1,nlstate);
2604:
1.218 brouard 2605: /* Starting with matrix unity */
1.126 brouard 2606: for (ii=1;ii<=nlstate+ndeath;ii++)
2607: for (j=1;j<=nlstate+ndeath;j++){
2608: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2609: }
1.169 brouard 2610:
2611: cov[1]=1.;
2612:
2613: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2614: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2615: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2616: ncvloop++;
1.126 brouard 2617: newm=savm;
2618: /* Covariates have to be included here again */
1.138 brouard 2619: cov[2]=agefin;
1.187 brouard 2620: if(nagesqr==1)
2621: cov[3]= agefin*agefin;;
1.234 brouard 2622: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2623: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2624: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2625: /* 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 2626: }
2627: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2628: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2629: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2630: /* 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 2631: }
1.237 brouard 2632: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2633: if(Dummy[Tvar[Tage[k]]]){
2634: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2635: } else{
1.235 brouard 2636: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2637: }
1.235 brouard 2638: /* 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 2639: }
1.237 brouard 2640: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2641: /* 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 2642: if(Dummy[Tvard[k][1]==0]){
2643: if(Dummy[Tvard[k][2]==0]){
2644: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2645: }else{
2646: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2647: }
2648: }else{
2649: if(Dummy[Tvard[k][2]==0]){
2650: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2651: }else{
2652: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2653: }
2654: }
1.234 brouard 2655: }
1.138 brouard 2656: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2657: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2658: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2659: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2660: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2661: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2662: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2663:
1.126 brouard 2664: savm=oldm;
2665: oldm=newm;
1.209 brouard 2666:
2667: for(j=1; j<=nlstate; j++){
2668: max[j]=0.;
2669: min[j]=1.;
2670: }
2671: for(i=1;i<=nlstate;i++){
2672: sumnew=0;
2673: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2674: for(j=1; j<=nlstate; j++){
2675: prlim[i][j]= newm[i][j]/(1-sumnew);
2676: max[j]=FMAX(max[j],prlim[i][j]);
2677: min[j]=FMIN(min[j],prlim[i][j]);
2678: }
2679: }
2680:
1.126 brouard 2681: maxmax=0.;
1.209 brouard 2682: for(j=1; j<=nlstate; j++){
2683: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2684: maxmax=FMAX(maxmax,meandiff[j]);
2685: /* 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 2686: } /* j loop */
1.203 brouard 2687: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2688: /* 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 2689: if(maxmax < ftolpl){
1.209 brouard 2690: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2691: free_vector(min,1,nlstate);
2692: free_vector(max,1,nlstate);
2693: free_vector(meandiff,1,nlstate);
1.126 brouard 2694: return prlim;
2695: }
1.288 brouard 2696: } /* agefin loop */
1.208 brouard 2697: /* After some age loop it doesn't converge */
1.288 brouard 2698: if(!first){
2699: first=1;
2700: printf("Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d). Others in log file only...\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
2701: }
2702: fprintf(ficlog, "Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d).\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
2703:
1.209 brouard 2704: /* 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); */
2705: free_vector(min,1,nlstate);
2706: free_vector(max,1,nlstate);
2707: free_vector(meandiff,1,nlstate);
1.208 brouard 2708:
1.169 brouard 2709: return prlim; /* should not reach here */
1.126 brouard 2710: }
2711:
1.217 brouard 2712:
2713: /**** Back Prevalence limit (stable or period prevalence) ****************/
2714:
1.218 brouard 2715: /* 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) */
2716: /* 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 2717: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2718: {
1.264 brouard 2719: /* 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 2720: matrix by transitions matrix until convergence is reached with precision ftolpl */
2721: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2722: /* Wx is row vector: population in state 1, population in state 2, population dead */
2723: /* or prevalence in state 1, prevalence in state 2, 0 */
2724: /* newm is the matrix after multiplications, its rows are identical at a factor */
2725: /* Initial matrix pimij */
2726: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2727: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2728: /* 0, 0 , 1} */
2729: /*
2730: * and after some iteration: */
2731: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2732: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2733: /* 0, 0 , 1} */
2734: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2735: /* {0.51571254859325999, 0.4842874514067399, */
2736: /* 0.51326036147820708, 0.48673963852179264} */
2737: /* If we start from prlim again, prlim tends to a constant matrix */
2738:
2739: int i, ii,j,k;
1.247 brouard 2740: int first=0;
1.217 brouard 2741: double *min, *max, *meandiff, maxmax,sumnew=0.;
2742: /* double **matprod2(); */ /* test */
2743: double **out, cov[NCOVMAX+1], **bmij();
2744: double **newm;
1.218 brouard 2745: double **dnewm, **doldm, **dsavm; /* for use */
2746: double **oldm, **savm; /* for use */
2747:
1.217 brouard 2748: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2749: int ncvloop=0;
2750:
2751: min=vector(1,nlstate);
2752: max=vector(1,nlstate);
2753: meandiff=vector(1,nlstate);
2754:
1.266 brouard 2755: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2756: oldm=oldms; savm=savms;
2757:
2758: /* Starting with matrix unity */
2759: for (ii=1;ii<=nlstate+ndeath;ii++)
2760: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2761: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2762: }
2763:
2764: cov[1]=1.;
2765:
2766: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2767: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2768: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2769: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2770: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2771: ncvloop++;
1.218 brouard 2772: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2773: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2774: /* Covariates have to be included here again */
2775: cov[2]=agefin;
2776: if(nagesqr==1)
2777: cov[3]= agefin*agefin;;
1.242 brouard 2778: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2779: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2780: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2781: /* 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 2782: }
2783: /* for (k=1; k<=cptcovn;k++) { */
2784: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2785: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2786: /* /\* 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])]); *\/ */
2787: /* } */
2788: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2789: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2790: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2791: /* 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]); */
2792: }
2793: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2794: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2795: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2796: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2797: for (k=1; k<=cptcovage;k++){ /* For product with age */
2798: if(Dummy[Tvar[Tage[k]]]){
2799: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2800: } else{
2801: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2802: }
2803: /* 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]); */
2804: }
2805: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2806: /* 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]); */
2807: if(Dummy[Tvard[k][1]==0]){
2808: if(Dummy[Tvard[k][2]==0]){
2809: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2810: }else{
2811: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2812: }
2813: }else{
2814: if(Dummy[Tvard[k][2]==0]){
2815: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2816: }else{
2817: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2818: }
2819: }
1.217 brouard 2820: }
2821:
2822: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2823: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2824: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2825: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2826: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2827: /* ij should be linked to the correct index of cov */
2828: /* age and covariate values ij are in 'cov', but we need to pass
2829: * ij for the observed prevalence at age and status and covariate
2830: * number: prevacurrent[(int)agefin][ii][ij]
2831: */
2832: /* 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 *\/ */
2833: /* 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 *\/ */
2834: 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 2835: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2836: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2837: /* for(i=1; i<=nlstate+ndeath; i++) { */
2838: /* printf("%d newm= ",i); */
2839: /* for(j=1;j<=nlstate+ndeath;j++) { */
2840: /* printf("%f ",newm[i][j]); */
2841: /* } */
2842: /* printf("oldm * "); */
2843: /* for(j=1;j<=nlstate+ndeath;j++) { */
2844: /* printf("%f ",oldm[i][j]); */
2845: /* } */
1.268 brouard 2846: /* printf(" bmmij "); */
1.266 brouard 2847: /* for(j=1;j<=nlstate+ndeath;j++) { */
2848: /* printf("%f ",pmmij[i][j]); */
2849: /* } */
2850: /* printf("\n"); */
2851: /* } */
2852: /* } */
1.217 brouard 2853: savm=oldm;
2854: oldm=newm;
1.266 brouard 2855:
1.217 brouard 2856: for(j=1; j<=nlstate; j++){
2857: max[j]=0.;
2858: min[j]=1.;
2859: }
2860: for(j=1; j<=nlstate; j++){
2861: for(i=1;i<=nlstate;i++){
1.234 brouard 2862: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2863: bprlim[i][j]= newm[i][j];
2864: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2865: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2866: }
2867: }
1.218 brouard 2868:
1.217 brouard 2869: maxmax=0.;
2870: for(i=1; i<=nlstate; i++){
2871: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2872: maxmax=FMAX(maxmax,meandiff[i]);
2873: /* 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 2874: } /* i loop */
1.217 brouard 2875: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2876: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2877: if(maxmax < ftolpl){
1.220 brouard 2878: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2879: free_vector(min,1,nlstate);
2880: free_vector(max,1,nlstate);
2881: free_vector(meandiff,1,nlstate);
2882: return bprlim;
2883: }
1.288 brouard 2884: } /* agefin loop */
1.217 brouard 2885: /* After some age loop it doesn't converge */
1.288 brouard 2886: if(!first){
1.247 brouard 2887: first=1;
2888: 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\
2889: 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);
2890: }
2891: 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 2892: 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);
2893: /* 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); */
2894: free_vector(min,1,nlstate);
2895: free_vector(max,1,nlstate);
2896: free_vector(meandiff,1,nlstate);
2897:
2898: return bprlim; /* should not reach here */
2899: }
2900:
1.126 brouard 2901: /*************** transition probabilities ***************/
2902:
2903: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2904: {
1.138 brouard 2905: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2906: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2907: model to the ncovmodel covariates (including constant and age).
2908: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2909: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2910: ncth covariate in the global vector x is given by the formula:
2911: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2912: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2913: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2914: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2915: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2916: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2917: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2918: */
2919: double s1, lnpijopii;
1.126 brouard 2920: /*double t34;*/
1.164 brouard 2921: int i,j, nc, ii, jj;
1.126 brouard 2922:
1.223 brouard 2923: for(i=1; i<= nlstate; i++){
2924: for(j=1; j<i;j++){
2925: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2926: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2927: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2928: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2929: }
2930: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2931: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2932: }
2933: for(j=i+1; j<=nlstate+ndeath;j++){
2934: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2935: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2936: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2937: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2938: }
2939: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2940: }
2941: }
1.218 brouard 2942:
1.223 brouard 2943: for(i=1; i<= nlstate; i++){
2944: s1=0;
2945: for(j=1; j<i; j++){
2946: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2947: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2948: }
2949: for(j=i+1; j<=nlstate+ndeath; j++){
2950: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2951: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2952: }
2953: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2954: ps[i][i]=1./(s1+1.);
2955: /* Computing other pijs */
2956: for(j=1; j<i; j++)
2957: ps[i][j]= exp(ps[i][j])*ps[i][i];
2958: for(j=i+1; j<=nlstate+ndeath; j++)
2959: ps[i][j]= exp(ps[i][j])*ps[i][i];
2960: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2961: } /* end i */
1.218 brouard 2962:
1.223 brouard 2963: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2964: for(jj=1; jj<= nlstate+ndeath; jj++){
2965: ps[ii][jj]=0;
2966: ps[ii][ii]=1;
2967: }
2968: }
1.293 ! brouard 2969: /* Added for backcast */ /* Transposed matrix too */
! 2970: for(jj=1; jj<= nlstate+ndeath; jj++){
! 2971: s1=0.;
! 2972: for(ii=1; ii<= nlstate+ndeath; ii++){
! 2973: s1+=ps[ii][jj];
! 2974: }
! 2975: for(ii=1; ii<= nlstate; ii++){
! 2976: ps[ii][jj]=ps[ii][jj]/s1;
! 2977: }
! 2978: }
! 2979: /* Transposition */
! 2980: for(jj=1; jj<= nlstate+ndeath; jj++){
! 2981: for(ii=jj; ii<= nlstate+ndeath; ii++){
! 2982: s1=ps[ii][jj];
! 2983: ps[ii][jj]=ps[jj][ii];
! 2984: ps[jj][ii]=s1;
! 2985: }
! 2986: }
1.223 brouard 2987: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2988: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2989: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2990: /* } */
2991: /* printf("\n "); */
2992: /* } */
2993: /* printf("\n ");printf("%lf ",cov[2]);*/
2994: /*
2995: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2996: goto end;*/
1.266 brouard 2997: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2998: }
2999:
1.218 brouard 3000: /*************** backward transition probabilities ***************/
3001:
3002: /* 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 ) */
3003: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3004: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3005: {
1.266 brouard 3006: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
3007: * 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 3008: */
1.218 brouard 3009: int i, ii, j,k;
1.222 brouard 3010:
3011: double **out, **pmij();
3012: double sumnew=0.;
1.218 brouard 3013: double agefin;
1.292 brouard 3014: double k3=0.; /* constant of the w_x diagonal matrix (in order for B to sum to 1 even for death state) */
1.222 brouard 3015: double **dnewm, **dsavm, **doldm;
3016: double **bbmij;
3017:
1.218 brouard 3018: doldm=ddoldms; /* global pointers */
1.222 brouard 3019: dnewm=ddnewms;
3020: dsavm=ddsavms;
3021:
3022: agefin=cov[2];
1.268 brouard 3023: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3024: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3025: the observed prevalence (with this covariate ij) at beginning of transition */
3026: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3027:
3028: /* P_x */
1.266 brouard 3029: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3030: /* outputs pmmij which is a stochastic matrix in row */
3031:
3032: /* Diag(w_x) */
1.292 brouard 3033: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3034: sumnew=0.;
1.269 brouard 3035: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3036: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 3037: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3038: sumnew+=prevacurrent[(int)agefin][ii][ij];
3039: }
3040: if(sumnew >0.01){ /* At least some value in the prevalence */
3041: for (ii=1;ii<=nlstate+ndeath;ii++){
3042: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3043: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3044: }
3045: }else{
3046: for (ii=1;ii<=nlstate+ndeath;ii++){
3047: for (j=1;j<=nlstate+ndeath;j++)
3048: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3049: }
3050: /* if(sumnew <0.9){ */
3051: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3052: /* } */
3053: }
3054: k3=0.0; /* We put the last diagonal to 0 */
3055: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3056: doldm[ii][ii]= k3;
3057: }
3058: /* End doldm, At the end doldm is diag[(w_i)] */
3059:
1.292 brouard 3060: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3061: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3062:
1.292 brouard 3063: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3064: /* 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 3065: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3066: sumnew=0.;
1.222 brouard 3067: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3068: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3069: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3070: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3071: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3072: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3073: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3074: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3075: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3076: /* }else */
1.268 brouard 3077: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3078: } /*End ii */
3079: } /* 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 */
3080:
1.292 brouard 3081: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3082: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3083: /* end bmij */
1.266 brouard 3084: return ps; /*pointer is unchanged */
1.218 brouard 3085: }
1.217 brouard 3086: /*************** transition probabilities ***************/
3087:
1.218 brouard 3088: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3089: {
3090: /* According to parameters values stored in x and the covariate's values stored in cov,
3091: computes the probability to be observed in state j being in state i by appying the
3092: model to the ncovmodel covariates (including constant and age).
3093: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3094: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3095: ncth covariate in the global vector x is given by the formula:
3096: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3097: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3098: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3099: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3100: Outputs ps[i][j] the probability to be observed in j being in j according to
3101: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3102: */
3103: double s1, lnpijopii;
3104: /*double t34;*/
3105: int i,j, nc, ii, jj;
3106:
1.234 brouard 3107: for(i=1; i<= nlstate; i++){
3108: for(j=1; j<i;j++){
3109: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3110: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3111: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3112: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3113: }
3114: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3115: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3116: }
3117: for(j=i+1; j<=nlstate+ndeath;j++){
3118: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3119: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3120: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3121: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3122: }
3123: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3124: }
3125: }
3126:
3127: for(i=1; i<= nlstate; i++){
3128: s1=0;
3129: for(j=1; j<i; j++){
3130: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3131: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3132: }
3133: for(j=i+1; j<=nlstate+ndeath; j++){
3134: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3135: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3136: }
3137: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3138: ps[i][i]=1./(s1+1.);
3139: /* Computing other pijs */
3140: for(j=1; j<i; j++)
3141: ps[i][j]= exp(ps[i][j])*ps[i][i];
3142: for(j=i+1; j<=nlstate+ndeath; j++)
3143: ps[i][j]= exp(ps[i][j])*ps[i][i];
3144: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3145: } /* end i */
3146:
3147: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3148: for(jj=1; jj<= nlstate+ndeath; jj++){
3149: ps[ii][jj]=0;
3150: ps[ii][ii]=1;
3151: }
3152: }
3153: /* Added for backcast */ /* Transposed matrix too */
3154: for(jj=1; jj<= nlstate+ndeath; jj++){
3155: s1=0.;
3156: for(ii=1; ii<= nlstate+ndeath; ii++){
3157: s1+=ps[ii][jj];
3158: }
3159: for(ii=1; ii<= nlstate; ii++){
3160: ps[ii][jj]=ps[ii][jj]/s1;
3161: }
3162: }
3163: /* Transposition */
3164: for(jj=1; jj<= nlstate+ndeath; jj++){
3165: for(ii=jj; ii<= nlstate+ndeath; ii++){
3166: s1=ps[ii][jj];
3167: ps[ii][jj]=ps[jj][ii];
3168: ps[jj][ii]=s1;
3169: }
3170: }
3171: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3172: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3173: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3174: /* } */
3175: /* printf("\n "); */
3176: /* } */
3177: /* printf("\n ");printf("%lf ",cov[2]);*/
3178: /*
3179: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3180: goto end;*/
3181: return ps;
1.217 brouard 3182: }
3183:
3184:
1.126 brouard 3185: /**************** Product of 2 matrices ******************/
3186:
1.145 brouard 3187: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3188: {
3189: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3190: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3191: /* in, b, out are matrice of pointers which should have been initialized
3192: before: only the contents of out is modified. The function returns
3193: a pointer to pointers identical to out */
1.145 brouard 3194: int i, j, k;
1.126 brouard 3195: for(i=nrl; i<= nrh; i++)
1.145 brouard 3196: for(k=ncolol; k<=ncoloh; k++){
3197: out[i][k]=0.;
3198: for(j=ncl; j<=nch; j++)
3199: out[i][k] +=in[i][j]*b[j][k];
3200: }
1.126 brouard 3201: return out;
3202: }
3203:
3204:
3205: /************* Higher Matrix Product ***************/
3206:
1.235 brouard 3207: 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 3208: {
1.218 brouard 3209: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3210: 'nhstepm*hstepm*stepm' months (i.e. until
3211: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3212: nhstepm*hstepm matrices.
3213: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3214: (typically every 2 years instead of every month which is too big
3215: for the memory).
3216: Model is determined by parameters x and covariates have to be
3217: included manually here.
3218:
3219: */
3220:
3221: int i, j, d, h, k;
1.131 brouard 3222: double **out, cov[NCOVMAX+1];
1.126 brouard 3223: double **newm;
1.187 brouard 3224: double agexact;
1.214 brouard 3225: double agebegin, ageend;
1.126 brouard 3226:
3227: /* Hstepm could be zero and should return the unit matrix */
3228: for (i=1;i<=nlstate+ndeath;i++)
3229: for (j=1;j<=nlstate+ndeath;j++){
3230: oldm[i][j]=(i==j ? 1.0 : 0.0);
3231: po[i][j][0]=(i==j ? 1.0 : 0.0);
3232: }
3233: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3234: for(h=1; h <=nhstepm; h++){
3235: for(d=1; d <=hstepm; d++){
3236: newm=savm;
3237: /* Covariates have to be included here again */
3238: cov[1]=1.;
1.214 brouard 3239: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3240: cov[2]=agexact;
3241: if(nagesqr==1)
1.227 brouard 3242: cov[3]= agexact*agexact;
1.235 brouard 3243: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3244: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3245: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3246: /* 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)); */
3247: }
3248: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3249: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3250: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3251: /* 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]); */
3252: }
3253: for (k=1; k<=cptcovage;k++){
3254: if(Dummy[Tvar[Tage[k]]]){
3255: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3256: } else{
3257: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3258: }
3259: /* 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]); */
3260: }
3261: for (k=1; k<=cptcovprod;k++){ /* */
3262: /* 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]); */
3263: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3264: }
3265: /* for (k=1; k<=cptcovn;k++) */
3266: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3267: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3268: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3269: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3270: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3271:
3272:
1.126 brouard 3273: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3274: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3275: /* right multiplication of oldm by the current matrix */
1.126 brouard 3276: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3277: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3278: /* if((int)age == 70){ */
3279: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3280: /* for(i=1; i<=nlstate+ndeath; i++) { */
3281: /* printf("%d pmmij ",i); */
3282: /* for(j=1;j<=nlstate+ndeath;j++) { */
3283: /* printf("%f ",pmmij[i][j]); */
3284: /* } */
3285: /* printf(" oldm "); */
3286: /* for(j=1;j<=nlstate+ndeath;j++) { */
3287: /* printf("%f ",oldm[i][j]); */
3288: /* } */
3289: /* printf("\n"); */
3290: /* } */
3291: /* } */
1.126 brouard 3292: savm=oldm;
3293: oldm=newm;
3294: }
3295: for(i=1; i<=nlstate+ndeath; i++)
3296: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3297: po[i][j][h]=newm[i][j];
3298: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3299: }
1.128 brouard 3300: /*printf("h=%d ",h);*/
1.126 brouard 3301: } /* end h */
1.267 brouard 3302: /* printf("\n H=%d \n",h); */
1.126 brouard 3303: return po;
3304: }
3305:
1.217 brouard 3306: /************* Higher Back Matrix Product ***************/
1.218 brouard 3307: /* 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 3308: 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 3309: {
1.266 brouard 3310: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3311: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3312: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3313: nhstepm*hstepm matrices.
3314: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3315: (typically every 2 years instead of every month which is too big
1.217 brouard 3316: for the memory).
1.218 brouard 3317: Model is determined by parameters x and covariates have to be
1.266 brouard 3318: included manually here. Then we use a call to bmij(x and cov)
3319: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3320: */
1.217 brouard 3321:
3322: int i, j, d, h, k;
1.266 brouard 3323: double **out, cov[NCOVMAX+1], **bmij();
3324: double **newm, ***newmm;
1.217 brouard 3325: double agexact;
3326: double agebegin, ageend;
1.222 brouard 3327: double **oldm, **savm;
1.217 brouard 3328:
1.266 brouard 3329: newmm=po; /* To be saved */
3330: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3331: /* Hstepm could be zero and should return the unit matrix */
3332: for (i=1;i<=nlstate+ndeath;i++)
3333: for (j=1;j<=nlstate+ndeath;j++){
3334: oldm[i][j]=(i==j ? 1.0 : 0.0);
3335: po[i][j][0]=(i==j ? 1.0 : 0.0);
3336: }
3337: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3338: for(h=1; h <=nhstepm; h++){
3339: for(d=1; d <=hstepm; d++){
3340: newm=savm;
3341: /* Covariates have to be included here again */
3342: cov[1]=1.;
1.271 brouard 3343: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3344: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3345: cov[2]=agexact;
3346: if(nagesqr==1)
1.222 brouard 3347: cov[3]= agexact*agexact;
1.266 brouard 3348: for (k=1; k<=cptcovn;k++){
3349: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3350: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3351: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3352: /* 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)); */
3353: }
1.267 brouard 3354: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3355: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3356: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3357: /* 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]); */
3358: }
3359: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3360: if(Dummy[Tvar[Tage[k]]]){
3361: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3362: } else{
3363: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3364: }
3365: /* 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]); */
3366: }
3367: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3368: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3369: }
1.217 brouard 3370: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3371: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3372:
1.218 brouard 3373: /* Careful transposed matrix */
1.266 brouard 3374: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3375: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3376: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3377: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3378: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3379: /* if((int)age == 70){ */
3380: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3381: /* for(i=1; i<=nlstate+ndeath; i++) { */
3382: /* printf("%d pmmij ",i); */
3383: /* for(j=1;j<=nlstate+ndeath;j++) { */
3384: /* printf("%f ",pmmij[i][j]); */
3385: /* } */
3386: /* printf(" oldm "); */
3387: /* for(j=1;j<=nlstate+ndeath;j++) { */
3388: /* printf("%f ",oldm[i][j]); */
3389: /* } */
3390: /* printf("\n"); */
3391: /* } */
3392: /* } */
3393: savm=oldm;
3394: oldm=newm;
3395: }
3396: for(i=1; i<=nlstate+ndeath; i++)
3397: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3398: po[i][j][h]=newm[i][j];
1.268 brouard 3399: /* if(h==nhstepm) */
3400: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3401: }
1.268 brouard 3402: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3403: } /* end h */
1.268 brouard 3404: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3405: return po;
3406: }
3407:
3408:
1.162 brouard 3409: #ifdef NLOPT
3410: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3411: double fret;
3412: double *xt;
3413: int j;
3414: myfunc_data *d2 = (myfunc_data *) pd;
3415: /* xt = (p1-1); */
3416: xt=vector(1,n);
3417: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3418:
3419: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3420: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3421: printf("Function = %.12lf ",fret);
3422: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3423: printf("\n");
3424: free_vector(xt,1,n);
3425: return fret;
3426: }
3427: #endif
1.126 brouard 3428:
3429: /*************** log-likelihood *************/
3430: double func( double *x)
3431: {
1.226 brouard 3432: int i, ii, j, k, mi, d, kk;
3433: int ioffset=0;
3434: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3435: double **out;
3436: double lli; /* Individual log likelihood */
3437: int s1, s2;
1.228 brouard 3438: 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 3439: double bbh, survp;
3440: long ipmx;
3441: double agexact;
3442: /*extern weight */
3443: /* We are differentiating ll according to initial status */
3444: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3445: /*for(i=1;i<imx;i++)
3446: printf(" %d\n",s[4][i]);
3447: */
1.162 brouard 3448:
1.226 brouard 3449: ++countcallfunc;
1.162 brouard 3450:
1.226 brouard 3451: cov[1]=1.;
1.126 brouard 3452:
1.226 brouard 3453: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3454: ioffset=0;
1.226 brouard 3455: if(mle==1){
3456: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3457: /* Computes the values of the ncovmodel covariates of the model
3458: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3459: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3460: to be observed in j being in i according to the model.
3461: */
1.243 brouard 3462: ioffset=2+nagesqr ;
1.233 brouard 3463: /* Fixed */
1.234 brouard 3464: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3465: 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)*/
3466: }
1.226 brouard 3467: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3468: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3469: has been calculated etc */
3470: /* For an individual i, wav[i] gives the number of effective waves */
3471: /* We compute the contribution to Likelihood of each effective transition
3472: mw[mi][i] is real wave of the mi th effectve wave */
3473: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3474: s2=s[mw[mi+1][i]][i];
3475: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3476: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3477: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3478: */
3479: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3480: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3481: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3482: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3483: }
3484: for (ii=1;ii<=nlstate+ndeath;ii++)
3485: for (j=1;j<=nlstate+ndeath;j++){
3486: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3487: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3488: }
3489: for(d=0; d<dh[mi][i]; d++){
3490: newm=savm;
3491: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3492: cov[2]=agexact;
3493: if(nagesqr==1)
3494: cov[3]= agexact*agexact; /* Should be changed here */
3495: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3496: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3497: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3498: else
3499: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3500: }
3501: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3502: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3503: savm=oldm;
3504: oldm=newm;
3505: } /* end mult */
3506:
3507: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3508: /* But now since version 0.9 we anticipate for bias at large stepm.
3509: * If stepm is larger than one month (smallest stepm) and if the exact delay
3510: * (in months) between two waves is not a multiple of stepm, we rounded to
3511: * the nearest (and in case of equal distance, to the lowest) interval but now
3512: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3513: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3514: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3515: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3516: * -stepm/2 to stepm/2 .
3517: * For stepm=1 the results are the same as for previous versions of Imach.
3518: * For stepm > 1 the results are less biased than in previous versions.
3519: */
1.234 brouard 3520: s1=s[mw[mi][i]][i];
3521: s2=s[mw[mi+1][i]][i];
3522: bbh=(double)bh[mi][i]/(double)stepm;
3523: /* bias bh is positive if real duration
3524: * is higher than the multiple of stepm and negative otherwise.
3525: */
3526: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3527: if( s2 > nlstate){
3528: /* i.e. if s2 is a death state and if the date of death is known
3529: then the contribution to the likelihood is the probability to
3530: die between last step unit time and current step unit time,
3531: which is also equal to probability to die before dh
3532: minus probability to die before dh-stepm .
3533: In version up to 0.92 likelihood was computed
3534: as if date of death was unknown. Death was treated as any other
3535: health state: the date of the interview describes the actual state
3536: and not the date of a change in health state. The former idea was
3537: to consider that at each interview the state was recorded
3538: (healthy, disable or death) and IMaCh was corrected; but when we
3539: introduced the exact date of death then we should have modified
3540: the contribution of an exact death to the likelihood. This new
3541: contribution is smaller and very dependent of the step unit
3542: stepm. It is no more the probability to die between last interview
3543: and month of death but the probability to survive from last
3544: interview up to one month before death multiplied by the
3545: probability to die within a month. Thanks to Chris
3546: Jackson for correcting this bug. Former versions increased
3547: mortality artificially. The bad side is that we add another loop
3548: which slows down the processing. The difference can be up to 10%
3549: lower mortality.
3550: */
3551: /* If, at the beginning of the maximization mostly, the
3552: cumulative probability or probability to be dead is
3553: constant (ie = 1) over time d, the difference is equal to
3554: 0. out[s1][3] = savm[s1][3]: probability, being at state
3555: s1 at precedent wave, to be dead a month before current
3556: wave is equal to probability, being at state s1 at
3557: precedent wave, to be dead at mont of the current
3558: wave. Then the observed probability (that this person died)
3559: is null according to current estimated parameter. In fact,
3560: it should be very low but not zero otherwise the log go to
3561: infinity.
3562: */
1.183 brouard 3563: /* #ifdef INFINITYORIGINAL */
3564: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3565: /* #else */
3566: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3567: /* lli=log(mytinydouble); */
3568: /* else */
3569: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3570: /* #endif */
1.226 brouard 3571: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3572:
1.226 brouard 3573: } else if ( s2==-1 ) { /* alive */
3574: for (j=1,survp=0. ; j<=nlstate; j++)
3575: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3576: /*survp += out[s1][j]; */
3577: lli= log(survp);
3578: }
3579: else if (s2==-4) {
3580: for (j=3,survp=0. ; j<=nlstate; j++)
3581: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3582: lli= log(survp);
3583: }
3584: else if (s2==-5) {
3585: for (j=1,survp=0. ; j<=2; j++)
3586: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3587: lli= log(survp);
3588: }
3589: else{
3590: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3591: /* 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 */
3592: }
3593: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3594: /*if(lli ==000.0)*/
3595: /*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); */
3596: ipmx +=1;
3597: sw += weight[i];
3598: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3599: /* if (lli < log(mytinydouble)){ */
3600: /* 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); */
3601: /* 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]); */
3602: /* } */
3603: } /* end of wave */
3604: } /* end of individual */
3605: } else if(mle==2){
3606: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3607: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3608: for(mi=1; mi<= wav[i]-1; mi++){
3609: for (ii=1;ii<=nlstate+ndeath;ii++)
3610: for (j=1;j<=nlstate+ndeath;j++){
3611: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3612: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3613: }
3614: for(d=0; d<=dh[mi][i]; d++){
3615: newm=savm;
3616: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3617: cov[2]=agexact;
3618: if(nagesqr==1)
3619: cov[3]= agexact*agexact;
3620: for (kk=1; kk<=cptcovage;kk++) {
3621: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3622: }
3623: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3624: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3625: savm=oldm;
3626: oldm=newm;
3627: } /* end mult */
3628:
3629: s1=s[mw[mi][i]][i];
3630: s2=s[mw[mi+1][i]][i];
3631: bbh=(double)bh[mi][i]/(double)stepm;
3632: 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 */
3633: ipmx +=1;
3634: sw += weight[i];
3635: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3636: } /* end of wave */
3637: } /* end of individual */
3638: } else if(mle==3){ /* exponential inter-extrapolation */
3639: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3640: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3641: for(mi=1; mi<= wav[i]-1; mi++){
3642: for (ii=1;ii<=nlstate+ndeath;ii++)
3643: for (j=1;j<=nlstate+ndeath;j++){
3644: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3645: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3646: }
3647: for(d=0; d<dh[mi][i]; d++){
3648: newm=savm;
3649: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3650: cov[2]=agexact;
3651: if(nagesqr==1)
3652: cov[3]= agexact*agexact;
3653: for (kk=1; kk<=cptcovage;kk++) {
3654: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3655: }
3656: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3657: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3658: savm=oldm;
3659: oldm=newm;
3660: } /* end mult */
3661:
3662: s1=s[mw[mi][i]][i];
3663: s2=s[mw[mi+1][i]][i];
3664: bbh=(double)bh[mi][i]/(double)stepm;
3665: 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 */
3666: ipmx +=1;
3667: sw += weight[i];
3668: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3669: } /* end of wave */
3670: } /* end of individual */
3671: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3672: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3673: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3674: for(mi=1; mi<= wav[i]-1; mi++){
3675: for (ii=1;ii<=nlstate+ndeath;ii++)
3676: for (j=1;j<=nlstate+ndeath;j++){
3677: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3678: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3679: }
3680: for(d=0; d<dh[mi][i]; d++){
3681: newm=savm;
3682: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3683: cov[2]=agexact;
3684: if(nagesqr==1)
3685: cov[3]= agexact*agexact;
3686: for (kk=1; kk<=cptcovage;kk++) {
3687: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3688: }
1.126 brouard 3689:
1.226 brouard 3690: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3691: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3692: savm=oldm;
3693: oldm=newm;
3694: } /* end mult */
3695:
3696: s1=s[mw[mi][i]][i];
3697: s2=s[mw[mi+1][i]][i];
3698: if( s2 > nlstate){
3699: lli=log(out[s1][s2] - savm[s1][s2]);
3700: } else if ( s2==-1 ) { /* alive */
3701: for (j=1,survp=0. ; j<=nlstate; j++)
3702: survp += out[s1][j];
3703: lli= log(survp);
3704: }else{
3705: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3706: }
3707: ipmx +=1;
3708: sw += weight[i];
3709: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3710: /* 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 3711: } /* end of wave */
3712: } /* end of individual */
3713: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3714: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3715: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3716: for(mi=1; mi<= wav[i]-1; mi++){
3717: for (ii=1;ii<=nlstate+ndeath;ii++)
3718: for (j=1;j<=nlstate+ndeath;j++){
3719: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3720: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3721: }
3722: for(d=0; d<dh[mi][i]; d++){
3723: newm=savm;
3724: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3725: cov[2]=agexact;
3726: if(nagesqr==1)
3727: cov[3]= agexact*agexact;
3728: for (kk=1; kk<=cptcovage;kk++) {
3729: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3730: }
1.126 brouard 3731:
1.226 brouard 3732: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3733: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3734: savm=oldm;
3735: oldm=newm;
3736: } /* end mult */
3737:
3738: s1=s[mw[mi][i]][i];
3739: s2=s[mw[mi+1][i]][i];
3740: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3741: ipmx +=1;
3742: sw += weight[i];
3743: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3744: /*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]);*/
3745: } /* end of wave */
3746: } /* end of individual */
3747: } /* End of if */
3748: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3749: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3750: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3751: return -l;
1.126 brouard 3752: }
3753:
3754: /*************** log-likelihood *************/
3755: double funcone( double *x)
3756: {
1.228 brouard 3757: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3758: int i, ii, j, k, mi, d, kk;
1.228 brouard 3759: int ioffset=0;
1.131 brouard 3760: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3761: double **out;
3762: double lli; /* Individual log likelihood */
3763: double llt;
3764: int s1, s2;
1.228 brouard 3765: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3766:
1.126 brouard 3767: double bbh, survp;
1.187 brouard 3768: double agexact;
1.214 brouard 3769: double agebegin, ageend;
1.126 brouard 3770: /*extern weight */
3771: /* We are differentiating ll according to initial status */
3772: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3773: /*for(i=1;i<imx;i++)
3774: printf(" %d\n",s[4][i]);
3775: */
3776: cov[1]=1.;
3777:
3778: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3779: ioffset=0;
3780: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3781: /* ioffset=2+nagesqr+cptcovage; */
3782: ioffset=2+nagesqr;
1.232 brouard 3783: /* Fixed */
1.224 brouard 3784: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3785: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3786: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3787: 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)*/
3788: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3789: /* cov[2+6]=covar[Tvar[6]][i]; */
3790: /* cov[2+6]=covar[2][i]; V2 */
3791: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3792: /* cov[2+7]=covar[Tvar[7]][i]; */
3793: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3794: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3795: /* cov[2+9]=covar[Tvar[9]][i]; */
3796: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3797: }
1.232 brouard 3798: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3799: /* 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?)*\/ */
3800: /* } */
1.231 brouard 3801: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3802: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3803: /* } */
1.225 brouard 3804:
1.233 brouard 3805:
3806: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3807: /* Wave varying (but not age varying) */
3808: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3809: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3810: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3811: }
1.232 brouard 3812: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3813: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3814: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3815: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3816: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3817: /* 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 3818: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3819: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3820: /* /\* 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]); *\/ */
3821: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3822: /* } */
1.126 brouard 3823: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3824: for (j=1;j<=nlstate+ndeath;j++){
3825: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3826: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3827: }
1.214 brouard 3828:
3829: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3830: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3831: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3832: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3833: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3834: and mw[mi+1][i]. dh depends on stepm.*/
3835: newm=savm;
1.247 brouard 3836: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3837: cov[2]=agexact;
3838: if(nagesqr==1)
3839: cov[3]= agexact*agexact;
3840: for (kk=1; kk<=cptcovage;kk++) {
3841: if(!FixedV[Tvar[Tage[kk]]])
3842: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3843: else
3844: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3845: }
3846: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3847: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3848: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3849: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3850: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3851: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3852: savm=oldm;
3853: oldm=newm;
1.126 brouard 3854: } /* end mult */
3855:
3856: s1=s[mw[mi][i]][i];
3857: s2=s[mw[mi+1][i]][i];
1.217 brouard 3858: /* if(s2==-1){ */
1.268 brouard 3859: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3860: /* /\* exit(1); *\/ */
3861: /* } */
1.126 brouard 3862: bbh=(double)bh[mi][i]/(double)stepm;
3863: /* bias is positive if real duration
3864: * is higher than the multiple of stepm and negative otherwise.
3865: */
3866: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3867: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3868: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3869: for (j=1,survp=0. ; j<=nlstate; j++)
3870: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3871: lli= log(survp);
1.126 brouard 3872: }else if (mle==1){
1.242 brouard 3873: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3874: } else if(mle==2){
1.242 brouard 3875: 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 3876: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3877: 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 3878: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3879: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3880: } else{ /* mle=0 back to 1 */
1.242 brouard 3881: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3882: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3883: } /* End of if */
3884: ipmx +=1;
3885: sw += weight[i];
3886: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3887: /*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 3888: if(globpr){
1.246 brouard 3889: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3890: %11.6f %11.6f %11.6f ", \
1.242 brouard 3891: 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 3892: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3893: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3894: llt +=ll[k]*gipmx/gsw;
3895: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3896: }
3897: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3898: }
1.232 brouard 3899: } /* end of wave */
3900: } /* end of individual */
3901: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3902: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3903: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3904: if(globpr==0){ /* First time we count the contributions and weights */
3905: gipmx=ipmx;
3906: gsw=sw;
3907: }
3908: return -l;
1.126 brouard 3909: }
3910:
3911:
3912: /*************** function likelione ***********/
1.292 brouard 3913: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 3914: {
3915: /* This routine should help understanding what is done with
3916: the selection of individuals/waves and
3917: to check the exact contribution to the likelihood.
3918: Plotting could be done.
3919: */
3920: int k;
3921:
3922: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3923: strcpy(fileresilk,"ILK_");
1.202 brouard 3924: strcat(fileresilk,fileresu);
1.126 brouard 3925: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3926: printf("Problem with resultfile: %s\n", fileresilk);
3927: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3928: }
1.214 brouard 3929: 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");
3930: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3931: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3932: for(k=1; k<=nlstate; k++)
3933: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3934: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3935: }
3936:
1.292 brouard 3937: *fretone=(*func)(p);
1.126 brouard 3938: if(*globpri !=0){
3939: fclose(ficresilk);
1.205 brouard 3940: if (mle ==0)
3941: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3942: else if(mle >=1)
3943: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3944: 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 3945: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3946:
3947: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3948: 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 3949: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3950: }
1.207 brouard 3951: 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 3952: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3953: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3954: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3955: fflush(fichtm);
1.205 brouard 3956: }
1.126 brouard 3957: return;
3958: }
3959:
3960:
3961: /*********** Maximum Likelihood Estimation ***************/
3962:
3963: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3964: {
1.165 brouard 3965: int i,j, iter=0;
1.126 brouard 3966: double **xi;
3967: double fret;
3968: double fretone; /* Only one call to likelihood */
3969: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3970:
3971: #ifdef NLOPT
3972: int creturn;
3973: nlopt_opt opt;
3974: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3975: double *lb;
3976: double minf; /* the minimum objective value, upon return */
3977: double * p1; /* Shifted parameters from 0 instead of 1 */
3978: myfunc_data dinst, *d = &dinst;
3979: #endif
3980:
3981:
1.126 brouard 3982: xi=matrix(1,npar,1,npar);
3983: for (i=1;i<=npar;i++)
3984: for (j=1;j<=npar;j++)
3985: xi[i][j]=(i==j ? 1.0 : 0.0);
3986: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3987: strcpy(filerespow,"POW_");
1.126 brouard 3988: strcat(filerespow,fileres);
3989: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3990: printf("Problem with resultfile: %s\n", filerespow);
3991: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3992: }
3993: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3994: for (i=1;i<=nlstate;i++)
3995: for(j=1;j<=nlstate+ndeath;j++)
3996: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3997: fprintf(ficrespow,"\n");
1.162 brouard 3998: #ifdef POWELL
1.126 brouard 3999: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4000: #endif
1.126 brouard 4001:
1.162 brouard 4002: #ifdef NLOPT
4003: #ifdef NEWUOA
4004: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4005: #else
4006: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4007: #endif
4008: lb=vector(0,npar-1);
4009: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4010: nlopt_set_lower_bounds(opt, lb);
4011: nlopt_set_initial_step1(opt, 0.1);
4012:
4013: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4014: d->function = func;
4015: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4016: nlopt_set_min_objective(opt, myfunc, d);
4017: nlopt_set_xtol_rel(opt, ftol);
4018: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4019: printf("nlopt failed! %d\n",creturn);
4020: }
4021: else {
4022: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4023: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4024: iter=1; /* not equal */
4025: }
4026: nlopt_destroy(opt);
4027: #endif
1.126 brouard 4028: free_matrix(xi,1,npar,1,npar);
4029: fclose(ficrespow);
1.203 brouard 4030: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4031: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4032: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4033:
4034: }
4035:
4036: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4037: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4038: {
4039: double **a,**y,*x,pd;
1.203 brouard 4040: /* double **hess; */
1.164 brouard 4041: int i, j;
1.126 brouard 4042: int *indx;
4043:
4044: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4045: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4046: void lubksb(double **a, int npar, int *indx, double b[]) ;
4047: void ludcmp(double **a, int npar, int *indx, double *d) ;
4048: double gompertz(double p[]);
1.203 brouard 4049: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4050:
4051: printf("\nCalculation of the hessian matrix. Wait...\n");
4052: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4053: for (i=1;i<=npar;i++){
1.203 brouard 4054: printf("%d-",i);fflush(stdout);
4055: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4056:
4057: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4058:
4059: /* printf(" %f ",p[i]);
4060: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4061: }
4062:
4063: for (i=1;i<=npar;i++) {
4064: for (j=1;j<=npar;j++) {
4065: if (j>i) {
1.203 brouard 4066: printf(".%d-%d",i,j);fflush(stdout);
4067: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4068: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4069:
4070: hess[j][i]=hess[i][j];
4071: /*printf(" %lf ",hess[i][j]);*/
4072: }
4073: }
4074: }
4075: printf("\n");
4076: fprintf(ficlog,"\n");
4077:
4078: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4079: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4080:
4081: a=matrix(1,npar,1,npar);
4082: y=matrix(1,npar,1,npar);
4083: x=vector(1,npar);
4084: indx=ivector(1,npar);
4085: for (i=1;i<=npar;i++)
4086: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4087: ludcmp(a,npar,indx,&pd);
4088:
4089: for (j=1;j<=npar;j++) {
4090: for (i=1;i<=npar;i++) x[i]=0;
4091: x[j]=1;
4092: lubksb(a,npar,indx,x);
4093: for (i=1;i<=npar;i++){
4094: matcov[i][j]=x[i];
4095: }
4096: }
4097:
4098: printf("\n#Hessian matrix#\n");
4099: fprintf(ficlog,"\n#Hessian matrix#\n");
4100: for (i=1;i<=npar;i++) {
4101: for (j=1;j<=npar;j++) {
1.203 brouard 4102: printf("%.6e ",hess[i][j]);
4103: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4104: }
4105: printf("\n");
4106: fprintf(ficlog,"\n");
4107: }
4108:
1.203 brouard 4109: /* printf("\n#Covariance matrix#\n"); */
4110: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4111: /* for (i=1;i<=npar;i++) { */
4112: /* for (j=1;j<=npar;j++) { */
4113: /* printf("%.6e ",matcov[i][j]); */
4114: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4115: /* } */
4116: /* printf("\n"); */
4117: /* fprintf(ficlog,"\n"); */
4118: /* } */
4119:
1.126 brouard 4120: /* Recompute Inverse */
1.203 brouard 4121: /* for (i=1;i<=npar;i++) */
4122: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4123: /* ludcmp(a,npar,indx,&pd); */
4124:
4125: /* printf("\n#Hessian matrix recomputed#\n"); */
4126:
4127: /* for (j=1;j<=npar;j++) { */
4128: /* for (i=1;i<=npar;i++) x[i]=0; */
4129: /* x[j]=1; */
4130: /* lubksb(a,npar,indx,x); */
4131: /* for (i=1;i<=npar;i++){ */
4132: /* y[i][j]=x[i]; */
4133: /* printf("%.3e ",y[i][j]); */
4134: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4135: /* } */
4136: /* printf("\n"); */
4137: /* fprintf(ficlog,"\n"); */
4138: /* } */
4139:
4140: /* Verifying the inverse matrix */
4141: #ifdef DEBUGHESS
4142: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4143:
1.203 brouard 4144: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4145: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4146:
4147: for (j=1;j<=npar;j++) {
4148: for (i=1;i<=npar;i++){
1.203 brouard 4149: printf("%.2f ",y[i][j]);
4150: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4151: }
4152: printf("\n");
4153: fprintf(ficlog,"\n");
4154: }
1.203 brouard 4155: #endif
1.126 brouard 4156:
4157: free_matrix(a,1,npar,1,npar);
4158: free_matrix(y,1,npar,1,npar);
4159: free_vector(x,1,npar);
4160: free_ivector(indx,1,npar);
1.203 brouard 4161: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4162:
4163:
4164: }
4165:
4166: /*************** hessian matrix ****************/
4167: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4168: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4169: int i;
4170: int l=1, lmax=20;
1.203 brouard 4171: double k1,k2, res, fx;
1.132 brouard 4172: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4173: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4174: int k=0,kmax=10;
4175: double l1;
4176:
4177: fx=func(x);
4178: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4179: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4180: l1=pow(10,l);
4181: delts=delt;
4182: for(k=1 ; k <kmax; k=k+1){
4183: delt = delta*(l1*k);
4184: p2[theta]=x[theta] +delt;
1.145 brouard 4185: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4186: p2[theta]=x[theta]-delt;
4187: k2=func(p2)-fx;
4188: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4189: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4190:
1.203 brouard 4191: #ifdef DEBUGHESSII
1.126 brouard 4192: 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);
4193: 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);
4194: #endif
4195: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4196: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4197: k=kmax;
4198: }
4199: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4200: k=kmax; l=lmax*10;
1.126 brouard 4201: }
4202: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4203: delts=delt;
4204: }
1.203 brouard 4205: } /* End loop k */
1.126 brouard 4206: }
4207: delti[theta]=delts;
4208: return res;
4209:
4210: }
4211:
1.203 brouard 4212: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4213: {
4214: int i;
1.164 brouard 4215: int l=1, lmax=20;
1.126 brouard 4216: double k1,k2,k3,k4,res,fx;
1.132 brouard 4217: double p2[MAXPARM+1];
1.203 brouard 4218: int k, kmax=1;
4219: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4220:
4221: int firstime=0;
1.203 brouard 4222:
1.126 brouard 4223: fx=func(x);
1.203 brouard 4224: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4225: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4226: p2[thetai]=x[thetai]+delti[thetai]*k;
4227: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4228: k1=func(p2)-fx;
4229:
1.203 brouard 4230: p2[thetai]=x[thetai]+delti[thetai]*k;
4231: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4232: k2=func(p2)-fx;
4233:
1.203 brouard 4234: p2[thetai]=x[thetai]-delti[thetai]*k;
4235: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4236: k3=func(p2)-fx;
4237:
1.203 brouard 4238: p2[thetai]=x[thetai]-delti[thetai]*k;
4239: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4240: k4=func(p2)-fx;
1.203 brouard 4241: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4242: if(k1*k2*k3*k4 <0.){
1.208 brouard 4243: firstime=1;
1.203 brouard 4244: kmax=kmax+10;
1.208 brouard 4245: }
4246: if(kmax >=10 || firstime ==1){
1.246 brouard 4247: 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);
4248: 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 4249: 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);
4250: 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);
4251: }
4252: #ifdef DEBUGHESSIJ
4253: v1=hess[thetai][thetai];
4254: v2=hess[thetaj][thetaj];
4255: cv12=res;
4256: /* Computing eigen value of Hessian matrix */
4257: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4258: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4259: if ((lc2 <0) || (lc1 <0) ){
4260: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4261: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4262: 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);
4263: 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);
4264: }
1.126 brouard 4265: #endif
4266: }
4267: return res;
4268: }
4269:
1.203 brouard 4270: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4271: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4272: /* { */
4273: /* int i; */
4274: /* int l=1, lmax=20; */
4275: /* double k1,k2,k3,k4,res,fx; */
4276: /* double p2[MAXPARM+1]; */
4277: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4278: /* int k=0,kmax=10; */
4279: /* double l1; */
4280:
4281: /* fx=func(x); */
4282: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4283: /* l1=pow(10,l); */
4284: /* delts=delt; */
4285: /* for(k=1 ; k <kmax; k=k+1){ */
4286: /* delt = delti*(l1*k); */
4287: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4288: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4289: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4290: /* k1=func(p2)-fx; */
4291:
4292: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4293: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4294: /* k2=func(p2)-fx; */
4295:
4296: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4297: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4298: /* k3=func(p2)-fx; */
4299:
4300: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4301: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4302: /* k4=func(p2)-fx; */
4303: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4304: /* #ifdef DEBUGHESSIJ */
4305: /* 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); */
4306: /* 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); */
4307: /* #endif */
4308: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4309: /* k=kmax; */
4310: /* } */
4311: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4312: /* k=kmax; l=lmax*10; */
4313: /* } */
4314: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4315: /* delts=delt; */
4316: /* } */
4317: /* } /\* End loop k *\/ */
4318: /* } */
4319: /* delti[theta]=delts; */
4320: /* return res; */
4321: /* } */
4322:
4323:
1.126 brouard 4324: /************** Inverse of matrix **************/
4325: void ludcmp(double **a, int n, int *indx, double *d)
4326: {
4327: int i,imax,j,k;
4328: double big,dum,sum,temp;
4329: double *vv;
4330:
4331: vv=vector(1,n);
4332: *d=1.0;
4333: for (i=1;i<=n;i++) {
4334: big=0.0;
4335: for (j=1;j<=n;j++)
4336: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4337: if (big == 0.0){
4338: printf(" Singular Hessian matrix at row %d:\n",i);
4339: for (j=1;j<=n;j++) {
4340: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4341: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4342: }
4343: fflush(ficlog);
4344: fclose(ficlog);
4345: nrerror("Singular matrix in routine ludcmp");
4346: }
1.126 brouard 4347: vv[i]=1.0/big;
4348: }
4349: for (j=1;j<=n;j++) {
4350: for (i=1;i<j;i++) {
4351: sum=a[i][j];
4352: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4353: a[i][j]=sum;
4354: }
4355: big=0.0;
4356: for (i=j;i<=n;i++) {
4357: sum=a[i][j];
4358: for (k=1;k<j;k++)
4359: sum -= a[i][k]*a[k][j];
4360: a[i][j]=sum;
4361: if ( (dum=vv[i]*fabs(sum)) >= big) {
4362: big=dum;
4363: imax=i;
4364: }
4365: }
4366: if (j != imax) {
4367: for (k=1;k<=n;k++) {
4368: dum=a[imax][k];
4369: a[imax][k]=a[j][k];
4370: a[j][k]=dum;
4371: }
4372: *d = -(*d);
4373: vv[imax]=vv[j];
4374: }
4375: indx[j]=imax;
4376: if (a[j][j] == 0.0) a[j][j]=TINY;
4377: if (j != n) {
4378: dum=1.0/(a[j][j]);
4379: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4380: }
4381: }
4382: free_vector(vv,1,n); /* Doesn't work */
4383: ;
4384: }
4385:
4386: void lubksb(double **a, int n, int *indx, double b[])
4387: {
4388: int i,ii=0,ip,j;
4389: double sum;
4390:
4391: for (i=1;i<=n;i++) {
4392: ip=indx[i];
4393: sum=b[ip];
4394: b[ip]=b[i];
4395: if (ii)
4396: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4397: else if (sum) ii=i;
4398: b[i]=sum;
4399: }
4400: for (i=n;i>=1;i--) {
4401: sum=b[i];
4402: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4403: b[i]=sum/a[i][i];
4404: }
4405: }
4406:
4407: void pstamp(FILE *fichier)
4408: {
1.196 brouard 4409: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4410: }
4411:
1.253 brouard 4412:
4413:
1.126 brouard 4414: /************ Frequencies ********************/
1.251 brouard 4415: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4416: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4417: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4418: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4419:
1.265 brouard 4420: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4421: int iind=0, iage=0;
4422: int mi; /* Effective wave */
4423: int first;
4424: double ***freq; /* Frequencies */
1.268 brouard 4425: 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 */
4426: 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 4427: double *meanq, *stdq, *idq;
1.226 brouard 4428: double **meanqt;
4429: double *pp, **prop, *posprop, *pospropt;
4430: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4431: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4432: double agebegin, ageend;
4433:
4434: pp=vector(1,nlstate);
1.251 brouard 4435: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4436: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4437: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4438: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4439: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4440: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4441: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4442: meanqt=matrix(1,lastpass,1,nqtveff);
4443: strcpy(fileresp,"P_");
4444: strcat(fileresp,fileresu);
4445: /*strcat(fileresphtm,fileresu);*/
4446: if((ficresp=fopen(fileresp,"w"))==NULL) {
4447: printf("Problem with prevalence resultfile: %s\n", fileresp);
4448: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4449: exit(0);
4450: }
1.240 brouard 4451:
1.226 brouard 4452: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4453: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4454: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4455: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4456: fflush(ficlog);
4457: exit(70);
4458: }
4459: else{
4460: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4461: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4462: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4463: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4464: }
1.237 brouard 4465: 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 4466:
1.226 brouard 4467: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4468: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4469: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4470: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4471: fflush(ficlog);
4472: exit(70);
1.240 brouard 4473: } else{
1.226 brouard 4474: 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 4475: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4476: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4477: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4478: }
1.240 brouard 4479: 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);
4480:
1.253 brouard 4481: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4482: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4483: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4484: j1=0;
1.126 brouard 4485:
1.227 brouard 4486: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4487: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4488: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4489:
4490:
1.226 brouard 4491: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4492: reference=low_education V1=0,V2=0
4493: med_educ V1=1 V2=0,
4494: high_educ V1=0 V2=1
4495: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4496: */
1.249 brouard 4497: dateintsum=0;
4498: k2cpt=0;
4499:
1.253 brouard 4500: if(cptcoveff == 0 )
1.265 brouard 4501: nl=1; /* Constant and age model only */
1.253 brouard 4502: else
4503: nl=2;
1.265 brouard 4504:
4505: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4506: /* Loop on nj=1 or 2 if dummy covariates j!=0
4507: * Loop on j1(1 to 2**cptcoveff) covariate combination
4508: * freq[s1][s2][iage] =0.
4509: * Loop on iind
4510: * ++freq[s1][s2][iage] weighted
4511: * end iind
4512: * if covariate and j!0
4513: * headers Variable on one line
4514: * endif cov j!=0
4515: * header of frequency table by age
4516: * Loop on age
4517: * pp[s1]+=freq[s1][s2][iage] weighted
4518: * pos+=freq[s1][s2][iage] weighted
4519: * Loop on s1 initial state
4520: * fprintf(ficresp
4521: * end s1
4522: * end age
4523: * if j!=0 computes starting values
4524: * end compute starting values
4525: * end j1
4526: * end nl
4527: */
1.253 brouard 4528: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4529: if(nj==1)
4530: j=0; /* First pass for the constant */
1.265 brouard 4531: else{
1.253 brouard 4532: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4533: }
1.251 brouard 4534: first=1;
1.265 brouard 4535: 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 4536: posproptt=0.;
4537: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4538: scanf("%d", i);*/
4539: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4540: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4541: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4542: freq[i][s2][m]=0;
1.251 brouard 4543:
4544: for (i=1; i<=nlstate; i++) {
1.240 brouard 4545: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4546: prop[i][m]=0;
4547: posprop[i]=0;
4548: pospropt[i]=0;
4549: }
1.283 brouard 4550: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4551: idq[z1]=0.;
4552: meanq[z1]=0.;
4553: stdq[z1]=0.;
1.283 brouard 4554: }
4555: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4556: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4557: /* meanqt[m][z1]=0.; */
4558: /* } */
4559: /* } */
1.251 brouard 4560: /* dateintsum=0; */
4561: /* k2cpt=0; */
4562:
1.265 brouard 4563: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4564: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4565: bool=1;
4566: if(j !=0){
4567: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4568: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4569: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4570: /* if(Tvaraff[z1] ==-20){ */
4571: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4572: /* }else if(Tvaraff[z1] ==-10){ */
4573: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4574: /* }else */
4575: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4576: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4577: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4578: /* 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",
4579: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4580: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4581: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4582: } /* Onlyf fixed */
4583: } /* end z1 */
4584: } /* cptcovn > 0 */
4585: } /* end any */
4586: }/* end j==0 */
1.265 brouard 4587: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4588: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4589: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4590: m=mw[mi][iind];
4591: if(j!=0){
4592: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4593: for (z1=1; z1<=cptcoveff; z1++) {
4594: if( Fixed[Tmodelind[z1]]==1){
4595: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4596: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4597: value is -1, we don't select. It differs from the
4598: constant and age model which counts them. */
4599: bool=0; /* not selected */
4600: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4601: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4602: bool=0;
4603: }
4604: }
4605: }
4606: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4607: } /* end j==0 */
4608: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4609: if(bool==1){ /*Selected */
1.251 brouard 4610: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4611: and mw[mi+1][iind]. dh depends on stepm. */
4612: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4613: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4614: if(m >=firstpass && m <=lastpass){
4615: k2=anint[m][iind]+(mint[m][iind]/12.);
4616: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4617: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4618: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4619: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4620: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4621: if (m<lastpass) {
4622: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4623: /* 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]); */
4624: if(s[m][iind]==-1)
4625: 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.));
4626: 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 4627: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean */
4628: idq[z1]=idq[z1]+weight[iind];
4629: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4630: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4631: }
1.251 brouard 4632: /* if((int)agev[m][iind] == 55) */
4633: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4634: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4635: 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 4636: }
1.251 brouard 4637: } /* end if between passes */
4638: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4639: dateintsum=dateintsum+k2; /* on all covariates ?*/
4640: k2cpt++;
4641: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4642: }
1.251 brouard 4643: }else{
4644: bool=1;
4645: }/* end bool 2 */
4646: } /* end m */
1.284 brouard 4647: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4648: /* idq[z1]=idq[z1]+weight[iind]; */
4649: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4650: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4651: /* } */
1.251 brouard 4652: } /* end bool */
4653: } /* end iind = 1 to imx */
4654: /* prop[s][age] is feeded for any initial and valid live state as well as
4655: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4656:
4657:
4658: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4659: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4660: pstamp(ficresp);
1.251 brouard 4661: if (cptcoveff>0 && j!=0){
1.265 brouard 4662: pstamp(ficresp);
1.251 brouard 4663: printf( "\n#********** Variable ");
4664: fprintf(ficresp, "\n#********** Variable ");
4665: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4666: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4667: fprintf(ficlog, "\n#********** Variable ");
4668: for (z1=1; z1<=cptcoveff; z1++){
4669: if(!FixedV[Tvaraff[z1]]){
4670: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4671: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4672: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4673: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4674: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4675: }else{
1.251 brouard 4676: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4677: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4678: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4679: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4680: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4681: }
4682: }
4683: printf( "**********\n#");
4684: fprintf(ficresp, "**********\n#");
4685: fprintf(ficresphtm, "**********</h3>\n");
4686: fprintf(ficresphtmfr, "**********</h3>\n");
4687: fprintf(ficlog, "**********\n");
4688: }
1.284 brouard 4689: /*
4690: Printing means of quantitative variables if any
4691: */
4692: for (z1=1; z1<= nqfveff; z1++) {
1.285 brouard 4693: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.0f individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.284 brouard 4694: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
4695: if(weightopt==1){
4696: printf(" Weighted mean and standard deviation of");
4697: fprintf(ficlog," Weighted mean and standard deviation of");
4698: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4699: }
1.285 brouard 4700: 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]));
4701: 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]));
4702: 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 4703: }
4704: /* for (z1=1; z1<= nqtveff; z1++) { */
4705: /* for(m=1;m<=lastpass;m++){ */
4706: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4707: /* } */
4708: /* } */
1.283 brouard 4709:
1.251 brouard 4710: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4711: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4712: fprintf(ficresp, " Age");
4713: 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 4714: for(i=1; i<=nlstate;i++) {
1.265 brouard 4715: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4716: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4717: }
1.265 brouard 4718: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4719: fprintf(ficresphtm, "\n");
4720:
4721: /* Header of frequency table by age */
4722: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4723: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4724: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4725: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4726: if(s2!=0 && m!=0)
4727: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4728: }
1.226 brouard 4729: }
1.251 brouard 4730: fprintf(ficresphtmfr, "\n");
4731:
4732: /* For each age */
4733: for(iage=iagemin; iage <= iagemax+3; iage++){
4734: fprintf(ficresphtm,"<tr>");
4735: if(iage==iagemax+1){
4736: fprintf(ficlog,"1");
4737: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4738: }else if(iage==iagemax+2){
4739: fprintf(ficlog,"0");
4740: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4741: }else if(iage==iagemax+3){
4742: fprintf(ficlog,"Total");
4743: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4744: }else{
1.240 brouard 4745: if(first==1){
1.251 brouard 4746: first=0;
4747: printf("See log file for details...\n");
4748: }
4749: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4750: fprintf(ficlog,"Age %d", iage);
4751: }
1.265 brouard 4752: for(s1=1; s1 <=nlstate ; s1++){
4753: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4754: pp[s1] += freq[s1][m][iage];
1.251 brouard 4755: }
1.265 brouard 4756: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4757: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4758: pos += freq[s1][m][iage];
4759: if(pp[s1]>=1.e-10){
1.251 brouard 4760: if(first==1){
1.265 brouard 4761: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4762: }
1.265 brouard 4763: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4764: }else{
4765: if(first==1)
1.265 brouard 4766: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4767: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4768: }
4769: }
4770:
1.265 brouard 4771: for(s1=1; s1 <=nlstate ; s1++){
4772: /* posprop[s1]=0; */
4773: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4774: pp[s1] += freq[s1][m][iage];
4775: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4776:
4777: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4778: pos += pp[s1]; /* pos is the total number of transitions until this age */
4779: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4780: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4781: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4782: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4783: }
4784:
4785: /* Writing ficresp */
4786: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4787: if( iage <= iagemax){
4788: fprintf(ficresp," %d",iage);
4789: }
4790: }else if( nj==2){
4791: if( iage <= iagemax){
4792: fprintf(ficresp," %d",iage);
4793: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4794: }
1.240 brouard 4795: }
1.265 brouard 4796: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4797: if(pos>=1.e-5){
1.251 brouard 4798: if(first==1)
1.265 brouard 4799: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4800: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4801: }else{
4802: if(first==1)
1.265 brouard 4803: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4804: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4805: }
4806: if( iage <= iagemax){
4807: if(pos>=1.e-5){
1.265 brouard 4808: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4809: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4810: }else if( nj==2){
4811: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4812: }
4813: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4814: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4815: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4816: } else{
4817: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4818: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4819: }
1.240 brouard 4820: }
1.265 brouard 4821: pospropt[s1] +=posprop[s1];
4822: } /* end loop s1 */
1.251 brouard 4823: /* pospropt=0.; */
1.265 brouard 4824: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4825: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4826: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4827: if(first==1){
1.265 brouard 4828: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4829: }
1.265 brouard 4830: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4831: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4832: }
1.265 brouard 4833: if(s1!=0 && m!=0)
4834: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4835: }
1.265 brouard 4836: } /* end loop s1 */
1.251 brouard 4837: posproptt=0.;
1.265 brouard 4838: for(s1=1; s1 <=nlstate; s1++){
4839: posproptt += pospropt[s1];
1.251 brouard 4840: }
4841: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4842: fprintf(ficresphtm,"</tr>\n");
4843: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4844: if(iage <= iagemax)
4845: fprintf(ficresp,"\n");
1.240 brouard 4846: }
1.251 brouard 4847: if(first==1)
4848: printf("Others in log...\n");
4849: fprintf(ficlog,"\n");
4850: } /* end loop age iage */
1.265 brouard 4851:
1.251 brouard 4852: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4853: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4854: if(posproptt < 1.e-5){
1.265 brouard 4855: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4856: }else{
1.265 brouard 4857: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4858: }
1.226 brouard 4859: }
1.251 brouard 4860: fprintf(ficresphtm,"</tr>\n");
4861: fprintf(ficresphtm,"</table>\n");
4862: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4863: if(posproptt < 1.e-5){
1.251 brouard 4864: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4865: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4866: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4867: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4868: invalidvarcomb[j1]=1;
1.226 brouard 4869: }else{
1.251 brouard 4870: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4871: invalidvarcomb[j1]=0;
1.226 brouard 4872: }
1.251 brouard 4873: fprintf(ficresphtmfr,"</table>\n");
4874: fprintf(ficlog,"\n");
4875: if(j!=0){
4876: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
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) {
1.265 brouard 4880: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4881: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4882: if(j1==1){ /* All dummy covariates to zero */
4883: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4884: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4885: printf("%d%d ",i,k);
4886: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4887: 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]));
4888: 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]));
4889: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4890: }
1.253 brouard 4891: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4892: for(iage=iagemin; iage <= iagemax+3; iage++){
4893: x[iage]= (double)iage;
4894: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4895: /* 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 4896: }
1.268 brouard 4897: /* Some are not finite, but linreg will ignore these ages */
4898: no=0;
1.253 brouard 4899: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4900: pstart[s1]=b;
4901: pstart[s1-1]=a;
1.252 brouard 4902: }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 */
4903: 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]);
4904: 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 4905: 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 4906: printf("%d%d ",i,k);
4907: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4908: 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 4909: }else{ /* Other cases, like quantitative fixed or varying covariates */
4910: ;
4911: }
4912: /* printf("%12.7f )", param[i][jj][k]); */
4913: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4914: s1++;
1.251 brouard 4915: } /* end jj */
4916: } /* end k!= i */
4917: } /* end k */
1.265 brouard 4918: } /* end i, s1 */
1.251 brouard 4919: } /* end j !=0 */
4920: } /* end selected combination of covariate j1 */
4921: if(j==0){ /* We can estimate starting values from the occurences in each case */
4922: printf("#Freqsummary: Starting values for the constants:\n");
4923: fprintf(ficlog,"\n");
1.265 brouard 4924: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4925: for(k=1; k <=(nlstate+ndeath); k++){
4926: if (k != i) {
4927: printf("%d%d ",i,k);
4928: fprintf(ficlog,"%d%d ",i,k);
4929: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4930: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4931: if(jj==1){ /* Age has to be done */
1.265 brouard 4932: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4933: 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]));
4934: 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 4935: }
4936: /* printf("%12.7f )", param[i][jj][k]); */
4937: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4938: s1++;
1.250 brouard 4939: }
1.251 brouard 4940: printf("\n");
4941: fprintf(ficlog,"\n");
1.250 brouard 4942: }
4943: }
1.284 brouard 4944: } /* end of state i */
1.251 brouard 4945: printf("#Freqsummary\n");
4946: fprintf(ficlog,"\n");
1.265 brouard 4947: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4948: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4949: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4950: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4951: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4952: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4953: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4954: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4955: /* } */
4956: }
1.265 brouard 4957: } /* end loop s1 */
1.251 brouard 4958:
4959: printf("\n");
4960: fprintf(ficlog,"\n");
4961: } /* end j=0 */
1.249 brouard 4962: } /* end j */
1.252 brouard 4963:
1.253 brouard 4964: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4965: for(i=1, jk=1; i <=nlstate; i++){
4966: for(j=1; j <=nlstate+ndeath; j++){
4967: if(j!=i){
4968: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4969: printf("%1d%1d",i,j);
4970: fprintf(ficparo,"%1d%1d",i,j);
4971: for(k=1; k<=ncovmodel;k++){
4972: /* printf(" %lf",param[i][j][k]); */
4973: /* fprintf(ficparo," %lf",param[i][j][k]); */
4974: p[jk]=pstart[jk];
4975: printf(" %f ",pstart[jk]);
4976: fprintf(ficparo," %f ",pstart[jk]);
4977: jk++;
4978: }
4979: printf("\n");
4980: fprintf(ficparo,"\n");
4981: }
4982: }
4983: }
4984: } /* end mle=-2 */
1.226 brouard 4985: dateintmean=dateintsum/k2cpt;
1.240 brouard 4986:
1.226 brouard 4987: fclose(ficresp);
4988: fclose(ficresphtm);
4989: fclose(ficresphtmfr);
1.283 brouard 4990: free_vector(idq,1,nqfveff);
1.226 brouard 4991: free_vector(meanq,1,nqfveff);
1.284 brouard 4992: free_vector(stdq,1,nqfveff);
1.226 brouard 4993: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4994: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4995: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4996: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4997: free_vector(pospropt,1,nlstate);
4998: free_vector(posprop,1,nlstate);
1.251 brouard 4999: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5000: free_vector(pp,1,nlstate);
5001: /* End of freqsummary */
5002: }
1.126 brouard 5003:
1.268 brouard 5004: /* Simple linear regression */
5005: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5006:
5007: /* y=a+bx regression */
5008: double sumx = 0.0; /* sum of x */
5009: double sumx2 = 0.0; /* sum of x**2 */
5010: double sumxy = 0.0; /* sum of x * y */
5011: double sumy = 0.0; /* sum of y */
5012: double sumy2 = 0.0; /* sum of y**2 */
5013: double sume2 = 0.0; /* sum of square or residuals */
5014: double yhat;
5015:
5016: double denom=0;
5017: int i;
5018: int ne=*no;
5019:
5020: for ( i=ifi, ne=0;i<=ila;i++) {
5021: if(!isfinite(x[i]) || !isfinite(y[i])){
5022: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5023: continue;
5024: }
5025: ne=ne+1;
5026: sumx += x[i];
5027: sumx2 += x[i]*x[i];
5028: sumxy += x[i] * y[i];
5029: sumy += y[i];
5030: sumy2 += y[i]*y[i];
5031: denom = (ne * sumx2 - sumx*sumx);
5032: /* 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); */
5033: }
5034:
5035: denom = (ne * sumx2 - sumx*sumx);
5036: if (denom == 0) {
5037: // vertical, slope m is infinity
5038: *b = INFINITY;
5039: *a = 0;
5040: if (r) *r = 0;
5041: return 1;
5042: }
5043:
5044: *b = (ne * sumxy - sumx * sumy) / denom;
5045: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5046: if (r!=NULL) {
5047: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5048: sqrt((sumx2 - sumx*sumx/ne) *
5049: (sumy2 - sumy*sumy/ne));
5050: }
5051: *no=ne;
5052: for ( i=ifi, ne=0;i<=ila;i++) {
5053: if(!isfinite(x[i]) || !isfinite(y[i])){
5054: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5055: continue;
5056: }
5057: ne=ne+1;
5058: yhat = y[i] - *a -*b* x[i];
5059: sume2 += yhat * yhat ;
5060:
5061: denom = (ne * sumx2 - sumx*sumx);
5062: /* 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); */
5063: }
5064: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5065: *sa= *sb * sqrt(sumx2/ne);
5066:
5067: return 0;
5068: }
5069:
1.126 brouard 5070: /************ Prevalence ********************/
1.227 brouard 5071: 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)
5072: {
5073: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5074: in each health status at the date of interview (if between dateprev1 and dateprev2).
5075: We still use firstpass and lastpass as another selection.
5076: */
1.126 brouard 5077:
1.227 brouard 5078: int i, m, jk, j1, bool, z1,j, iv;
5079: int mi; /* Effective wave */
5080: int iage;
5081: double agebegin, ageend;
5082:
5083: double **prop;
5084: double posprop;
5085: double y2; /* in fractional years */
5086: int iagemin, iagemax;
5087: int first; /** to stop verbosity which is redirected to log file */
5088:
5089: iagemin= (int) agemin;
5090: iagemax= (int) agemax;
5091: /*pp=vector(1,nlstate);*/
1.251 brouard 5092: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5093: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5094: j1=0;
1.222 brouard 5095:
1.227 brouard 5096: /*j=cptcoveff;*/
5097: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5098:
1.288 brouard 5099: first=0;
1.227 brouard 5100: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5101: for (i=1; i<=nlstate; i++)
1.251 brouard 5102: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5103: prop[i][iage]=0.0;
5104: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5105: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5106: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5107:
5108: for (i=1; i<=imx; i++) { /* Each individual */
5109: bool=1;
5110: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5111: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5112: m=mw[mi][i];
5113: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5114: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5115: for (z1=1; z1<=cptcoveff; z1++){
5116: if( Fixed[Tmodelind[z1]]==1){
5117: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5118: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5119: bool=0;
5120: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5121: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5122: bool=0;
5123: }
5124: }
5125: if(bool==1){ /* Otherwise we skip that wave/person */
5126: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5127: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5128: if(m >=firstpass && m <=lastpass){
5129: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5130: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5131: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5132: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5133: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5134: 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);
5135: exit(1);
5136: }
5137: if (s[m][i]>0 && s[m][i]<=nlstate) {
5138: /*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]]);*/
5139: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5140: prop[s[m][i]][iagemax+3] += weight[i];
5141: } /* end valid statuses */
5142: } /* end selection of dates */
5143: } /* end selection of waves */
5144: } /* end bool */
5145: } /* end wave */
5146: } /* end individual */
5147: for(i=iagemin; i <= iagemax+3; i++){
5148: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5149: posprop += prop[jk][i];
5150: }
5151:
5152: for(jk=1; jk <=nlstate ; jk++){
5153: if( i <= iagemax){
5154: if(posprop>=1.e-5){
5155: probs[i][jk][j1]= prop[jk][i]/posprop;
5156: } else{
1.288 brouard 5157: if(!first){
5158: first=1;
1.266 brouard 5159: 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]);
5160: }else{
1.288 brouard 5161: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases.\n",jk,i,jk, j1,probs[i][jk][j1]);
1.227 brouard 5162: }
5163: }
5164: }
5165: }/* end jk */
5166: }/* end i */
1.222 brouard 5167: /*} *//* end i1 */
1.227 brouard 5168: } /* end j1 */
1.222 brouard 5169:
1.227 brouard 5170: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5171: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5172: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5173: } /* End of prevalence */
1.126 brouard 5174:
5175: /************* Waves Concatenation ***************/
5176:
5177: 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)
5178: {
5179: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5180: Death is a valid wave (if date is known).
5181: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5182: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5183: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5184: */
1.126 brouard 5185:
1.224 brouard 5186: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5187: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5188: double sum=0., jmean=0.;*/
1.224 brouard 5189: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5190: int j, k=0,jk, ju, jl;
5191: double sum=0.;
5192: first=0;
1.214 brouard 5193: firstwo=0;
1.217 brouard 5194: firsthree=0;
1.218 brouard 5195: firstfour=0;
1.164 brouard 5196: jmin=100000;
1.126 brouard 5197: jmax=-1;
5198: jmean=0.;
1.224 brouard 5199:
5200: /* Treating live states */
1.214 brouard 5201: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5202: mi=0; /* First valid wave */
1.227 brouard 5203: mli=0; /* Last valid wave */
1.126 brouard 5204: m=firstpass;
1.214 brouard 5205: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5206: 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 */
5207: mli=m-1;/* mw[++mi][i]=m-1; */
5208: }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 */
5209: mw[++mi][i]=m;
5210: mli=m;
1.224 brouard 5211: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5212: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5213: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5214: }
1.227 brouard 5215: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5216: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5217: break;
1.224 brouard 5218: #else
1.227 brouard 5219: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5220: if(firsthree == 0){
1.262 brouard 5221: 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 5222: firsthree=1;
5223: }
1.262 brouard 5224: 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 5225: mw[++mi][i]=m;
5226: mli=m;
5227: }
5228: if(s[m][i]==-2){ /* Vital status is really unknown */
5229: nbwarn++;
5230: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5231: 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);
5232: 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);
5233: }
5234: break;
5235: }
5236: break;
1.224 brouard 5237: #endif
1.227 brouard 5238: }/* End m >= lastpass */
1.126 brouard 5239: }/* end while */
1.224 brouard 5240:
1.227 brouard 5241: /* 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 5242: /* After last pass */
1.224 brouard 5243: /* Treating death states */
1.214 brouard 5244: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5245: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5246: /* } */
1.126 brouard 5247: mi++; /* Death is another wave */
5248: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5249: /* Only death is a correct wave */
1.126 brouard 5250: mw[mi][i]=m;
1.257 brouard 5251: } /* else not in a death state */
1.224 brouard 5252: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5253: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5254: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5255: 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 */
5256: nbwarn++;
5257: if(firstfiv==0){
5258: 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 );
5259: firstfiv=1;
5260: }else{
5261: 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 );
5262: }
5263: }else{ /* Death occured afer last wave potential bias */
5264: nberr++;
5265: if(firstwo==0){
1.257 brouard 5266: 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 5267: firstwo=1;
5268: }
1.257 brouard 5269: 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 5270: }
1.257 brouard 5271: }else{ /* if date of interview is unknown */
1.227 brouard 5272: /* death is known but not confirmed by death status at any wave */
5273: if(firstfour==0){
5274: 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 );
5275: firstfour=1;
5276: }
5277: 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 5278: }
1.224 brouard 5279: } /* end if date of death is known */
5280: #endif
5281: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5282: /* wav[i]=mw[mi][i]; */
1.126 brouard 5283: if(mi==0){
5284: nbwarn++;
5285: if(first==0){
1.227 brouard 5286: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5287: first=1;
1.126 brouard 5288: }
5289: if(first==1){
1.227 brouard 5290: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5291: }
5292: } /* end mi==0 */
5293: } /* End individuals */
1.214 brouard 5294: /* wav and mw are no more changed */
1.223 brouard 5295:
1.214 brouard 5296:
1.126 brouard 5297: for(i=1; i<=imx; i++){
5298: for(mi=1; mi<wav[i];mi++){
5299: if (stepm <=0)
1.227 brouard 5300: dh[mi][i]=1;
1.126 brouard 5301: else{
1.260 brouard 5302: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5303: if (agedc[i] < 2*AGESUP) {
5304: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5305: if(j==0) j=1; /* Survives at least one month after exam */
5306: else if(j<0){
5307: nberr++;
5308: 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]);
5309: j=1; /* Temporary Dangerous patch */
5310: 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);
5311: 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]);
5312: 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);
5313: }
5314: k=k+1;
5315: if (j >= jmax){
5316: jmax=j;
5317: ijmax=i;
5318: }
5319: if (j <= jmin){
5320: jmin=j;
5321: ijmin=i;
5322: }
5323: sum=sum+j;
5324: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5325: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5326: }
5327: }
5328: else{
5329: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5330: /* 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 5331:
1.227 brouard 5332: k=k+1;
5333: if (j >= jmax) {
5334: jmax=j;
5335: ijmax=i;
5336: }
5337: else if (j <= jmin){
5338: jmin=j;
5339: ijmin=i;
5340: }
5341: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5342: /*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]);*/
5343: if(j<0){
5344: nberr++;
5345: 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]);
5346: 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]);
5347: }
5348: sum=sum+j;
5349: }
5350: jk= j/stepm;
5351: jl= j -jk*stepm;
5352: ju= j -(jk+1)*stepm;
5353: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5354: if(jl==0){
5355: dh[mi][i]=jk;
5356: bh[mi][i]=0;
5357: }else{ /* We want a negative bias in order to only have interpolation ie
5358: * to avoid the price of an extra matrix product in likelihood */
5359: dh[mi][i]=jk+1;
5360: bh[mi][i]=ju;
5361: }
5362: }else{
5363: if(jl <= -ju){
5364: dh[mi][i]=jk;
5365: bh[mi][i]=jl; /* bias is positive if real duration
5366: * is higher than the multiple of stepm and negative otherwise.
5367: */
5368: }
5369: else{
5370: dh[mi][i]=jk+1;
5371: bh[mi][i]=ju;
5372: }
5373: if(dh[mi][i]==0){
5374: dh[mi][i]=1; /* At least one step */
5375: bh[mi][i]=ju; /* At least one step */
5376: /* 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);*/
5377: }
5378: } /* end if mle */
1.126 brouard 5379: }
5380: } /* end wave */
5381: }
5382: jmean=sum/k;
5383: 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 5384: 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 5385: }
1.126 brouard 5386:
5387: /*********** Tricode ****************************/
1.220 brouard 5388: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5389: {
5390: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5391: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5392: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5393: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5394: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5395: */
1.130 brouard 5396:
1.242 brouard 5397: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5398: int modmaxcovj=0; /* Modality max of covariates j */
5399: int cptcode=0; /* Modality max of covariates j */
5400: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5401:
5402:
1.242 brouard 5403: /* cptcoveff=0; */
5404: /* *cptcov=0; */
1.126 brouard 5405:
1.242 brouard 5406: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5407: for (k=1; k <= maxncov; k++)
5408: for(j=1; j<=2; j++)
5409: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5410:
1.242 brouard 5411: /* Loop on covariates without age and products and no quantitative variable */
5412: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5413: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5414: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5415: switch(Fixed[k]) {
5416: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5417: 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*/
5418: ij=(int)(covar[Tvar[k]][i]);
5419: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5420: * If product of Vn*Vm, still boolean *:
5421: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5422: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5423: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5424: modality of the nth covariate of individual i. */
5425: if (ij > modmaxcovj)
5426: modmaxcovj=ij;
5427: else if (ij < modmincovj)
5428: modmincovj=ij;
1.287 brouard 5429: if (ij <0 || ij >1 ){
5430: printf("Information, IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5431: fprintf(ficlog,"Information, currently IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5432: }
5433: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5434: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5435: exit(1);
5436: }else
5437: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5438: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5439: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5440: /* getting the maximum value of the modality of the covariate
5441: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5442: female ies 1, then modmaxcovj=1.
5443: */
5444: } /* end for loop on individuals i */
5445: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5446: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5447: cptcode=modmaxcovj;
5448: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5449: /*for (i=0; i<=cptcode; i++) {*/
5450: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5451: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5452: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5453: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5454: if( j != -1){
5455: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5456: covariate for which somebody answered excluding
5457: undefined. Usually 2: 0 and 1. */
5458: }
5459: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5460: covariate for which somebody answered including
5461: undefined. Usually 3: -1, 0 and 1. */
5462: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5463: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5464: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5465:
1.242 brouard 5466: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5467: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5468: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5469: /* modmincovj=3; modmaxcovj = 7; */
5470: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5471: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5472: /* defining two dummy variables: variables V1_1 and V1_2.*/
5473: /* nbcode[Tvar[j]][ij]=k; */
5474: /* nbcode[Tvar[j]][1]=0; */
5475: /* nbcode[Tvar[j]][2]=1; */
5476: /* nbcode[Tvar[j]][3]=2; */
5477: /* To be continued (not working yet). */
5478: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5479:
5480: /* 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*/
5481: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5482: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5483: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5484: /*, could be restored in the future */
5485: for (i=0; i<=1; i++) { /* i= 1 to 2 for dichotomous, or from 1 to 3 or from -1 or 0 to 1 currently*/
1.242 brouard 5486: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5487: break;
5488: }
5489: ij++;
1.287 brouard 5490: nbcode[Tvar[k]][ij]=i; /* stores the original value of modality i in an array nbcode, ij modality from 1 to last non-nul modality. nbcode[1][1]=0 nbcode[1][2]=1 . Could be -1*/
1.242 brouard 5491: cptcode = ij; /* New max modality for covar j */
5492: } /* end of loop on modality i=-1 to 1 or more */
5493: break;
5494: case 1: /* Testing on varying covariate, could be simple and
5495: * should look at waves or product of fixed *
5496: * varying. No time to test -1, assuming 0 and 1 only */
5497: ij=0;
5498: for(i=0; i<=1;i++){
5499: nbcode[Tvar[k]][++ij]=i;
5500: }
5501: break;
5502: default:
5503: break;
5504: } /* end switch */
5505: } /* end dummy test */
1.287 brouard 5506: } /* end of loop on model-covariate k. nbcode[Tvark][1]=-1, nbcode[Tvark][1]=0 and nbcode[Tvark][2]=1 sets the value of covariate k*/
1.242 brouard 5507:
5508: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5509: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5510: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5511: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5512: 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 */
5513: 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 */
5514: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5515: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5516:
5517: ij=0;
5518: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5519: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5520: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5521: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5522: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5523: /* If product not in single variable we don't print results */
5524: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5525: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5526: 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*/
5527: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5528: 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 */
5529: if(Fixed[k]!=0)
5530: anyvaryingduminmodel=1;
5531: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5532: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5533: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5534: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5535: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5536: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5537: }
5538: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5539: /* ij--; */
5540: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5541: *cptcov=ij; /*Number of total real effective covariates: effective
5542: * because they can be excluded from the model and real
5543: * if in the model but excluded because missing values, but how to get k from ij?*/
5544: for(j=ij+1; j<= cptcovt; j++){
5545: Tvaraff[j]=0;
5546: Tmodelind[j]=0;
5547: }
5548: for(j=ntveff+1; j<= cptcovt; j++){
5549: TmodelInvind[j]=0;
5550: }
5551: /* To be sorted */
5552: ;
5553: }
1.126 brouard 5554:
1.145 brouard 5555:
1.126 brouard 5556: /*********** Health Expectancies ****************/
5557:
1.235 brouard 5558: 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 5559:
5560: {
5561: /* Health expectancies, no variances */
1.164 brouard 5562: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5563: int nhstepma, nstepma; /* Decreasing with age */
5564: double age, agelim, hf;
5565: double ***p3mat;
5566: double eip;
5567:
1.238 brouard 5568: /* pstamp(ficreseij); */
1.126 brouard 5569: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5570: fprintf(ficreseij,"# Age");
5571: for(i=1; i<=nlstate;i++){
5572: for(j=1; j<=nlstate;j++){
5573: fprintf(ficreseij," e%1d%1d ",i,j);
5574: }
5575: fprintf(ficreseij," e%1d. ",i);
5576: }
5577: fprintf(ficreseij,"\n");
5578:
5579:
5580: if(estepm < stepm){
5581: printf ("Problem %d lower than %d\n",estepm, stepm);
5582: }
5583: else hstepm=estepm;
5584: /* We compute the life expectancy from trapezoids spaced every estepm months
5585: * This is mainly to measure the difference between two models: for example
5586: * if stepm=24 months pijx are given only every 2 years and by summing them
5587: * we are calculating an estimate of the Life Expectancy assuming a linear
5588: * progression in between and thus overestimating or underestimating according
5589: * to the curvature of the survival function. If, for the same date, we
5590: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5591: * to compare the new estimate of Life expectancy with the same linear
5592: * hypothesis. A more precise result, taking into account a more precise
5593: * curvature will be obtained if estepm is as small as stepm. */
5594:
5595: /* For example we decided to compute the life expectancy with the smallest unit */
5596: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5597: nhstepm is the number of hstepm from age to agelim
5598: nstepm is the number of stepm from age to agelin.
1.270 brouard 5599: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5600: and note for a fixed period like estepm months */
5601: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5602: survival function given by stepm (the optimization length). Unfortunately it
5603: means that if the survival funtion is printed only each two years of age and if
5604: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5605: results. So we changed our mind and took the option of the best precision.
5606: */
5607: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5608:
5609: agelim=AGESUP;
5610: /* If stepm=6 months */
5611: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5612: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5613:
5614: /* nhstepm age range expressed in number of stepm */
5615: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5616: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5617: /* if (stepm >= YEARM) hstepm=1;*/
5618: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5619: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5620:
5621: for (age=bage; age<=fage; age ++){
5622: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5623: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5624: /* if (stepm >= YEARM) hstepm=1;*/
5625: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5626:
5627: /* If stepm=6 months */
5628: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5629: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5630:
1.235 brouard 5631: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5632:
5633: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5634:
5635: printf("%d|",(int)age);fflush(stdout);
5636: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5637:
5638: /* Computing expectancies */
5639: for(i=1; i<=nlstate;i++)
5640: for(j=1; j<=nlstate;j++)
5641: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5642: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5643:
5644: /* 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]);*/
5645:
5646: }
5647:
5648: fprintf(ficreseij,"%3.0f",age );
5649: for(i=1; i<=nlstate;i++){
5650: eip=0;
5651: for(j=1; j<=nlstate;j++){
5652: eip +=eij[i][j][(int)age];
5653: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5654: }
5655: fprintf(ficreseij,"%9.4f", eip );
5656: }
5657: fprintf(ficreseij,"\n");
5658:
5659: }
5660: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5661: printf("\n");
5662: fprintf(ficlog,"\n");
5663:
5664: }
5665:
1.235 brouard 5666: 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 5667:
5668: {
5669: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5670: to initial status i, ei. .
1.126 brouard 5671: */
5672: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5673: int nhstepma, nstepma; /* Decreasing with age */
5674: double age, agelim, hf;
5675: double ***p3matp, ***p3matm, ***varhe;
5676: double **dnewm,**doldm;
5677: double *xp, *xm;
5678: double **gp, **gm;
5679: double ***gradg, ***trgradg;
5680: int theta;
5681:
5682: double eip, vip;
5683:
5684: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5685: xp=vector(1,npar);
5686: xm=vector(1,npar);
5687: dnewm=matrix(1,nlstate*nlstate,1,npar);
5688: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5689:
5690: pstamp(ficresstdeij);
5691: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5692: fprintf(ficresstdeij,"# Age");
5693: for(i=1; i<=nlstate;i++){
5694: for(j=1; j<=nlstate;j++)
5695: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5696: fprintf(ficresstdeij," e%1d. ",i);
5697: }
5698: fprintf(ficresstdeij,"\n");
5699:
5700: pstamp(ficrescveij);
5701: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5702: fprintf(ficrescveij,"# Age");
5703: for(i=1; i<=nlstate;i++)
5704: for(j=1; j<=nlstate;j++){
5705: cptj= (j-1)*nlstate+i;
5706: for(i2=1; i2<=nlstate;i2++)
5707: for(j2=1; j2<=nlstate;j2++){
5708: cptj2= (j2-1)*nlstate+i2;
5709: if(cptj2 <= cptj)
5710: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5711: }
5712: }
5713: fprintf(ficrescveij,"\n");
5714:
5715: if(estepm < stepm){
5716: printf ("Problem %d lower than %d\n",estepm, stepm);
5717: }
5718: else hstepm=estepm;
5719: /* We compute the life expectancy from trapezoids spaced every estepm months
5720: * This is mainly to measure the difference between two models: for example
5721: * if stepm=24 months pijx are given only every 2 years and by summing them
5722: * we are calculating an estimate of the Life Expectancy assuming a linear
5723: * progression in between and thus overestimating or underestimating according
5724: * to the curvature of the survival function. If, for the same date, we
5725: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5726: * to compare the new estimate of Life expectancy with the same linear
5727: * hypothesis. A more precise result, taking into account a more precise
5728: * curvature will be obtained if estepm is as small as stepm. */
5729:
5730: /* For example we decided to compute the life expectancy with the smallest unit */
5731: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5732: nhstepm is the number of hstepm from age to agelim
5733: nstepm is the number of stepm from age to agelin.
5734: Look at hpijx to understand the reason of that which relies in memory size
5735: and note for a fixed period like estepm months */
5736: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5737: survival function given by stepm (the optimization length). Unfortunately it
5738: means that if the survival funtion is printed only each two years of age and if
5739: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5740: results. So we changed our mind and took the option of the best precision.
5741: */
5742: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5743:
5744: /* If stepm=6 months */
5745: /* nhstepm age range expressed in number of stepm */
5746: agelim=AGESUP;
5747: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5748: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5749: /* if (stepm >= YEARM) hstepm=1;*/
5750: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5751:
5752: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5753: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5754: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5755: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5756: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5757: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5758:
5759: for (age=bage; age<=fage; age ++){
5760: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5761: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5762: /* if (stepm >= YEARM) hstepm=1;*/
5763: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5764:
1.126 brouard 5765: /* If stepm=6 months */
5766: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5767: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5768:
5769: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5770:
1.126 brouard 5771: /* Computing Variances of health expectancies */
5772: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5773: decrease memory allocation */
5774: for(theta=1; theta <=npar; theta++){
5775: for(i=1; i<=npar; i++){
1.222 brouard 5776: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5777: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5778: }
1.235 brouard 5779: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5780: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5781:
1.126 brouard 5782: for(j=1; j<= nlstate; j++){
1.222 brouard 5783: for(i=1; i<=nlstate; i++){
5784: for(h=0; h<=nhstepm-1; h++){
5785: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5786: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5787: }
5788: }
1.126 brouard 5789: }
1.218 brouard 5790:
1.126 brouard 5791: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5792: for(h=0; h<=nhstepm-1; h++){
5793: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5794: }
1.126 brouard 5795: }/* End theta */
5796:
5797:
5798: for(h=0; h<=nhstepm-1; h++)
5799: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5800: for(theta=1; theta <=npar; theta++)
5801: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5802:
1.218 brouard 5803:
1.222 brouard 5804: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5805: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5806: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5807:
1.222 brouard 5808: printf("%d|",(int)age);fflush(stdout);
5809: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5810: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5811: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5812: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5813: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5814: for(ij=1;ij<=nlstate*nlstate;ij++)
5815: for(ji=1;ji<=nlstate*nlstate;ji++)
5816: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5817: }
5818: }
1.218 brouard 5819:
1.126 brouard 5820: /* Computing expectancies */
1.235 brouard 5821: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5822: for(i=1; i<=nlstate;i++)
5823: for(j=1; j<=nlstate;j++)
1.222 brouard 5824: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5825: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5826:
1.222 brouard 5827: /* 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 5828:
1.222 brouard 5829: }
1.269 brouard 5830:
5831: /* Standard deviation of expectancies ij */
1.126 brouard 5832: fprintf(ficresstdeij,"%3.0f",age );
5833: for(i=1; i<=nlstate;i++){
5834: eip=0.;
5835: vip=0.;
5836: for(j=1; j<=nlstate;j++){
1.222 brouard 5837: eip += eij[i][j][(int)age];
5838: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5839: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5840: 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 5841: }
5842: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5843: }
5844: fprintf(ficresstdeij,"\n");
1.218 brouard 5845:
1.269 brouard 5846: /* Variance of expectancies ij */
1.126 brouard 5847: fprintf(ficrescveij,"%3.0f",age );
5848: for(i=1; i<=nlstate;i++)
5849: for(j=1; j<=nlstate;j++){
1.222 brouard 5850: cptj= (j-1)*nlstate+i;
5851: for(i2=1; i2<=nlstate;i2++)
5852: for(j2=1; j2<=nlstate;j2++){
5853: cptj2= (j2-1)*nlstate+i2;
5854: if(cptj2 <= cptj)
5855: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5856: }
1.126 brouard 5857: }
5858: fprintf(ficrescveij,"\n");
1.218 brouard 5859:
1.126 brouard 5860: }
5861: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5862: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5863: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5864: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5865: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5866: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5867: printf("\n");
5868: fprintf(ficlog,"\n");
1.218 brouard 5869:
1.126 brouard 5870: free_vector(xm,1,npar);
5871: free_vector(xp,1,npar);
5872: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5873: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5874: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5875: }
1.218 brouard 5876:
1.126 brouard 5877: /************ Variance ******************/
1.235 brouard 5878: 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 5879: {
1.279 brouard 5880: /** Variance of health expectancies
5881: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5882: * double **newm;
5883: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5884: */
1.218 brouard 5885:
5886: /* int movingaverage(); */
5887: double **dnewm,**doldm;
5888: double **dnewmp,**doldmp;
5889: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 5890: int first=0;
1.218 brouard 5891: int k;
5892: double *xp;
1.279 brouard 5893: double **gp, **gm; /**< for var eij */
5894: double ***gradg, ***trgradg; /**< for var eij */
5895: double **gradgp, **trgradgp; /**< for var p point j */
5896: double *gpp, *gmp; /**< for var p point j */
5897: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5898: double ***p3mat;
5899: double age,agelim, hf;
5900: /* double ***mobaverage; */
5901: int theta;
5902: char digit[4];
5903: char digitp[25];
5904:
5905: char fileresprobmorprev[FILENAMELENGTH];
5906:
5907: if(popbased==1){
5908: if(mobilav!=0)
5909: strcpy(digitp,"-POPULBASED-MOBILAV_");
5910: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5911: }
5912: else
5913: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5914:
1.218 brouard 5915: /* if (mobilav!=0) { */
5916: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5917: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5918: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5919: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5920: /* } */
5921: /* } */
5922:
5923: strcpy(fileresprobmorprev,"PRMORPREV-");
5924: sprintf(digit,"%-d",ij);
5925: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5926: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5927: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5928: strcat(fileresprobmorprev,fileresu);
5929: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5930: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5931: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5932: }
5933: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5934: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5935: pstamp(ficresprobmorprev);
5936: 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 5937: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5938: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5939: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5940: }
5941: for(j=1;j<=cptcoveff;j++)
5942: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5943: fprintf(ficresprobmorprev,"\n");
5944:
1.218 brouard 5945: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5946: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5947: fprintf(ficresprobmorprev," p.%-d SE",j);
5948: for(i=1; i<=nlstate;i++)
5949: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5950: }
5951: fprintf(ficresprobmorprev,"\n");
5952:
5953: fprintf(ficgp,"\n# Routine varevsij");
5954: fprintf(ficgp,"\nunset title \n");
5955: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5956: 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");
5957: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 5958:
1.218 brouard 5959: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5960: pstamp(ficresvij);
5961: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5962: if(popbased==1)
5963: 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);
5964: else
5965: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5966: fprintf(ficresvij,"# Age");
5967: for(i=1; i<=nlstate;i++)
5968: for(j=1; j<=nlstate;j++)
5969: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5970: fprintf(ficresvij,"\n");
5971:
5972: xp=vector(1,npar);
5973: dnewm=matrix(1,nlstate,1,npar);
5974: doldm=matrix(1,nlstate,1,nlstate);
5975: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5976: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5977:
5978: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5979: gpp=vector(nlstate+1,nlstate+ndeath);
5980: gmp=vector(nlstate+1,nlstate+ndeath);
5981: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5982:
1.218 brouard 5983: if(estepm < stepm){
5984: printf ("Problem %d lower than %d\n",estepm, stepm);
5985: }
5986: else hstepm=estepm;
5987: /* For example we decided to compute the life expectancy with the smallest unit */
5988: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5989: nhstepm is the number of hstepm from age to agelim
5990: nstepm is the number of stepm from age to agelim.
5991: Look at function hpijx to understand why because of memory size limitations,
5992: we decided (b) to get a life expectancy respecting the most precise curvature of the
5993: survival function given by stepm (the optimization length). Unfortunately it
5994: means that if the survival funtion is printed every two years of age and if
5995: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5996: results. So we changed our mind and took the option of the best precision.
5997: */
5998: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5999: agelim = AGESUP;
6000: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6001: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6002: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6003: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6004: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6005: gp=matrix(0,nhstepm,1,nlstate);
6006: gm=matrix(0,nhstepm,1,nlstate);
6007:
6008:
6009: for(theta=1; theta <=npar; theta++){
6010: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6011: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6012: }
1.279 brouard 6013: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6014: * returns into prlim .
1.288 brouard 6015: */
1.242 brouard 6016: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6017:
6018: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6019: if (popbased==1) {
6020: if(mobilav ==0){
6021: for(i=1; i<=nlstate;i++)
6022: prlim[i][i]=probs[(int)age][i][ij];
6023: }else{ /* mobilav */
6024: for(i=1; i<=nlstate;i++)
6025: prlim[i][i]=mobaverage[(int)age][i][ij];
6026: }
6027: }
1.279 brouard 6028: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}_x\f$ at horizon h.
6029: */
6030: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=0 to nhstepm */
1.292 brouard 6031: /**< And for each alive state j, sums over i \f$ w^i_x {}{h}_p^{ij}x\f$, which are the probability
1.279 brouard 6032: * at horizon h in state j including mortality.
6033: */
1.218 brouard 6034: for(j=1; j<= nlstate; j++){
6035: for(h=0; h<=nhstepm; h++){
6036: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6037: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6038: }
6039: }
1.279 brouard 6040: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6041: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6042: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6043: */
6044: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6045: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6046: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6047: }
6048:
6049: /* Again with minus shift */
1.218 brouard 6050:
6051: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6052: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6053:
1.242 brouard 6054: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6055:
6056: if (popbased==1) {
6057: if(mobilav ==0){
6058: for(i=1; i<=nlstate;i++)
6059: prlim[i][i]=probs[(int)age][i][ij];
6060: }else{ /* mobilav */
6061: for(i=1; i<=nlstate;i++)
6062: prlim[i][i]=mobaverage[(int)age][i][ij];
6063: }
6064: }
6065:
1.235 brouard 6066: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6067:
6068: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6069: for(h=0; h<=nhstepm; h++){
6070: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6071: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6072: }
6073: }
6074: /* This for computing probability of death (h=1 means
6075: computed over hstepm matrices product = hstepm*stepm months)
6076: as a weighted average of prlim.
6077: */
6078: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6079: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6080: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6081: }
1.279 brouard 6082: /* end shifting computations */
6083:
6084: /**< Computing gradient matrix at horizon h
6085: */
1.218 brouard 6086: for(j=1; j<= nlstate; j++) /* vareij */
6087: for(h=0; h<=nhstepm; h++){
6088: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6089: }
1.279 brouard 6090: /**< Gradient of overall mortality p.3 (or p.j)
6091: */
6092: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6093: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6094: }
6095:
6096: } /* End theta */
1.279 brouard 6097:
6098: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6099: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6100:
6101: for(h=0; h<=nhstepm; h++) /* veij */
6102: for(j=1; j<=nlstate;j++)
6103: for(theta=1; theta <=npar; theta++)
6104: trgradg[h][j][theta]=gradg[h][theta][j];
6105:
6106: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6107: for(theta=1; theta <=npar; theta++)
6108: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6109: /**< as well as its transposed matrix
6110: */
1.218 brouard 6111:
6112: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6113: for(i=1;i<=nlstate;i++)
6114: for(j=1;j<=nlstate;j++)
6115: vareij[i][j][(int)age] =0.;
1.279 brouard 6116:
6117: /* Computing trgradg by matcov by gradg at age and summing over h
6118: * and k (nhstepm) formula 15 of article
6119: * Lievre-Brouard-Heathcote
6120: */
6121:
1.218 brouard 6122: for(h=0;h<=nhstepm;h++){
6123: for(k=0;k<=nhstepm;k++){
6124: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6125: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6126: for(i=1;i<=nlstate;i++)
6127: for(j=1;j<=nlstate;j++)
6128: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6129: }
6130: }
6131:
1.279 brouard 6132: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6133: * p.j overall mortality formula 49 but computed directly because
6134: * we compute the grad (wix pijx) instead of grad (pijx),even if
6135: * wix is independent of theta.
6136: */
1.218 brouard 6137: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6138: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6139: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6140: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6141: varppt[j][i]=doldmp[j][i];
6142: /* end ppptj */
6143: /* x centered again */
6144:
1.242 brouard 6145: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6146:
6147: if (popbased==1) {
6148: if(mobilav ==0){
6149: for(i=1; i<=nlstate;i++)
6150: prlim[i][i]=probs[(int)age][i][ij];
6151: }else{ /* mobilav */
6152: for(i=1; i<=nlstate;i++)
6153: prlim[i][i]=mobaverage[(int)age][i][ij];
6154: }
6155: }
6156:
6157: /* This for computing probability of death (h=1 means
6158: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6159: as a weighted average of prlim.
6160: */
1.235 brouard 6161: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6162: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6163: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6164: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6165: }
6166: /* end probability of death */
6167:
6168: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6169: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6170: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6171: for(i=1; i<=nlstate;i++){
6172: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6173: }
6174: }
6175: fprintf(ficresprobmorprev,"\n");
6176:
6177: fprintf(ficresvij,"%.0f ",age );
6178: for(i=1; i<=nlstate;i++)
6179: for(j=1; j<=nlstate;j++){
6180: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6181: }
6182: fprintf(ficresvij,"\n");
6183: free_matrix(gp,0,nhstepm,1,nlstate);
6184: free_matrix(gm,0,nhstepm,1,nlstate);
6185: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6186: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6187: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6188: } /* End age */
6189: free_vector(gpp,nlstate+1,nlstate+ndeath);
6190: free_vector(gmp,nlstate+1,nlstate+ndeath);
6191: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6192: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6193: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6194: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6195: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6196: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6197: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6198: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6199: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6200: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6201: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6202: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6203: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6204: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6205: 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);
6206: /* 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 6207: */
1.218 brouard 6208: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6209: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6210:
1.218 brouard 6211: free_vector(xp,1,npar);
6212: free_matrix(doldm,1,nlstate,1,nlstate);
6213: free_matrix(dnewm,1,nlstate,1,npar);
6214: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6215: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6216: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6217: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6218: fclose(ficresprobmorprev);
6219: fflush(ficgp);
6220: fflush(fichtm);
6221: } /* end varevsij */
1.126 brouard 6222:
6223: /************ Variance of prevlim ******************/
1.269 brouard 6224: 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 6225: {
1.205 brouard 6226: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6227: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6228:
1.268 brouard 6229: double **dnewmpar,**doldm;
1.126 brouard 6230: int i, j, nhstepm, hstepm;
6231: double *xp;
6232: double *gp, *gm;
6233: double **gradg, **trgradg;
1.208 brouard 6234: double **mgm, **mgp;
1.126 brouard 6235: double age,agelim;
6236: int theta;
6237:
6238: pstamp(ficresvpl);
1.288 brouard 6239: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6240: fprintf(ficresvpl,"# Age ");
6241: if(nresult >=1)
6242: fprintf(ficresvpl," Result# ");
1.126 brouard 6243: for(i=1; i<=nlstate;i++)
6244: fprintf(ficresvpl," %1d-%1d",i,i);
6245: fprintf(ficresvpl,"\n");
6246:
6247: xp=vector(1,npar);
1.268 brouard 6248: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6249: doldm=matrix(1,nlstate,1,nlstate);
6250:
6251: hstepm=1*YEARM; /* Every year of age */
6252: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6253: agelim = AGESUP;
6254: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6255: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6256: if (stepm >= YEARM) hstepm=1;
6257: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6258: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6259: mgp=matrix(1,npar,1,nlstate);
6260: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6261: gp=vector(1,nlstate);
6262: gm=vector(1,nlstate);
6263:
6264: for(theta=1; theta <=npar; theta++){
6265: for(i=1; i<=npar; i++){ /* Computes gradient */
6266: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6267: }
1.288 brouard 6268: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6269: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6270: /* else */
6271: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6272: for(i=1;i<=nlstate;i++){
1.126 brouard 6273: gp[i] = prlim[i][i];
1.208 brouard 6274: mgp[theta][i] = prlim[i][i];
6275: }
1.126 brouard 6276: for(i=1; i<=npar; i++) /* Computes gradient */
6277: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6278: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6279: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6280: /* else */
6281: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6282: for(i=1;i<=nlstate;i++){
1.126 brouard 6283: gm[i] = prlim[i][i];
1.208 brouard 6284: mgm[theta][i] = prlim[i][i];
6285: }
1.126 brouard 6286: for(i=1;i<=nlstate;i++)
6287: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6288: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6289: } /* End theta */
6290:
6291: trgradg =matrix(1,nlstate,1,npar);
6292:
6293: for(j=1; j<=nlstate;j++)
6294: for(theta=1; theta <=npar; theta++)
6295: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6296: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6297: /* printf("\nmgm mgp %d ",(int)age); */
6298: /* for(j=1; j<=nlstate;j++){ */
6299: /* printf(" %d ",j); */
6300: /* for(theta=1; theta <=npar; theta++) */
6301: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6302: /* printf("\n "); */
6303: /* } */
6304: /* } */
6305: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6306: /* printf("\n gradg %d ",(int)age); */
6307: /* for(j=1; j<=nlstate;j++){ */
6308: /* printf("%d ",j); */
6309: /* for(theta=1; theta <=npar; theta++) */
6310: /* printf("%d %lf ",theta,gradg[theta][j]); */
6311: /* printf("\n "); */
6312: /* } */
6313: /* } */
1.126 brouard 6314:
6315: for(i=1;i<=nlstate;i++)
6316: varpl[i][(int)age] =0.;
1.209 brouard 6317: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6318: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6319: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6320: }else{
1.268 brouard 6321: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6322: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6323: }
1.126 brouard 6324: for(i=1;i<=nlstate;i++)
6325: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6326:
6327: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6328: if(nresult >=1)
6329: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6330: for(i=1; i<=nlstate;i++){
1.126 brouard 6331: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6332: /* for(j=1;j<=nlstate;j++) */
6333: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6334: }
1.126 brouard 6335: fprintf(ficresvpl,"\n");
6336: free_vector(gp,1,nlstate);
6337: free_vector(gm,1,nlstate);
1.208 brouard 6338: free_matrix(mgm,1,npar,1,nlstate);
6339: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6340: free_matrix(gradg,1,npar,1,nlstate);
6341: free_matrix(trgradg,1,nlstate,1,npar);
6342: } /* End age */
6343:
6344: free_vector(xp,1,npar);
6345: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6346: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6347:
6348: }
6349:
6350:
6351: /************ Variance of backprevalence limit ******************/
1.269 brouard 6352: 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 6353: {
6354: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6355: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6356:
6357: double **dnewmpar,**doldm;
6358: int i, j, nhstepm, hstepm;
6359: double *xp;
6360: double *gp, *gm;
6361: double **gradg, **trgradg;
6362: double **mgm, **mgp;
6363: double age,agelim;
6364: int theta;
6365:
6366: pstamp(ficresvbl);
6367: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6368: fprintf(ficresvbl,"# Age ");
6369: if(nresult >=1)
6370: fprintf(ficresvbl," Result# ");
6371: for(i=1; i<=nlstate;i++)
6372: fprintf(ficresvbl," %1d-%1d",i,i);
6373: fprintf(ficresvbl,"\n");
6374:
6375: xp=vector(1,npar);
6376: dnewmpar=matrix(1,nlstate,1,npar);
6377: doldm=matrix(1,nlstate,1,nlstate);
6378:
6379: hstepm=1*YEARM; /* Every year of age */
6380: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6381: agelim = AGEINF;
6382: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6383: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6384: if (stepm >= YEARM) hstepm=1;
6385: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6386: gradg=matrix(1,npar,1,nlstate);
6387: mgp=matrix(1,npar,1,nlstate);
6388: mgm=matrix(1,npar,1,nlstate);
6389: gp=vector(1,nlstate);
6390: gm=vector(1,nlstate);
6391:
6392: for(theta=1; theta <=npar; theta++){
6393: for(i=1; i<=npar; i++){ /* Computes gradient */
6394: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6395: }
6396: if(mobilavproj > 0 )
6397: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6398: else
6399: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6400: for(i=1;i<=nlstate;i++){
6401: gp[i] = bprlim[i][i];
6402: mgp[theta][i] = bprlim[i][i];
6403: }
6404: for(i=1; i<=npar; i++) /* Computes gradient */
6405: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6406: if(mobilavproj > 0 )
6407: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6408: else
6409: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6410: for(i=1;i<=nlstate;i++){
6411: gm[i] = bprlim[i][i];
6412: mgm[theta][i] = bprlim[i][i];
6413: }
6414: for(i=1;i<=nlstate;i++)
6415: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6416: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6417: } /* End theta */
6418:
6419: trgradg =matrix(1,nlstate,1,npar);
6420:
6421: for(j=1; j<=nlstate;j++)
6422: for(theta=1; theta <=npar; theta++)
6423: trgradg[j][theta]=gradg[theta][j];
6424: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6425: /* printf("\nmgm mgp %d ",(int)age); */
6426: /* for(j=1; j<=nlstate;j++){ */
6427: /* printf(" %d ",j); */
6428: /* for(theta=1; theta <=npar; theta++) */
6429: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6430: /* printf("\n "); */
6431: /* } */
6432: /* } */
6433: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6434: /* printf("\n gradg %d ",(int)age); */
6435: /* for(j=1; j<=nlstate;j++){ */
6436: /* printf("%d ",j); */
6437: /* for(theta=1; theta <=npar; theta++) */
6438: /* printf("%d %lf ",theta,gradg[theta][j]); */
6439: /* printf("\n "); */
6440: /* } */
6441: /* } */
6442:
6443: for(i=1;i<=nlstate;i++)
6444: varbpl[i][(int)age] =0.;
6445: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6446: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6447: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6448: }else{
6449: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6450: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6451: }
6452: for(i=1;i<=nlstate;i++)
6453: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6454:
6455: fprintf(ficresvbl,"%.0f ",age );
6456: if(nresult >=1)
6457: fprintf(ficresvbl,"%d ",nres );
6458: for(i=1; i<=nlstate;i++)
6459: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6460: fprintf(ficresvbl,"\n");
6461: free_vector(gp,1,nlstate);
6462: free_vector(gm,1,nlstate);
6463: free_matrix(mgm,1,npar,1,nlstate);
6464: free_matrix(mgp,1,npar,1,nlstate);
6465: free_matrix(gradg,1,npar,1,nlstate);
6466: free_matrix(trgradg,1,nlstate,1,npar);
6467: } /* End age */
6468:
6469: free_vector(xp,1,npar);
6470: free_matrix(doldm,1,nlstate,1,npar);
6471: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6472:
6473: }
6474:
6475: /************ Variance of one-step probabilities ******************/
6476: 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 6477: {
6478: int i, j=0, k1, l1, tj;
6479: int k2, l2, j1, z1;
6480: int k=0, l;
6481: int first=1, first1, first2;
6482: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6483: double **dnewm,**doldm;
6484: double *xp;
6485: double *gp, *gm;
6486: double **gradg, **trgradg;
6487: double **mu;
6488: double age, cov[NCOVMAX+1];
6489: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6490: int theta;
6491: char fileresprob[FILENAMELENGTH];
6492: char fileresprobcov[FILENAMELENGTH];
6493: char fileresprobcor[FILENAMELENGTH];
6494: double ***varpij;
6495:
6496: strcpy(fileresprob,"PROB_");
6497: strcat(fileresprob,fileres);
6498: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6499: printf("Problem with resultfile: %s\n", fileresprob);
6500: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6501: }
6502: strcpy(fileresprobcov,"PROBCOV_");
6503: strcat(fileresprobcov,fileresu);
6504: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6505: printf("Problem with resultfile: %s\n", fileresprobcov);
6506: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6507: }
6508: strcpy(fileresprobcor,"PROBCOR_");
6509: strcat(fileresprobcor,fileresu);
6510: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6511: printf("Problem with resultfile: %s\n", fileresprobcor);
6512: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6513: }
6514: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6515: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6516: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6517: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6518: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6519: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6520: pstamp(ficresprob);
6521: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6522: fprintf(ficresprob,"# Age");
6523: pstamp(ficresprobcov);
6524: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6525: fprintf(ficresprobcov,"# Age");
6526: pstamp(ficresprobcor);
6527: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6528: fprintf(ficresprobcor,"# Age");
1.126 brouard 6529:
6530:
1.222 brouard 6531: for(i=1; i<=nlstate;i++)
6532: for(j=1; j<=(nlstate+ndeath);j++){
6533: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6534: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6535: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6536: }
6537: /* fprintf(ficresprob,"\n");
6538: fprintf(ficresprobcov,"\n");
6539: fprintf(ficresprobcor,"\n");
6540: */
6541: xp=vector(1,npar);
6542: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6543: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6544: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6545: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6546: first=1;
6547: fprintf(ficgp,"\n# Routine varprob");
6548: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6549: fprintf(fichtm,"\n");
6550:
1.288 brouard 6551: fprintf(fichtm,"\n<li><h4> <a href=\"%s\">Matrix of variance-covariance of one-step probabilities (drawings)</a></h4> this page is important in order to visualize confidence intervals and especially correlation between disability and recovery, or more generally, way in and way back. File %s</li>\n",optionfilehtmcov,optionfilehtmcov);
1.222 brouard 6552: 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);
6553: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6554: and drawn. It helps understanding how is the covariance between two incidences.\
6555: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6556: 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 6557: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6558: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6559: standard deviations wide on each axis. <br>\
6560: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6561: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6562: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6563:
1.222 brouard 6564: cov[1]=1;
6565: /* tj=cptcoveff; */
1.225 brouard 6566: tj = (int) pow(2,cptcoveff);
1.222 brouard 6567: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6568: j1=0;
1.224 brouard 6569: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6570: if (cptcovn>0) {
6571: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6572: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6573: fprintf(ficresprob, "**********\n#\n");
6574: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6575: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6576: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6577:
1.222 brouard 6578: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6579: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6580: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6581:
6582:
1.222 brouard 6583: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6584: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6585: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6586:
1.222 brouard 6587: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6588: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6589: fprintf(ficresprobcor, "**********\n#");
6590: if(invalidvarcomb[j1]){
6591: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6592: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6593: continue;
6594: }
6595: }
6596: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6597: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6598: gp=vector(1,(nlstate)*(nlstate+ndeath));
6599: gm=vector(1,(nlstate)*(nlstate+ndeath));
6600: for (age=bage; age<=fage; age ++){
6601: cov[2]=age;
6602: if(nagesqr==1)
6603: cov[3]= age*age;
6604: for (k=1; k<=cptcovn;k++) {
6605: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6606: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6607: * 1 1 1 1 1
6608: * 2 2 1 1 1
6609: * 3 1 2 1 1
6610: */
6611: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6612: }
6613: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6614: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6615: for (k=1; k<=cptcovprod;k++)
6616: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6617:
6618:
1.222 brouard 6619: for(theta=1; theta <=npar; theta++){
6620: for(i=1; i<=npar; i++)
6621: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6622:
1.222 brouard 6623: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6624:
1.222 brouard 6625: k=0;
6626: for(i=1; i<= (nlstate); i++){
6627: for(j=1; j<=(nlstate+ndeath);j++){
6628: k=k+1;
6629: gp[k]=pmmij[i][j];
6630: }
6631: }
1.220 brouard 6632:
1.222 brouard 6633: for(i=1; i<=npar; i++)
6634: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6635:
1.222 brouard 6636: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6637: k=0;
6638: for(i=1; i<=(nlstate); i++){
6639: for(j=1; j<=(nlstate+ndeath);j++){
6640: k=k+1;
6641: gm[k]=pmmij[i][j];
6642: }
6643: }
1.220 brouard 6644:
1.222 brouard 6645: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6646: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6647: }
1.126 brouard 6648:
1.222 brouard 6649: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6650: for(theta=1; theta <=npar; theta++)
6651: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6652:
1.222 brouard 6653: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6654: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6655:
1.222 brouard 6656: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6657:
1.222 brouard 6658: k=0;
6659: for(i=1; i<=(nlstate); i++){
6660: for(j=1; j<=(nlstate+ndeath);j++){
6661: k=k+1;
6662: mu[k][(int) age]=pmmij[i][j];
6663: }
6664: }
6665: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6666: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6667: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6668:
1.222 brouard 6669: /*printf("\n%d ",(int)age);
6670: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6671: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6672: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6673: }*/
1.220 brouard 6674:
1.222 brouard 6675: fprintf(ficresprob,"\n%d ",(int)age);
6676: fprintf(ficresprobcov,"\n%d ",(int)age);
6677: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6678:
1.222 brouard 6679: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6680: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6681: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6682: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6683: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6684: }
6685: i=0;
6686: for (k=1; k<=(nlstate);k++){
6687: for (l=1; l<=(nlstate+ndeath);l++){
6688: i++;
6689: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6690: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6691: for (j=1; j<=i;j++){
6692: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6693: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6694: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6695: }
6696: }
6697: }/* end of loop for state */
6698: } /* end of loop for age */
6699: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6700: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6701: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6702: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6703:
6704: /* Confidence intervalle of pij */
6705: /*
6706: fprintf(ficgp,"\nunset parametric;unset label");
6707: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6708: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6709: 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);
6710: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6711: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6712: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6713: */
6714:
6715: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6716: first1=1;first2=2;
6717: for (k2=1; k2<=(nlstate);k2++){
6718: for (l2=1; l2<=(nlstate+ndeath);l2++){
6719: if(l2==k2) continue;
6720: j=(k2-1)*(nlstate+ndeath)+l2;
6721: for (k1=1; k1<=(nlstate);k1++){
6722: for (l1=1; l1<=(nlstate+ndeath);l1++){
6723: if(l1==k1) continue;
6724: i=(k1-1)*(nlstate+ndeath)+l1;
6725: if(i<=j) continue;
6726: for (age=bage; age<=fage; age ++){
6727: if ((int)age %5==0){
6728: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6729: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6730: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6731: mu1=mu[i][(int) age]/stepm*YEARM ;
6732: mu2=mu[j][(int) age]/stepm*YEARM;
6733: c12=cv12/sqrt(v1*v2);
6734: /* Computing eigen value of matrix of covariance */
6735: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6736: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6737: if ((lc2 <0) || (lc1 <0) ){
6738: if(first2==1){
6739: first1=0;
6740: 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);
6741: }
6742: 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);
6743: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6744: /* lc2=fabs(lc2); */
6745: }
1.220 brouard 6746:
1.222 brouard 6747: /* Eigen vectors */
1.280 brouard 6748: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6749: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6750: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6751: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6752: }else
6753: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6754: /*v21=sqrt(1.-v11*v11); *//* error */
6755: v21=(lc1-v1)/cv12*v11;
6756: v12=-v21;
6757: v22=v11;
6758: tnalp=v21/v11;
6759: if(first1==1){
6760: first1=0;
6761: 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);
6762: }
6763: 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);
6764: /*printf(fignu*/
6765: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6766: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6767: if(first==1){
6768: first=0;
6769: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6770: fprintf(ficgp,"\nset parametric;unset label");
6771: 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);
6772: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6773: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6774: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6775: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6776: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6777: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6778: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6779: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6780: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6781: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6782: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6783: 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 6784: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6785: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6786: }else{
6787: first=0;
6788: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6789: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6790: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6791: 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 6792: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6793: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6794: }/* if first */
6795: } /* age mod 5 */
6796: } /* end loop age */
6797: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6798: first=1;
6799: } /*l12 */
6800: } /* k12 */
6801: } /*l1 */
6802: }/* k1 */
6803: } /* loop on combination of covariates j1 */
6804: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6805: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6806: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6807: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6808: free_vector(xp,1,npar);
6809: fclose(ficresprob);
6810: fclose(ficresprobcov);
6811: fclose(ficresprobcor);
6812: fflush(ficgp);
6813: fflush(fichtmcov);
6814: }
1.126 brouard 6815:
6816:
6817: /******************* Printing html file ***********/
1.201 brouard 6818: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6819: int lastpass, int stepm, int weightopt, char model[],\
6820: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6821: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6822: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6823: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6824: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6825:
6826: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6827: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6828: </ul>");
1.237 brouard 6829: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6830: </ul>", model);
1.214 brouard 6831: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6832: 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",
6833: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6834: 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 6835: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6836: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6837: fprintf(fichtm,"\
6838: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6839: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6840: fprintf(fichtm,"\
1.217 brouard 6841: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6842: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6843: fprintf(fichtm,"\
1.288 brouard 6844: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6845: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6846: fprintf(fichtm,"\
1.288 brouard 6847: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6848: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6849: fprintf(fichtm,"\
1.211 brouard 6850: - (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 6851: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6852: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6853: if(prevfcast==1){
6854: fprintf(fichtm,"\
6855: - Prevalence projections by age and states: \
1.201 brouard 6856: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6857: }
1.126 brouard 6858:
6859:
1.225 brouard 6860: m=pow(2,cptcoveff);
1.222 brouard 6861: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6862:
1.264 brouard 6863: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6864:
6865: jj1=0;
6866:
6867: fprintf(fichtm," \n<ul>");
6868: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6869: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6870: if(m != 1 && TKresult[nres]!= k1)
6871: continue;
6872: jj1++;
6873: if (cptcovn > 0) {
6874: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6875: for (cpt=1; cpt<=cptcoveff;cpt++){
6876: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6877: }
6878: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6879: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6880: }
6881: fprintf(fichtm,"\">");
6882:
6883: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6884: fprintf(fichtm,"************ Results for covariates");
6885: for (cpt=1; cpt<=cptcoveff;cpt++){
6886: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6887: }
6888: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6889: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6890: }
6891: if(invalidvarcomb[k1]){
6892: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6893: continue;
6894: }
6895: fprintf(fichtm,"</a></li>");
6896: } /* cptcovn >0 */
6897: }
6898: fprintf(fichtm," \n</ul>");
6899:
1.222 brouard 6900: jj1=0;
1.237 brouard 6901:
6902: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6903: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6904: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6905: continue;
1.220 brouard 6906:
1.222 brouard 6907: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6908: jj1++;
6909: if (cptcovn > 0) {
1.264 brouard 6910: fprintf(fichtm,"\n<p><a name=\"rescov");
6911: for (cpt=1; cpt<=cptcoveff;cpt++){
6912: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6913: }
6914: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6915: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6916: }
6917: fprintf(fichtm,"\"</a>");
6918:
1.222 brouard 6919: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6920: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6921: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6922: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6923: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6924: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6925: }
1.237 brouard 6926: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6927: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6928: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6929: }
6930:
1.230 brouard 6931: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6932: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6933: if(invalidvarcomb[k1]){
6934: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6935: printf("\nCombination (%d) ignored because no cases \n",k1);
6936: continue;
6937: }
6938: }
6939: /* aij, bij */
1.259 brouard 6940: 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 6941: <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 6942: /* Pij */
1.241 brouard 6943: 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> \
6944: <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 6945: /* Quasi-incidences */
6946: 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 6947: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6948: 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 6949: 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> \
6950: <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 6951: /* Survival functions (period) in state j */
6952: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6953: fprintf(fichtm,"<br>\n- Survival functions in state %d. And probability to be observed in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 6954: <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 6955: }
6956: /* State specific survival functions (period) */
6957: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6958: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
6959: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 6960: <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 6961: }
1.288 brouard 6962: /* Period (forward stable) prevalence in each health state */
1.222 brouard 6963: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6964: 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> \
6965: <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 6966: }
6967: if(backcast==1){
1.288 brouard 6968: /* Backward prevalence in each health state */
1.222 brouard 6969: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6970: 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 6971: <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 6972: }
1.217 brouard 6973: }
1.222 brouard 6974: if(prevfcast==1){
1.288 brouard 6975: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 6976: for(cpt=1; cpt<=nlstate;cpt++){
1.288 brouard 6977: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), from year %.1f up to year %.1f tending to period (stable) forward prevalence in state %d. Or probability to be in state %d being in an observed weighted state (from 1 to %d). <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.273 brouard 6978: <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 6979: }
6980: }
1.268 brouard 6981: if(backcast==1){
6982: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6983: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6984: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6985: 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 \
6986: 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) \
6987: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6988: <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 6989: }
6990: }
1.220 brouard 6991:
1.222 brouard 6992: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6993: 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> \
6994: <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 6995: }
6996: /* } /\* end i1 *\/ */
6997: }/* End k1 */
6998: fprintf(fichtm,"</ul>");
1.126 brouard 6999:
1.222 brouard 7000: fprintf(fichtm,"\
1.126 brouard 7001: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7002: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7003: - 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 7004: But because parameters are usually highly correlated (a higher incidence of disability \
7005: and a higher incidence of recovery can give very close observed transition) it might \
7006: be very useful to look not only at linear confidence intervals estimated from the \
7007: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7008: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7009: covariance matrix of the one-step probabilities. \
7010: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7011:
1.222 brouard 7012: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7013: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7014: fprintf(fichtm,"\
1.126 brouard 7015: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7016: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7017:
1.222 brouard 7018: fprintf(fichtm,"\
1.126 brouard 7019: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7020: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7021: fprintf(fichtm,"\
1.126 brouard 7022: - 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): \
7023: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7024: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7025: fprintf(fichtm,"\
1.126 brouard 7026: - (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): \
7027: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7028: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7029: fprintf(fichtm,"\
1.288 brouard 7030: - Variances and covariances of health expectancies by age. Status (i) based health expectancies (in state j), e<sup>ij</sup> are weighted by the forward (period) prevalences in each state i (if popbased=1, an additional computation is done using the cross-sectional prevalences, i.e population based) (estepm=%d months): <a href=\"%s\">%s</a><br>\n",
1.222 brouard 7031: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7032: fprintf(fichtm,"\
1.128 brouard 7033: - 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 7034: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7035: fprintf(fichtm,"\
1.288 brouard 7036: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7037: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7038:
7039: /* if(popforecast==1) fprintf(fichtm,"\n */
7040: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7041: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7042: /* <br>",fileres,fileres,fileres,fileres); */
7043: /* else */
7044: /* 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 7045: fflush(fichtm);
7046: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7047:
1.225 brouard 7048: m=pow(2,cptcoveff);
1.222 brouard 7049: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7050:
1.222 brouard 7051: jj1=0;
1.237 brouard 7052:
1.241 brouard 7053: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7054: for(k1=1; k1<=m;k1++){
1.253 brouard 7055: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7056: continue;
1.222 brouard 7057: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7058: jj1++;
1.126 brouard 7059: if (cptcovn > 0) {
7060: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7061: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7062: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7063: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7064: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7065: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7066: }
7067:
1.126 brouard 7068: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7069:
1.222 brouard 7070: if(invalidvarcomb[k1]){
7071: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7072: continue;
7073: }
1.126 brouard 7074: }
7075: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7076: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 7077: 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 7078: <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 7079: }
7080: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 7081: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
7082: true period expectancies (those weighted with period prevalences are also\
7083: drawn in addition to the population based expectancies computed using\
1.241 brouard 7084: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
7085: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7086: /* } /\* end i1 *\/ */
7087: }/* End k1 */
1.241 brouard 7088: }/* End nres */
1.222 brouard 7089: fprintf(fichtm,"</ul>");
7090: fflush(fichtm);
1.126 brouard 7091: }
7092:
7093: /******************* Gnuplot file **************/
1.270 brouard 7094: 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 7095:
7096: char dirfileres[132],optfileres[132];
1.264 brouard 7097: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7098: 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 7099: int lv=0, vlv=0, kl=0;
1.130 brouard 7100: int ng=0;
1.201 brouard 7101: int vpopbased;
1.223 brouard 7102: int ioffset; /* variable offset for columns */
1.270 brouard 7103: int iyearc=1; /* variable column for year of projection */
7104: int iagec=1; /* variable column for age of projection */
1.235 brouard 7105: int nres=0; /* Index of resultline */
1.266 brouard 7106: int istart=1; /* For starting graphs in projections */
1.219 brouard 7107:
1.126 brouard 7108: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7109: /* printf("Problem with file %s",optionfilegnuplot); */
7110: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7111: /* } */
7112:
7113: /*#ifdef windows */
7114: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7115: /*#endif */
1.225 brouard 7116: m=pow(2,cptcoveff);
1.126 brouard 7117:
1.274 brouard 7118: /* diagram of the model */
7119: fprintf(ficgp,"\n#Diagram of the model \n");
7120: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7121: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7122: 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);
7123:
7124: 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);
7125: fprintf(ficgp,"\n#show arrow\nunset label\n");
7126: 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);
7127: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7128: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7129: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7130: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7131:
1.202 brouard 7132: /* Contribution to likelihood */
7133: /* Plot the probability implied in the likelihood */
1.223 brouard 7134: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7135: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7136: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7137: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7138: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7139: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7140: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7141: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7142: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7143: 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));
7144: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7145: 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));
7146: for (i=1; i<= nlstate ; i ++) {
7147: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7148: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7149: 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);
7150: for (j=2; j<= nlstate+ndeath ; j ++) {
7151: 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);
7152: }
7153: fprintf(ficgp,";\nset out; unset ylabel;\n");
7154: }
7155: /* 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 */
7156: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7157: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7158: fprintf(ficgp,"\nset out;unset log\n");
7159: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7160:
1.126 brouard 7161: strcpy(dirfileres,optionfilefiname);
7162: strcpy(optfileres,"vpl");
1.223 brouard 7163: /* 1eme*/
1.238 brouard 7164: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7165: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7166: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7167: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7168: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7169: continue;
7170: /* We are interested in selected combination by the resultline */
1.246 brouard 7171: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7172: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7173: strcpy(gplotlabel,"(");
1.238 brouard 7174: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7175: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7176: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7177: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7178: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7179: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7180: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7181: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7182: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7183: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7184: }
7185: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7186: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7187: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7188: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7189: }
7190: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7191: /* printf("\n#\n"); */
1.238 brouard 7192: fprintf(ficgp,"\n#\n");
7193: if(invalidvarcomb[k1]){
1.260 brouard 7194: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7195: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7196: continue;
7197: }
1.235 brouard 7198:
1.241 brouard 7199: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7200: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7201: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7202: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7203: 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);
7204: /* 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); */
7205: /* k1-1 error should be nres-1*/
1.238 brouard 7206: for (i=1; i<= nlstate ; i ++) {
7207: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7208: else fprintf(ficgp," %%*lf (%%*lf)");
7209: }
1.288 brouard 7210: fprintf(ficgp,"\" t\"Forward prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7211: for (i=1; i<= nlstate ; i ++) {
7212: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7213: else fprintf(ficgp," %%*lf (%%*lf)");
7214: }
1.260 brouard 7215: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7216: for (i=1; i<= nlstate ; i ++) {
7217: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7218: else fprintf(ficgp," %%*lf (%%*lf)");
7219: }
1.265 brouard 7220: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7221:
7222: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7223: if(cptcoveff ==0){
1.271 brouard 7224: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7225: }else{
7226: kl=0;
7227: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7228: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7229: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7230: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7231: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7232: vlv= nbcode[Tvaraff[k]][lv];
7233: kl++;
7234: /* 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 *\/ */
7235: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7236: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7237: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7238: if(k==cptcoveff){
7239: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7240: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7241: }else{
7242: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7243: kl++;
7244: }
7245: } /* end covariate */
7246: } /* end if no covariate */
7247:
1.238 brouard 7248: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7249: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7250: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7251: if(cptcoveff ==0){
1.245 brouard 7252: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7253: }else{
7254: kl=0;
7255: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7256: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7257: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7258: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7259: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7260: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7261: kl++;
1.238 brouard 7262: /* 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 *\/ */
7263: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7264: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7265: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7266: if(k==cptcoveff){
1.245 brouard 7267: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7268: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7269: }else{
7270: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7271: kl++;
7272: }
7273: } /* end covariate */
7274: } /* end if no covariate */
1.268 brouard 7275: if(backcast == 1){
7276: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7277: /* k1-1 error should be nres-1*/
7278: for (i=1; i<= nlstate ; i ++) {
7279: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7280: else fprintf(ficgp," %%*lf (%%*lf)");
7281: }
1.271 brouard 7282: 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 7283: for (i=1; i<= nlstate ; i ++) {
7284: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7285: else fprintf(ficgp," %%*lf (%%*lf)");
7286: }
1.276 brouard 7287: 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 7288: for (i=1; i<= nlstate ; i ++) {
7289: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7290: else fprintf(ficgp," %%*lf (%%*lf)");
7291: }
1.274 brouard 7292: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7293: } /* end if backprojcast */
1.238 brouard 7294: } /* end if backcast */
1.276 brouard 7295: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7296: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7297: } /* nres */
1.201 brouard 7298: } /* k1 */
7299: } /* cpt */
1.235 brouard 7300:
7301:
1.126 brouard 7302: /*2 eme*/
1.238 brouard 7303: for (k1=1; k1<= m ; k1 ++){
7304: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7305: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7306: continue;
7307: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7308: strcpy(gplotlabel,"(");
1.238 brouard 7309: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7310: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7311: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7312: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7313: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7314: vlv= nbcode[Tvaraff[k]][lv];
7315: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7316: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7317: }
1.237 brouard 7318: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7319: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7320: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7321: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7322: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7323: }
1.264 brouard 7324: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7325: fprintf(ficgp,"\n#\n");
1.223 brouard 7326: if(invalidvarcomb[k1]){
7327: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7328: continue;
7329: }
1.219 brouard 7330:
1.241 brouard 7331: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7332: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7333: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7334: if(vpopbased==0){
1.238 brouard 7335: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7336: }else
1.238 brouard 7337: fprintf(ficgp,"\nreplot ");
7338: for (i=1; i<= nlstate+1 ; i ++) {
7339: k=2*i;
1.261 brouard 7340: 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 7341: for (j=1; j<= nlstate+1 ; j ++) {
7342: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7343: else fprintf(ficgp," %%*lf (%%*lf)");
7344: }
7345: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7346: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7347: 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 7348: for (j=1; j<= nlstate+1 ; j ++) {
7349: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7350: else fprintf(ficgp," %%*lf (%%*lf)");
7351: }
7352: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7353: 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 7354: for (j=1; j<= nlstate+1 ; j ++) {
7355: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7356: else fprintf(ficgp," %%*lf (%%*lf)");
7357: }
7358: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7359: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7360: } /* state */
7361: } /* vpopbased */
1.264 brouard 7362: 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 7363: } /* end nres */
7364: } /* k1 end 2 eme*/
7365:
7366:
7367: /*3eme*/
7368: for (k1=1; k1<= m ; k1 ++){
7369: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7370: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7371: continue;
7372:
7373: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7374: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7375: strcpy(gplotlabel,"(");
1.238 brouard 7376: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7377: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7378: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7379: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7380: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7381: vlv= nbcode[Tvaraff[k]][lv];
7382: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7383: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7384: }
7385: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7386: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7387: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7388: }
1.264 brouard 7389: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7390: fprintf(ficgp,"\n#\n");
7391: if(invalidvarcomb[k1]){
7392: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7393: continue;
7394: }
7395:
7396: /* k=2+nlstate*(2*cpt-2); */
7397: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7398: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7399: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7400: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7401: 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 7402: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7403: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7404: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7405: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7406: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7407: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7408:
1.238 brouard 7409: */
7410: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7411: 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 7412: /* 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 7413:
1.238 brouard 7414: }
1.261 brouard 7415: 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 7416: }
1.264 brouard 7417: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7418: } /* end nres */
7419: } /* end kl 3eme */
1.126 brouard 7420:
1.223 brouard 7421: /* 4eme */
1.201 brouard 7422: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7423: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7424: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7425: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7426: continue;
1.238 brouard 7427: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7428: strcpy(gplotlabel,"(");
1.238 brouard 7429: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7430: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7431: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7432: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7433: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7434: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7435: vlv= nbcode[Tvaraff[k]][lv];
7436: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7437: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7438: }
7439: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7440: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7441: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7442: }
1.264 brouard 7443: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7444: fprintf(ficgp,"\n#\n");
7445: if(invalidvarcomb[k1]){
7446: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7447: continue;
1.223 brouard 7448: }
1.238 brouard 7449:
1.241 brouard 7450: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7451: 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 7452: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7453: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7454: k=3;
7455: for (i=1; i<= nlstate ; i ++){
7456: if(i==1){
7457: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7458: }else{
7459: fprintf(ficgp,", '' ");
7460: }
7461: l=(nlstate+ndeath)*(i-1)+1;
7462: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7463: for (j=2; j<= nlstate+ndeath ; j ++)
7464: fprintf(ficgp,"+$%d",k+l+j-1);
7465: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7466: } /* nlstate */
1.264 brouard 7467: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7468: } /* end cpt state*/
7469: } /* end nres */
7470: } /* end covariate k1 */
7471:
1.220 brouard 7472: /* 5eme */
1.201 brouard 7473: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7474: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7475: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7476: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7477: continue;
1.238 brouard 7478: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7479: strcpy(gplotlabel,"(");
1.238 brouard 7480: 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);
7481: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7482: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7483: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7484: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7485: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7486: vlv= nbcode[Tvaraff[k]][lv];
7487: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7488: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7489: }
7490: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7491: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7492: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7493: }
1.264 brouard 7494: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7495: fprintf(ficgp,"\n#\n");
7496: if(invalidvarcomb[k1]){
7497: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7498: continue;
7499: }
1.227 brouard 7500:
1.241 brouard 7501: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7502: 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 7503: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7504: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7505: k=3;
7506: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7507: if(j==1)
7508: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7509: else
7510: fprintf(ficgp,", '' ");
7511: l=(nlstate+ndeath)*(cpt-1) +j;
7512: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7513: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7514: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7515: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7516: } /* nlstate */
7517: fprintf(ficgp,", '' ");
7518: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7519: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7520: l=(nlstate+ndeath)*(cpt-1) +j;
7521: if(j < nlstate)
7522: fprintf(ficgp,"$%d +",k+l);
7523: else
7524: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7525: }
1.264 brouard 7526: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7527: } /* end cpt state*/
7528: } /* end covariate */
7529: } /* end nres */
1.227 brouard 7530:
1.220 brouard 7531: /* 6eme */
1.202 brouard 7532: /* CV preval stable (period) for each covariate */
1.237 brouard 7533: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7534: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7535: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7536: continue;
1.255 brouard 7537: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7538: strcpy(gplotlabel,"(");
1.288 brouard 7539: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7540: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7541: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7542: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7543: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7544: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7545: vlv= nbcode[Tvaraff[k]][lv];
7546: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7547: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7548: }
1.237 brouard 7549: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7550: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7551: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7552: }
1.264 brouard 7553: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7554: fprintf(ficgp,"\n#\n");
1.223 brouard 7555: if(invalidvarcomb[k1]){
1.227 brouard 7556: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7557: continue;
1.223 brouard 7558: }
1.227 brouard 7559:
1.241 brouard 7560: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7561: 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 7562: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7563: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7564: k=3; /* Offset */
1.255 brouard 7565: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7566: if(i==1)
7567: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7568: else
7569: fprintf(ficgp,", '' ");
1.255 brouard 7570: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7571: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7572: for (j=2; j<= nlstate ; j ++)
7573: fprintf(ficgp,"+$%d",k+l+j-1);
7574: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7575: } /* nlstate */
1.264 brouard 7576: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7577: } /* end cpt state*/
7578: } /* end covariate */
1.227 brouard 7579:
7580:
1.220 brouard 7581: /* 7eme */
1.218 brouard 7582: if(backcast == 1){
1.288 brouard 7583: /* CV backward prevalence for each covariate */
1.237 brouard 7584: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7585: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7586: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7587: continue;
1.268 brouard 7588: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7589: strcpy(gplotlabel,"(");
1.288 brouard 7590: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7591: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7592: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7593: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7594: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7595: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7596: vlv= nbcode[Tvaraff[k]][lv];
7597: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7598: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7599: }
1.237 brouard 7600: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7601: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7602: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7603: }
1.264 brouard 7604: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7605: fprintf(ficgp,"\n#\n");
7606: if(invalidvarcomb[k1]){
7607: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7608: continue;
7609: }
7610:
1.241 brouard 7611: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7612: 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 7613: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7614: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7615: k=3; /* Offset */
1.268 brouard 7616: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7617: if(i==1)
7618: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7619: else
7620: fprintf(ficgp,", '' ");
7621: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7622: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7623: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7624: /* 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 7625: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7626: /* for (j=2; j<= nlstate ; j ++) */
7627: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7628: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7629: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7630: } /* nlstate */
1.264 brouard 7631: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7632: } /* end cpt state*/
7633: } /* end covariate */
7634: } /* End if backcast */
7635:
1.223 brouard 7636: /* 8eme */
1.218 brouard 7637: if(prevfcast==1){
1.288 brouard 7638: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7639:
1.237 brouard 7640: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7641: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7642: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7643: continue;
1.211 brouard 7644: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7645: strcpy(gplotlabel,"(");
1.288 brouard 7646: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7647: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7648: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7649: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7650: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7651: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7652: vlv= nbcode[Tvaraff[k]][lv];
7653: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7654: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7655: }
1.237 brouard 7656: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7657: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7658: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7659: }
1.264 brouard 7660: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7661: fprintf(ficgp,"\n#\n");
7662: if(invalidvarcomb[k1]){
7663: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7664: continue;
7665: }
7666:
7667: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7668: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7669: 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 7670: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7671: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7672:
7673: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7674: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7675: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7676: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7677: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7678: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7679: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7680: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7681: if(i==istart){
1.227 brouard 7682: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7683: }else{
7684: fprintf(ficgp,",\\\n '' ");
7685: }
7686: if(cptcoveff ==0){ /* No covariate */
7687: ioffset=2; /* Age is in 2 */
7688: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7689: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7690: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7691: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7692: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7693: if(i==nlstate+1){
1.270 brouard 7694: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7695: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7696: fprintf(ficgp,",\\\n '' ");
7697: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7698: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7699: offyear, \
1.268 brouard 7700: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7701: }else
1.227 brouard 7702: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7703: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7704: }else{ /* more than 2 covariates */
1.270 brouard 7705: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7706: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7707: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7708: iyearc=ioffset-1;
7709: iagec=ioffset;
1.227 brouard 7710: fprintf(ficgp," u %d:(",ioffset);
7711: kl=0;
7712: strcpy(gplotcondition,"(");
7713: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7714: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7715: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7716: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7717: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7718: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7719: kl++;
7720: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7721: kl++;
7722: if(k <cptcoveff && cptcoveff>1)
7723: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7724: }
7725: strcpy(gplotcondition+strlen(gplotcondition),")");
7726: /* 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 *\/ */
7727: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7728: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7729: /* '' 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*/
7730: if(i==nlstate+1){
1.270 brouard 7731: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7732: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7733: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7734: fprintf(ficgp," u %d:(",iagec);
7735: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7736: iyearc, iagec, offyear, \
7737: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7738: /* '' 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 7739: }else{
7740: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7741: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7742: }
7743: } /* end if covariate */
7744: } /* nlstate */
1.264 brouard 7745: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7746: } /* end cpt state*/
7747: } /* end covariate */
7748: } /* End if prevfcast */
1.227 brouard 7749:
1.268 brouard 7750: if(backcast==1){
7751: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7752:
7753: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7754: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7755: if(m != 1 && TKresult[nres]!= k1)
7756: continue;
7757: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7758: strcpy(gplotlabel,"(");
7759: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7760: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7761: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7762: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7763: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7764: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7765: vlv= nbcode[Tvaraff[k]][lv];
7766: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7767: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7768: }
7769: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7770: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7771: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7772: }
7773: strcpy(gplotlabel+strlen(gplotlabel),")");
7774: fprintf(ficgp,"\n#\n");
7775: if(invalidvarcomb[k1]){
7776: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7777: continue;
7778: }
7779:
7780: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7781: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7782: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7783: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7784: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7785:
7786: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7787: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7788: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7789: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7790: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7791: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7792: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7793: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7794: if(i==istart){
7795: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7796: }else{
7797: fprintf(ficgp,",\\\n '' ");
7798: }
7799: if(cptcoveff ==0){ /* No covariate */
7800: ioffset=2; /* Age is in 2 */
7801: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7802: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7803: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7804: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7805: fprintf(ficgp," u %d:(", ioffset);
7806: if(i==nlstate+1){
1.270 brouard 7807: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7808: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7809: fprintf(ficgp,",\\\n '' ");
7810: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7811: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7812: offbyear, \
7813: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7814: }else
7815: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7816: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7817: }else{ /* more than 2 covariates */
1.270 brouard 7818: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7819: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7820: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7821: iyearc=ioffset-1;
7822: iagec=ioffset;
1.268 brouard 7823: fprintf(ficgp," u %d:(",ioffset);
7824: kl=0;
7825: strcpy(gplotcondition,"(");
7826: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7827: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7828: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7829: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7830: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7831: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7832: kl++;
7833: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7834: kl++;
7835: if(k <cptcoveff && cptcoveff>1)
7836: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7837: }
7838: strcpy(gplotcondition+strlen(gplotcondition),")");
7839: /* 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 *\/ */
7840: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7841: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7842: /* '' 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*/
7843: if(i==nlstate+1){
1.270 brouard 7844: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7845: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7846: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7847: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7848: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7849: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7850: iyearc,iagec,offbyear, \
7851: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7852: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7853: }else{
7854: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7855: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7856: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7857: }
7858: } /* end if covariate */
7859: } /* nlstate */
7860: fprintf(ficgp,"\nset out; unset label;\n");
7861: } /* end cpt state*/
7862: } /* end covariate */
7863: } /* End if backcast */
7864:
1.227 brouard 7865:
1.238 brouard 7866: /* 9eme writing MLE parameters */
7867: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7868: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7869: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7870: for(k=1; k <=(nlstate+ndeath); k++){
7871: if (k != i) {
1.227 brouard 7872: fprintf(ficgp,"# current state %d\n",k);
7873: for(j=1; j <=ncovmodel; j++){
7874: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7875: jk++;
7876: }
7877: fprintf(ficgp,"\n");
1.126 brouard 7878: }
7879: }
1.223 brouard 7880: }
1.187 brouard 7881: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7882:
1.145 brouard 7883: /*goto avoid;*/
1.238 brouard 7884: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7885: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7886: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7887: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7888: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7889: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7890: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7891: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7892: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7893: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7894: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7895: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7896: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7897: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7898: fprintf(ficgp,"#\n");
1.223 brouard 7899: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7900: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7901: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7902: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7903: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7904: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7905: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7906: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7907: continue;
1.264 brouard 7908: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7909: strcpy(gplotlabel,"(");
1.276 brouard 7910: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7911: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7912: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7913: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7914: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7915: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7916: vlv= nbcode[Tvaraff[k]][lv];
7917: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7918: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7919: }
1.237 brouard 7920: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7921: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7922: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7923: }
1.264 brouard 7924: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7925: fprintf(ficgp,"\n#\n");
1.264 brouard 7926: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7927: fprintf(ficgp,"\nset key outside ");
7928: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7929: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7930: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7931: if (ng==1){
7932: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7933: fprintf(ficgp,"\nunset log y");
7934: }else if (ng==2){
7935: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7936: fprintf(ficgp,"\nset log y");
7937: }else if (ng==3){
7938: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7939: fprintf(ficgp,"\nset log y");
7940: }else
7941: fprintf(ficgp,"\nunset title ");
7942: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7943: i=1;
7944: for(k2=1; k2<=nlstate; k2++) {
7945: k3=i;
7946: for(k=1; k<=(nlstate+ndeath); k++) {
7947: if (k != k2){
7948: switch( ng) {
7949: case 1:
7950: if(nagesqr==0)
7951: fprintf(ficgp," p%d+p%d*x",i,i+1);
7952: else /* nagesqr =1 */
7953: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7954: break;
7955: case 2: /* ng=2 */
7956: if(nagesqr==0)
7957: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7958: else /* nagesqr =1 */
7959: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7960: break;
7961: case 3:
7962: if(nagesqr==0)
7963: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7964: else /* nagesqr =1 */
7965: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7966: break;
7967: }
7968: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7969: ijp=1; /* product no age */
7970: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7971: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7972: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7973: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7974: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7975: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7976: if(DummyV[j]==0){
7977: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7978: }else{ /* quantitative */
7979: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7980: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7981: }
7982: ij++;
1.237 brouard 7983: }
1.268 brouard 7984: }
7985: }else if(cptcovprod >0){
7986: if(j==Tprod[ijp]) { /* */
7987: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7988: if(ijp <=cptcovprod) { /* Product */
7989: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7990: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7991: /* 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)]); */
7992: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7993: }else{ /* Vn is dummy and Vm is quanti */
7994: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7995: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7996: }
7997: }else{ /* Vn*Vm Vn is quanti */
7998: if(DummyV[Tvard[ijp][2]]==0){
7999: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8000: }else{ /* Both quanti */
8001: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8002: }
1.237 brouard 8003: }
1.268 brouard 8004: ijp++;
1.237 brouard 8005: }
1.268 brouard 8006: } /* end Tprod */
1.237 brouard 8007: } else{ /* simple covariate */
1.264 brouard 8008: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8009: if(Dummy[j]==0){
8010: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8011: }else{ /* quantitative */
8012: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8013: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8014: }
1.237 brouard 8015: } /* end simple */
8016: } /* end j */
1.223 brouard 8017: }else{
8018: i=i-ncovmodel;
8019: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8020: fprintf(ficgp," (1.");
8021: }
1.227 brouard 8022:
1.223 brouard 8023: if(ng != 1){
8024: fprintf(ficgp,")/(1");
1.227 brouard 8025:
1.264 brouard 8026: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8027: if(nagesqr==0)
1.264 brouard 8028: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8029: else /* nagesqr =1 */
1.264 brouard 8030: 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 8031:
1.223 brouard 8032: ij=1;
8033: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8034: if(cptcovage >0){
8035: if((j-2)==Tage[ij]) { /* Bug valgrind */
8036: if(ij <=cptcovage) { /* Bug valgrind */
8037: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8038: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8039: ij++;
8040: }
8041: }
8042: }else
8043: 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 8044: }
8045: fprintf(ficgp,")");
8046: }
8047: fprintf(ficgp,")");
8048: if(ng ==2)
1.276 brouard 8049: 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 8050: else /* ng= 3 */
1.276 brouard 8051: 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 8052: }else{ /* end ng <> 1 */
8053: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8054: 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 8055: }
8056: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8057: fprintf(ficgp,",");
8058: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8059: fprintf(ficgp,",");
8060: i=i+ncovmodel;
8061: } /* end k */
8062: } /* end k2 */
1.276 brouard 8063: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8064: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8065: } /* end k1 */
1.223 brouard 8066: } /* end ng */
8067: /* avoid: */
8068: fflush(ficgp);
1.126 brouard 8069: } /* end gnuplot */
8070:
8071:
8072: /*************** Moving average **************/
1.219 brouard 8073: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8074: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8075:
1.222 brouard 8076: int i, cpt, cptcod;
8077: int modcovmax =1;
8078: int mobilavrange, mob;
8079: int iage=0;
1.288 brouard 8080: int firstA1=0, firstA2=0;
1.222 brouard 8081:
1.266 brouard 8082: double sum=0., sumr=0.;
1.222 brouard 8083: double age;
1.266 brouard 8084: double *sumnewp, *sumnewm, *sumnewmr;
8085: double *agemingood, *agemaxgood;
8086: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8087:
8088:
1.278 brouard 8089: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8090: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8091:
8092: sumnewp = vector(1,ncovcombmax);
8093: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8094: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8095: agemingood = vector(1,ncovcombmax);
1.266 brouard 8096: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8097: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8098: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8099:
8100: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8101: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8102: sumnewp[cptcod]=0.;
1.266 brouard 8103: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8104: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8105: }
8106: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8107:
1.266 brouard 8108: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8109: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8110: else mobilavrange=mobilav;
8111: for (age=bage; age<=fage; age++)
8112: for (i=1; i<=nlstate;i++)
8113: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8114: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8115: /* We keep the original values on the extreme ages bage, fage and for
8116: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8117: we use a 5 terms etc. until the borders are no more concerned.
8118: */
8119: for (mob=3;mob <=mobilavrange;mob=mob+2){
8120: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8121: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8122: sumnewm[cptcod]=0.;
8123: for (i=1; i<=nlstate;i++){
1.222 brouard 8124: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8125: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8126: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8127: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8128: }
8129: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8130: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8131: } /* end i */
8132: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8133: } /* end cptcod */
1.222 brouard 8134: }/* end age */
8135: }/* end mob */
1.266 brouard 8136: }else{
8137: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8138: return -1;
1.266 brouard 8139: }
8140:
8141: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8142: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8143: if(invalidvarcomb[cptcod]){
8144: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8145: continue;
8146: }
1.219 brouard 8147:
1.266 brouard 8148: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8149: sumnewm[cptcod]=0.;
8150: sumnewmr[cptcod]=0.;
8151: for (i=1; i<=nlstate;i++){
8152: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8153: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8154: }
8155: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8156: agemingoodr[cptcod]=age;
8157: }
8158: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8159: agemingood[cptcod]=age;
8160: }
8161: } /* age */
8162: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8163: sumnewm[cptcod]=0.;
1.266 brouard 8164: sumnewmr[cptcod]=0.;
1.222 brouard 8165: for (i=1; i<=nlstate;i++){
8166: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8167: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8168: }
8169: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8170: agemaxgoodr[cptcod]=age;
1.222 brouard 8171: }
8172: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8173: agemaxgood[cptcod]=age;
8174: }
8175: } /* age */
8176: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8177: /* but they will change */
1.288 brouard 8178: firstA1=0;firstA2=0;
1.266 brouard 8179: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8180: sumnewm[cptcod]=0.;
8181: sumnewmr[cptcod]=0.;
8182: for (i=1; i<=nlstate;i++){
8183: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8184: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8185: }
8186: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8187: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8188: agemaxgoodr[cptcod]=age; /* age min */
8189: for (i=1; i<=nlstate;i++)
8190: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8191: }else{ /* bad we change the value with the values of good ages */
8192: for (i=1; i<=nlstate;i++){
8193: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8194: } /* i */
8195: } /* end bad */
8196: }else{
8197: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8198: agemaxgood[cptcod]=age;
8199: }else{ /* bad we change the value with the values of good ages */
8200: for (i=1; i<=nlstate;i++){
8201: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8202: } /* i */
8203: } /* end bad */
8204: }/* end else */
8205: sum=0.;sumr=0.;
8206: for (i=1; i<=nlstate;i++){
8207: sum+=mobaverage[(int)age][i][cptcod];
8208: sumr+=probs[(int)age][i][cptcod];
8209: }
8210: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8211: if(!firstA1){
8212: firstA1=1;
8213: printf("Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d. Others in log file...\n",cptcod,sumr, (int)age, (int)bage);
8214: }
8215: fprintf(ficlog,"Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.266 brouard 8216: } /* end bad */
8217: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8218: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8219: if(!firstA2){
8220: firstA2=1;
8221: printf("Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d. Others in log file...\n",cptcod,sumr, (int)age, (int)bage);
8222: }
8223: fprintf(ficlog,"Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.222 brouard 8224: } /* end bad */
8225: }/* age */
1.266 brouard 8226:
8227: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8228: sumnewm[cptcod]=0.;
1.266 brouard 8229: sumnewmr[cptcod]=0.;
1.222 brouard 8230: for (i=1; i<=nlstate;i++){
8231: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8232: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8233: }
8234: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8235: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8236: agemingoodr[cptcod]=age;
8237: for (i=1; i<=nlstate;i++)
8238: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8239: }else{ /* bad we change the value with the values of good ages */
8240: for (i=1; i<=nlstate;i++){
8241: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8242: } /* i */
8243: } /* end bad */
8244: }else{
8245: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8246: agemingood[cptcod]=age;
8247: }else{ /* bad */
8248: for (i=1; i<=nlstate;i++){
8249: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8250: } /* i */
8251: } /* end bad */
8252: }/* end else */
8253: sum=0.;sumr=0.;
8254: for (i=1; i<=nlstate;i++){
8255: sum+=mobaverage[(int)age][i][cptcod];
8256: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8257: }
1.266 brouard 8258: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8259: 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 8260: } /* end bad */
8261: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8262: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8263: 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 8264: } /* end bad */
8265: }/* age */
1.266 brouard 8266:
1.222 brouard 8267:
8268: for (age=bage; age<=fage; age++){
1.235 brouard 8269: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8270: sumnewp[cptcod]=0.;
8271: sumnewm[cptcod]=0.;
8272: for (i=1; i<=nlstate;i++){
8273: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8274: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8275: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8276: }
8277: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8278: }
8279: /* printf("\n"); */
8280: /* } */
1.266 brouard 8281:
1.222 brouard 8282: /* brutal averaging */
1.266 brouard 8283: /* for (i=1; i<=nlstate;i++){ */
8284: /* for (age=1; age<=bage; age++){ */
8285: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8286: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8287: /* } */
8288: /* for (age=fage; age<=AGESUP; age++){ */
8289: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8290: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8291: /* } */
8292: /* } /\* end i status *\/ */
8293: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8294: /* for (age=1; age<=AGESUP; age++){ */
8295: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8296: /* mobaverage[(int)age][i][cptcod]=0.; */
8297: /* } */
8298: /* } */
1.222 brouard 8299: }/* end cptcod */
1.266 brouard 8300: free_vector(agemaxgoodr,1, ncovcombmax);
8301: free_vector(agemaxgood,1, ncovcombmax);
8302: free_vector(agemingood,1, ncovcombmax);
8303: free_vector(agemingoodr,1, ncovcombmax);
8304: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8305: free_vector(sumnewm,1, ncovcombmax);
8306: free_vector(sumnewp,1, ncovcombmax);
8307: return 0;
8308: }/* End movingaverage */
1.218 brouard 8309:
1.126 brouard 8310:
8311: /************** Forecasting ******************/
1.269 brouard 8312: 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 8313: /* proj1, year, month, day of starting projection
8314: agemin, agemax range of age
8315: dateprev1 dateprev2 range of dates during which prevalence is computed
8316: anproj2 year of en of projection (same day and month as proj1).
8317: */
1.267 brouard 8318: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8319: double agec; /* generic age */
8320: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8321: double *popeffectif,*popcount;
8322: double ***p3mat;
1.218 brouard 8323: /* double ***mobaverage; */
1.126 brouard 8324: char fileresf[FILENAMELENGTH];
8325:
8326: agelim=AGESUP;
1.211 brouard 8327: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8328: in each health status at the date of interview (if between dateprev1 and dateprev2).
8329: We still use firstpass and lastpass as another selection.
8330: */
1.214 brouard 8331: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8332: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8333:
1.201 brouard 8334: strcpy(fileresf,"F_");
8335: strcat(fileresf,fileresu);
1.126 brouard 8336: if((ficresf=fopen(fileresf,"w"))==NULL) {
8337: printf("Problem with forecast resultfile: %s\n", fileresf);
8338: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8339: }
1.235 brouard 8340: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8341: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8342:
1.225 brouard 8343: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8344:
8345:
8346: stepsize=(int) (stepm+YEARM-1)/YEARM;
8347: if (stepm<=12) stepsize=1;
8348: if(estepm < stepm){
8349: printf ("Problem %d lower than %d\n",estepm, stepm);
8350: }
1.270 brouard 8351: else{
8352: hstepm=estepm;
8353: }
8354: if(estepm > stepm){ /* Yes every two year */
8355: stepsize=2;
8356: }
1.126 brouard 8357:
8358: hstepm=hstepm/stepm;
8359: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8360: fractional in yp1 */
8361: anprojmean=yp;
8362: yp2=modf((yp1*12),&yp);
8363: mprojmean=yp;
8364: yp1=modf((yp2*30.5),&yp);
8365: jprojmean=yp;
8366: if(jprojmean==0) jprojmean=1;
8367: if(mprojmean==0) jprojmean=1;
8368:
1.227 brouard 8369: i1=pow(2,cptcoveff);
1.126 brouard 8370: if (cptcovn < 1){i1=1;}
8371:
8372: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8373:
8374: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8375:
1.126 brouard 8376: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8377: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8378: for(k=1; k<=i1;k++){
1.253 brouard 8379: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8380: continue;
1.227 brouard 8381: if(invalidvarcomb[k]){
8382: printf("\nCombination (%d) projection ignored because no cases \n",k);
8383: continue;
8384: }
8385: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8386: for(j=1;j<=cptcoveff;j++) {
8387: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8388: }
1.235 brouard 8389: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8390: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8391: }
1.227 brouard 8392: fprintf(ficresf," yearproj age");
8393: for(j=1; j<=nlstate+ndeath;j++){
8394: for(i=1; i<=nlstate;i++)
8395: fprintf(ficresf," p%d%d",i,j);
8396: fprintf(ficresf," wp.%d",j);
8397: }
8398: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8399: fprintf(ficresf,"\n");
8400: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8401: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8402: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8403: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8404: nhstepm = nhstepm/hstepm;
8405: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8406: oldm=oldms;savm=savms;
1.268 brouard 8407: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8408: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8409: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8410: for (h=0; h<=nhstepm; h++){
8411: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8412: break;
8413: }
8414: }
8415: fprintf(ficresf,"\n");
8416: for(j=1;j<=cptcoveff;j++)
8417: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8418: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8419:
8420: for(j=1; j<=nlstate+ndeath;j++) {
8421: ppij=0.;
8422: for(i=1; i<=nlstate;i++) {
1.278 brouard 8423: if (mobilav>=1)
8424: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8425: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8426: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8427: }
1.268 brouard 8428: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8429: } /* end i */
8430: fprintf(ficresf," %.3f", ppij);
8431: }/* end j */
1.227 brouard 8432: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8433: } /* end agec */
1.266 brouard 8434: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8435: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8436: } /* end yearp */
8437: } /* end k */
1.219 brouard 8438:
1.126 brouard 8439: fclose(ficresf);
1.215 brouard 8440: printf("End of Computing forecasting \n");
8441: fprintf(ficlog,"End of Computing forecasting\n");
8442:
1.126 brouard 8443: }
8444:
1.269 brouard 8445: /************** Back Forecasting ******************/
8446: 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 8447: /* back1, year, month, day of starting backection
8448: agemin, agemax range of age
8449: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8450: anback2 year of end of backprojection (same day and month as back1).
8451: prevacurrent and prev are prevalences.
1.267 brouard 8452: */
8453: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8454: double agec; /* generic age */
1.268 brouard 8455: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8456: double *popeffectif,*popcount;
8457: double ***p3mat;
8458: /* double ***mobaverage; */
8459: char fileresfb[FILENAMELENGTH];
8460:
1.268 brouard 8461: agelim=AGEINF;
1.267 brouard 8462: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8463: in each health status at the date of interview (if between dateprev1 and dateprev2).
8464: We still use firstpass and lastpass as another selection.
8465: */
8466: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8467: /* firstpass, lastpass, stepm, weightopt, model); */
8468:
8469: /*Do we need to compute prevalence again?*/
8470:
8471: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8472:
8473: strcpy(fileresfb,"FB_");
8474: strcat(fileresfb,fileresu);
8475: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8476: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8477: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8478: }
8479: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8480: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8481:
8482: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8483:
8484:
8485: stepsize=(int) (stepm+YEARM-1)/YEARM;
8486: if (stepm<=12) stepsize=1;
8487: if(estepm < stepm){
8488: printf ("Problem %d lower than %d\n",estepm, stepm);
8489: }
1.270 brouard 8490: else{
8491: hstepm=estepm;
8492: }
8493: if(estepm >= stepm){ /* Yes every two year */
8494: stepsize=2;
8495: }
1.267 brouard 8496:
8497: hstepm=hstepm/stepm;
8498: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8499: fractional in yp1 */
8500: anprojmean=yp;
8501: yp2=modf((yp1*12),&yp);
8502: mprojmean=yp;
8503: yp1=modf((yp2*30.5),&yp);
8504: jprojmean=yp;
8505: if(jprojmean==0) jprojmean=1;
8506: if(mprojmean==0) jprojmean=1;
8507:
8508: i1=pow(2,cptcoveff);
8509: if (cptcovn < 1){i1=1;}
8510:
8511: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8512: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8513:
8514: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8515:
8516: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8517: for(k=1; k<=i1;k++){
8518: if(i1 != 1 && TKresult[nres]!= k)
8519: continue;
8520: if(invalidvarcomb[k]){
8521: printf("\nCombination (%d) projection ignored because no cases \n",k);
8522: continue;
8523: }
1.268 brouard 8524: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8525: for(j=1;j<=cptcoveff;j++) {
8526: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8527: }
8528: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8529: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8530: }
8531: fprintf(ficresfb," yearbproj age");
8532: for(j=1; j<=nlstate+ndeath;j++){
8533: for(i=1; i<=nlstate;i++)
1.268 brouard 8534: fprintf(ficresfb," b%d%d",i,j);
8535: fprintf(ficresfb," b.%d",j);
1.267 brouard 8536: }
8537: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8538: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8539: fprintf(ficresfb,"\n");
8540: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8541: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8542: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8543: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8544: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8545: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8546: nhstepm = nhstepm/hstepm;
8547: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8548: oldm=oldms;savm=savms;
1.268 brouard 8549: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8550: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8551: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8552: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8553: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8554: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8555: for (h=0; h<=nhstepm; h++){
1.268 brouard 8556: if (h*hstepm/YEARM*stepm ==-yearp) {
8557: break;
8558: }
8559: }
8560: fprintf(ficresfb,"\n");
8561: for(j=1;j<=cptcoveff;j++)
8562: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8563: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8564: for(i=1; i<=nlstate+ndeath;i++) {
8565: ppij=0.;ppi=0.;
8566: for(j=1; j<=nlstate;j++) {
8567: /* if (mobilav==1) */
1.269 brouard 8568: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8569: ppi=ppi+prevacurrent[(int)agec][j][k];
8570: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8571: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8572: /* else { */
8573: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8574: /* } */
1.268 brouard 8575: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8576: } /* end j */
8577: if(ppi <0.99){
8578: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8579: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8580: }
8581: fprintf(ficresfb," %.3f", ppij);
8582: }/* end j */
1.267 brouard 8583: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8584: } /* end agec */
8585: } /* end yearp */
8586: } /* end k */
1.217 brouard 8587:
1.267 brouard 8588: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8589:
1.267 brouard 8590: fclose(ficresfb);
8591: printf("End of Computing Back forecasting \n");
8592: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8593:
1.267 brouard 8594: }
1.217 brouard 8595:
1.269 brouard 8596: /* Variance of prevalence limit: varprlim */
8597: void varprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **prlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
1.288 brouard 8598: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8599:
8600: char fileresvpl[FILENAMELENGTH];
8601: FILE *ficresvpl;
8602: double **oldm, **savm;
8603: double **varpl; /* Variances of prevalence limits by age */
8604: int i1, k, nres, j ;
8605:
8606: strcpy(fileresvpl,"VPL_");
8607: strcat(fileresvpl,fileresu);
8608: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8609: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8610: exit(0);
8611: }
1.288 brouard 8612: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8613: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8614:
8615: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8616: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8617:
8618: i1=pow(2,cptcoveff);
8619: if (cptcovn < 1){i1=1;}
8620:
8621: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8622: for(k=1; k<=i1;k++){
8623: if(i1 != 1 && TKresult[nres]!= k)
8624: continue;
8625: fprintf(ficresvpl,"\n#****** ");
8626: printf("\n#****** ");
8627: fprintf(ficlog,"\n#****** ");
8628: for(j=1;j<=cptcoveff;j++) {
8629: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8630: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8631: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8632: }
8633: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8634: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8635: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8636: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8637: }
8638: fprintf(ficresvpl,"******\n");
8639: printf("******\n");
8640: fprintf(ficlog,"******\n");
8641:
8642: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8643: oldm=oldms;savm=savms;
8644: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8645: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8646: /*}*/
8647: }
8648:
8649: fclose(ficresvpl);
1.288 brouard 8650: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8651: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8652:
8653: }
8654: /* Variance of back prevalence: varbprlim */
8655: 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){
8656: /*------- Variance of back (stable) prevalence------*/
8657:
8658: char fileresvbl[FILENAMELENGTH];
8659: FILE *ficresvbl;
8660:
8661: double **oldm, **savm;
8662: double **varbpl; /* Variances of back prevalence limits by age */
8663: int i1, k, nres, j ;
8664:
8665: strcpy(fileresvbl,"VBL_");
8666: strcat(fileresvbl,fileresu);
8667: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8668: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8669: exit(0);
8670: }
8671: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8672: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8673:
8674:
8675: i1=pow(2,cptcoveff);
8676: if (cptcovn < 1){i1=1;}
8677:
8678: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8679: for(k=1; k<=i1;k++){
8680: if(i1 != 1 && TKresult[nres]!= k)
8681: continue;
8682: fprintf(ficresvbl,"\n#****** ");
8683: printf("\n#****** ");
8684: fprintf(ficlog,"\n#****** ");
8685: for(j=1;j<=cptcoveff;j++) {
8686: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8687: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8688: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8689: }
8690: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8691: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8692: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8693: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8694: }
8695: fprintf(ficresvbl,"******\n");
8696: printf("******\n");
8697: fprintf(ficlog,"******\n");
8698:
8699: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8700: oldm=oldms;savm=savms;
8701:
8702: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8703: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8704: /*}*/
8705: }
8706:
8707: fclose(ficresvbl);
8708: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8709: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8710:
8711: } /* End of varbprlim */
8712:
1.126 brouard 8713: /************** Forecasting *****not tested NB*************/
1.227 brouard 8714: /* 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 8715:
1.227 brouard 8716: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8717: /* int *popage; */
8718: /* double calagedatem, agelim, kk1, kk2; */
8719: /* double *popeffectif,*popcount; */
8720: /* double ***p3mat,***tabpop,***tabpopprev; */
8721: /* /\* double ***mobaverage; *\/ */
8722: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8723:
1.227 brouard 8724: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8725: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8726: /* agelim=AGESUP; */
8727: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8728:
1.227 brouard 8729: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8730:
8731:
1.227 brouard 8732: /* strcpy(filerespop,"POP_"); */
8733: /* strcat(filerespop,fileresu); */
8734: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8735: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8736: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8737: /* } */
8738: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8739: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8740:
1.227 brouard 8741: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8742:
1.227 brouard 8743: /* /\* if (mobilav!=0) { *\/ */
8744: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8745: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8746: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8747: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8748: /* /\* } *\/ */
8749: /* /\* } *\/ */
1.126 brouard 8750:
1.227 brouard 8751: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8752: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8753:
1.227 brouard 8754: /* agelim=AGESUP; */
1.126 brouard 8755:
1.227 brouard 8756: /* hstepm=1; */
8757: /* hstepm=hstepm/stepm; */
1.218 brouard 8758:
1.227 brouard 8759: /* if (popforecast==1) { */
8760: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8761: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8762: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8763: /* } */
8764: /* popage=ivector(0,AGESUP); */
8765: /* popeffectif=vector(0,AGESUP); */
8766: /* popcount=vector(0,AGESUP); */
1.126 brouard 8767:
1.227 brouard 8768: /* i=1; */
8769: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8770:
1.227 brouard 8771: /* imx=i; */
8772: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8773: /* } */
1.218 brouard 8774:
1.227 brouard 8775: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8776: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8777: /* k=k+1; */
8778: /* fprintf(ficrespop,"\n#******"); */
8779: /* for(j=1;j<=cptcoveff;j++) { */
8780: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8781: /* } */
8782: /* fprintf(ficrespop,"******\n"); */
8783: /* fprintf(ficrespop,"# Age"); */
8784: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8785: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8786:
1.227 brouard 8787: /* for (cpt=0; cpt<=0;cpt++) { */
8788: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8789:
1.227 brouard 8790: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8791: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8792: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8793:
1.227 brouard 8794: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8795: /* oldm=oldms;savm=savms; */
8796: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8797:
1.227 brouard 8798: /* for (h=0; h<=nhstepm; h++){ */
8799: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8800: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8801: /* } */
8802: /* for(j=1; j<=nlstate+ndeath;j++) { */
8803: /* kk1=0.;kk2=0; */
8804: /* for(i=1; i<=nlstate;i++) { */
8805: /* if (mobilav==1) */
8806: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8807: /* else { */
8808: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8809: /* } */
8810: /* } */
8811: /* if (h==(int)(calagedatem+12*cpt)){ */
8812: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8813: /* /\*fprintf(ficrespop," %.3f", kk1); */
8814: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8815: /* } */
8816: /* } */
8817: /* for(i=1; i<=nlstate;i++){ */
8818: /* kk1=0.; */
8819: /* for(j=1; j<=nlstate;j++){ */
8820: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8821: /* } */
8822: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8823: /* } */
1.218 brouard 8824:
1.227 brouard 8825: /* if (h==(int)(calagedatem+12*cpt)) */
8826: /* for(j=1; j<=nlstate;j++) */
8827: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8828: /* } */
8829: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8830: /* } */
8831: /* } */
1.218 brouard 8832:
1.227 brouard 8833: /* /\******\/ */
1.218 brouard 8834:
1.227 brouard 8835: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8836: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8837: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8838: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8839: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8840:
1.227 brouard 8841: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8842: /* oldm=oldms;savm=savms; */
8843: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8844: /* for (h=0; h<=nhstepm; h++){ */
8845: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8846: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8847: /* } */
8848: /* for(j=1; j<=nlstate+ndeath;j++) { */
8849: /* kk1=0.;kk2=0; */
8850: /* for(i=1; i<=nlstate;i++) { */
8851: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8852: /* } */
8853: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8854: /* } */
8855: /* } */
8856: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8857: /* } */
8858: /* } */
8859: /* } */
8860: /* } */
1.218 brouard 8861:
1.227 brouard 8862: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8863:
1.227 brouard 8864: /* if (popforecast==1) { */
8865: /* free_ivector(popage,0,AGESUP); */
8866: /* free_vector(popeffectif,0,AGESUP); */
8867: /* free_vector(popcount,0,AGESUP); */
8868: /* } */
8869: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8870: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8871: /* fclose(ficrespop); */
8872: /* } /\* End of popforecast *\/ */
1.218 brouard 8873:
1.126 brouard 8874: int fileappend(FILE *fichier, char *optionfich)
8875: {
8876: if((fichier=fopen(optionfich,"a"))==NULL) {
8877: printf("Problem with file: %s\n", optionfich);
8878: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8879: return (0);
8880: }
8881: fflush(fichier);
8882: return (1);
8883: }
8884:
8885:
8886: /**************** function prwizard **********************/
8887: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8888: {
8889:
8890: /* Wizard to print covariance matrix template */
8891:
1.164 brouard 8892: char ca[32], cb[32];
8893: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8894: int numlinepar;
8895:
8896: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8897: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8898: for(i=1; i <=nlstate; i++){
8899: jj=0;
8900: for(j=1; j <=nlstate+ndeath; j++){
8901: if(j==i) continue;
8902: jj++;
8903: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8904: printf("%1d%1d",i,j);
8905: fprintf(ficparo,"%1d%1d",i,j);
8906: for(k=1; k<=ncovmodel;k++){
8907: /* printf(" %lf",param[i][j][k]); */
8908: /* fprintf(ficparo," %lf",param[i][j][k]); */
8909: printf(" 0.");
8910: fprintf(ficparo," 0.");
8911: }
8912: printf("\n");
8913: fprintf(ficparo,"\n");
8914: }
8915: }
8916: printf("# Scales (for hessian or gradient estimation)\n");
8917: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8918: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8919: for(i=1; i <=nlstate; i++){
8920: jj=0;
8921: for(j=1; j <=nlstate+ndeath; j++){
8922: if(j==i) continue;
8923: jj++;
8924: fprintf(ficparo,"%1d%1d",i,j);
8925: printf("%1d%1d",i,j);
8926: fflush(stdout);
8927: for(k=1; k<=ncovmodel;k++){
8928: /* printf(" %le",delti3[i][j][k]); */
8929: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8930: printf(" 0.");
8931: fprintf(ficparo," 0.");
8932: }
8933: numlinepar++;
8934: printf("\n");
8935: fprintf(ficparo,"\n");
8936: }
8937: }
8938: printf("# Covariance matrix\n");
8939: /* # 121 Var(a12)\n\ */
8940: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8941: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8942: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8943: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8944: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8945: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8946: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8947: fflush(stdout);
8948: fprintf(ficparo,"# Covariance matrix\n");
8949: /* # 121 Var(a12)\n\ */
8950: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8951: /* # ...\n\ */
8952: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8953:
8954: for(itimes=1;itimes<=2;itimes++){
8955: jj=0;
8956: for(i=1; i <=nlstate; i++){
8957: for(j=1; j <=nlstate+ndeath; j++){
8958: if(j==i) continue;
8959: for(k=1; k<=ncovmodel;k++){
8960: jj++;
8961: ca[0]= k+'a'-1;ca[1]='\0';
8962: if(itimes==1){
8963: printf("#%1d%1d%d",i,j,k);
8964: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8965: }else{
8966: printf("%1d%1d%d",i,j,k);
8967: fprintf(ficparo,"%1d%1d%d",i,j,k);
8968: /* printf(" %.5le",matcov[i][j]); */
8969: }
8970: ll=0;
8971: for(li=1;li <=nlstate; li++){
8972: for(lj=1;lj <=nlstate+ndeath; lj++){
8973: if(lj==li) continue;
8974: for(lk=1;lk<=ncovmodel;lk++){
8975: ll++;
8976: if(ll<=jj){
8977: cb[0]= lk +'a'-1;cb[1]='\0';
8978: if(ll<jj){
8979: if(itimes==1){
8980: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8981: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8982: }else{
8983: printf(" 0.");
8984: fprintf(ficparo," 0.");
8985: }
8986: }else{
8987: if(itimes==1){
8988: printf(" Var(%s%1d%1d)",ca,i,j);
8989: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8990: }else{
8991: printf(" 0.");
8992: fprintf(ficparo," 0.");
8993: }
8994: }
8995: }
8996: } /* end lk */
8997: } /* end lj */
8998: } /* end li */
8999: printf("\n");
9000: fprintf(ficparo,"\n");
9001: numlinepar++;
9002: } /* end k*/
9003: } /*end j */
9004: } /* end i */
9005: } /* end itimes */
9006:
9007: } /* end of prwizard */
9008: /******************* Gompertz Likelihood ******************************/
9009: double gompertz(double x[])
9010: {
9011: double A,B,L=0.0,sump=0.,num=0.;
9012: int i,n=0; /* n is the size of the sample */
9013:
1.220 brouard 9014: for (i=1;i<=imx ; i++) {
1.126 brouard 9015: sump=sump+weight[i];
9016: /* sump=sump+1;*/
9017: num=num+1;
9018: }
9019:
9020:
9021: /* for (i=0; i<=imx; i++)
9022: 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]);*/
9023:
9024: for (i=1;i<=imx ; i++)
9025: {
9026: if (cens[i] == 1 && wav[i]>1)
9027: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9028:
9029: if (cens[i] == 0 && wav[i]>1)
9030: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
9031: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
9032:
9033: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9034: if (wav[i] > 1 ) { /* ??? */
9035: L=L+A*weight[i];
9036: /* 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]);*/
9037: }
9038: }
9039:
9040: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9041:
9042: return -2*L*num/sump;
9043: }
9044:
1.136 brouard 9045: #ifdef GSL
9046: /******************* Gompertz_f Likelihood ******************************/
9047: double gompertz_f(const gsl_vector *v, void *params)
9048: {
9049: double A,B,LL=0.0,sump=0.,num=0.;
9050: double *x= (double *) v->data;
9051: int i,n=0; /* n is the size of the sample */
9052:
9053: for (i=0;i<=imx-1 ; i++) {
9054: sump=sump+weight[i];
9055: /* sump=sump+1;*/
9056: num=num+1;
9057: }
9058:
9059:
9060: /* for (i=0; i<=imx; i++)
9061: 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]);*/
9062: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9063: for (i=1;i<=imx ; i++)
9064: {
9065: if (cens[i] == 1 && wav[i]>1)
9066: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9067:
9068: if (cens[i] == 0 && wav[i]>1)
9069: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9070: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9071:
9072: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9073: if (wav[i] > 1 ) { /* ??? */
9074: LL=LL+A*weight[i];
9075: /* 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]);*/
9076: }
9077: }
9078:
9079: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9080: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9081:
9082: return -2*LL*num/sump;
9083: }
9084: #endif
9085:
1.126 brouard 9086: /******************* Printing html file ***********/
1.201 brouard 9087: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9088: int lastpass, int stepm, int weightopt, char model[],\
9089: int imx, double p[],double **matcov,double agemortsup){
9090: int i,k;
9091:
9092: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9093: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9094: for (i=1;i<=2;i++)
9095: 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 9096: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9097: fprintf(fichtm,"</ul>");
9098:
9099: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9100:
9101: 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>");
9102:
9103: for (k=agegomp;k<(agemortsup-2);k++)
9104: 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]);
9105:
9106:
9107: fflush(fichtm);
9108: }
9109:
9110: /******************* Gnuplot file **************/
1.201 brouard 9111: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9112:
9113: char dirfileres[132],optfileres[132];
1.164 brouard 9114:
1.126 brouard 9115: int ng;
9116:
9117:
9118: /*#ifdef windows */
9119: fprintf(ficgp,"cd \"%s\" \n",pathc);
9120: /*#endif */
9121:
9122:
9123: strcpy(dirfileres,optionfilefiname);
9124: strcpy(optfileres,"vpl");
1.199 brouard 9125: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9126: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9127: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9128: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9129: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9130:
9131: }
9132:
1.136 brouard 9133: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9134: {
1.126 brouard 9135:
1.136 brouard 9136: /*-------- data file ----------*/
9137: FILE *fic;
9138: char dummy[]=" ";
1.240 brouard 9139: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9140: int lstra;
1.136 brouard 9141: int linei, month, year,iout;
9142: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9143: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9144: char *stratrunc;
1.223 brouard 9145:
1.240 brouard 9146: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9147: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9148:
1.240 brouard 9149: for(v=1; v <=ncovcol;v++){
9150: DummyV[v]=0;
9151: FixedV[v]=0;
9152: }
9153: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9154: DummyV[v]=1;
9155: FixedV[v]=0;
9156: }
9157: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9158: DummyV[v]=0;
9159: FixedV[v]=1;
9160: }
9161: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9162: DummyV[v]=1;
9163: FixedV[v]=1;
9164: }
9165: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9166: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9167: 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]);
9168: }
1.126 brouard 9169:
1.136 brouard 9170: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9171: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9172: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9173: }
1.126 brouard 9174:
1.136 brouard 9175: i=1;
9176: linei=0;
9177: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9178: linei=linei+1;
9179: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9180: if(line[j] == '\t')
9181: line[j] = ' ';
9182: }
9183: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9184: ;
9185: };
9186: line[j+1]=0; /* Trims blanks at end of line */
9187: if(line[0]=='#'){
9188: fprintf(ficlog,"Comment line\n%s\n",line);
9189: printf("Comment line\n%s\n",line);
9190: continue;
9191: }
9192: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9193: strcpy(line, linetmp);
1.223 brouard 9194:
9195: /* Loops on waves */
9196: for (j=maxwav;j>=1;j--){
9197: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9198: cutv(stra, strb, line, ' ');
9199: if(strb[0]=='.') { /* Missing value */
9200: lval=-1;
9201: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9202: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9203: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9204: 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);
9205: 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);
9206: return 1;
9207: }
9208: }else{
9209: errno=0;
9210: /* what_kind_of_number(strb); */
9211: dval=strtod(strb,&endptr);
9212: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9213: /* if(strb != endptr && *endptr == '\0') */
9214: /* dval=dlval; */
9215: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9216: if( strb[0]=='\0' || (*endptr != '\0')){
9217: 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);
9218: 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);
9219: return 1;
9220: }
9221: cotqvar[j][iv][i]=dval;
9222: cotvar[j][ntv+iv][i]=dval;
9223: }
9224: strcpy(line,stra);
1.223 brouard 9225: }/* end loop ntqv */
1.225 brouard 9226:
1.223 brouard 9227: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9228: cutv(stra, strb, line, ' ');
9229: if(strb[0]=='.') { /* Missing value */
9230: lval=-1;
9231: }else{
9232: errno=0;
9233: lval=strtol(strb,&endptr,10);
9234: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9235: if( strb[0]=='\0' || (*endptr != '\0')){
9236: 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);
9237: 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);
9238: return 1;
9239: }
9240: }
9241: if(lval <-1 || lval >1){
9242: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9243: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9244: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9245: For example, for multinomial values like 1, 2 and 3,\n \
9246: build V1=0 V2=0 for the reference value (1),\n \
9247: V1=1 V2=0 for (2) \n \
1.223 brouard 9248: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9249: output of IMaCh is often meaningless.\n \
1.223 brouard 9250: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9251: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9252: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9253: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9254: For example, for multinomial values like 1, 2 and 3,\n \
9255: build V1=0 V2=0 for the reference value (1),\n \
9256: V1=1 V2=0 for (2) \n \
1.223 brouard 9257: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9258: output of IMaCh is often meaningless.\n \
1.223 brouard 9259: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9260: return 1;
9261: }
9262: cotvar[j][iv][i]=(double)(lval);
9263: strcpy(line,stra);
1.223 brouard 9264: }/* end loop ntv */
1.225 brouard 9265:
1.223 brouard 9266: /* Statuses at wave */
1.137 brouard 9267: cutv(stra, strb, line, ' ');
1.223 brouard 9268: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9269: lval=-1;
1.136 brouard 9270: }else{
1.238 brouard 9271: errno=0;
9272: lval=strtol(strb,&endptr,10);
9273: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9274: if( strb[0]=='\0' || (*endptr != '\0')){
9275: 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);
9276: 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);
9277: return 1;
9278: }
1.136 brouard 9279: }
1.225 brouard 9280:
1.136 brouard 9281: s[j][i]=lval;
1.225 brouard 9282:
1.223 brouard 9283: /* Date of Interview */
1.136 brouard 9284: strcpy(line,stra);
9285: cutv(stra, strb,line,' ');
1.169 brouard 9286: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9287: }
1.169 brouard 9288: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9289: month=99;
9290: year=9999;
1.136 brouard 9291: }else{
1.225 brouard 9292: 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);
9293: 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);
9294: return 1;
1.136 brouard 9295: }
9296: anint[j][i]= (double) year;
9297: mint[j][i]= (double)month;
9298: strcpy(line,stra);
1.223 brouard 9299: } /* End loop on waves */
1.225 brouard 9300:
1.223 brouard 9301: /* Date of death */
1.136 brouard 9302: cutv(stra, strb,line,' ');
1.169 brouard 9303: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9304: }
1.169 brouard 9305: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9306: month=99;
9307: year=9999;
9308: }else{
1.141 brouard 9309: 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 9310: 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);
9311: return 1;
1.136 brouard 9312: }
9313: andc[i]=(double) year;
9314: moisdc[i]=(double) month;
9315: strcpy(line,stra);
9316:
1.223 brouard 9317: /* Date of birth */
1.136 brouard 9318: cutv(stra, strb,line,' ');
1.169 brouard 9319: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9320: }
1.169 brouard 9321: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9322: month=99;
9323: year=9999;
9324: }else{
1.141 brouard 9325: 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);
9326: 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 9327: return 1;
1.136 brouard 9328: }
9329: if (year==9999) {
1.141 brouard 9330: 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);
9331: 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 9332: return 1;
9333:
1.136 brouard 9334: }
9335: annais[i]=(double)(year);
9336: moisnais[i]=(double)(month);
9337: strcpy(line,stra);
1.225 brouard 9338:
1.223 brouard 9339: /* Sample weight */
1.136 brouard 9340: cutv(stra, strb,line,' ');
9341: errno=0;
9342: dval=strtod(strb,&endptr);
9343: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9344: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9345: 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 9346: fflush(ficlog);
9347: return 1;
9348: }
9349: weight[i]=dval;
9350: strcpy(line,stra);
1.225 brouard 9351:
1.223 brouard 9352: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9353: cutv(stra, strb, line, ' ');
9354: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9355: lval=-1;
1.223 brouard 9356: }else{
1.225 brouard 9357: errno=0;
9358: /* what_kind_of_number(strb); */
9359: dval=strtod(strb,&endptr);
9360: /* if(strb != endptr && *endptr == '\0') */
9361: /* dval=dlval; */
9362: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9363: if( strb[0]=='\0' || (*endptr != '\0')){
9364: 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);
9365: 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);
9366: return 1;
9367: }
9368: coqvar[iv][i]=dval;
1.226 brouard 9369: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9370: }
9371: strcpy(line,stra);
9372: }/* end loop nqv */
1.136 brouard 9373:
1.223 brouard 9374: /* Covariate values */
1.136 brouard 9375: for (j=ncovcol;j>=1;j--){
9376: cutv(stra, strb,line,' ');
1.223 brouard 9377: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9378: lval=-1;
1.136 brouard 9379: }else{
1.225 brouard 9380: errno=0;
9381: lval=strtol(strb,&endptr,10);
9382: if( strb[0]=='\0' || (*endptr != '\0')){
9383: 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);
9384: 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);
9385: return 1;
9386: }
1.136 brouard 9387: }
9388: if(lval <-1 || lval >1){
1.225 brouard 9389: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9390: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9391: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9392: For example, for multinomial values like 1, 2 and 3,\n \
9393: build V1=0 V2=0 for the reference value (1),\n \
9394: V1=1 V2=0 for (2) \n \
1.136 brouard 9395: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9396: output of IMaCh is often meaningless.\n \
1.136 brouard 9397: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9398: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9399: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9400: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9401: For example, for multinomial values like 1, 2 and 3,\n \
9402: build V1=0 V2=0 for the reference value (1),\n \
9403: V1=1 V2=0 for (2) \n \
1.136 brouard 9404: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9405: output of IMaCh is often meaningless.\n \
1.136 brouard 9406: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9407: return 1;
1.136 brouard 9408: }
9409: covar[j][i]=(double)(lval);
9410: strcpy(line,stra);
9411: }
9412: lstra=strlen(stra);
1.225 brouard 9413:
1.136 brouard 9414: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9415: stratrunc = &(stra[lstra-9]);
9416: num[i]=atol(stratrunc);
9417: }
9418: else
9419: num[i]=atol(stra);
9420: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9421: 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;}*/
9422:
9423: i=i+1;
9424: } /* End loop reading data */
1.225 brouard 9425:
1.136 brouard 9426: *imax=i-1; /* Number of individuals */
9427: fclose(fic);
1.225 brouard 9428:
1.136 brouard 9429: return (0);
1.164 brouard 9430: /* endread: */
1.225 brouard 9431: printf("Exiting readdata: ");
9432: fclose(fic);
9433: return (1);
1.223 brouard 9434: }
1.126 brouard 9435:
1.234 brouard 9436: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9437: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9438: while (*p2 == ' ')
1.234 brouard 9439: p2++;
9440: /* while ((*p1++ = *p2++) !=0) */
9441: /* ; */
9442: /* do */
9443: /* while (*p2 == ' ') */
9444: /* p2++; */
9445: /* while (*p1++ == *p2++); */
9446: *stri=p2;
1.145 brouard 9447: }
9448:
1.235 brouard 9449: int decoderesult ( char resultline[], int nres)
1.230 brouard 9450: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9451: {
1.235 brouard 9452: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9453: char resultsav[MAXLINE];
1.234 brouard 9454: int resultmodel[MAXLINE];
9455: int modelresult[MAXLINE];
1.230 brouard 9456: char stra[80], strb[80], strc[80], strd[80],stre[80];
9457:
1.234 brouard 9458: removefirstspace(&resultline);
1.233 brouard 9459: printf("decoderesult:%s\n",resultline);
1.230 brouard 9460:
9461: if (strstr(resultline,"v") !=0){
9462: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9463: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9464: return 1;
9465: }
9466: trimbb(resultsav, resultline);
9467: if (strlen(resultsav) >1){
9468: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9469: }
1.253 brouard 9470: if(j == 0){ /* Resultline but no = */
9471: TKresult[nres]=0; /* Combination for the nresult and the model */
9472: return (0);
9473: }
9474:
1.234 brouard 9475: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9476: 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);
9477: 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);
9478: }
9479: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9480: if(nbocc(resultsav,'=') >1){
9481: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9482: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9483: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9484: }else
9485: cutl(strc,strd,resultsav,'=');
1.230 brouard 9486: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9487:
1.230 brouard 9488: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9489: Tvarsel[k]=atoi(strc);
9490: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9491: /* cptcovsel++; */
9492: if (nbocc(stra,'=') >0)
9493: strcpy(resultsav,stra); /* and analyzes it */
9494: }
1.235 brouard 9495: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9496: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9497: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9498: match=0;
1.236 brouard 9499: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9500: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9501: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9502: match=1;
9503: break;
9504: }
9505: }
9506: if(match == 0){
9507: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9508: }
9509: }
9510: }
1.235 brouard 9511: /* Checking for missing or useless values in comparison of current model needs */
9512: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9513: match=0;
1.235 brouard 9514: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9515: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9516: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9517: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9518: ++match;
9519: }
9520: }
9521: }
9522: if(match == 0){
9523: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9524: }else if(match > 1){
9525: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9526: }
9527: }
1.235 brouard 9528:
1.234 brouard 9529: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9530: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9531: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9532: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9533: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9534: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9535: /* 1 0 0 0 */
9536: /* 2 1 0 0 */
9537: /* 3 0 1 0 */
9538: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9539: /* 5 0 0 1 */
9540: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9541: /* 7 0 1 1 */
9542: /* 8 1 1 1 */
1.237 brouard 9543: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9544: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9545: /* V5*age V5 known which value for nres? */
9546: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9547: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9548: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9549: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9550: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9551: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9552: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9553: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9554: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9555: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9556: k4++;;
9557: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9558: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9559: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9560: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9561: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9562: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9563: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9564: k4q++;;
9565: }
9566: }
1.234 brouard 9567:
1.235 brouard 9568: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9569: return (0);
9570: }
1.235 brouard 9571:
1.230 brouard 9572: int decodemodel( char model[], int lastobs)
9573: /**< This routine decodes the model and returns:
1.224 brouard 9574: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9575: * - nagesqr = 1 if age*age in the model, otherwise 0.
9576: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9577: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9578: * - cptcovage number of covariates with age*products =2
9579: * - cptcovs number of simple covariates
9580: * - 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
9581: * which is a new column after the 9 (ncovcol) variables.
9582: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9583: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9584: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9585: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9586: */
1.136 brouard 9587: {
1.238 brouard 9588: int i, j, k, ks, v;
1.227 brouard 9589: int j1, k1, k2, k3, k4;
1.136 brouard 9590: char modelsav[80];
1.145 brouard 9591: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9592: char *strpt;
1.136 brouard 9593:
1.145 brouard 9594: /*removespace(model);*/
1.136 brouard 9595: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9596: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9597: if (strstr(model,"AGE") !=0){
1.192 brouard 9598: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9599: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9600: return 1;
9601: }
1.141 brouard 9602: if (strstr(model,"v") !=0){
9603: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9604: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9605: return 1;
9606: }
1.187 brouard 9607: strcpy(modelsav,model);
9608: if ((strpt=strstr(model,"age*age")) !=0){
9609: printf(" strpt=%s, model=%s\n",strpt, model);
9610: if(strpt != model){
1.234 brouard 9611: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9612: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9613: corresponding column of parameters.\n",model);
1.234 brouard 9614: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9615: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9616: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9617: return 1;
1.225 brouard 9618: }
1.187 brouard 9619: nagesqr=1;
9620: if (strstr(model,"+age*age") !=0)
1.234 brouard 9621: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9622: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9623: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9624: else
1.234 brouard 9625: substrchaine(modelsav, model, "age*age");
1.187 brouard 9626: }else
9627: nagesqr=0;
9628: if (strlen(modelsav) >1){
9629: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9630: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9631: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9632: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9633: * cst, age and age*age
9634: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9635: /* including age products which are counted in cptcovage.
9636: * but the covariates which are products must be treated
9637: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9638: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9639: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9640:
9641:
1.187 brouard 9642: /* Design
9643: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9644: * < ncovcol=8 >
9645: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9646: * k= 1 2 3 4 5 6 7 8
9647: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9648: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9649: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9650: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9651: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9652: * Tage[++cptcovage]=k
9653: * if products, new covar are created after ncovcol with k1
9654: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9655: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9656: * 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
9657: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9658: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9659: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9660: * < ncovcol=8 >
9661: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9662: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9663: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9664: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9665: * p Tprod[1]@2={ 6, 5}
9666: *p Tvard[1][1]@4= {7, 8, 5, 6}
9667: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9668: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9669: *How to reorganize?
9670: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9671: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9672: * {2, 1, 4, 8, 5, 6, 3, 7}
9673: * Struct []
9674: */
1.225 brouard 9675:
1.187 brouard 9676: /* This loop fills the array Tvar from the string 'model'.*/
9677: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9678: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9679: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9680: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9681: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9682: /* k=1 Tvar[1]=2 (from V2) */
9683: /* k=5 Tvar[5] */
9684: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9685: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9686: /* } */
1.198 brouard 9687: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9688: /*
9689: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9690: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9691: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9692: }
1.187 brouard 9693: cptcovage=0;
9694: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9695: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9696: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9697: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9698: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9699: /*scanf("%d",i);*/
9700: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9701: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9702: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9703: /* covar is not filled and then is empty */
9704: cptcovprod--;
9705: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9706: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9707: Typevar[k]=1; /* 1 for age product */
9708: cptcovage++; /* Sums the number of covariates which include age as a product */
9709: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9710: /*printf("stre=%s ", stre);*/
9711: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9712: cptcovprod--;
9713: cutl(stre,strb,strc,'V');
9714: Tvar[k]=atoi(stre);
9715: Typevar[k]=1; /* 1 for age product */
9716: cptcovage++;
9717: Tage[cptcovage]=k;
9718: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9719: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9720: cptcovn++;
9721: cptcovprodnoage++;k1++;
9722: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9723: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9724: because this model-covariate is a construction we invent a new column
9725: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9726: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9727: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9728: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9729: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9730: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9731: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9732: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9733: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9734: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9735: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9736: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9737: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9738: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9739: for (i=1; i<=lastobs;i++){
9740: /* Computes the new covariate which is a product of
9741: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9742: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9743: }
9744: } /* End age is not in the model */
9745: } /* End if model includes a product */
9746: else { /* no more sum */
9747: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9748: /* scanf("%d",i);*/
9749: cutl(strd,strc,strb,'V');
9750: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9751: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9752: Tvar[k]=atoi(strd);
9753: Typevar[k]=0; /* 0 for simple covariates */
9754: }
9755: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9756: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9757: scanf("%d",i);*/
1.187 brouard 9758: } /* end of loop + on total covariates */
9759: } /* end if strlen(modelsave == 0) age*age might exist */
9760: } /* end if strlen(model == 0) */
1.136 brouard 9761:
9762: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9763: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9764:
1.136 brouard 9765: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9766: printf("cptcovprod=%d ", cptcovprod);
9767: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9768: scanf("%d ",i);*/
9769:
9770:
1.230 brouard 9771: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9772: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9773: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9774: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9775: k = 1 2 3 4 5 6 7 8 9
9776: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9777: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9778: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9779: Dummy[k] 1 0 0 0 3 1 1 2 3
9780: Tmodelind[combination of covar]=k;
1.225 brouard 9781: */
9782: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9783: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9784: /* 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 9785: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9786: printf("Model=%s\n\
9787: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9788: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9789: 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);
9790: fprintf(ficlog,"Model=%s\n\
9791: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9792: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9793: 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 9794: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9795: 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 */
9796: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9797: Fixed[k]= 0;
9798: Dummy[k]= 0;
1.225 brouard 9799: ncoveff++;
1.232 brouard 9800: ncovf++;
1.234 brouard 9801: nsd++;
9802: modell[k].maintype= FTYPE;
9803: TvarsD[nsd]=Tvar[k];
9804: TvarsDind[nsd]=k;
9805: TvarF[ncovf]=Tvar[k];
9806: TvarFind[ncovf]=k;
9807: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9808: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9809: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9810: Fixed[k]= 0;
9811: Dummy[k]= 0;
9812: ncoveff++;
9813: ncovf++;
9814: modell[k].maintype= FTYPE;
9815: TvarF[ncovf]=Tvar[k];
9816: TvarFind[ncovf]=k;
1.230 brouard 9817: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9818: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9819: }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 9820: Fixed[k]= 0;
9821: Dummy[k]= 1;
1.230 brouard 9822: nqfveff++;
1.234 brouard 9823: modell[k].maintype= FTYPE;
9824: modell[k].subtype= FQ;
9825: nsq++;
9826: TvarsQ[nsq]=Tvar[k];
9827: TvarsQind[nsq]=k;
1.232 brouard 9828: ncovf++;
1.234 brouard 9829: TvarF[ncovf]=Tvar[k];
9830: TvarFind[ncovf]=k;
1.231 brouard 9831: 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 9832: 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 9833: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9834: Fixed[k]= 1;
9835: Dummy[k]= 0;
1.225 brouard 9836: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9837: modell[k].maintype= VTYPE;
9838: modell[k].subtype= VD;
9839: nsd++;
9840: TvarsD[nsd]=Tvar[k];
9841: TvarsDind[nsd]=k;
9842: ncovv++; /* Only simple time varying variables */
9843: TvarV[ncovv]=Tvar[k];
1.242 brouard 9844: 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 9845: 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 */
9846: 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 9847: 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);
9848: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9849: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9850: Fixed[k]= 1;
9851: Dummy[k]= 1;
9852: nqtveff++;
9853: modell[k].maintype= VTYPE;
9854: modell[k].subtype= VQ;
9855: ncovv++; /* Only simple time varying variables */
9856: nsq++;
9857: TvarsQ[nsq]=Tvar[k];
9858: TvarsQind[nsq]=k;
9859: TvarV[ncovv]=Tvar[k];
1.242 brouard 9860: 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 9861: 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 */
9862: 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 9863: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9864: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9865: 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 9866: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9867: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9868: ncova++;
9869: TvarA[ncova]=Tvar[k];
9870: TvarAind[ncova]=k;
1.231 brouard 9871: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9872: Fixed[k]= 2;
9873: Dummy[k]= 2;
9874: modell[k].maintype= ATYPE;
9875: modell[k].subtype= APFD;
9876: /* ncoveff++; */
1.227 brouard 9877: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9878: Fixed[k]= 2;
9879: Dummy[k]= 3;
9880: modell[k].maintype= ATYPE;
9881: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9882: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9883: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9884: Fixed[k]= 3;
9885: Dummy[k]= 2;
9886: modell[k].maintype= ATYPE;
9887: modell[k].subtype= APVD; /* Product age * varying dummy */
9888: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9889: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9890: Fixed[k]= 3;
9891: Dummy[k]= 3;
9892: modell[k].maintype= ATYPE;
9893: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9894: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9895: }
9896: }else if (Typevar[k] == 2) { /* product without age */
9897: k1=Tposprod[k];
9898: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9899: if(Tvard[k1][2] <=ncovcol){
9900: Fixed[k]= 1;
9901: Dummy[k]= 0;
9902: modell[k].maintype= FTYPE;
9903: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9904: ncovf++; /* Fixed variables without age */
9905: TvarF[ncovf]=Tvar[k];
9906: TvarFind[ncovf]=k;
9907: }else if(Tvard[k1][2] <=ncovcol+nqv){
9908: Fixed[k]= 0; /* or 2 ?*/
9909: Dummy[k]= 1;
9910: modell[k].maintype= FTYPE;
9911: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9912: ncovf++; /* Varying variables without age */
9913: TvarF[ncovf]=Tvar[k];
9914: TvarFind[ncovf]=k;
9915: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9916: Fixed[k]= 1;
9917: Dummy[k]= 0;
9918: modell[k].maintype= VTYPE;
9919: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9920: ncovv++; /* Varying variables without age */
9921: TvarV[ncovv]=Tvar[k];
9922: TvarVind[ncovv]=k;
9923: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9924: Fixed[k]= 1;
9925: Dummy[k]= 1;
9926: modell[k].maintype= VTYPE;
9927: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9928: ncovv++; /* Varying variables without age */
9929: TvarV[ncovv]=Tvar[k];
9930: TvarVind[ncovv]=k;
9931: }
1.227 brouard 9932: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9933: if(Tvard[k1][2] <=ncovcol){
9934: Fixed[k]= 0; /* or 2 ?*/
9935: Dummy[k]= 1;
9936: modell[k].maintype= FTYPE;
9937: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9938: ncovf++; /* Fixed variables without age */
9939: TvarF[ncovf]=Tvar[k];
9940: TvarFind[ncovf]=k;
9941: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9942: Fixed[k]= 1;
9943: Dummy[k]= 1;
9944: modell[k].maintype= VTYPE;
9945: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9946: ncovv++; /* Varying variables without age */
9947: TvarV[ncovv]=Tvar[k];
9948: TvarVind[ncovv]=k;
9949: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9950: Fixed[k]= 1;
9951: Dummy[k]= 1;
9952: modell[k].maintype= VTYPE;
9953: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9954: ncovv++; /* Varying variables without age */
9955: TvarV[ncovv]=Tvar[k];
9956: TvarVind[ncovv]=k;
9957: ncovv++; /* Varying variables without age */
9958: TvarV[ncovv]=Tvar[k];
9959: TvarVind[ncovv]=k;
9960: }
1.227 brouard 9961: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9962: if(Tvard[k1][2] <=ncovcol){
9963: Fixed[k]= 1;
9964: Dummy[k]= 1;
9965: modell[k].maintype= VTYPE;
9966: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9967: ncovv++; /* Varying variables without age */
9968: TvarV[ncovv]=Tvar[k];
9969: TvarVind[ncovv]=k;
9970: }else if(Tvard[k1][2] <=ncovcol+nqv){
9971: Fixed[k]= 1;
9972: Dummy[k]= 1;
9973: modell[k].maintype= VTYPE;
9974: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9975: ncovv++; /* Varying variables without age */
9976: TvarV[ncovv]=Tvar[k];
9977: TvarVind[ncovv]=k;
9978: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9979: Fixed[k]= 1;
9980: Dummy[k]= 0;
9981: modell[k].maintype= VTYPE;
9982: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9983: ncovv++; /* Varying variables without age */
9984: TvarV[ncovv]=Tvar[k];
9985: TvarVind[ncovv]=k;
9986: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9987: Fixed[k]= 1;
9988: Dummy[k]= 1;
9989: modell[k].maintype= VTYPE;
9990: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9991: ncovv++; /* Varying variables without age */
9992: TvarV[ncovv]=Tvar[k];
9993: TvarVind[ncovv]=k;
9994: }
1.227 brouard 9995: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9996: if(Tvard[k1][2] <=ncovcol){
9997: Fixed[k]= 1;
9998: Dummy[k]= 1;
9999: modell[k].maintype= VTYPE;
10000: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10001: ncovv++; /* Varying variables without age */
10002: TvarV[ncovv]=Tvar[k];
10003: TvarVind[ncovv]=k;
10004: }else if(Tvard[k1][2] <=ncovcol+nqv){
10005: Fixed[k]= 1;
10006: Dummy[k]= 1;
10007: modell[k].maintype= VTYPE;
10008: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10009: ncovv++; /* Varying variables without age */
10010: TvarV[ncovv]=Tvar[k];
10011: TvarVind[ncovv]=k;
10012: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10013: Fixed[k]= 1;
10014: Dummy[k]= 1;
10015: modell[k].maintype= VTYPE;
10016: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10017: ncovv++; /* Varying variables without age */
10018: TvarV[ncovv]=Tvar[k];
10019: TvarVind[ncovv]=k;
10020: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10021: Fixed[k]= 1;
10022: Dummy[k]= 1;
10023: modell[k].maintype= VTYPE;
10024: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10025: ncovv++; /* Varying variables without age */
10026: TvarV[ncovv]=Tvar[k];
10027: TvarVind[ncovv]=k;
10028: }
1.227 brouard 10029: }else{
1.240 brouard 10030: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10031: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10032: } /*end k1*/
1.225 brouard 10033: }else{
1.226 brouard 10034: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10035: 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 10036: }
1.227 brouard 10037: 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 10038: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10039: 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]);
10040: }
10041: /* Searching for doublons in the model */
10042: for(k1=1; k1<= cptcovt;k1++){
10043: for(k2=1; k2 <k1;k2++){
1.285 brouard 10044: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10045: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10046: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10047: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10048: 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]);
10049: 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 10050: return(1);
10051: }
10052: }else if (Typevar[k1] ==2){
10053: k3=Tposprod[k1];
10054: k4=Tposprod[k2];
10055: 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])) ){
10056: 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]]);
10057: 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);
10058: return(1);
10059: }
10060: }
1.227 brouard 10061: }
10062: }
1.225 brouard 10063: }
10064: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10065: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10066: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10067: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10068: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10069: /*endread:*/
1.225 brouard 10070: printf("Exiting decodemodel: ");
10071: return (1);
1.136 brouard 10072: }
10073:
1.169 brouard 10074: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10075: {/* Check ages at death */
1.136 brouard 10076: int i, m;
1.218 brouard 10077: int firstone=0;
10078:
1.136 brouard 10079: for (i=1; i<=imx; i++) {
10080: for(m=2; (m<= maxwav); m++) {
10081: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10082: anint[m][i]=9999;
1.216 brouard 10083: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10084: s[m][i]=-1;
1.136 brouard 10085: }
10086: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10087: *nberr = *nberr + 1;
1.218 brouard 10088: if(firstone == 0){
10089: firstone=1;
1.260 brouard 10090: 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 10091: }
1.262 brouard 10092: 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 10093: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10094: }
10095: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10096: (*nberr)++;
1.259 brouard 10097: 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 10098: 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 10099: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10100: }
10101: }
10102: }
10103:
10104: for (i=1; i<=imx; i++) {
10105: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10106: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10107: 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 10108: if (s[m][i] >= nlstate+1) {
1.169 brouard 10109: if(agedc[i]>0){
10110: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10111: agev[m][i]=agedc[i];
1.214 brouard 10112: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10113: }else {
1.136 brouard 10114: if ((int)andc[i]!=9999){
10115: nbwarn++;
10116: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10117: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10118: agev[m][i]=-1;
10119: }
10120: }
1.169 brouard 10121: } /* agedc > 0 */
1.214 brouard 10122: } /* end if */
1.136 brouard 10123: else if(s[m][i] !=9){ /* Standard case, age in fractional
10124: years but with the precision of a month */
10125: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10126: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10127: agev[m][i]=1;
10128: else if(agev[m][i] < *agemin){
10129: *agemin=agev[m][i];
10130: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10131: }
10132: else if(agev[m][i] >*agemax){
10133: *agemax=agev[m][i];
1.156 brouard 10134: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10135: }
10136: /*agev[m][i]=anint[m][i]-annais[i];*/
10137: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10138: } /* en if 9*/
1.136 brouard 10139: else { /* =9 */
1.214 brouard 10140: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10141: agev[m][i]=1;
10142: s[m][i]=-1;
10143: }
10144: }
1.214 brouard 10145: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10146: agev[m][i]=1;
1.214 brouard 10147: else{
10148: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10149: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10150: agev[m][i]=0;
10151: }
10152: } /* End for lastpass */
10153: }
1.136 brouard 10154:
10155: for (i=1; i<=imx; i++) {
10156: for(m=firstpass; (m<=lastpass); m++){
10157: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10158: (*nberr)++;
1.136 brouard 10159: 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);
10160: 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);
10161: return 1;
10162: }
10163: }
10164: }
10165:
10166: /*for (i=1; i<=imx; i++){
10167: for (m=firstpass; (m<lastpass); m++){
10168: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10169: }
10170:
10171: }*/
10172:
10173:
1.139 brouard 10174: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10175: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10176:
10177: return (0);
1.164 brouard 10178: /* endread:*/
1.136 brouard 10179: printf("Exiting calandcheckages: ");
10180: return (1);
10181: }
10182:
1.172 brouard 10183: #if defined(_MSC_VER)
10184: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10185: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10186: //#include "stdafx.h"
10187: //#include <stdio.h>
10188: //#include <tchar.h>
10189: //#include <windows.h>
10190: //#include <iostream>
10191: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10192:
10193: LPFN_ISWOW64PROCESS fnIsWow64Process;
10194:
10195: BOOL IsWow64()
10196: {
10197: BOOL bIsWow64 = FALSE;
10198:
10199: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10200: // (HANDLE, PBOOL);
10201:
10202: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10203:
10204: HMODULE module = GetModuleHandle(_T("kernel32"));
10205: const char funcName[] = "IsWow64Process";
10206: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10207: GetProcAddress(module, funcName);
10208:
10209: if (NULL != fnIsWow64Process)
10210: {
10211: if (!fnIsWow64Process(GetCurrentProcess(),
10212: &bIsWow64))
10213: //throw std::exception("Unknown error");
10214: printf("Unknown error\n");
10215: }
10216: return bIsWow64 != FALSE;
10217: }
10218: #endif
1.177 brouard 10219:
1.191 brouard 10220: void syscompilerinfo(int logged)
1.292 brouard 10221: {
10222: #include <stdint.h>
10223:
10224: /* #include "syscompilerinfo.h"*/
1.185 brouard 10225: /* command line Intel compiler 32bit windows, XP compatible:*/
10226: /* /GS /W3 /Gy
10227: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10228: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10229: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10230: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10231: */
10232: /* 64 bits */
1.185 brouard 10233: /*
10234: /GS /W3 /Gy
10235: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10236: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10237: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10238: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10239: /* Optimization are useless and O3 is slower than O2 */
10240: /*
10241: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10242: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10243: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10244: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10245: */
1.186 brouard 10246: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10247: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10248: /PDB:"visual studio
10249: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10250: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10251: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10252: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10253: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10254: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10255: uiAccess='false'"
10256: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10257: /NOLOGO /TLBID:1
10258: */
1.292 brouard 10259:
10260:
1.177 brouard 10261: #if defined __INTEL_COMPILER
1.178 brouard 10262: #if defined(__GNUC__)
10263: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10264: #endif
1.177 brouard 10265: #elif defined(__GNUC__)
1.179 brouard 10266: #ifndef __APPLE__
1.174 brouard 10267: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10268: #endif
1.177 brouard 10269: struct utsname sysInfo;
1.178 brouard 10270: int cross = CROSS;
10271: if (cross){
10272: printf("Cross-");
1.191 brouard 10273: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10274: }
1.174 brouard 10275: #endif
10276:
1.191 brouard 10277: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10278: #if defined(__clang__)
1.191 brouard 10279: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10280: #endif
10281: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10282: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10283: #endif
10284: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10285: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10286: #endif
10287: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10288: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10289: #endif
10290: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10291: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10292: #endif
10293: #if defined(_MSC_VER)
1.191 brouard 10294: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10295: #endif
10296: #if defined(__PGI)
1.191 brouard 10297: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10298: #endif
10299: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10300: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10301: #endif
1.191 brouard 10302: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10303:
1.167 brouard 10304: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10305: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10306: // Windows (x64 and x86)
1.191 brouard 10307: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10308: #elif __unix__ // all unices, not all compilers
10309: // Unix
1.191 brouard 10310: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10311: #elif __linux__
10312: // linux
1.191 brouard 10313: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10314: #elif __APPLE__
1.174 brouard 10315: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10316: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10317: #endif
10318:
10319: /* __MINGW32__ */
10320: /* __CYGWIN__ */
10321: /* __MINGW64__ */
10322: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10323: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10324: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10325: /* _WIN64 // Defined for applications for Win64. */
10326: /* _M_X64 // Defined for compilations that target x64 processors. */
10327: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10328:
1.167 brouard 10329: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10330: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10331: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10332: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10333: #else
1.191 brouard 10334: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10335: #endif
10336:
1.169 brouard 10337: #if defined(__GNUC__)
10338: # if defined(__GNUC_PATCHLEVEL__)
10339: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10340: + __GNUC_MINOR__ * 100 \
10341: + __GNUC_PATCHLEVEL__)
10342: # else
10343: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10344: + __GNUC_MINOR__ * 100)
10345: # endif
1.174 brouard 10346: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10347: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10348:
10349: if (uname(&sysInfo) != -1) {
10350: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10351: 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 10352: }
10353: else
10354: perror("uname() error");
1.179 brouard 10355: //#ifndef __INTEL_COMPILER
10356: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10357: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10358: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10359: #endif
1.169 brouard 10360: #endif
1.172 brouard 10361:
1.286 brouard 10362: // void main ()
1.172 brouard 10363: // {
1.169 brouard 10364: #if defined(_MSC_VER)
1.174 brouard 10365: if (IsWow64()){
1.191 brouard 10366: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10367: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10368: }
10369: else{
1.191 brouard 10370: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10371: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10372: }
1.172 brouard 10373: // printf("\nPress Enter to continue...");
10374: // getchar();
10375: // }
10376:
1.169 brouard 10377: #endif
10378:
1.167 brouard 10379:
1.219 brouard 10380: }
1.136 brouard 10381:
1.219 brouard 10382: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10383: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10384: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10385: /* double ftolpl = 1.e-10; */
1.180 brouard 10386: double age, agebase, agelim;
1.203 brouard 10387: double tot;
1.180 brouard 10388:
1.202 brouard 10389: strcpy(filerespl,"PL_");
10390: strcat(filerespl,fileresu);
10391: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10392: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10393: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10394: }
1.288 brouard 10395: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10396: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10397: pstamp(ficrespl);
1.288 brouard 10398: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10399: fprintf(ficrespl,"#Age ");
10400: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10401: fprintf(ficrespl,"\n");
1.180 brouard 10402:
1.219 brouard 10403: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10404:
1.219 brouard 10405: agebase=ageminpar;
10406: agelim=agemaxpar;
1.180 brouard 10407:
1.227 brouard 10408: /* i1=pow(2,ncoveff); */
1.234 brouard 10409: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10410: if (cptcovn < 1){i1=1;}
1.180 brouard 10411:
1.238 brouard 10412: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10413: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10414: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10415: continue;
1.235 brouard 10416:
1.238 brouard 10417: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10418: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10419: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10420: /* k=k+1; */
10421: /* to clean */
10422: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10423: fprintf(ficrespl,"#******");
10424: printf("#******");
10425: fprintf(ficlog,"#******");
10426: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10427: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10428: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10429: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10430: }
10431: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10432: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10433: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10434: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10435: }
10436: fprintf(ficrespl,"******\n");
10437: printf("******\n");
10438: fprintf(ficlog,"******\n");
10439: if(invalidvarcomb[k]){
10440: printf("\nCombination (%d) ignored because no case \n",k);
10441: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10442: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10443: continue;
10444: }
1.219 brouard 10445:
1.238 brouard 10446: fprintf(ficrespl,"#Age ");
10447: for(j=1;j<=cptcoveff;j++) {
10448: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10449: }
10450: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10451: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10452:
1.238 brouard 10453: for (age=agebase; age<=agelim; age++){
10454: /* for (age=agebase; age<=agebase; age++){ */
10455: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10456: fprintf(ficrespl,"%.0f ",age );
10457: for(j=1;j<=cptcoveff;j++)
10458: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10459: tot=0.;
10460: for(i=1; i<=nlstate;i++){
10461: tot += prlim[i][i];
10462: fprintf(ficrespl," %.5f", prlim[i][i]);
10463: }
10464: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10465: } /* Age */
10466: /* was end of cptcod */
10467: } /* cptcov */
10468: } /* nres */
1.219 brouard 10469: return 0;
1.180 brouard 10470: }
10471:
1.218 brouard 10472: int back_prevalence_limit(double *p, double **bprlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp, double dateprev1,double dateprev2, int firstpass, int lastpass, int mobilavproj){
1.288 brouard 10473: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10474:
10475: /* Computes the back prevalence limit for any combination of covariate values
10476: * at any age between ageminpar and agemaxpar
10477: */
1.235 brouard 10478: int i, j, k, i1, nres=0 ;
1.217 brouard 10479: /* double ftolpl = 1.e-10; */
10480: double age, agebase, agelim;
10481: double tot;
1.218 brouard 10482: /* double ***mobaverage; */
10483: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10484:
10485: strcpy(fileresplb,"PLB_");
10486: strcat(fileresplb,fileresu);
10487: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10488: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10489: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10490: }
1.288 brouard 10491: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10492: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10493: pstamp(ficresplb);
1.288 brouard 10494: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10495: fprintf(ficresplb,"#Age ");
10496: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10497: fprintf(ficresplb,"\n");
10498:
1.218 brouard 10499:
10500: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10501:
10502: agebase=ageminpar;
10503: agelim=agemaxpar;
10504:
10505:
1.227 brouard 10506: i1=pow(2,cptcoveff);
1.218 brouard 10507: if (cptcovn < 1){i1=1;}
1.227 brouard 10508:
1.238 brouard 10509: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10510: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10511: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10512: continue;
10513: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10514: fprintf(ficresplb,"#******");
10515: printf("#******");
10516: fprintf(ficlog,"#******");
10517: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10518: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10519: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10520: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10521: }
10522: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10523: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10524: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10525: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10526: }
10527: fprintf(ficresplb,"******\n");
10528: printf("******\n");
10529: fprintf(ficlog,"******\n");
10530: if(invalidvarcomb[k]){
10531: printf("\nCombination (%d) ignored because no cases \n",k);
10532: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10533: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10534: continue;
10535: }
1.218 brouard 10536:
1.238 brouard 10537: fprintf(ficresplb,"#Age ");
10538: for(j=1;j<=cptcoveff;j++) {
10539: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10540: }
10541: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10542: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10543:
10544:
1.238 brouard 10545: for (age=agebase; age<=agelim; age++){
10546: /* for (age=agebase; age<=agebase; age++){ */
10547: if(mobilavproj > 0){
10548: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10549: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10550: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10551: }else if (mobilavproj == 0){
10552: 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);
10553: 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);
10554: exit(1);
10555: }else{
10556: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10557: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10558: /* printf("TOTOT\n"); */
10559: /* exit(1); */
1.238 brouard 10560: }
10561: fprintf(ficresplb,"%.0f ",age );
10562: for(j=1;j<=cptcoveff;j++)
10563: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10564: tot=0.;
10565: for(i=1; i<=nlstate;i++){
10566: tot += bprlim[i][i];
10567: fprintf(ficresplb," %.5f", bprlim[i][i]);
10568: }
10569: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10570: } /* Age */
10571: /* was end of cptcod */
1.255 brouard 10572: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10573: } /* end of any combination */
10574: } /* end of nres */
1.218 brouard 10575: /* hBijx(p, bage, fage); */
10576: /* fclose(ficrespijb); */
10577:
10578: return 0;
1.217 brouard 10579: }
1.218 brouard 10580:
1.180 brouard 10581: int hPijx(double *p, int bage, int fage){
10582: /*------------- h Pij x at various ages ------------*/
10583:
10584: int stepsize;
10585: int agelim;
10586: int hstepm;
10587: int nhstepm;
1.235 brouard 10588: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10589:
10590: double agedeb;
10591: double ***p3mat;
10592:
1.201 brouard 10593: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10594: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10595: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10596: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10597: }
10598: printf("Computing pij: result on file '%s' \n", filerespij);
10599: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10600:
10601: stepsize=(int) (stepm+YEARM-1)/YEARM;
10602: /*if (stepm<=24) stepsize=2;*/
10603:
10604: agelim=AGESUP;
10605: hstepm=stepsize*YEARM; /* Every year of age */
10606: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10607:
1.180 brouard 10608: /* hstepm=1; aff par mois*/
10609: pstamp(ficrespij);
10610: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10611: i1= pow(2,cptcoveff);
1.218 brouard 10612: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10613: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10614: /* k=k+1; */
1.235 brouard 10615: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10616: for(k=1; k<=i1;k++){
1.253 brouard 10617: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10618: continue;
1.183 brouard 10619: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10620: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10621: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10622: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10623: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10624: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10625: }
1.183 brouard 10626: fprintf(ficrespij,"******\n");
10627:
10628: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10629: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10630: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10631:
10632: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10633:
1.183 brouard 10634: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10635: oldm=oldms;savm=savms;
1.235 brouard 10636: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10637: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10638: for(i=1; i<=nlstate;i++)
10639: for(j=1; j<=nlstate+ndeath;j++)
10640: fprintf(ficrespij," %1d-%1d",i,j);
10641: fprintf(ficrespij,"\n");
10642: for (h=0; h<=nhstepm; h++){
10643: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10644: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10645: for(i=1; i<=nlstate;i++)
10646: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10647: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10648: fprintf(ficrespij,"\n");
10649: }
1.183 brouard 10650: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10651: fprintf(ficrespij,"\n");
10652: }
1.180 brouard 10653: /*}*/
10654: }
1.218 brouard 10655: return 0;
1.180 brouard 10656: }
1.218 brouard 10657:
10658: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10659: /*------------- h Bij x at various ages ------------*/
10660:
10661: int stepsize;
1.218 brouard 10662: /* int agelim; */
10663: int ageminl;
1.217 brouard 10664: int hstepm;
10665: int nhstepm;
1.238 brouard 10666: int h, i, i1, j, k, nres;
1.218 brouard 10667:
1.217 brouard 10668: double agedeb;
10669: double ***p3mat;
1.218 brouard 10670:
10671: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10672: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10673: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10674: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10675: }
10676: printf("Computing pij back: result on file '%s' \n", filerespijb);
10677: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10678:
10679: stepsize=(int) (stepm+YEARM-1)/YEARM;
10680: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10681:
1.218 brouard 10682: /* agelim=AGESUP; */
1.289 brouard 10683: ageminl=AGEINF; /* was 30 */
1.218 brouard 10684: hstepm=stepsize*YEARM; /* Every year of age */
10685: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10686:
10687: /* hstepm=1; aff par mois*/
10688: pstamp(ficrespijb);
1.255 brouard 10689: 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 10690: i1= pow(2,cptcoveff);
1.218 brouard 10691: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10692: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10693: /* k=k+1; */
1.238 brouard 10694: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10695: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10696: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10697: continue;
10698: fprintf(ficrespijb,"\n#****** ");
10699: for(j=1;j<=cptcoveff;j++)
10700: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10701: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10702: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10703: }
10704: fprintf(ficrespijb,"******\n");
1.264 brouard 10705: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10706: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10707: continue;
10708: }
10709:
10710: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10711: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10712: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10713: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10714: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10715:
10716: /* nhstepm=nhstepm*YEARM; aff par mois*/
10717:
1.266 brouard 10718: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10719: /* and memory limitations if stepm is small */
10720:
1.238 brouard 10721: /* oldm=oldms;savm=savms; */
10722: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10723: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10724: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10725: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10726: for(i=1; i<=nlstate;i++)
10727: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10728: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10729: fprintf(ficrespijb,"\n");
1.238 brouard 10730: for (h=0; h<=nhstepm; h++){
10731: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10732: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10733: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10734: for(i=1; i<=nlstate;i++)
10735: for(j=1; j<=nlstate+ndeath;j++)
10736: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10737: fprintf(ficrespijb,"\n");
10738: }
10739: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10740: fprintf(ficrespijb,"\n");
10741: } /* end age deb */
10742: } /* end combination */
10743: } /* end nres */
1.218 brouard 10744: return 0;
10745: } /* hBijx */
1.217 brouard 10746:
1.180 brouard 10747:
1.136 brouard 10748: /***********************************************/
10749: /**************** Main Program *****************/
10750: /***********************************************/
10751:
10752: int main(int argc, char *argv[])
10753: {
10754: #ifdef GSL
10755: const gsl_multimin_fminimizer_type *T;
10756: size_t iteri = 0, it;
10757: int rval = GSL_CONTINUE;
10758: int status = GSL_SUCCESS;
10759: double ssval;
10760: #endif
10761: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 10762: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
10763: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 10764: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10765: int jj, ll, li, lj, lk;
1.136 brouard 10766: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10767: int num_filled;
1.136 brouard 10768: int itimes;
10769: int NDIM=2;
10770: int vpopbased=0;
1.235 brouard 10771: int nres=0;
1.258 brouard 10772: int endishere=0;
1.277 brouard 10773: int noffset=0;
1.274 brouard 10774: int ncurrv=0; /* Temporary variable */
10775:
1.164 brouard 10776: char ca[32], cb[32];
1.136 brouard 10777: /* FILE *fichtm; *//* Html File */
10778: /* FILE *ficgp;*/ /*Gnuplot File */
10779: struct stat info;
1.191 brouard 10780: double agedeb=0.;
1.194 brouard 10781:
10782: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10783: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10784:
1.165 brouard 10785: double fret;
1.191 brouard 10786: double dum=0.; /* Dummy variable */
1.136 brouard 10787: double ***p3mat;
1.218 brouard 10788: /* double ***mobaverage; */
1.164 brouard 10789:
10790: char line[MAXLINE];
1.197 brouard 10791: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10792:
1.234 brouard 10793: char modeltemp[MAXLINE];
1.230 brouard 10794: char resultline[MAXLINE];
10795:
1.136 brouard 10796: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10797: char *tok, *val; /* pathtot */
1.290 brouard 10798: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 10799: int c, h , cpt, c2;
1.191 brouard 10800: int jl=0;
10801: int i1, j1, jk, stepsize=0;
1.194 brouard 10802: int count=0;
10803:
1.164 brouard 10804: int *tab;
1.136 brouard 10805: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.293 ! brouard 10806: int backcast=0; /* defined as global for mlikeli and mle*/
1.136 brouard 10807: int mobilav=0,popforecast=0;
1.191 brouard 10808: int hstepm=0, nhstepm=0;
1.136 brouard 10809: int agemortsup;
10810: float sumlpop=0.;
10811: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10812: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10813:
1.191 brouard 10814: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10815: double ftolpl=FTOL;
10816: double **prlim;
1.217 brouard 10817: double **bprlim;
1.136 brouard 10818: double ***param; /* Matrix of parameters */
1.251 brouard 10819: double ***paramstart; /* Matrix of starting parameter values */
10820: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10821: double **matcov; /* Matrix of covariance */
1.203 brouard 10822: double **hess; /* Hessian matrix */
1.136 brouard 10823: double ***delti3; /* Scale */
10824: double *delti; /* Scale */
10825: double ***eij, ***vareij;
10826: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10827:
1.136 brouard 10828: double *epj, vepp;
1.164 brouard 10829:
1.273 brouard 10830: double dateprev1, dateprev2;
10831: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10832: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10833:
1.136 brouard 10834: double **ximort;
1.145 brouard 10835: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10836: int *dcwave;
10837:
1.164 brouard 10838: char z[1]="c";
1.136 brouard 10839:
10840: /*char *strt;*/
10841: char strtend[80];
1.126 brouard 10842:
1.164 brouard 10843:
1.126 brouard 10844: /* setlocale (LC_ALL, ""); */
10845: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10846: /* textdomain (PACKAGE); */
10847: /* setlocale (LC_CTYPE, ""); */
10848: /* setlocale (LC_MESSAGES, ""); */
10849:
10850: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10851: rstart_time = time(NULL);
10852: /* (void) gettimeofday(&start_time,&tzp);*/
10853: start_time = *localtime(&rstart_time);
1.126 brouard 10854: curr_time=start_time;
1.157 brouard 10855: /*tml = *localtime(&start_time.tm_sec);*/
10856: /* strcpy(strstart,asctime(&tml)); */
10857: strcpy(strstart,asctime(&start_time));
1.126 brouard 10858:
10859: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10860: /* tp.tm_sec = tp.tm_sec +86400; */
10861: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10862: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10863: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10864: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10865: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10866: /* strt=asctime(&tmg); */
10867: /* printf("Time(after) =%s",strstart); */
10868: /* (void) time (&time_value);
10869: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10870: * tm = *localtime(&time_value);
10871: * strstart=asctime(&tm);
10872: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10873: */
10874:
10875: nberr=0; /* Number of errors and warnings */
10876: nbwarn=0;
1.184 brouard 10877: #ifdef WIN32
10878: _getcwd(pathcd, size);
10879: #else
1.126 brouard 10880: getcwd(pathcd, size);
1.184 brouard 10881: #endif
1.191 brouard 10882: syscompilerinfo(0);
1.196 brouard 10883: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10884: if(argc <=1){
10885: printf("\nEnter the parameter file name: ");
1.205 brouard 10886: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10887: printf("ERROR Empty parameter file name\n");
10888: goto end;
10889: }
1.126 brouard 10890: i=strlen(pathr);
10891: if(pathr[i-1]=='\n')
10892: pathr[i-1]='\0';
1.156 brouard 10893: i=strlen(pathr);
1.205 brouard 10894: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10895: pathr[i-1]='\0';
1.205 brouard 10896: }
10897: i=strlen(pathr);
10898: if( i==0 ){
10899: printf("ERROR Empty parameter file name\n");
10900: goto end;
10901: }
10902: for (tok = pathr; tok != NULL; ){
1.126 brouard 10903: printf("Pathr |%s|\n",pathr);
10904: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10905: printf("val= |%s| pathr=%s\n",val,pathr);
10906: strcpy (pathtot, val);
10907: if(pathr[0] == '\0') break; /* Dirty */
10908: }
10909: }
1.281 brouard 10910: else if (argc<=2){
10911: strcpy(pathtot,argv[1]);
10912: }
1.126 brouard 10913: else{
10914: strcpy(pathtot,argv[1]);
1.281 brouard 10915: strcpy(z,argv[2]);
10916: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 10917: }
10918: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10919: /*cygwin_split_path(pathtot,path,optionfile);
10920: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10921: /* cutv(path,optionfile,pathtot,'\\');*/
10922:
10923: /* Split argv[0], imach program to get pathimach */
10924: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10925: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10926: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10927: /* strcpy(pathimach,argv[0]); */
10928: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10929: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10930: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10931: #ifdef WIN32
10932: _chdir(path); /* Can be a relative path */
10933: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10934: #else
1.126 brouard 10935: chdir(path); /* Can be a relative path */
1.184 brouard 10936: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10937: #endif
10938: printf("Current directory %s!\n",pathcd);
1.126 brouard 10939: strcpy(command,"mkdir ");
10940: strcat(command,optionfilefiname);
10941: if((outcmd=system(command)) != 0){
1.169 brouard 10942: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10943: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10944: /* fclose(ficlog); */
10945: /* exit(1); */
10946: }
10947: /* if((imk=mkdir(optionfilefiname))<0){ */
10948: /* perror("mkdir"); */
10949: /* } */
10950:
10951: /*-------- arguments in the command line --------*/
10952:
1.186 brouard 10953: /* Main Log file */
1.126 brouard 10954: strcat(filelog, optionfilefiname);
10955: strcat(filelog,".log"); /* */
10956: if((ficlog=fopen(filelog,"w"))==NULL) {
10957: printf("Problem with logfile %s\n",filelog);
10958: goto end;
10959: }
10960: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10961: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10962: fprintf(ficlog,"\nEnter the parameter file name: \n");
10963: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10964: path=%s \n\
10965: optionfile=%s\n\
10966: optionfilext=%s\n\
1.156 brouard 10967: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10968:
1.197 brouard 10969: syscompilerinfo(1);
1.167 brouard 10970:
1.126 brouard 10971: printf("Local time (at start):%s",strstart);
10972: fprintf(ficlog,"Local time (at start): %s",strstart);
10973: fflush(ficlog);
10974: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10975: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10976:
10977: /* */
10978: strcpy(fileres,"r");
10979: strcat(fileres, optionfilefiname);
1.201 brouard 10980: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10981: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10982: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10983:
1.186 brouard 10984: /* Main ---------arguments file --------*/
1.126 brouard 10985:
10986: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10987: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10988: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10989: fflush(ficlog);
1.149 brouard 10990: /* goto end; */
10991: exit(70);
1.126 brouard 10992: }
10993:
10994: strcpy(filereso,"o");
1.201 brouard 10995: strcat(filereso,fileresu);
1.126 brouard 10996: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10997: printf("Problem with Output resultfile: %s\n", filereso);
10998: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10999: fflush(ficlog);
11000: goto end;
11001: }
1.278 brouard 11002: /*-------- Rewriting parameter file ----------*/
11003: strcpy(rfileres,"r"); /* "Rparameterfile */
11004: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11005: strcat(rfileres,"."); /* */
11006: strcat(rfileres,optionfilext); /* Other files have txt extension */
11007: if((ficres =fopen(rfileres,"w"))==NULL) {
11008: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11009: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11010: fflush(ficlog);
11011: goto end;
11012: }
11013: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11014:
1.278 brouard 11015:
1.126 brouard 11016: /* Reads comments: lines beginning with '#' */
11017: numlinepar=0;
1.277 brouard 11018: /* Is it a BOM UTF-8 Windows file? */
11019: /* First parameter line */
1.197 brouard 11020: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11021: noffset=0;
11022: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11023: {
11024: noffset=noffset+3;
11025: printf("# File is an UTF8 Bom.\n"); // 0xBF
11026: }
11027: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
11028: {
11029: noffset=noffset+2;
11030: printf("# File is an UTF16BE BOM file\n");
11031: }
11032: else if( line[0] == 0 && line[1] == 0)
11033: {
11034: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11035: noffset=noffset+4;
11036: printf("# File is an UTF16BE BOM file\n");
11037: }
11038: } else{
11039: ;/*printf(" Not a BOM file\n");*/
11040: }
11041:
1.197 brouard 11042: /* If line starts with a # it is a comment */
1.277 brouard 11043: if (line[noffset] == '#') {
1.197 brouard 11044: numlinepar++;
11045: fputs(line,stdout);
11046: fputs(line,ficparo);
1.278 brouard 11047: fputs(line,ficres);
1.197 brouard 11048: fputs(line,ficlog);
11049: continue;
11050: }else
11051: break;
11052: }
11053: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11054: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11055: if (num_filled != 5) {
11056: printf("Should be 5 parameters\n");
1.283 brouard 11057: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11058: }
1.126 brouard 11059: numlinepar++;
1.197 brouard 11060: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11061: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11062: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11063: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11064: }
11065: /* Second parameter line */
11066: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11067: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11068: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11069: if (line[0] == '#') {
11070: numlinepar++;
1.283 brouard 11071: printf("%s",line);
11072: fprintf(ficres,"%s",line);
11073: fprintf(ficparo,"%s",line);
11074: fprintf(ficlog,"%s",line);
1.197 brouard 11075: continue;
11076: }else
11077: break;
11078: }
1.223 brouard 11079: 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", \
11080: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11081: if (num_filled != 11) {
11082: 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 11083: printf("but line=%s\n",line);
1.283 brouard 11084: 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");
11085: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11086: }
1.286 brouard 11087: if( lastpass > maxwav){
11088: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11089: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11090: fflush(ficlog);
11091: goto end;
11092: }
11093: 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 11094: 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 11095: 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 11096: 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 11097: }
1.203 brouard 11098: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11099: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11100: /* Third parameter line */
11101: while(fgets(line, MAXLINE, ficpar)) {
11102: /* If line starts with a # it is a comment */
11103: if (line[0] == '#') {
11104: numlinepar++;
1.283 brouard 11105: printf("%s",line);
11106: fprintf(ficres,"%s",line);
11107: fprintf(ficparo,"%s",line);
11108: fprintf(ficlog,"%s",line);
1.197 brouard 11109: continue;
11110: }else
11111: break;
11112: }
1.201 brouard 11113: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11114: if (num_filled != 1){
11115: printf("ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
11116: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
1.197 brouard 11117: model[0]='\0';
11118: goto end;
11119: }
11120: else{
11121: if (model[0]=='+'){
11122: for(i=1; i<=strlen(model);i++)
11123: modeltemp[i-1]=model[i];
1.201 brouard 11124: strcpy(model,modeltemp);
1.197 brouard 11125: }
11126: }
1.199 brouard 11127: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11128: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11129: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11130: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11131: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11132: }
11133: /* 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); */
11134: /* numlinepar=numlinepar+3; /\* In general *\/ */
11135: /* 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 11136: /* 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); */
11137: /* 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 11138: fflush(ficlog);
1.190 brouard 11139: /* if(model[0]=='#'|| model[0]== '\0'){ */
11140: if(model[0]=='#'){
1.279 brouard 11141: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11142: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11143: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11144: if(mle != -1){
1.279 brouard 11145: 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 11146: exit(1);
11147: }
11148: }
1.126 brouard 11149: while((c=getc(ficpar))=='#' && c!= EOF){
11150: ungetc(c,ficpar);
11151: fgets(line, MAXLINE, ficpar);
11152: numlinepar++;
1.195 brouard 11153: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11154: z[0]=line[1];
11155: }
11156: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11157: fputs(line, stdout);
11158: //puts(line);
1.126 brouard 11159: fputs(line,ficparo);
11160: fputs(line,ficlog);
11161: }
11162: ungetc(c,ficpar);
11163:
11164:
1.290 brouard 11165: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11166: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11167: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11168: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11169: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11170: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11171: v1+v2*age+v2*v3 makes cptcovn = 3
11172: */
11173: if (strlen(model)>1)
1.187 brouard 11174: 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 11175: else
1.187 brouard 11176: ncovmodel=2; /* Constant and age */
1.133 brouard 11177: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11178: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11179: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11180: 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);
11181: 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);
11182: fflush(stdout);
11183: fclose (ficlog);
11184: goto end;
11185: }
1.126 brouard 11186: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11187: delti=delti3[1][1];
11188: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11189: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11190: /* We could also provide initial parameters values giving by simple logistic regression
11191: * only one way, that is without matrix product. We will have nlstate maximizations */
11192: /* for(i=1;i<nlstate;i++){ */
11193: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11194: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11195: /* } */
1.126 brouard 11196: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11197: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11198: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11199: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11200: fclose (ficparo);
11201: fclose (ficlog);
11202: goto end;
11203: exit(0);
1.220 brouard 11204: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11205: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11206: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11207: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11208: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11209: matcov=matrix(1,npar,1,npar);
1.203 brouard 11210: hess=matrix(1,npar,1,npar);
1.220 brouard 11211: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11212: /* Read guessed parameters */
1.126 brouard 11213: /* Reads comments: lines beginning with '#' */
11214: while((c=getc(ficpar))=='#' && c!= EOF){
11215: ungetc(c,ficpar);
11216: fgets(line, MAXLINE, ficpar);
11217: numlinepar++;
1.141 brouard 11218: fputs(line,stdout);
1.126 brouard 11219: fputs(line,ficparo);
11220: fputs(line,ficlog);
11221: }
11222: ungetc(c,ficpar);
11223:
11224: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11225: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11226: for(i=1; i <=nlstate; i++){
1.234 brouard 11227: j=0;
1.126 brouard 11228: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11229: if(jj==i) continue;
11230: j++;
1.292 brouard 11231: while((c=getc(ficpar))=='#' && c!= EOF){
11232: ungetc(c,ficpar);
11233: fgets(line, MAXLINE, ficpar);
11234: numlinepar++;
11235: fputs(line,stdout);
11236: fputs(line,ficparo);
11237: fputs(line,ficlog);
11238: }
11239: ungetc(c,ficpar);
1.234 brouard 11240: fscanf(ficpar,"%1d%1d",&i1,&j1);
11241: if ((i1 != i) || (j1 != jj)){
11242: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11243: It might be a problem of design; if ncovcol and the model are correct\n \
11244: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11245: exit(1);
11246: }
11247: fprintf(ficparo,"%1d%1d",i1,j1);
11248: if(mle==1)
11249: printf("%1d%1d",i,jj);
11250: fprintf(ficlog,"%1d%1d",i,jj);
11251: for(k=1; k<=ncovmodel;k++){
11252: fscanf(ficpar," %lf",¶m[i][j][k]);
11253: if(mle==1){
11254: printf(" %lf",param[i][j][k]);
11255: fprintf(ficlog," %lf",param[i][j][k]);
11256: }
11257: else
11258: fprintf(ficlog," %lf",param[i][j][k]);
11259: fprintf(ficparo," %lf",param[i][j][k]);
11260: }
11261: fscanf(ficpar,"\n");
11262: numlinepar++;
11263: if(mle==1)
11264: printf("\n");
11265: fprintf(ficlog,"\n");
11266: fprintf(ficparo,"\n");
1.126 brouard 11267: }
11268: }
11269: fflush(ficlog);
1.234 brouard 11270:
1.251 brouard 11271: /* Reads parameters values */
1.126 brouard 11272: p=param[1][1];
1.251 brouard 11273: pstart=paramstart[1][1];
1.126 brouard 11274:
11275: /* Reads comments: lines beginning with '#' */
11276: while((c=getc(ficpar))=='#' && c!= EOF){
11277: ungetc(c,ficpar);
11278: fgets(line, MAXLINE, ficpar);
11279: numlinepar++;
1.141 brouard 11280: fputs(line,stdout);
1.126 brouard 11281: fputs(line,ficparo);
11282: fputs(line,ficlog);
11283: }
11284: ungetc(c,ficpar);
11285:
11286: for(i=1; i <=nlstate; i++){
11287: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11288: fscanf(ficpar,"%1d%1d",&i1,&j1);
11289: if ( (i1-i) * (j1-j) != 0){
11290: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11291: exit(1);
11292: }
11293: printf("%1d%1d",i,j);
11294: fprintf(ficparo,"%1d%1d",i1,j1);
11295: fprintf(ficlog,"%1d%1d",i1,j1);
11296: for(k=1; k<=ncovmodel;k++){
11297: fscanf(ficpar,"%le",&delti3[i][j][k]);
11298: printf(" %le",delti3[i][j][k]);
11299: fprintf(ficparo," %le",delti3[i][j][k]);
11300: fprintf(ficlog," %le",delti3[i][j][k]);
11301: }
11302: fscanf(ficpar,"\n");
11303: numlinepar++;
11304: printf("\n");
11305: fprintf(ficparo,"\n");
11306: fprintf(ficlog,"\n");
1.126 brouard 11307: }
11308: }
11309: fflush(ficlog);
1.234 brouard 11310:
1.145 brouard 11311: /* Reads covariance matrix */
1.126 brouard 11312: delti=delti3[1][1];
1.220 brouard 11313:
11314:
1.126 brouard 11315: /* 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 11316:
1.126 brouard 11317: /* Reads comments: lines beginning with '#' */
11318: while((c=getc(ficpar))=='#' && c!= EOF){
11319: ungetc(c,ficpar);
11320: fgets(line, MAXLINE, ficpar);
11321: numlinepar++;
1.141 brouard 11322: fputs(line,stdout);
1.126 brouard 11323: fputs(line,ficparo);
11324: fputs(line,ficlog);
11325: }
11326: ungetc(c,ficpar);
1.220 brouard 11327:
1.126 brouard 11328: matcov=matrix(1,npar,1,npar);
1.203 brouard 11329: hess=matrix(1,npar,1,npar);
1.131 brouard 11330: for(i=1; i <=npar; i++)
11331: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11332:
1.194 brouard 11333: /* Scans npar lines */
1.126 brouard 11334: for(i=1; i <=npar; i++){
1.226 brouard 11335: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11336: if(count != 3){
1.226 brouard 11337: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11338: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11339: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11340: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11341: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11342: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11343: exit(1);
1.220 brouard 11344: }else{
1.226 brouard 11345: if(mle==1)
11346: printf("%1d%1d%d",i1,j1,jk);
11347: }
11348: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11349: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11350: for(j=1; j <=i; j++){
1.226 brouard 11351: fscanf(ficpar," %le",&matcov[i][j]);
11352: if(mle==1){
11353: printf(" %.5le",matcov[i][j]);
11354: }
11355: fprintf(ficlog," %.5le",matcov[i][j]);
11356: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11357: }
11358: fscanf(ficpar,"\n");
11359: numlinepar++;
11360: if(mle==1)
1.220 brouard 11361: printf("\n");
1.126 brouard 11362: fprintf(ficlog,"\n");
11363: fprintf(ficparo,"\n");
11364: }
1.194 brouard 11365: /* End of read covariance matrix npar lines */
1.126 brouard 11366: for(i=1; i <=npar; i++)
11367: for(j=i+1;j<=npar;j++)
1.226 brouard 11368: matcov[i][j]=matcov[j][i];
1.126 brouard 11369:
11370: if(mle==1)
11371: printf("\n");
11372: fprintf(ficlog,"\n");
11373:
11374: fflush(ficlog);
11375:
11376: } /* End of mle != -3 */
1.218 brouard 11377:
1.186 brouard 11378: /* Main data
11379: */
1.290 brouard 11380: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11381: /* num=lvector(1,n); */
11382: /* moisnais=vector(1,n); */
11383: /* annais=vector(1,n); */
11384: /* moisdc=vector(1,n); */
11385: /* andc=vector(1,n); */
11386: /* weight=vector(1,n); */
11387: /* agedc=vector(1,n); */
11388: /* cod=ivector(1,n); */
11389: /* for(i=1;i<=n;i++){ */
11390: num=lvector(firstobs,lastobs);
11391: moisnais=vector(firstobs,lastobs);
11392: annais=vector(firstobs,lastobs);
11393: moisdc=vector(firstobs,lastobs);
11394: andc=vector(firstobs,lastobs);
11395: weight=vector(firstobs,lastobs);
11396: agedc=vector(firstobs,lastobs);
11397: cod=ivector(firstobs,lastobs);
11398: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11399: num[i]=0;
11400: moisnais[i]=0;
11401: annais[i]=0;
11402: moisdc[i]=0;
11403: andc[i]=0;
11404: agedc[i]=0;
11405: cod[i]=0;
11406: weight[i]=1.0; /* Equal weights, 1 by default */
11407: }
1.290 brouard 11408: mint=matrix(1,maxwav,firstobs,lastobs);
11409: anint=matrix(1,maxwav,firstobs,lastobs);
11410: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11411: tab=ivector(1,NCOVMAX);
1.144 brouard 11412: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11413: 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 11414:
1.136 brouard 11415: /* Reads data from file datafile */
11416: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11417: goto end;
11418:
11419: /* Calculation of the number of parameters from char model */
1.234 brouard 11420: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11421: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11422: k=3 V4 Tvar[k=3]= 4 (from V4)
11423: k=2 V1 Tvar[k=2]= 1 (from V1)
11424: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11425: */
11426:
11427: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11428: TvarsDind=ivector(1,NCOVMAX); /* */
11429: TvarsD=ivector(1,NCOVMAX); /* */
11430: TvarsQind=ivector(1,NCOVMAX); /* */
11431: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11432: TvarF=ivector(1,NCOVMAX); /* */
11433: TvarFind=ivector(1,NCOVMAX); /* */
11434: TvarV=ivector(1,NCOVMAX); /* */
11435: TvarVind=ivector(1,NCOVMAX); /* */
11436: TvarA=ivector(1,NCOVMAX); /* */
11437: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11438: TvarFD=ivector(1,NCOVMAX); /* */
11439: TvarFDind=ivector(1,NCOVMAX); /* */
11440: TvarFQ=ivector(1,NCOVMAX); /* */
11441: TvarFQind=ivector(1,NCOVMAX); /* */
11442: TvarVD=ivector(1,NCOVMAX); /* */
11443: TvarVDind=ivector(1,NCOVMAX); /* */
11444: TvarVQ=ivector(1,NCOVMAX); /* */
11445: TvarVQind=ivector(1,NCOVMAX); /* */
11446:
1.230 brouard 11447: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11448: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11449: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11450: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11451: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11452: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11453: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11454: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11455: */
11456: /* For model-covariate k tells which data-covariate to use but
11457: because this model-covariate is a construction we invent a new column
11458: ncovcol + k1
11459: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11460: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11461: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11462: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11463: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11464: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11465: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11466: */
1.145 brouard 11467: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11468: 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 11469: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11470: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11471: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11472: 4 covariates (3 plus signs)
11473: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11474: */
1.230 brouard 11475: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11476: * individual dummy, fixed or varying:
11477: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11478: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11479: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11480: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11481: * Tmodelind[1]@9={9,0,3,2,}*/
11482: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11483: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11484: * individual quantitative, fixed or varying:
11485: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11486: * 3, 1, 0, 0, 0, 0, 0, 0},
11487: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11488: /* Main decodemodel */
11489:
1.187 brouard 11490:
1.223 brouard 11491: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11492: goto end;
11493:
1.137 brouard 11494: if((double)(lastobs-imx)/(double)imx > 1.10){
11495: nbwarn++;
11496: 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);
11497: 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);
11498: }
1.136 brouard 11499: /* if(mle==1){*/
1.137 brouard 11500: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11501: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11502: }
11503:
11504: /*-calculation of age at interview from date of interview and age at death -*/
11505: agev=matrix(1,maxwav,1,imx);
11506:
11507: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11508: goto end;
11509:
1.126 brouard 11510:
1.136 brouard 11511: agegomp=(int)agemin;
1.290 brouard 11512: free_vector(moisnais,firstobs,lastobs);
11513: free_vector(annais,firstobs,lastobs);
1.126 brouard 11514: /* free_matrix(mint,1,maxwav,1,n);
11515: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11516: /* free_vector(moisdc,1,n); */
11517: /* free_vector(andc,1,n); */
1.145 brouard 11518: /* */
11519:
1.126 brouard 11520: wav=ivector(1,imx);
1.214 brouard 11521: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11522: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11523: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11524: 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.*/
11525: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11526: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11527:
11528: /* Concatenates waves */
1.214 brouard 11529: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11530: Death is a valid wave (if date is known).
11531: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11532: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11533: and mw[mi+1][i]. dh depends on stepm.
11534: */
11535:
1.126 brouard 11536: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11537: /* Concatenates waves */
1.145 brouard 11538:
1.290 brouard 11539: free_vector(moisdc,firstobs,lastobs);
11540: free_vector(andc,firstobs,lastobs);
1.215 brouard 11541:
1.126 brouard 11542: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11543: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11544: ncodemax[1]=1;
1.145 brouard 11545: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11546: cptcoveff=0;
1.220 brouard 11547: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11548: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11549: }
11550:
11551: ncovcombmax=pow(2,cptcoveff);
11552: invalidvarcomb=ivector(1, ncovcombmax);
11553: for(i=1;i<ncovcombmax;i++)
11554: invalidvarcomb[i]=0;
11555:
1.211 brouard 11556: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11557: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11558: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11559:
1.200 brouard 11560: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11561: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11562: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11563: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11564: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11565: * (currently 0 or 1) in the data.
11566: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11567: * corresponding modality (h,j).
11568: */
11569:
1.145 brouard 11570: h=0;
11571: /*if (cptcovn > 0) */
1.126 brouard 11572: m=pow(2,cptcoveff);
11573:
1.144 brouard 11574: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11575: * For k=4 covariates, h goes from 1 to m=2**k
11576: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11577: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11578: * h\k 1 2 3 4
1.143 brouard 11579: *______________________________
11580: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11581: * 2 2 1 1 1
11582: * 3 i=2 1 2 1 1
11583: * 4 2 2 1 1
11584: * 5 i=3 1 i=2 1 2 1
11585: * 6 2 1 2 1
11586: * 7 i=4 1 2 2 1
11587: * 8 2 2 2 1
1.197 brouard 11588: * 9 i=5 1 i=3 1 i=2 1 2
11589: * 10 2 1 1 2
11590: * 11 i=6 1 2 1 2
11591: * 12 2 2 1 2
11592: * 13 i=7 1 i=4 1 2 2
11593: * 14 2 1 2 2
11594: * 15 i=8 1 2 2 2
11595: * 16 2 2 2 2
1.143 brouard 11596: */
1.212 brouard 11597: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11598: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11599: * and the value of each covariate?
11600: * V1=1, V2=1, V3=2, V4=1 ?
11601: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11602: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11603: * In order to get the real value in the data, we use nbcode
11604: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11605: * We are keeping this crazy system in order to be able (in the future?)
11606: * to have more than 2 values (0 or 1) for a covariate.
11607: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11608: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11609: * bbbbbbbb
11610: * 76543210
11611: * h-1 00000101 (6-1=5)
1.219 brouard 11612: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11613: * &
11614: * 1 00000001 (1)
1.219 brouard 11615: * 00000000 = 1 & ((h-1) >> (k-1))
11616: * +1= 00000001 =1
1.211 brouard 11617: *
11618: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11619: * h' 1101 =2^3+2^2+0x2^1+2^0
11620: * >>k' 11
11621: * & 00000001
11622: * = 00000001
11623: * +1 = 00000010=2 = codtabm(14,3)
11624: * Reverse h=6 and m=16?
11625: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11626: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11627: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11628: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11629: * V3=decodtabm(14,3,2**4)=2
11630: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11631: *(h-1) >> (j-1) 0011 =13 >> 2
11632: * &1 000000001
11633: * = 000000001
11634: * +1= 000000010 =2
11635: * 2211
11636: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11637: * V3=2
1.220 brouard 11638: * codtabm and decodtabm are identical
1.211 brouard 11639: */
11640:
1.145 brouard 11641:
11642: free_ivector(Ndum,-1,NCOVMAX);
11643:
11644:
1.126 brouard 11645:
1.186 brouard 11646: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11647: strcpy(optionfilegnuplot,optionfilefiname);
11648: if(mle==-3)
1.201 brouard 11649: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11650: strcat(optionfilegnuplot,".gp");
11651:
11652: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11653: printf("Problem with file %s",optionfilegnuplot);
11654: }
11655: else{
1.204 brouard 11656: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11657: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11658: //fprintf(ficgp,"set missing 'NaNq'\n");
11659: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11660: }
11661: /* fclose(ficgp);*/
1.186 brouard 11662:
11663:
11664: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11665:
11666: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11667: if(mle==-3)
1.201 brouard 11668: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11669: strcat(optionfilehtm,".htm");
11670: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11671: printf("Problem with %s \n",optionfilehtm);
11672: exit(0);
1.126 brouard 11673: }
11674:
11675: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11676: strcat(optionfilehtmcov,"-cov.htm");
11677: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11678: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11679: }
11680: else{
11681: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11682: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11683: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11684: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11685: }
11686:
1.213 brouard 11687: 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 11688: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11689: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11690: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11691: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11692: \n\
11693: <hr size=\"2\" color=\"#EC5E5E\">\
11694: <ul><li><h4>Parameter files</h4>\n\
11695: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11696: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11697: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11698: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11699: - Date and time at start: %s</ul>\n",\
11700: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11701: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11702: fileres,fileres,\
11703: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11704: fflush(fichtm);
11705:
11706: strcpy(pathr,path);
11707: strcat(pathr,optionfilefiname);
1.184 brouard 11708: #ifdef WIN32
11709: _chdir(optionfilefiname); /* Move to directory named optionfile */
11710: #else
1.126 brouard 11711: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11712: #endif
11713:
1.126 brouard 11714:
1.220 brouard 11715: /* Calculates basic frequencies. Computes observed prevalence at single age
11716: and for any valid combination of covariates
1.126 brouard 11717: and prints on file fileres'p'. */
1.251 brouard 11718: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11719: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11720:
11721: fprintf(fichtm,"\n");
1.286 brouard 11722: 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 11723: ftol, stepm);
11724: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11725: ncurrv=1;
11726: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11727: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11728: ncurrv=i;
11729: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11730: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 11731: ncurrv=i;
11732: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11733: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 11734: ncurrv=i;
11735: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11736: 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", \
11737: nlstate, ndeath, maxwav, mle, weightopt);
11738:
11739: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11740: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11741:
11742:
11743: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11744: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11745: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11746: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11747: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11748: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11749: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11750: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11751: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11752:
1.126 brouard 11753: /* For Powell, parameters are in a vector p[] starting at p[1]
11754: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11755: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11756:
11757: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11758: /* For mortality only */
1.126 brouard 11759: if (mle==-3){
1.136 brouard 11760: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11761: for(i=1;i<=NDIM;i++)
11762: for(j=1;j<=NDIM;j++)
11763: ximort[i][j]=0.;
1.186 brouard 11764: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 11765: cens=ivector(firstobs,lastobs);
11766: ageexmed=vector(firstobs,lastobs);
11767: agecens=vector(firstobs,lastobs);
11768: dcwave=ivector(firstobs,lastobs);
1.223 brouard 11769:
1.126 brouard 11770: for (i=1; i<=imx; i++){
11771: dcwave[i]=-1;
11772: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11773: if (s[m][i]>nlstate) {
11774: dcwave[i]=m;
11775: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11776: break;
11777: }
1.126 brouard 11778: }
1.226 brouard 11779:
1.126 brouard 11780: for (i=1; i<=imx; i++) {
11781: if (wav[i]>0){
1.226 brouard 11782: ageexmed[i]=agev[mw[1][i]][i];
11783: j=wav[i];
11784: agecens[i]=1.;
11785:
11786: if (ageexmed[i]> 1 && wav[i] > 0){
11787: agecens[i]=agev[mw[j][i]][i];
11788: cens[i]= 1;
11789: }else if (ageexmed[i]< 1)
11790: cens[i]= -1;
11791: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11792: cens[i]=0 ;
1.126 brouard 11793: }
11794: else cens[i]=-1;
11795: }
11796:
11797: for (i=1;i<=NDIM;i++) {
11798: for (j=1;j<=NDIM;j++)
1.226 brouard 11799: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11800: }
11801:
1.145 brouard 11802: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11803: /*printf("%lf %lf", p[1], p[2]);*/
11804:
11805:
1.136 brouard 11806: #ifdef GSL
11807: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11808: #else
1.126 brouard 11809: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11810: #endif
1.201 brouard 11811: strcpy(filerespow,"POW-MORT_");
11812: strcat(filerespow,fileresu);
1.126 brouard 11813: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11814: printf("Problem with resultfile: %s\n", filerespow);
11815: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11816: }
1.136 brouard 11817: #ifdef GSL
11818: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11819: #else
1.126 brouard 11820: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11821: #endif
1.126 brouard 11822: /* for (i=1;i<=nlstate;i++)
11823: for(j=1;j<=nlstate+ndeath;j++)
11824: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11825: */
11826: fprintf(ficrespow,"\n");
1.136 brouard 11827: #ifdef GSL
11828: /* gsl starts here */
11829: T = gsl_multimin_fminimizer_nmsimplex;
11830: gsl_multimin_fminimizer *sfm = NULL;
11831: gsl_vector *ss, *x;
11832: gsl_multimin_function minex_func;
11833:
11834: /* Initial vertex size vector */
11835: ss = gsl_vector_alloc (NDIM);
11836:
11837: if (ss == NULL){
11838: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11839: }
11840: /* Set all step sizes to 1 */
11841: gsl_vector_set_all (ss, 0.001);
11842:
11843: /* Starting point */
1.126 brouard 11844:
1.136 brouard 11845: x = gsl_vector_alloc (NDIM);
11846:
11847: if (x == NULL){
11848: gsl_vector_free(ss);
11849: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11850: }
11851:
11852: /* Initialize method and iterate */
11853: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11854: /* gsl_vector_set(x, 0, 0.0268); */
11855: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11856: gsl_vector_set(x, 0, p[1]);
11857: gsl_vector_set(x, 1, p[2]);
11858:
11859: minex_func.f = &gompertz_f;
11860: minex_func.n = NDIM;
11861: minex_func.params = (void *)&p; /* ??? */
11862:
11863: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11864: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11865:
11866: printf("Iterations beginning .....\n\n");
11867: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11868:
11869: iteri=0;
11870: while (rval == GSL_CONTINUE){
11871: iteri++;
11872: status = gsl_multimin_fminimizer_iterate(sfm);
11873:
11874: if (status) printf("error: %s\n", gsl_strerror (status));
11875: fflush(0);
11876:
11877: if (status)
11878: break;
11879:
11880: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11881: ssval = gsl_multimin_fminimizer_size (sfm);
11882:
11883: if (rval == GSL_SUCCESS)
11884: printf ("converged to a local maximum at\n");
11885:
11886: printf("%5d ", iteri);
11887: for (it = 0; it < NDIM; it++){
11888: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11889: }
11890: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11891: }
11892:
11893: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11894:
11895: gsl_vector_free(x); /* initial values */
11896: gsl_vector_free(ss); /* inital step size */
11897: for (it=0; it<NDIM; it++){
11898: p[it+1]=gsl_vector_get(sfm->x,it);
11899: fprintf(ficrespow," %.12lf", p[it]);
11900: }
11901: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11902: #endif
11903: #ifdef POWELL
11904: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11905: #endif
1.126 brouard 11906: fclose(ficrespow);
11907:
1.203 brouard 11908: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11909:
11910: for(i=1; i <=NDIM; i++)
11911: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11912: matcov[i][j]=matcov[j][i];
1.126 brouard 11913:
11914: printf("\nCovariance matrix\n ");
1.203 brouard 11915: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11916: for(i=1; i <=NDIM; i++) {
11917: for(j=1;j<=NDIM;j++){
1.220 brouard 11918: printf("%f ",matcov[i][j]);
11919: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11920: }
1.203 brouard 11921: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11922: }
11923:
11924: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11925: for (i=1;i<=NDIM;i++) {
1.126 brouard 11926: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11927: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11928: }
1.126 brouard 11929: lsurv=vector(1,AGESUP);
11930: lpop=vector(1,AGESUP);
11931: tpop=vector(1,AGESUP);
11932: lsurv[agegomp]=100000;
11933:
11934: for (k=agegomp;k<=AGESUP;k++) {
11935: agemortsup=k;
11936: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11937: }
11938:
11939: for (k=agegomp;k<agemortsup;k++)
11940: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11941:
11942: for (k=agegomp;k<agemortsup;k++){
11943: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11944: sumlpop=sumlpop+lpop[k];
11945: }
11946:
11947: tpop[agegomp]=sumlpop;
11948: for (k=agegomp;k<(agemortsup-3);k++){
11949: /* tpop[k+1]=2;*/
11950: tpop[k+1]=tpop[k]-lpop[k];
11951: }
11952:
11953:
11954: printf("\nAge lx qx dx Lx Tx e(x)\n");
11955: for (k=agegomp;k<(agemortsup-2);k++)
11956: 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]);
11957:
11958:
11959: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11960: ageminpar=50;
11961: agemaxpar=100;
1.194 brouard 11962: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11963: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11964: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11965: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11966: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11967: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11968: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11969: }else{
11970: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11971: 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 11972: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11973: }
1.201 brouard 11974: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11975: stepm, weightopt,\
11976: model,imx,p,matcov,agemortsup);
11977:
11978: free_vector(lsurv,1,AGESUP);
11979: free_vector(lpop,1,AGESUP);
11980: free_vector(tpop,1,AGESUP);
1.220 brouard 11981: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 11982: free_ivector(dcwave,firstobs,lastobs);
11983: free_vector(agecens,firstobs,lastobs);
11984: free_vector(ageexmed,firstobs,lastobs);
11985: free_ivector(cens,firstobs,lastobs);
1.220 brouard 11986: #ifdef GSL
1.136 brouard 11987: #endif
1.186 brouard 11988: } /* Endof if mle==-3 mortality only */
1.205 brouard 11989: /* Standard */
11990: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11991: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11992: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11993: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11994: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11995: for (k=1; k<=npar;k++)
11996: printf(" %d %8.5f",k,p[k]);
11997: printf("\n");
1.205 brouard 11998: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11999: /* mlikeli uses func not funcone */
1.247 brouard 12000: /* for(i=1;i<nlstate;i++){ */
12001: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12002: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12003: /* } */
1.205 brouard 12004: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12005: }
12006: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12007: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12008: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12009: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12010: }
12011: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12012: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12013: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12014: for (k=1; k<=npar;k++)
12015: printf(" %d %8.5f",k,p[k]);
12016: printf("\n");
12017:
12018: /*--------- results files --------------*/
1.283 brouard 12019: /* 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 12020:
12021:
12022: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12023: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12024: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12025: for(i=1,jk=1; i <=nlstate; i++){
12026: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12027: if (k != i) {
12028: printf("%d%d ",i,k);
12029: fprintf(ficlog,"%d%d ",i,k);
12030: fprintf(ficres,"%1d%1d ",i,k);
12031: for(j=1; j <=ncovmodel; j++){
12032: printf("%12.7f ",p[jk]);
12033: fprintf(ficlog,"%12.7f ",p[jk]);
12034: fprintf(ficres,"%12.7f ",p[jk]);
12035: jk++;
12036: }
12037: printf("\n");
12038: fprintf(ficlog,"\n");
12039: fprintf(ficres,"\n");
12040: }
1.126 brouard 12041: }
12042: }
1.203 brouard 12043: if(mle != 0){
12044: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12045: ftolhess=ftol; /* Usually correct */
1.203 brouard 12046: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12047: 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");
12048: 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");
12049: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12050: for(k=1; k <=(nlstate+ndeath); k++){
12051: if (k != i) {
12052: printf("%d%d ",i,k);
12053: fprintf(ficlog,"%d%d ",i,k);
12054: for(j=1; j <=ncovmodel; j++){
12055: 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]));
12056: 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]));
12057: jk++;
12058: }
12059: printf("\n");
12060: fprintf(ficlog,"\n");
12061: }
12062: }
1.193 brouard 12063: }
1.203 brouard 12064: } /* end of hesscov and Wald tests */
1.225 brouard 12065:
1.203 brouard 12066: /* */
1.126 brouard 12067: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12068: printf("# Scales (for hessian or gradient estimation)\n");
12069: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12070: for(i=1,jk=1; i <=nlstate; i++){
12071: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12072: if (j!=i) {
12073: fprintf(ficres,"%1d%1d",i,j);
12074: printf("%1d%1d",i,j);
12075: fprintf(ficlog,"%1d%1d",i,j);
12076: for(k=1; k<=ncovmodel;k++){
12077: printf(" %.5e",delti[jk]);
12078: fprintf(ficlog," %.5e",delti[jk]);
12079: fprintf(ficres," %.5e",delti[jk]);
12080: jk++;
12081: }
12082: printf("\n");
12083: fprintf(ficlog,"\n");
12084: fprintf(ficres,"\n");
12085: }
1.126 brouard 12086: }
12087: }
12088:
12089: 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 12090: if(mle >= 1) /* To big for the screen */
1.126 brouard 12091: 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");
12092: 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");
12093: /* # 121 Var(a12)\n\ */
12094: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12095: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12096: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12097: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12098: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12099: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12100: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12101:
12102:
12103: /* Just to have a covariance matrix which will be more understandable
12104: even is we still don't want to manage dictionary of variables
12105: */
12106: for(itimes=1;itimes<=2;itimes++){
12107: jj=0;
12108: for(i=1; i <=nlstate; i++){
1.225 brouard 12109: for(j=1; j <=nlstate+ndeath; j++){
12110: if(j==i) continue;
12111: for(k=1; k<=ncovmodel;k++){
12112: jj++;
12113: ca[0]= k+'a'-1;ca[1]='\0';
12114: if(itimes==1){
12115: if(mle>=1)
12116: printf("#%1d%1d%d",i,j,k);
12117: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12118: fprintf(ficres,"#%1d%1d%d",i,j,k);
12119: }else{
12120: if(mle>=1)
12121: printf("%1d%1d%d",i,j,k);
12122: fprintf(ficlog,"%1d%1d%d",i,j,k);
12123: fprintf(ficres,"%1d%1d%d",i,j,k);
12124: }
12125: ll=0;
12126: for(li=1;li <=nlstate; li++){
12127: for(lj=1;lj <=nlstate+ndeath; lj++){
12128: if(lj==li) continue;
12129: for(lk=1;lk<=ncovmodel;lk++){
12130: ll++;
12131: if(ll<=jj){
12132: cb[0]= lk +'a'-1;cb[1]='\0';
12133: if(ll<jj){
12134: if(itimes==1){
12135: if(mle>=1)
12136: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12137: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12138: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12139: }else{
12140: if(mle>=1)
12141: printf(" %.5e",matcov[jj][ll]);
12142: fprintf(ficlog," %.5e",matcov[jj][ll]);
12143: fprintf(ficres," %.5e",matcov[jj][ll]);
12144: }
12145: }else{
12146: if(itimes==1){
12147: if(mle>=1)
12148: printf(" Var(%s%1d%1d)",ca,i,j);
12149: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12150: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12151: }else{
12152: if(mle>=1)
12153: printf(" %.7e",matcov[jj][ll]);
12154: fprintf(ficlog," %.7e",matcov[jj][ll]);
12155: fprintf(ficres," %.7e",matcov[jj][ll]);
12156: }
12157: }
12158: }
12159: } /* end lk */
12160: } /* end lj */
12161: } /* end li */
12162: if(mle>=1)
12163: printf("\n");
12164: fprintf(ficlog,"\n");
12165: fprintf(ficres,"\n");
12166: numlinepar++;
12167: } /* end k*/
12168: } /*end j */
1.126 brouard 12169: } /* end i */
12170: } /* end itimes */
12171:
12172: fflush(ficlog);
12173: fflush(ficres);
1.225 brouard 12174: while(fgets(line, MAXLINE, ficpar)) {
12175: /* If line starts with a # it is a comment */
12176: if (line[0] == '#') {
12177: numlinepar++;
12178: fputs(line,stdout);
12179: fputs(line,ficparo);
12180: fputs(line,ficlog);
12181: continue;
12182: }else
12183: break;
12184: }
12185:
1.209 brouard 12186: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12187: /* ungetc(c,ficpar); */
12188: /* fgets(line, MAXLINE, ficpar); */
12189: /* fputs(line,stdout); */
12190: /* fputs(line,ficparo); */
12191: /* } */
12192: /* ungetc(c,ficpar); */
1.126 brouard 12193:
12194: estepm=0;
1.209 brouard 12195: 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 12196:
12197: if (num_filled != 6) {
12198: 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);
12199: 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);
12200: goto end;
12201: }
12202: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12203: }
12204: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12205: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12206:
1.209 brouard 12207: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12208: if (estepm==0 || estepm < stepm) estepm=stepm;
12209: if (fage <= 2) {
12210: bage = ageminpar;
12211: fage = agemaxpar;
12212: }
12213:
12214: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12215: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12216: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12217:
1.186 brouard 12218: /* Other stuffs, more or less useful */
1.254 brouard 12219: while(fgets(line, MAXLINE, ficpar)) {
12220: /* If line starts with a # it is a comment */
12221: if (line[0] == '#') {
12222: numlinepar++;
12223: fputs(line,stdout);
12224: fputs(line,ficparo);
12225: fputs(line,ficlog);
12226: continue;
12227: }else
12228: break;
12229: }
12230:
12231: 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){
12232:
12233: if (num_filled != 7) {
12234: 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);
12235: 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);
12236: goto end;
12237: }
12238: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12239: 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);
12240: 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);
12241: 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 12242: }
1.254 brouard 12243:
12244: while(fgets(line, MAXLINE, ficpar)) {
12245: /* If line starts with a # it is a comment */
12246: if (line[0] == '#') {
12247: numlinepar++;
12248: fputs(line,stdout);
12249: fputs(line,ficparo);
12250: fputs(line,ficlog);
12251: continue;
12252: }else
12253: break;
1.126 brouard 12254: }
12255:
12256:
12257: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12258: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12259:
1.254 brouard 12260: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12261: if (num_filled != 1) {
12262: 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);
12263: 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);
12264: goto end;
12265: }
12266: printf("pop_based=%d\n",popbased);
12267: fprintf(ficlog,"pop_based=%d\n",popbased);
12268: fprintf(ficparo,"pop_based=%d\n",popbased);
12269: fprintf(ficres,"pop_based=%d\n",popbased);
12270: }
12271:
1.258 brouard 12272: /* Results */
12273: nresult=0;
12274: do{
12275: if(!fgets(line, MAXLINE, ficpar)){
12276: endishere=1;
12277: parameterline=14;
12278: }else if (line[0] == '#') {
12279: /* If line starts with a # it is a comment */
1.254 brouard 12280: numlinepar++;
12281: fputs(line,stdout);
12282: fputs(line,ficparo);
12283: fputs(line,ficlog);
12284: continue;
1.258 brouard 12285: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12286: parameterline=11;
12287: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12288: parameterline=12;
12289: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12290: parameterline=13;
12291: else{
12292: parameterline=14;
1.254 brouard 12293: }
1.258 brouard 12294: switch (parameterline){
12295: case 11:
12296: 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){
12297: if (num_filled != 8) {
12298: 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);
12299: 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);
12300: goto end;
12301: }
12302: 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);
12303: 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);
12304: 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);
12305: 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);
12306: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12307: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12308: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12309:
1.258 brouard 12310: }
1.254 brouard 12311: break;
1.258 brouard 12312: case 12:
12313: /*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);*/
12314: 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){
12315: if (num_filled != 8) {
1.262 brouard 12316: 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);
12317: 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 12318: goto end;
12319: }
12320: 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);
12321: 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);
12322: 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);
12323: 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);
12324: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12325: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12326: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12327: }
1.230 brouard 12328: break;
1.258 brouard 12329: case 13:
12330: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12331: if (num_filled == 0){
12332: resultline[0]='\0';
12333: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12334: 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);
12335: break;
12336: } else if (num_filled != 1){
12337: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12338: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12339: }
12340: nresult++; /* Sum of resultlines */
12341: printf("Result %d: result=%s\n",nresult, resultline);
12342: if(nresult > MAXRESULTLINES){
12343: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12344: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12345: goto end;
12346: }
12347: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12348: fprintf(ficparo,"result: %s\n",resultline);
12349: fprintf(ficres,"result: %s\n",resultline);
12350: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12351: break;
1.258 brouard 12352: case 14:
1.259 brouard 12353: if(ncovmodel >2 && nresult==0 ){
12354: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12355: goto end;
12356: }
1.259 brouard 12357: break;
1.258 brouard 12358: default:
12359: nresult=1;
12360: decoderesult(".",nresult ); /* No covariate */
12361: }
12362: } /* End switch parameterline */
12363: }while(endishere==0); /* End do */
1.126 brouard 12364:
1.230 brouard 12365: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12366: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12367:
12368: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12369: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12370: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12371: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12372: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12373: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12374: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12375: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12376: }else{
1.270 brouard 12377: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12378: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12379: }
12380: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12381: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12382: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12383:
1.225 brouard 12384: /*------------ free_vector -------------*/
12385: /* chdir(path); */
1.220 brouard 12386:
1.215 brouard 12387: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12388: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12389: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12390: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12391: free_lvector(num,firstobs,lastobs);
12392: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12393: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12394: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12395: fclose(ficparo);
12396: fclose(ficres);
1.220 brouard 12397:
12398:
1.186 brouard 12399: /* Other results (useful)*/
1.220 brouard 12400:
12401:
1.126 brouard 12402: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12403: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12404: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12405: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12406: fclose(ficrespl);
12407:
12408: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12409: /*#include "hpijx.h"*/
12410: hPijx(p, bage, fage);
1.145 brouard 12411: fclose(ficrespij);
1.227 brouard 12412:
1.220 brouard 12413: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12414: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12415: k=1;
1.126 brouard 12416: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12417:
1.269 brouard 12418: /* Prevalence for each covariate combination in probs[age][status][cov] */
12419: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12420: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12421: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12422: for(k=1;k<=ncovcombmax;k++)
12423: probs[i][j][k]=0.;
1.269 brouard 12424: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12425: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12426: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12427: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12428: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12429: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12430: for(k=1;k<=ncovcombmax;k++)
12431: mobaverages[i][j][k]=0.;
1.219 brouard 12432: mobaverage=mobaverages;
12433: if (mobilav!=0) {
1.235 brouard 12434: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12435: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12436: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12437: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12438: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12439: }
1.269 brouard 12440: } else if (mobilavproj !=0) {
1.235 brouard 12441: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12442: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12443: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12444: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12445: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12446: }
1.269 brouard 12447: }else{
12448: printf("Internal error moving average\n");
12449: fflush(stdout);
12450: exit(1);
1.219 brouard 12451: }
12452: }/* end if moving average */
1.227 brouard 12453:
1.126 brouard 12454: /*---------- Forecasting ------------------*/
12455: if(prevfcast==1){
12456: /* if(stepm ==1){*/
1.269 brouard 12457: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12458: }
1.269 brouard 12459:
12460: /* Backcasting */
1.217 brouard 12461: if(backcast==1){
1.219 brouard 12462: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12463: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12464: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12465:
12466: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12467:
12468: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12469:
1.219 brouard 12470: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12471: fclose(ficresplb);
12472:
1.222 brouard 12473: hBijx(p, bage, fage, mobaverage);
12474: fclose(ficrespijb);
1.219 brouard 12475:
1.269 brouard 12476: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12477: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12478: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12479:
12480:
1.269 brouard 12481: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12482: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12483: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12484: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12485: } /* end Backcasting */
1.268 brouard 12486:
1.186 brouard 12487:
12488: /* ------ Other prevalence ratios------------ */
1.126 brouard 12489:
1.215 brouard 12490: free_ivector(wav,1,imx);
12491: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12492: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12493: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12494:
12495:
1.127 brouard 12496: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12497:
1.201 brouard 12498: strcpy(filerese,"E_");
12499: strcat(filerese,fileresu);
1.126 brouard 12500: if((ficreseij=fopen(filerese,"w"))==NULL) {
12501: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12502: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12503: }
1.208 brouard 12504: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12505: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12506:
12507: pstamp(ficreseij);
1.219 brouard 12508:
1.235 brouard 12509: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12510: if (cptcovn < 1){i1=1;}
12511:
12512: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12513: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12514: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12515: continue;
1.219 brouard 12516: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12517: printf("\n#****** ");
1.225 brouard 12518: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12519: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12520: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12521: }
12522: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12523: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12524: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12525: }
12526: fprintf(ficreseij,"******\n");
1.235 brouard 12527: printf("******\n");
1.219 brouard 12528:
12529: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12530: oldm=oldms;savm=savms;
1.235 brouard 12531: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12532:
1.219 brouard 12533: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12534: }
12535: fclose(ficreseij);
1.208 brouard 12536: printf("done evsij\n");fflush(stdout);
12537: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12538:
1.218 brouard 12539:
1.227 brouard 12540: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12541:
1.201 brouard 12542: strcpy(filerest,"T_");
12543: strcat(filerest,fileresu);
1.127 brouard 12544: if((ficrest=fopen(filerest,"w"))==NULL) {
12545: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12546: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12547: }
1.208 brouard 12548: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12549: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12550: strcpy(fileresstde,"STDE_");
12551: strcat(fileresstde,fileresu);
1.126 brouard 12552: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12553: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12554: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12555: }
1.227 brouard 12556: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12557: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12558:
1.201 brouard 12559: strcpy(filerescve,"CVE_");
12560: strcat(filerescve,fileresu);
1.126 brouard 12561: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12562: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12563: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12564: }
1.227 brouard 12565: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12566: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12567:
1.201 brouard 12568: strcpy(fileresv,"V_");
12569: strcat(fileresv,fileresu);
1.126 brouard 12570: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12571: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12572: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12573: }
1.227 brouard 12574: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12575: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12576:
1.235 brouard 12577: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12578: if (cptcovn < 1){i1=1;}
12579:
12580: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12581: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12582: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12583: continue;
1.242 brouard 12584: printf("\n#****** Result for:");
12585: fprintf(ficrest,"\n#****** Result for:");
12586: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12587: for(j=1;j<=cptcoveff;j++){
12588: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12589: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12590: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12591: }
1.235 brouard 12592: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12593: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12594: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12595: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12596: }
1.208 brouard 12597: fprintf(ficrest,"******\n");
1.227 brouard 12598: fprintf(ficlog,"******\n");
12599: printf("******\n");
1.208 brouard 12600:
12601: fprintf(ficresstdeij,"\n#****** ");
12602: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12603: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12604: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12605: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12606: }
1.235 brouard 12607: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12608: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12609: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12610: }
1.208 brouard 12611: fprintf(ficresstdeij,"******\n");
12612: fprintf(ficrescveij,"******\n");
12613:
12614: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12615: /* pstamp(ficresvij); */
1.225 brouard 12616: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12617: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12618: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12619: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12620: }
1.208 brouard 12621: fprintf(ficresvij,"******\n");
12622:
12623: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12624: oldm=oldms;savm=savms;
1.235 brouard 12625: printf(" cvevsij ");
12626: fprintf(ficlog, " cvevsij ");
12627: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12628: printf(" end cvevsij \n ");
12629: fprintf(ficlog, " end cvevsij \n ");
12630:
12631: /*
12632: */
12633: /* goto endfree; */
12634:
12635: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12636: pstamp(ficrest);
12637:
1.269 brouard 12638: epj=vector(1,nlstate+1);
1.208 brouard 12639: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12640: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12641: cptcod= 0; /* To be deleted */
12642: printf("varevsij vpopbased=%d \n",vpopbased);
12643: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12644: 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 12645: 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 ");
12646: if(vpopbased==1)
12647: 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);
12648: else
1.288 brouard 12649: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12650: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12651: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12652: fprintf(ficrest,"\n");
12653: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 12654: printf("Computing age specific forward period (stable) prevalences in each health state \n");
12655: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12656: for(age=bage; age <=fage ;age++){
1.235 brouard 12657: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12658: if (vpopbased==1) {
12659: if(mobilav ==0){
12660: for(i=1; i<=nlstate;i++)
12661: prlim[i][i]=probs[(int)age][i][k];
12662: }else{ /* mobilav */
12663: for(i=1; i<=nlstate;i++)
12664: prlim[i][i]=mobaverage[(int)age][i][k];
12665: }
12666: }
1.219 brouard 12667:
1.227 brouard 12668: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12669: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12670: /* printf(" age %4.0f ",age); */
12671: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12672: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12673: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12674: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12675: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12676: }
12677: epj[nlstate+1] +=epj[j];
12678: }
12679: /* printf(" age %4.0f \n",age); */
1.219 brouard 12680:
1.227 brouard 12681: for(i=1, vepp=0.;i <=nlstate;i++)
12682: for(j=1;j <=nlstate;j++)
12683: vepp += vareij[i][j][(int)age];
12684: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12685: for(j=1;j <=nlstate;j++){
12686: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12687: }
12688: fprintf(ficrest,"\n");
12689: }
1.208 brouard 12690: } /* End vpopbased */
1.269 brouard 12691: free_vector(epj,1,nlstate+1);
1.208 brouard 12692: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12693: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12694: printf("done selection\n");fflush(stdout);
12695: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12696:
1.235 brouard 12697: } /* End k selection */
1.227 brouard 12698:
12699: printf("done State-specific expectancies\n");fflush(stdout);
12700: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12701:
1.288 brouard 12702: /* variance-covariance of forward period prevalence*/
1.269 brouard 12703: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12704:
1.227 brouard 12705:
1.290 brouard 12706: free_vector(weight,firstobs,lastobs);
1.227 brouard 12707: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 12708: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
12709: free_matrix(anint,1,maxwav,firstobs,lastobs);
12710: free_matrix(mint,1,maxwav,firstobs,lastobs);
12711: free_ivector(cod,firstobs,lastobs);
1.227 brouard 12712: free_ivector(tab,1,NCOVMAX);
12713: fclose(ficresstdeij);
12714: fclose(ficrescveij);
12715: fclose(ficresvij);
12716: fclose(ficrest);
12717: fclose(ficpar);
12718:
12719:
1.126 brouard 12720: /*---------- End : free ----------------*/
1.219 brouard 12721: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12722: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12723: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12724: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12725: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12726: } /* mle==-3 arrives here for freeing */
1.227 brouard 12727: /* endfree:*/
12728: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12729: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12730: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 12731: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
12732: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
12733: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
12734: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 12735: free_matrix(matcov,1,npar,1,npar);
12736: free_matrix(hess,1,npar,1,npar);
12737: /*free_vector(delti,1,npar);*/
12738: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12739: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12740: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12741: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12742:
12743: free_ivector(ncodemax,1,NCOVMAX);
12744: free_ivector(ncodemaxwundef,1,NCOVMAX);
12745: free_ivector(Dummy,-1,NCOVMAX);
12746: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12747: free_ivector(DummyV,1,NCOVMAX);
12748: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12749: free_ivector(Typevar,-1,NCOVMAX);
12750: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12751: free_ivector(TvarsQ,1,NCOVMAX);
12752: free_ivector(TvarsQind,1,NCOVMAX);
12753: free_ivector(TvarsD,1,NCOVMAX);
12754: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12755: free_ivector(TvarFD,1,NCOVMAX);
12756: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12757: free_ivector(TvarF,1,NCOVMAX);
12758: free_ivector(TvarFind,1,NCOVMAX);
12759: free_ivector(TvarV,1,NCOVMAX);
12760: free_ivector(TvarVind,1,NCOVMAX);
12761: free_ivector(TvarA,1,NCOVMAX);
12762: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12763: free_ivector(TvarFQ,1,NCOVMAX);
12764: free_ivector(TvarFQind,1,NCOVMAX);
12765: free_ivector(TvarVD,1,NCOVMAX);
12766: free_ivector(TvarVDind,1,NCOVMAX);
12767: free_ivector(TvarVQ,1,NCOVMAX);
12768: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12769: free_ivector(Tvarsel,1,NCOVMAX);
12770: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12771: free_ivector(Tposprod,1,NCOVMAX);
12772: free_ivector(Tprod,1,NCOVMAX);
12773: free_ivector(Tvaraff,1,NCOVMAX);
12774: free_ivector(invalidvarcomb,1,ncovcombmax);
12775: free_ivector(Tage,1,NCOVMAX);
12776: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12777: free_ivector(TmodelInvind,1,NCOVMAX);
12778: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12779:
12780: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12781: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12782: fflush(fichtm);
12783: fflush(ficgp);
12784:
1.227 brouard 12785:
1.126 brouard 12786: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12787: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12788: 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 12789: }else{
12790: printf("End of Imach\n");
12791: fprintf(ficlog,"End of Imach\n");
12792: }
12793: printf("See log file on %s\n",filelog);
12794: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12795: /*(void) gettimeofday(&end_time,&tzp);*/
12796: rend_time = time(NULL);
12797: end_time = *localtime(&rend_time);
12798: /* tml = *localtime(&end_time.tm_sec); */
12799: strcpy(strtend,asctime(&end_time));
1.126 brouard 12800: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12801: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12802: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12803:
1.157 brouard 12804: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12805: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12806: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12807: /* printf("Total time was %d uSec.\n", total_usecs);*/
12808: /* if(fileappend(fichtm,optionfilehtm)){ */
12809: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12810: fclose(fichtm);
12811: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12812: fclose(fichtmcov);
12813: fclose(ficgp);
12814: fclose(ficlog);
12815: /*------ End -----------*/
1.227 brouard 12816:
1.281 brouard 12817:
12818: /* Executes gnuplot */
1.227 brouard 12819:
12820: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12821: #ifdef WIN32
1.227 brouard 12822: if (_chdir(pathcd) != 0)
12823: printf("Can't move to directory %s!\n",path);
12824: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12825: #else
1.227 brouard 12826: if(chdir(pathcd) != 0)
12827: printf("Can't move to directory %s!\n", path);
12828: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12829: #endif
1.126 brouard 12830: printf("Current directory %s!\n",pathcd);
12831: /*strcat(plotcmd,CHARSEPARATOR);*/
12832: sprintf(plotcmd,"gnuplot");
1.157 brouard 12833: #ifdef _WIN32
1.126 brouard 12834: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12835: #endif
12836: if(!stat(plotcmd,&info)){
1.158 brouard 12837: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12838: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12839: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12840: }else
12841: strcpy(pplotcmd,plotcmd);
1.157 brouard 12842: #ifdef __unix
1.126 brouard 12843: strcpy(plotcmd,GNUPLOTPROGRAM);
12844: if(!stat(plotcmd,&info)){
1.158 brouard 12845: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12846: }else
12847: strcpy(pplotcmd,plotcmd);
12848: #endif
12849: }else
12850: strcpy(pplotcmd,plotcmd);
12851:
12852: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12853: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 12854: strcpy(pplotcmd,plotcmd);
1.227 brouard 12855:
1.126 brouard 12856: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 12857: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12858: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12859: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 12860: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 12861: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 12862: strcpy(plotcmd,pplotcmd);
12863: }
1.126 brouard 12864: }
1.158 brouard 12865: printf(" Successful, please wait...");
1.126 brouard 12866: while (z[0] != 'q') {
12867: /* chdir(path); */
1.154 brouard 12868: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12869: scanf("%s",z);
12870: /* if (z[0] == 'c') system("./imach"); */
12871: if (z[0] == 'e') {
1.158 brouard 12872: #ifdef __APPLE__
1.152 brouard 12873: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12874: #elif __linux
12875: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12876: #else
1.152 brouard 12877: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12878: #endif
12879: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12880: system(pplotcmd);
1.126 brouard 12881: }
12882: else if (z[0] == 'g') system(plotcmd);
12883: else if (z[0] == 'q') exit(0);
12884: }
1.227 brouard 12885: end:
1.126 brouard 12886: while (z[0] != 'q') {
1.195 brouard 12887: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12888: scanf("%s",z);
12889: }
1.283 brouard 12890: printf("End\n");
1.282 brouard 12891: exit(0);
1.126 brouard 12892: }
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