Annotation of imach/src/imach.c, revision 1.298
1.298 ! brouard 1: /* $Id: imach.c,v 1.297 2019/05/22 17:56:10 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.298 ! brouard 4: Revision 1.297 2019/05/22 17:56:10 brouard
! 5: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
! 6:
1.297 brouard 7: Revision 1.296 2019/05/20 13:03:18 brouard
8: Summary: Projection syntax simplified
9:
10:
11: We can now start projections, forward or backward, from the mean date
12: of inteviews up to or down to a number of years of projection:
13: prevforecast=1 yearsfproj=15.3 mobil_average=0
14: or
15: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
16: or
17: prevbackcast=1 yearsbproj=12.3 mobil_average=1
18: or
19: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
20:
1.296 brouard 21: Revision 1.295 2019/05/18 09:52:50 brouard
22: Summary: doxygen tex bug
23:
1.295 brouard 24: Revision 1.294 2019/05/16 14:54:33 brouard
25: Summary: There was some wrong lines added
26:
1.294 brouard 27: Revision 1.293 2019/05/09 15:17:34 brouard
28: *** empty log message ***
29:
1.293 brouard 30: Revision 1.292 2019/05/09 14:17:20 brouard
31: Summary: Some updates
32:
1.292 brouard 33: Revision 1.291 2019/05/09 13:44:18 brouard
34: Summary: Before ncovmax
35:
1.291 brouard 36: Revision 1.290 2019/05/09 13:39:37 brouard
37: Summary: 0.99r18 unlimited number of individuals
38:
39: 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.
40:
1.290 brouard 41: Revision 1.289 2018/12/13 09:16:26 brouard
42: Summary: Bug for young ages (<-30) will be in r17
43:
1.289 brouard 44: Revision 1.288 2018/05/02 20:58:27 brouard
45: Summary: Some bugs fixed
46:
1.288 brouard 47: Revision 1.287 2018/05/01 17:57:25 brouard
48: Summary: Bug fixed by providing frequencies only for non missing covariates
49:
1.287 brouard 50: Revision 1.286 2018/04/27 14:27:04 brouard
51: Summary: some minor bugs
52:
1.286 brouard 53: Revision 1.285 2018/04/21 21:02:16 brouard
54: Summary: Some bugs fixed, valgrind tested
55:
1.285 brouard 56: Revision 1.284 2018/04/20 05:22:13 brouard
57: Summary: Computing mean and stdeviation of fixed quantitative variables
58:
1.284 brouard 59: Revision 1.283 2018/04/19 14:49:16 brouard
60: Summary: Some minor bugs fixed
61:
1.283 brouard 62: Revision 1.282 2018/02/27 22:50:02 brouard
63: *** empty log message ***
64:
1.282 brouard 65: Revision 1.281 2018/02/27 19:25:23 brouard
66: Summary: Adding second argument for quitting
67:
1.281 brouard 68: Revision 1.280 2018/02/21 07:58:13 brouard
69: Summary: 0.99r15
70:
71: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
72:
1.280 brouard 73: Revision 1.279 2017/07/20 13:35:01 brouard
74: Summary: temporary working
75:
1.279 brouard 76: Revision 1.278 2017/07/19 14:09:02 brouard
77: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
78:
1.278 brouard 79: Revision 1.277 2017/07/17 08:53:49 brouard
80: Summary: BOM files can be read now
81:
1.277 brouard 82: Revision 1.276 2017/06/30 15:48:31 brouard
83: Summary: Graphs improvements
84:
1.276 brouard 85: Revision 1.275 2017/06/30 13:39:33 brouard
86: Summary: Saito's color
87:
1.275 brouard 88: Revision 1.274 2017/06/29 09:47:08 brouard
89: Summary: Version 0.99r14
90:
1.274 brouard 91: Revision 1.273 2017/06/27 11:06:02 brouard
92: Summary: More documentation on projections
93:
1.273 brouard 94: Revision 1.272 2017/06/27 10:22:40 brouard
95: Summary: Color of backprojection changed from 6 to 5(yellow)
96:
1.272 brouard 97: Revision 1.271 2017/06/27 10:17:50 brouard
98: Summary: Some bug with rint
99:
1.271 brouard 100: Revision 1.270 2017/05/24 05:45:29 brouard
101: *** empty log message ***
102:
1.270 brouard 103: Revision 1.269 2017/05/23 08:39:25 brouard
104: Summary: Code into subroutine, cleanings
105:
1.269 brouard 106: Revision 1.268 2017/05/18 20:09:32 brouard
107: Summary: backprojection and confidence intervals of backprevalence
108:
1.268 brouard 109: Revision 1.267 2017/05/13 10:25:05 brouard
110: Summary: temporary save for backprojection
111:
1.267 brouard 112: Revision 1.266 2017/05/13 07:26:12 brouard
113: Summary: Version 0.99r13 (improvements and bugs fixed)
114:
1.266 brouard 115: Revision 1.265 2017/04/26 16:22:11 brouard
116: Summary: imach 0.99r13 Some bugs fixed
117:
1.265 brouard 118: Revision 1.264 2017/04/26 06:01:29 brouard
119: Summary: Labels in graphs
120:
1.264 brouard 121: Revision 1.263 2017/04/24 15:23:15 brouard
122: Summary: to save
123:
1.263 brouard 124: Revision 1.262 2017/04/18 16:48:12 brouard
125: *** empty log message ***
126:
1.262 brouard 127: Revision 1.261 2017/04/05 10:14:09 brouard
128: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
129:
1.261 brouard 130: Revision 1.260 2017/04/04 17:46:59 brouard
131: Summary: Gnuplot indexations fixed (humm)
132:
1.260 brouard 133: Revision 1.259 2017/04/04 13:01:16 brouard
134: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
135:
1.259 brouard 136: Revision 1.258 2017/04/03 10:17:47 brouard
137: Summary: Version 0.99r12
138:
139: Some cleanings, conformed with updated documentation.
140:
1.258 brouard 141: Revision 1.257 2017/03/29 16:53:30 brouard
142: Summary: Temp
143:
1.257 brouard 144: Revision 1.256 2017/03/27 05:50:23 brouard
145: Summary: Temporary
146:
1.256 brouard 147: Revision 1.255 2017/03/08 16:02:28 brouard
148: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
149:
1.255 brouard 150: Revision 1.254 2017/03/08 07:13:00 brouard
151: Summary: Fixing data parameter line
152:
1.254 brouard 153: Revision 1.253 2016/12/15 11:59:41 brouard
154: Summary: 0.99 in progress
155:
1.253 brouard 156: Revision 1.252 2016/09/15 21:15:37 brouard
157: *** empty log message ***
158:
1.252 brouard 159: Revision 1.251 2016/09/15 15:01:13 brouard
160: Summary: not working
161:
1.251 brouard 162: Revision 1.250 2016/09/08 16:07:27 brouard
163: Summary: continue
164:
1.250 brouard 165: Revision 1.249 2016/09/07 17:14:18 brouard
166: Summary: Starting values from frequencies
167:
1.249 brouard 168: Revision 1.248 2016/09/07 14:10:18 brouard
169: *** empty log message ***
170:
1.248 brouard 171: Revision 1.247 2016/09/02 11:11:21 brouard
172: *** empty log message ***
173:
1.247 brouard 174: Revision 1.246 2016/09/02 08:49:22 brouard
175: *** empty log message ***
176:
1.246 brouard 177: Revision 1.245 2016/09/02 07:25:01 brouard
178: *** empty log message ***
179:
1.245 brouard 180: Revision 1.244 2016/09/02 07:17:34 brouard
181: *** empty log message ***
182:
1.244 brouard 183: Revision 1.243 2016/09/02 06:45:35 brouard
184: *** empty log message ***
185:
1.243 brouard 186: Revision 1.242 2016/08/30 15:01:20 brouard
187: Summary: Fixing a lots
188:
1.242 brouard 189: Revision 1.241 2016/08/29 17:17:25 brouard
190: Summary: gnuplot problem in Back projection to fix
191:
1.241 brouard 192: Revision 1.240 2016/08/29 07:53:18 brouard
193: Summary: Better
194:
1.240 brouard 195: Revision 1.239 2016/08/26 15:51:03 brouard
196: Summary: Improvement in Powell output in order to copy and paste
197:
198: Author:
199:
1.239 brouard 200: Revision 1.238 2016/08/26 14:23:35 brouard
201: Summary: Starting tests of 0.99
202:
1.238 brouard 203: Revision 1.237 2016/08/26 09:20:19 brouard
204: Summary: to valgrind
205:
1.237 brouard 206: Revision 1.236 2016/08/25 10:50:18 brouard
207: *** empty log message ***
208:
1.236 brouard 209: Revision 1.235 2016/08/25 06:59:23 brouard
210: *** empty log message ***
211:
1.235 brouard 212: Revision 1.234 2016/08/23 16:51:20 brouard
213: *** empty log message ***
214:
1.234 brouard 215: Revision 1.233 2016/08/23 07:40:50 brouard
216: Summary: not working
217:
1.233 brouard 218: Revision 1.232 2016/08/22 14:20:21 brouard
219: Summary: not working
220:
1.232 brouard 221: Revision 1.231 2016/08/22 07:17:15 brouard
222: Summary: not working
223:
1.231 brouard 224: Revision 1.230 2016/08/22 06:55:53 brouard
225: Summary: Not working
226:
1.230 brouard 227: Revision 1.229 2016/07/23 09:45:53 brouard
228: Summary: Completing for func too
229:
1.229 brouard 230: Revision 1.228 2016/07/22 17:45:30 brouard
231: Summary: Fixing some arrays, still debugging
232:
1.227 brouard 233: Revision 1.226 2016/07/12 18:42:34 brouard
234: Summary: temp
235:
1.226 brouard 236: Revision 1.225 2016/07/12 08:40:03 brouard
237: Summary: saving but not running
238:
1.225 brouard 239: Revision 1.224 2016/07/01 13:16:01 brouard
240: Summary: Fixes
241:
1.224 brouard 242: Revision 1.223 2016/02/19 09:23:35 brouard
243: Summary: temporary
244:
1.223 brouard 245: Revision 1.222 2016/02/17 08:14:50 brouard
246: Summary: Probably last 0.98 stable version 0.98r6
247:
1.222 brouard 248: Revision 1.221 2016/02/15 23:35:36 brouard
249: Summary: minor bug
250:
1.220 brouard 251: Revision 1.219 2016/02/15 00:48:12 brouard
252: *** empty log message ***
253:
1.219 brouard 254: Revision 1.218 2016/02/12 11:29:23 brouard
255: Summary: 0.99 Back projections
256:
1.218 brouard 257: Revision 1.217 2015/12/23 17:18:31 brouard
258: Summary: Experimental backcast
259:
1.217 brouard 260: Revision 1.216 2015/12/18 17:32:11 brouard
261: Summary: 0.98r4 Warning and status=-2
262:
263: Version 0.98r4 is now:
264: - displaying an error when status is -1, date of interview unknown and date of death known;
265: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
266: Older changes concerning s=-2, dating from 2005 have been supersed.
267:
1.216 brouard 268: Revision 1.215 2015/12/16 08:52:24 brouard
269: Summary: 0.98r4 working
270:
1.215 brouard 271: Revision 1.214 2015/12/16 06:57:54 brouard
272: Summary: temporary not working
273:
1.214 brouard 274: Revision 1.213 2015/12/11 18:22:17 brouard
275: Summary: 0.98r4
276:
1.213 brouard 277: Revision 1.212 2015/11/21 12:47:24 brouard
278: Summary: minor typo
279:
1.212 brouard 280: Revision 1.211 2015/11/21 12:41:11 brouard
281: Summary: 0.98r3 with some graph of projected cross-sectional
282:
283: Author: Nicolas Brouard
284:
1.211 brouard 285: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 286: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 287: Summary: Adding ftolpl parameter
288: Author: N Brouard
289:
290: We had difficulties to get smoothed confidence intervals. It was due
291: to the period prevalence which wasn't computed accurately. The inner
292: parameter ftolpl is now an outer parameter of the .imach parameter
293: file after estepm. If ftolpl is small 1.e-4 and estepm too,
294: computation are long.
295:
1.209 brouard 296: Revision 1.208 2015/11/17 14:31:57 brouard
297: Summary: temporary
298:
1.208 brouard 299: Revision 1.207 2015/10/27 17:36:57 brouard
300: *** empty log message ***
301:
1.207 brouard 302: Revision 1.206 2015/10/24 07:14:11 brouard
303: *** empty log message ***
304:
1.206 brouard 305: Revision 1.205 2015/10/23 15:50:53 brouard
306: Summary: 0.98r3 some clarification for graphs on likelihood contributions
307:
1.205 brouard 308: Revision 1.204 2015/10/01 16:20:26 brouard
309: Summary: Some new graphs of contribution to likelihood
310:
1.204 brouard 311: Revision 1.203 2015/09/30 17:45:14 brouard
312: Summary: looking at better estimation of the hessian
313:
314: Also a better criteria for convergence to the period prevalence And
315: therefore adding the number of years needed to converge. (The
316: prevalence in any alive state shold sum to one
317:
1.203 brouard 318: Revision 1.202 2015/09/22 19:45:16 brouard
319: Summary: Adding some overall graph on contribution to likelihood. Might change
320:
1.202 brouard 321: Revision 1.201 2015/09/15 17:34:58 brouard
322: Summary: 0.98r0
323:
324: - Some new graphs like suvival functions
325: - Some bugs fixed like model=1+age+V2.
326:
1.201 brouard 327: Revision 1.200 2015/09/09 16:53:55 brouard
328: Summary: Big bug thanks to Flavia
329:
330: Even model=1+age+V2. did not work anymore
331:
1.200 brouard 332: Revision 1.199 2015/09/07 14:09:23 brouard
333: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
334:
1.199 brouard 335: Revision 1.198 2015/09/03 07:14:39 brouard
336: Summary: 0.98q5 Flavia
337:
1.198 brouard 338: Revision 1.197 2015/09/01 18:24:39 brouard
339: *** empty log message ***
340:
1.197 brouard 341: Revision 1.196 2015/08/18 23:17:52 brouard
342: Summary: 0.98q5
343:
1.196 brouard 344: Revision 1.195 2015/08/18 16:28:39 brouard
345: Summary: Adding a hack for testing purpose
346:
347: After reading the title, ftol and model lines, if the comment line has
348: a q, starting with #q, the answer at the end of the run is quit. It
349: permits to run test files in batch with ctest. The former workaround was
350: $ echo q | imach foo.imach
351:
1.195 brouard 352: Revision 1.194 2015/08/18 13:32:00 brouard
353: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
354:
1.194 brouard 355: Revision 1.193 2015/08/04 07:17:42 brouard
356: Summary: 0.98q4
357:
1.193 brouard 358: Revision 1.192 2015/07/16 16:49:02 brouard
359: Summary: Fixing some outputs
360:
1.192 brouard 361: Revision 1.191 2015/07/14 10:00:33 brouard
362: Summary: Some fixes
363:
1.191 brouard 364: Revision 1.190 2015/05/05 08:51:13 brouard
365: Summary: Adding digits in output parameters (7 digits instead of 6)
366:
367: Fix 1+age+.
368:
1.190 brouard 369: Revision 1.189 2015/04/30 14:45:16 brouard
370: Summary: 0.98q2
371:
1.189 brouard 372: Revision 1.188 2015/04/30 08:27:53 brouard
373: *** empty log message ***
374:
1.188 brouard 375: Revision 1.187 2015/04/29 09:11:15 brouard
376: *** empty log message ***
377:
1.187 brouard 378: Revision 1.186 2015/04/23 12:01:52 brouard
379: Summary: V1*age is working now, version 0.98q1
380:
381: Some codes had been disabled in order to simplify and Vn*age was
382: working in the optimization phase, ie, giving correct MLE parameters,
383: but, as usual, outputs were not correct and program core dumped.
384:
1.186 brouard 385: Revision 1.185 2015/03/11 13:26:42 brouard
386: Summary: Inclusion of compile and links command line for Intel Compiler
387:
1.185 brouard 388: Revision 1.184 2015/03/11 11:52:39 brouard
389: Summary: Back from Windows 8. Intel Compiler
390:
1.184 brouard 391: Revision 1.183 2015/03/10 20:34:32 brouard
392: Summary: 0.98q0, trying with directest, mnbrak fixed
393:
394: We use directest instead of original Powell test; probably no
395: incidence on the results, but better justifications;
396: We fixed Numerical Recipes mnbrak routine which was wrong and gave
397: wrong results.
398:
1.183 brouard 399: Revision 1.182 2015/02/12 08:19:57 brouard
400: Summary: Trying to keep directest which seems simpler and more general
401: Author: Nicolas Brouard
402:
1.182 brouard 403: Revision 1.181 2015/02/11 23:22:24 brouard
404: Summary: Comments on Powell added
405:
406: Author:
407:
1.181 brouard 408: Revision 1.180 2015/02/11 17:33:45 brouard
409: Summary: Finishing move from main to function (hpijx and prevalence_limit)
410:
1.180 brouard 411: Revision 1.179 2015/01/04 09:57:06 brouard
412: Summary: back to OS/X
413:
1.179 brouard 414: Revision 1.178 2015/01/04 09:35:48 brouard
415: *** empty log message ***
416:
1.178 brouard 417: Revision 1.177 2015/01/03 18:40:56 brouard
418: Summary: Still testing ilc32 on OSX
419:
1.177 brouard 420: Revision 1.176 2015/01/03 16:45:04 brouard
421: *** empty log message ***
422:
1.176 brouard 423: Revision 1.175 2015/01/03 16:33:42 brouard
424: *** empty log message ***
425:
1.175 brouard 426: Revision 1.174 2015/01/03 16:15:49 brouard
427: Summary: Still in cross-compilation
428:
1.174 brouard 429: Revision 1.173 2015/01/03 12:06:26 brouard
430: Summary: trying to detect cross-compilation
431:
1.173 brouard 432: Revision 1.172 2014/12/27 12:07:47 brouard
433: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
434:
1.172 brouard 435: Revision 1.171 2014/12/23 13:26:59 brouard
436: Summary: Back from Visual C
437:
438: Still problem with utsname.h on Windows
439:
1.171 brouard 440: Revision 1.170 2014/12/23 11:17:12 brouard
441: Summary: Cleaning some \%% back to %%
442:
443: The escape was mandatory for a specific compiler (which one?), but too many warnings.
444:
1.170 brouard 445: Revision 1.169 2014/12/22 23:08:31 brouard
446: Summary: 0.98p
447:
448: Outputs some informations on compiler used, OS etc. Testing on different platforms.
449:
1.169 brouard 450: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 451: Summary: update
1.169 brouard 452:
1.168 brouard 453: Revision 1.167 2014/12/22 13:50:56 brouard
454: Summary: Testing uname and compiler version and if compiled 32 or 64
455:
456: Testing on Linux 64
457:
1.167 brouard 458: Revision 1.166 2014/12/22 11:40:47 brouard
459: *** empty log message ***
460:
1.166 brouard 461: Revision 1.165 2014/12/16 11:20:36 brouard
462: Summary: After compiling on Visual C
463:
464: * imach.c (Module): Merging 1.61 to 1.162
465:
1.165 brouard 466: Revision 1.164 2014/12/16 10:52:11 brouard
467: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
468:
469: * imach.c (Module): Merging 1.61 to 1.162
470:
1.164 brouard 471: Revision 1.163 2014/12/16 10:30:11 brouard
472: * imach.c (Module): Merging 1.61 to 1.162
473:
1.163 brouard 474: Revision 1.162 2014/09/25 11:43:39 brouard
475: Summary: temporary backup 0.99!
476:
1.162 brouard 477: Revision 1.1 2014/09/16 11:06:58 brouard
478: Summary: With some code (wrong) for nlopt
479:
480: Author:
481:
482: Revision 1.161 2014/09/15 20:41:41 brouard
483: Summary: Problem with macro SQR on Intel compiler
484:
1.161 brouard 485: Revision 1.160 2014/09/02 09:24:05 brouard
486: *** empty log message ***
487:
1.160 brouard 488: Revision 1.159 2014/09/01 10:34:10 brouard
489: Summary: WIN32
490: Author: Brouard
491:
1.159 brouard 492: Revision 1.158 2014/08/27 17:11:51 brouard
493: *** empty log message ***
494:
1.158 brouard 495: Revision 1.157 2014/08/27 16:26:55 brouard
496: Summary: Preparing windows Visual studio version
497: Author: Brouard
498:
499: In order to compile on Visual studio, time.h is now correct and time_t
500: and tm struct should be used. difftime should be used but sometimes I
501: just make the differences in raw time format (time(&now).
502: Trying to suppress #ifdef LINUX
503: Add xdg-open for __linux in order to open default browser.
504:
1.157 brouard 505: Revision 1.156 2014/08/25 20:10:10 brouard
506: *** empty log message ***
507:
1.156 brouard 508: Revision 1.155 2014/08/25 18:32:34 brouard
509: Summary: New compile, minor changes
510: Author: Brouard
511:
1.155 brouard 512: Revision 1.154 2014/06/20 17:32:08 brouard
513: Summary: Outputs now all graphs of convergence to period prevalence
514:
1.154 brouard 515: Revision 1.153 2014/06/20 16:45:46 brouard
516: Summary: If 3 live state, convergence to period prevalence on same graph
517: Author: Brouard
518:
1.153 brouard 519: Revision 1.152 2014/06/18 17:54:09 brouard
520: Summary: open browser, use gnuplot on same dir than imach if not found in the path
521:
1.152 brouard 522: Revision 1.151 2014/06/18 16:43:30 brouard
523: *** empty log message ***
524:
1.151 brouard 525: Revision 1.150 2014/06/18 16:42:35 brouard
526: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
527: Author: brouard
528:
1.150 brouard 529: Revision 1.149 2014/06/18 15:51:14 brouard
530: Summary: Some fixes in parameter files errors
531: Author: Nicolas Brouard
532:
1.149 brouard 533: Revision 1.148 2014/06/17 17:38:48 brouard
534: Summary: Nothing new
535: Author: Brouard
536:
537: Just a new packaging for OS/X version 0.98nS
538:
1.148 brouard 539: Revision 1.147 2014/06/16 10:33:11 brouard
540: *** empty log message ***
541:
1.147 brouard 542: Revision 1.146 2014/06/16 10:20:28 brouard
543: Summary: Merge
544: Author: Brouard
545:
546: Merge, before building revised version.
547:
1.146 brouard 548: Revision 1.145 2014/06/10 21:23:15 brouard
549: Summary: Debugging with valgrind
550: Author: Nicolas Brouard
551:
552: Lot of changes in order to output the results with some covariates
553: After the Edimburgh REVES conference 2014, it seems mandatory to
554: improve the code.
555: No more memory valgrind error but a lot has to be done in order to
556: continue the work of splitting the code into subroutines.
557: Also, decodemodel has been improved. Tricode is still not
558: optimal. nbcode should be improved. Documentation has been added in
559: the source code.
560:
1.144 brouard 561: Revision 1.143 2014/01/26 09:45:38 brouard
562: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
563:
564: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
565: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
566:
1.143 brouard 567: Revision 1.142 2014/01/26 03:57:36 brouard
568: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
569:
570: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
571:
1.142 brouard 572: Revision 1.141 2014/01/26 02:42:01 brouard
573: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
574:
1.141 brouard 575: Revision 1.140 2011/09/02 10:37:54 brouard
576: Summary: times.h is ok with mingw32 now.
577:
1.140 brouard 578: Revision 1.139 2010/06/14 07:50:17 brouard
579: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
580: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
581:
1.139 brouard 582: Revision 1.138 2010/04/30 18:19:40 brouard
583: *** empty log message ***
584:
1.138 brouard 585: Revision 1.137 2010/04/29 18:11:38 brouard
586: (Module): Checking covariates for more complex models
587: than V1+V2. A lot of change to be done. Unstable.
588:
1.137 brouard 589: Revision 1.136 2010/04/26 20:30:53 brouard
590: (Module): merging some libgsl code. Fixing computation
591: of likelione (using inter/intrapolation if mle = 0) in order to
592: get same likelihood as if mle=1.
593: Some cleaning of code and comments added.
594:
1.136 brouard 595: Revision 1.135 2009/10/29 15:33:14 brouard
596: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
597:
1.135 brouard 598: Revision 1.134 2009/10/29 13:18:53 brouard
599: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
600:
1.134 brouard 601: Revision 1.133 2009/07/06 10:21:25 brouard
602: just nforces
603:
1.133 brouard 604: Revision 1.132 2009/07/06 08:22:05 brouard
605: Many tings
606:
1.132 brouard 607: Revision 1.131 2009/06/20 16:22:47 brouard
608: Some dimensions resccaled
609:
1.131 brouard 610: Revision 1.130 2009/05/26 06:44:34 brouard
611: (Module): Max Covariate is now set to 20 instead of 8. A
612: lot of cleaning with variables initialized to 0. Trying to make
613: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
614:
1.130 brouard 615: Revision 1.129 2007/08/31 13:49:27 lievre
616: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
617:
1.129 lievre 618: Revision 1.128 2006/06/30 13:02:05 brouard
619: (Module): Clarifications on computing e.j
620:
1.128 brouard 621: Revision 1.127 2006/04/28 18:11:50 brouard
622: (Module): Yes the sum of survivors was wrong since
623: imach-114 because nhstepm was no more computed in the age
624: loop. Now we define nhstepma in the age loop.
625: (Module): In order to speed up (in case of numerous covariates) we
626: compute health expectancies (without variances) in a first step
627: and then all the health expectancies with variances or standard
628: deviation (needs data from the Hessian matrices) which slows the
629: computation.
630: In the future we should be able to stop the program is only health
631: expectancies and graph are needed without standard deviations.
632:
1.127 brouard 633: Revision 1.126 2006/04/28 17:23:28 brouard
634: (Module): Yes the sum of survivors was wrong since
635: imach-114 because nhstepm was no more computed in the age
636: loop. Now we define nhstepma in the age loop.
637: Version 0.98h
638:
1.126 brouard 639: Revision 1.125 2006/04/04 15:20:31 lievre
640: Errors in calculation of health expectancies. Age was not initialized.
641: Forecasting file added.
642:
643: Revision 1.124 2006/03/22 17:13:53 lievre
644: Parameters are printed with %lf instead of %f (more numbers after the comma).
645: The log-likelihood is printed in the log file
646:
647: Revision 1.123 2006/03/20 10:52:43 brouard
648: * imach.c (Module): <title> changed, corresponds to .htm file
649: name. <head> headers where missing.
650:
651: * imach.c (Module): Weights can have a decimal point as for
652: English (a comma might work with a correct LC_NUMERIC environment,
653: otherwise the weight is truncated).
654: Modification of warning when the covariates values are not 0 or
655: 1.
656: Version 0.98g
657:
658: Revision 1.122 2006/03/20 09:45:41 brouard
659: (Module): Weights can have a decimal point as for
660: English (a comma might work with a correct LC_NUMERIC environment,
661: otherwise the weight is truncated).
662: Modification of warning when the covariates values are not 0 or
663: 1.
664: Version 0.98g
665:
666: Revision 1.121 2006/03/16 17:45:01 lievre
667: * imach.c (Module): Comments concerning covariates added
668:
669: * imach.c (Module): refinements in the computation of lli if
670: status=-2 in order to have more reliable computation if stepm is
671: not 1 month. Version 0.98f
672:
673: Revision 1.120 2006/03/16 15:10:38 lievre
674: (Module): refinements in the computation of lli if
675: status=-2 in order to have more reliable computation if stepm is
676: not 1 month. Version 0.98f
677:
678: Revision 1.119 2006/03/15 17:42:26 brouard
679: (Module): Bug if status = -2, the loglikelihood was
680: computed as likelihood omitting the logarithm. Version O.98e
681:
682: Revision 1.118 2006/03/14 18:20:07 brouard
683: (Module): varevsij Comments added explaining the second
684: table of variances if popbased=1 .
685: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
686: (Module): Function pstamp added
687: (Module): Version 0.98d
688:
689: Revision 1.117 2006/03/14 17:16:22 brouard
690: (Module): varevsij Comments added explaining the second
691: table of variances if popbased=1 .
692: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
693: (Module): Function pstamp added
694: (Module): Version 0.98d
695:
696: Revision 1.116 2006/03/06 10:29:27 brouard
697: (Module): Variance-covariance wrong links and
698: varian-covariance of ej. is needed (Saito).
699:
700: Revision 1.115 2006/02/27 12:17:45 brouard
701: (Module): One freematrix added in mlikeli! 0.98c
702:
703: Revision 1.114 2006/02/26 12:57:58 brouard
704: (Module): Some improvements in processing parameter
705: filename with strsep.
706:
707: Revision 1.113 2006/02/24 14:20:24 brouard
708: (Module): Memory leaks checks with valgrind and:
709: datafile was not closed, some imatrix were not freed and on matrix
710: allocation too.
711:
712: Revision 1.112 2006/01/30 09:55:26 brouard
713: (Module): Back to gnuplot.exe instead of wgnuplot.exe
714:
715: Revision 1.111 2006/01/25 20:38:18 brouard
716: (Module): Lots of cleaning and bugs added (Gompertz)
717: (Module): Comments can be added in data file. Missing date values
718: can be a simple dot '.'.
719:
720: Revision 1.110 2006/01/25 00:51:50 brouard
721: (Module): Lots of cleaning and bugs added (Gompertz)
722:
723: Revision 1.109 2006/01/24 19:37:15 brouard
724: (Module): Comments (lines starting with a #) are allowed in data.
725:
726: Revision 1.108 2006/01/19 18:05:42 lievre
727: Gnuplot problem appeared...
728: To be fixed
729:
730: Revision 1.107 2006/01/19 16:20:37 brouard
731: Test existence of gnuplot in imach path
732:
733: Revision 1.106 2006/01/19 13:24:36 brouard
734: Some cleaning and links added in html output
735:
736: Revision 1.105 2006/01/05 20:23:19 lievre
737: *** empty log message ***
738:
739: Revision 1.104 2005/09/30 16:11:43 lievre
740: (Module): sump fixed, loop imx fixed, and simplifications.
741: (Module): If the status is missing at the last wave but we know
742: that the person is alive, then we can code his/her status as -2
743: (instead of missing=-1 in earlier versions) and his/her
744: contributions to the likelihood is 1 - Prob of dying from last
745: health status (= 1-p13= p11+p12 in the easiest case of somebody in
746: the healthy state at last known wave). Version is 0.98
747:
748: Revision 1.103 2005/09/30 15:54:49 lievre
749: (Module): sump fixed, loop imx fixed, and simplifications.
750:
751: Revision 1.102 2004/09/15 17:31:30 brouard
752: Add the possibility to read data file including tab characters.
753:
754: Revision 1.101 2004/09/15 10:38:38 brouard
755: Fix on curr_time
756:
757: Revision 1.100 2004/07/12 18:29:06 brouard
758: Add version for Mac OS X. Just define UNIX in Makefile
759:
760: Revision 1.99 2004/06/05 08:57:40 brouard
761: *** empty log message ***
762:
763: Revision 1.98 2004/05/16 15:05:56 brouard
764: New version 0.97 . First attempt to estimate force of mortality
765: directly from the data i.e. without the need of knowing the health
766: state at each age, but using a Gompertz model: log u =a + b*age .
767: This is the basic analysis of mortality and should be done before any
768: other analysis, in order to test if the mortality estimated from the
769: cross-longitudinal survey is different from the mortality estimated
770: from other sources like vital statistic data.
771:
772: The same imach parameter file can be used but the option for mle should be -3.
773:
1.133 brouard 774: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 775: former routines in order to include the new code within the former code.
776:
777: The output is very simple: only an estimate of the intercept and of
778: the slope with 95% confident intervals.
779:
780: Current limitations:
781: A) Even if you enter covariates, i.e. with the
782: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
783: B) There is no computation of Life Expectancy nor Life Table.
784:
785: Revision 1.97 2004/02/20 13:25:42 lievre
786: Version 0.96d. Population forecasting command line is (temporarily)
787: suppressed.
788:
789: Revision 1.96 2003/07/15 15:38:55 brouard
790: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
791: rewritten within the same printf. Workaround: many printfs.
792:
793: Revision 1.95 2003/07/08 07:54:34 brouard
794: * imach.c (Repository):
795: (Repository): Using imachwizard code to output a more meaningful covariance
796: matrix (cov(a12,c31) instead of numbers.
797:
798: Revision 1.94 2003/06/27 13:00:02 brouard
799: Just cleaning
800:
801: Revision 1.93 2003/06/25 16:33:55 brouard
802: (Module): On windows (cygwin) function asctime_r doesn't
803: exist so I changed back to asctime which exists.
804: (Module): Version 0.96b
805:
806: Revision 1.92 2003/06/25 16:30:45 brouard
807: (Module): On windows (cygwin) function asctime_r doesn't
808: exist so I changed back to asctime which exists.
809:
810: Revision 1.91 2003/06/25 15:30:29 brouard
811: * imach.c (Repository): Duplicated warning errors corrected.
812: (Repository): Elapsed time after each iteration is now output. It
813: helps to forecast when convergence will be reached. Elapsed time
814: is stamped in powell. We created a new html file for the graphs
815: concerning matrix of covariance. It has extension -cov.htm.
816:
817: Revision 1.90 2003/06/24 12:34:15 brouard
818: (Module): Some bugs corrected for windows. Also, when
819: mle=-1 a template is output in file "or"mypar.txt with the design
820: of the covariance matrix to be input.
821:
822: Revision 1.89 2003/06/24 12:30:52 brouard
823: (Module): Some bugs corrected for windows. Also, when
824: mle=-1 a template is output in file "or"mypar.txt with the design
825: of the covariance matrix to be input.
826:
827: Revision 1.88 2003/06/23 17:54:56 brouard
828: * 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.
829:
830: Revision 1.87 2003/06/18 12:26:01 brouard
831: Version 0.96
832:
833: Revision 1.86 2003/06/17 20:04:08 brouard
834: (Module): Change position of html and gnuplot routines and added
835: routine fileappend.
836:
837: Revision 1.85 2003/06/17 13:12:43 brouard
838: * imach.c (Repository): Check when date of death was earlier that
839: current date of interview. It may happen when the death was just
840: prior to the death. In this case, dh was negative and likelihood
841: was wrong (infinity). We still send an "Error" but patch by
842: assuming that the date of death was just one stepm after the
843: interview.
844: (Repository): Because some people have very long ID (first column)
845: we changed int to long in num[] and we added a new lvector for
846: memory allocation. But we also truncated to 8 characters (left
847: truncation)
848: (Repository): No more line truncation errors.
849:
850: Revision 1.84 2003/06/13 21:44:43 brouard
851: * imach.c (Repository): Replace "freqsummary" at a correct
852: place. It differs from routine "prevalence" which may be called
853: many times. Probs is memory consuming and must be used with
854: parcimony.
855: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
856:
857: Revision 1.83 2003/06/10 13:39:11 lievre
858: *** empty log message ***
859:
860: Revision 1.82 2003/06/05 15:57:20 brouard
861: Add log in imach.c and fullversion number is now printed.
862:
863: */
864: /*
865: Interpolated Markov Chain
866:
867: Short summary of the programme:
868:
1.227 brouard 869: This program computes Healthy Life Expectancies or State-specific
870: (if states aren't health statuses) Expectancies from
871: cross-longitudinal data. Cross-longitudinal data consist in:
872:
873: -1- a first survey ("cross") where individuals from different ages
874: are interviewed on their health status or degree of disability (in
875: the case of a health survey which is our main interest)
876:
877: -2- at least a second wave of interviews ("longitudinal") which
878: measure each change (if any) in individual health status. Health
879: expectancies are computed from the time spent in each health state
880: according to a model. More health states you consider, more time is
881: necessary to reach the Maximum Likelihood of the parameters involved
882: in the model. The simplest model is the multinomial logistic model
883: where pij is the probability to be observed in state j at the second
884: wave conditional to be observed in state i at the first
885: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
886: etc , where 'age' is age and 'sex' is a covariate. If you want to
887: have a more complex model than "constant and age", you should modify
888: the program where the markup *Covariates have to be included here
889: again* invites you to do it. More covariates you add, slower the
1.126 brouard 890: convergence.
891:
892: The advantage of this computer programme, compared to a simple
893: multinomial logistic model, is clear when the delay between waves is not
894: identical for each individual. Also, if a individual missed an
895: intermediate interview, the information is lost, but taken into
896: account using an interpolation or extrapolation.
897:
898: hPijx is the probability to be observed in state i at age x+h
899: conditional to the observed state i at age x. The delay 'h' can be
900: split into an exact number (nh*stepm) of unobserved intermediate
901: states. This elementary transition (by month, quarter,
902: semester or year) is modelled as a multinomial logistic. The hPx
903: matrix is simply the matrix product of nh*stepm elementary matrices
904: and the contribution of each individual to the likelihood is simply
905: hPijx.
906:
907: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 908: of the life expectancies. It also computes the period (stable) prevalence.
909:
910: Back prevalence and projections:
1.227 brouard 911:
912: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
913: double agemaxpar, double ftolpl, int *ncvyearp, double
914: dateprev1,double dateprev2, int firstpass, int lastpass, int
915: mobilavproj)
916:
917: Computes the back prevalence limit for any combination of
918: covariate values k at any age between ageminpar and agemaxpar and
919: returns it in **bprlim. In the loops,
920:
921: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
922: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
923:
924: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 925: Computes for any combination of covariates k and any age between bage and fage
926: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
927: oldm=oldms;savm=savms;
1.227 brouard 928:
1.267 brouard 929: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 930: Computes the transition matrix starting at age 'age' over
931: 'nhstepm*hstepm*stepm' months (i.e. until
932: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 933: nhstepm*hstepm matrices.
934:
935: Returns p3mat[i][j][h] after calling
936: p3mat[i][j][h]=matprod2(newm,
937: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
938: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
939: oldm);
1.226 brouard 940:
941: Important routines
942:
943: - func (or funcone), computes logit (pij) distinguishing
944: o fixed variables (single or product dummies or quantitative);
945: o varying variables by:
946: (1) wave (single, product dummies, quantitative),
947: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
948: % fixed dummy (treated) or quantitative (not done because time-consuming);
949: % varying dummy (not done) or quantitative (not done);
950: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
951: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
952: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
953: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
954: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 955:
1.226 brouard 956:
957:
1.133 brouard 958: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
959: Institut national d'études démographiques, Paris.
1.126 brouard 960: This software have been partly granted by Euro-REVES, a concerted action
961: from the European Union.
962: It is copyrighted identically to a GNU software product, ie programme and
963: software can be distributed freely for non commercial use. Latest version
964: can be accessed at http://euroreves.ined.fr/imach .
965:
966: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
967: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
968:
969: **********************************************************************/
970: /*
971: main
972: read parameterfile
973: read datafile
974: concatwav
975: freqsummary
976: if (mle >= 1)
977: mlikeli
978: print results files
979: if mle==1
980: computes hessian
981: read end of parameter file: agemin, agemax, bage, fage, estepm
982: begin-prev-date,...
983: open gnuplot file
984: open html file
1.145 brouard 985: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
986: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
987: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
988: freexexit2 possible for memory heap.
989:
990: h Pij x | pij_nom ficrestpij
991: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
992: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
993: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
994:
995: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
996: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
997: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
998: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
999: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1000:
1.126 brouard 1001: forecasting if prevfcast==1 prevforecast call prevalence()
1002: health expectancies
1003: Variance-covariance of DFLE
1004: prevalence()
1005: movingaverage()
1006: varevsij()
1007: if popbased==1 varevsij(,popbased)
1008: total life expectancies
1009: Variance of period (stable) prevalence
1010: end
1011: */
1012:
1.187 brouard 1013: /* #define DEBUG */
1014: /* #define DEBUGBRENT */
1.203 brouard 1015: /* #define DEBUGLINMIN */
1016: /* #define DEBUGHESS */
1017: #define DEBUGHESSIJ
1.224 brouard 1018: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1019: #define POWELL /* Instead of NLOPT */
1.224 brouard 1020: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1021: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1022: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 1023:
1024: #include <math.h>
1025: #include <stdio.h>
1026: #include <stdlib.h>
1027: #include <string.h>
1.226 brouard 1028: #include <ctype.h>
1.159 brouard 1029:
1030: #ifdef _WIN32
1031: #include <io.h>
1.172 brouard 1032: #include <windows.h>
1033: #include <tchar.h>
1.159 brouard 1034: #else
1.126 brouard 1035: #include <unistd.h>
1.159 brouard 1036: #endif
1.126 brouard 1037:
1038: #include <limits.h>
1039: #include <sys/types.h>
1.171 brouard 1040:
1041: #if defined(__GNUC__)
1042: #include <sys/utsname.h> /* Doesn't work on Windows */
1043: #endif
1044:
1.126 brouard 1045: #include <sys/stat.h>
1046: #include <errno.h>
1.159 brouard 1047: /* extern int errno; */
1.126 brouard 1048:
1.157 brouard 1049: /* #ifdef LINUX */
1050: /* #include <time.h> */
1051: /* #include "timeval.h" */
1052: /* #else */
1053: /* #include <sys/time.h> */
1054: /* #endif */
1055:
1.126 brouard 1056: #include <time.h>
1057:
1.136 brouard 1058: #ifdef GSL
1059: #include <gsl/gsl_errno.h>
1060: #include <gsl/gsl_multimin.h>
1061: #endif
1062:
1.167 brouard 1063:
1.162 brouard 1064: #ifdef NLOPT
1065: #include <nlopt.h>
1066: typedef struct {
1067: double (* function)(double [] );
1068: } myfunc_data ;
1069: #endif
1070:
1.126 brouard 1071: /* #include <libintl.h> */
1072: /* #define _(String) gettext (String) */
1073:
1.251 brouard 1074: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1075:
1076: #define GNUPLOTPROGRAM "gnuplot"
1077: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1078: #define FILENAMELENGTH 132
1079:
1080: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1081: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1082:
1.144 brouard 1083: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1084: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1085:
1086: #define NINTERVMAX 8
1.144 brouard 1087: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1088: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1089: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1090: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1091: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1092: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1093: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1094: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1095: /* #define AGESUP 130 */
1.288 brouard 1096: /* #define AGESUP 150 */
1097: #define AGESUP 200
1.268 brouard 1098: #define AGEINF 0
1.218 brouard 1099: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1100: #define AGEBASE 40
1.194 brouard 1101: #define AGEOVERFLOW 1.e20
1.164 brouard 1102: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1103: #ifdef _WIN32
1104: #define DIRSEPARATOR '\\'
1105: #define CHARSEPARATOR "\\"
1106: #define ODIRSEPARATOR '/'
1107: #else
1.126 brouard 1108: #define DIRSEPARATOR '/'
1109: #define CHARSEPARATOR "/"
1110: #define ODIRSEPARATOR '\\'
1111: #endif
1112:
1.298 ! brouard 1113: /* $Id: imach.c,v 1.297 2019/05/22 17:56:10 brouard Exp $ */
1.126 brouard 1114: /* $State: Exp $ */
1.196 brouard 1115: #include "version.h"
1116: char version[]=__IMACH_VERSION__;
1.283 brouard 1117: 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";
1.298 ! brouard 1118: char fullversion[]="$Revision: 1.297 $ $Date: 2019/05/22 17:56:10 $";
1.126 brouard 1119: char strstart[80];
1120: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1121: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1122: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1123: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1124: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1125: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1126: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1127: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1128: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1129: int cptcovprodnoage=0; /**< Number of covariate products without age */
1130: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1131: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1132: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1133: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1134: int nsd=0; /**< Total number of single dummy variables (output) */
1135: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1136: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1137: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1138: int ntveff=0; /**< ntveff number of effective time varying variables */
1139: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1140: int cptcov=0; /* Working variable */
1.290 brouard 1141: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1142: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1143: int npar=NPARMAX;
1144: int nlstate=2; /* Number of live states */
1145: int ndeath=1; /* Number of dead states */
1.130 brouard 1146: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1147: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1148: int popbased=0;
1149:
1150: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1151: int maxwav=0; /* Maxim number of waves */
1152: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1153: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1154: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1155: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1156: int mle=1, weightopt=0;
1.126 brouard 1157: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1158: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1159: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1160: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1161: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1162: int selected(int kvar); /* Is covariate kvar selected for printing results */
1163:
1.130 brouard 1164: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1165: double **matprod2(); /* test */
1.126 brouard 1166: double **oldm, **newm, **savm; /* Working pointers to matrices */
1167: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1168: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1169:
1.136 brouard 1170: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1171: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1172: FILE *ficlog, *ficrespow;
1.130 brouard 1173: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1174: double fretone; /* Only one call to likelihood */
1.130 brouard 1175: long ipmx=0; /* Number of contributions */
1.126 brouard 1176: double sw; /* Sum of weights */
1177: char filerespow[FILENAMELENGTH];
1178: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1179: FILE *ficresilk;
1180: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1181: FILE *ficresprobmorprev;
1182: FILE *fichtm, *fichtmcov; /* Html File */
1183: FILE *ficreseij;
1184: char filerese[FILENAMELENGTH];
1185: FILE *ficresstdeij;
1186: char fileresstde[FILENAMELENGTH];
1187: FILE *ficrescveij;
1188: char filerescve[FILENAMELENGTH];
1189: FILE *ficresvij;
1190: char fileresv[FILENAMELENGTH];
1.269 brouard 1191:
1.126 brouard 1192: char title[MAXLINE];
1.234 brouard 1193: char model[MAXLINE]; /**< The model line */
1.217 brouard 1194: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1195: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1196: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1197: char command[FILENAMELENGTH];
1198: int outcmd=0;
1199:
1.217 brouard 1200: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1201: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1202: char filelog[FILENAMELENGTH]; /* Log file */
1203: char filerest[FILENAMELENGTH];
1204: char fileregp[FILENAMELENGTH];
1205: char popfile[FILENAMELENGTH];
1206:
1207: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1208:
1.157 brouard 1209: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1210: /* struct timezone tzp; */
1211: /* extern int gettimeofday(); */
1212: struct tm tml, *gmtime(), *localtime();
1213:
1214: extern time_t time();
1215:
1216: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1217: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1218: struct tm tm;
1219:
1.126 brouard 1220: char strcurr[80], strfor[80];
1221:
1222: char *endptr;
1223: long lval;
1224: double dval;
1225:
1226: #define NR_END 1
1227: #define FREE_ARG char*
1228: #define FTOL 1.0e-10
1229:
1230: #define NRANSI
1.240 brouard 1231: #define ITMAX 200
1232: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1233:
1234: #define TOL 2.0e-4
1235:
1236: #define CGOLD 0.3819660
1237: #define ZEPS 1.0e-10
1238: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1239:
1240: #define GOLD 1.618034
1241: #define GLIMIT 100.0
1242: #define TINY 1.0e-20
1243:
1244: static double maxarg1,maxarg2;
1245: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1246: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1247:
1248: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1249: #define rint(a) floor(a+0.5)
1.166 brouard 1250: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1251: #define mytinydouble 1.0e-16
1.166 brouard 1252: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1253: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1254: /* static double dsqrarg; */
1255: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1256: static double sqrarg;
1257: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1258: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1259: int agegomp= AGEGOMP;
1260:
1261: int imx;
1262: int stepm=1;
1263: /* Stepm, step in month: minimum step interpolation*/
1264:
1265: int estepm;
1266: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1267:
1268: int m,nb;
1269: long *num;
1.197 brouard 1270: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1271: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1272: covariate for which somebody answered excluding
1273: undefined. Usually 2: 0 and 1. */
1274: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1275: covariate for which somebody answered including
1276: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1277: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1278: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1279: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1280: double *ageexmed,*agecens;
1281: double dateintmean=0;
1.296 brouard 1282: double anprojd, mprojd, jprojd; /* For eventual projections */
1283: double anprojf, mprojf, jprojf;
1.126 brouard 1284:
1.296 brouard 1285: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1286: double anbackf, mbackf, jbackf;
1287: double jintmean,mintmean,aintmean;
1.126 brouard 1288: double *weight;
1289: int **s; /* Status */
1.141 brouard 1290: double *agedc;
1.145 brouard 1291: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1292: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1293: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1294: double **coqvar; /* Fixed quantitative covariate nqv */
1295: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1296: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1297: double idx;
1298: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1299: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1300: /*k 1 2 3 4 5 6 7 8 9 */
1301: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1302: /* Tndvar[k] 1 2 3 4 5 */
1303: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1304: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1305: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1306: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1307: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1308: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1309: /* Tprod[i]=k 4 7 */
1310: /* Tage[i]=k 5 8 */
1311: /* */
1312: /* Type */
1313: /* V 1 2 3 4 5 */
1314: /* F F V V V */
1315: /* D Q D D Q */
1316: /* */
1317: int *TvarsD;
1318: int *TvarsDind;
1319: int *TvarsQ;
1320: int *TvarsQind;
1321:
1.235 brouard 1322: #define MAXRESULTLINES 10
1323: int nresult=0;
1.258 brouard 1324: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1325: int TKresult[MAXRESULTLINES];
1.237 brouard 1326: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1327: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1328: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1329: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1330: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1331: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1332:
1.234 brouard 1333: /* 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 1334: 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 */
1335: 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 */
1336: 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 */
1337: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1338: 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 */
1339: 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 1340: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1341: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1342: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1343: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1344: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1345: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1346: 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 */
1347: 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 */
1348:
1.230 brouard 1349: int *Tvarsel; /**< Selected covariates for output */
1350: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1351: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1352: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1353: 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 1354: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1355: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1356: int *Tage;
1.227 brouard 1357: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1358: 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 1359: 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*/
1360: 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 1361: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1362: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1363: int **Tvard;
1364: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1365: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1366: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1367: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1368: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1369: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1370: double *lsurv, *lpop, *tpop;
1371:
1.231 brouard 1372: #define FD 1; /* Fixed dummy covariate */
1373: #define FQ 2; /* Fixed quantitative covariate */
1374: #define FP 3; /* Fixed product covariate */
1375: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1376: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1377: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1378: #define VD 10; /* Varying dummy covariate */
1379: #define VQ 11; /* Varying quantitative covariate */
1380: #define VP 12; /* Varying product covariate */
1381: #define VPDD 13; /* Varying product dummy*dummy covariate */
1382: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1383: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1384: #define APFD 16; /* Age product * fixed dummy covariate */
1385: #define APFQ 17; /* Age product * fixed quantitative covariate */
1386: #define APVD 18; /* Age product * varying dummy covariate */
1387: #define APVQ 19; /* Age product * varying quantitative covariate */
1388:
1389: #define FTYPE 1; /* Fixed covariate */
1390: #define VTYPE 2; /* Varying covariate (loop in wave) */
1391: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1392:
1393: struct kmodel{
1394: int maintype; /* main type */
1395: int subtype; /* subtype */
1396: };
1397: struct kmodel modell[NCOVMAX];
1398:
1.143 brouard 1399: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1400: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1401:
1402: /**************** split *************************/
1403: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1404: {
1405: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1406: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1407: */
1408: char *ss; /* pointer */
1.186 brouard 1409: int l1=0, l2=0; /* length counters */
1.126 brouard 1410:
1411: l1 = strlen(path ); /* length of path */
1412: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1413: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1414: if ( ss == NULL ) { /* no directory, so determine current directory */
1415: strcpy( name, path ); /* we got the fullname name because no directory */
1416: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1417: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1418: /* get current working directory */
1419: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1420: #ifdef WIN32
1421: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1422: #else
1423: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1424: #endif
1.126 brouard 1425: return( GLOCK_ERROR_GETCWD );
1426: }
1427: /* got dirc from getcwd*/
1428: printf(" DIRC = %s \n",dirc);
1.205 brouard 1429: } else { /* strip directory from path */
1.126 brouard 1430: ss++; /* after this, the filename */
1431: l2 = strlen( ss ); /* length of filename */
1432: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1433: strcpy( name, ss ); /* save file name */
1434: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1435: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1436: printf(" DIRC2 = %s \n",dirc);
1437: }
1438: /* We add a separator at the end of dirc if not exists */
1439: l1 = strlen( dirc ); /* length of directory */
1440: if( dirc[l1-1] != DIRSEPARATOR ){
1441: dirc[l1] = DIRSEPARATOR;
1442: dirc[l1+1] = 0;
1443: printf(" DIRC3 = %s \n",dirc);
1444: }
1445: ss = strrchr( name, '.' ); /* find last / */
1446: if (ss >0){
1447: ss++;
1448: strcpy(ext,ss); /* save extension */
1449: l1= strlen( name);
1450: l2= strlen(ss)+1;
1451: strncpy( finame, name, l1-l2);
1452: finame[l1-l2]= 0;
1453: }
1454:
1455: return( 0 ); /* we're done */
1456: }
1457:
1458:
1459: /******************************************/
1460:
1461: void replace_back_to_slash(char *s, char*t)
1462: {
1463: int i;
1464: int lg=0;
1465: i=0;
1466: lg=strlen(t);
1467: for(i=0; i<= lg; i++) {
1468: (s[i] = t[i]);
1469: if (t[i]== '\\') s[i]='/';
1470: }
1471: }
1472:
1.132 brouard 1473: char *trimbb(char *out, char *in)
1.137 brouard 1474: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1475: char *s;
1476: s=out;
1477: while (*in != '\0'){
1.137 brouard 1478: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1479: in++;
1480: }
1481: *out++ = *in++;
1482: }
1483: *out='\0';
1484: return s;
1485: }
1486:
1.187 brouard 1487: /* char *substrchaine(char *out, char *in, char *chain) */
1488: /* { */
1489: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1490: /* char *s, *t; */
1491: /* t=in;s=out; */
1492: /* while ((*in != *chain) && (*in != '\0')){ */
1493: /* *out++ = *in++; */
1494: /* } */
1495:
1496: /* /\* *in matches *chain *\/ */
1497: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1498: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1499: /* } */
1500: /* in--; chain--; */
1501: /* while ( (*in != '\0')){ */
1502: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1503: /* *out++ = *in++; */
1504: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1505: /* } */
1506: /* *out='\0'; */
1507: /* out=s; */
1508: /* return out; */
1509: /* } */
1510: char *substrchaine(char *out, char *in, char *chain)
1511: {
1512: /* Substract chain 'chain' from 'in', return and output 'out' */
1513: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1514:
1515: char *strloc;
1516:
1517: strcpy (out, in);
1518: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1519: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1520: if(strloc != NULL){
1521: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1522: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1523: /* strcpy (strloc, strloc +strlen(chain));*/
1524: }
1525: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1526: return out;
1527: }
1528:
1529:
1.145 brouard 1530: char *cutl(char *blocc, char *alocc, char *in, char occ)
1531: {
1.187 brouard 1532: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1533: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1534: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1535: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1536: */
1.160 brouard 1537: char *s, *t;
1.145 brouard 1538: t=in;s=in;
1539: while ((*in != occ) && (*in != '\0')){
1540: *alocc++ = *in++;
1541: }
1542: if( *in == occ){
1543: *(alocc)='\0';
1544: s=++in;
1545: }
1546:
1547: if (s == t) {/* occ not found */
1548: *(alocc-(in-s))='\0';
1549: in=s;
1550: }
1551: while ( *in != '\0'){
1552: *blocc++ = *in++;
1553: }
1554:
1555: *blocc='\0';
1556: return t;
1557: }
1.137 brouard 1558: char *cutv(char *blocc, char *alocc, char *in, char occ)
1559: {
1.187 brouard 1560: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1561: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1562: gives blocc="abcdef2ghi" and alocc="j".
1563: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1564: */
1565: char *s, *t;
1566: t=in;s=in;
1567: while (*in != '\0'){
1568: while( *in == occ){
1569: *blocc++ = *in++;
1570: s=in;
1571: }
1572: *blocc++ = *in++;
1573: }
1574: if (s == t) /* occ not found */
1575: *(blocc-(in-s))='\0';
1576: else
1577: *(blocc-(in-s)-1)='\0';
1578: in=s;
1579: while ( *in != '\0'){
1580: *alocc++ = *in++;
1581: }
1582:
1583: *alocc='\0';
1584: return s;
1585: }
1586:
1.126 brouard 1587: int nbocc(char *s, char occ)
1588: {
1589: int i,j=0;
1590: int lg=20;
1591: i=0;
1592: lg=strlen(s);
1593: for(i=0; i<= lg; i++) {
1.234 brouard 1594: if (s[i] == occ ) j++;
1.126 brouard 1595: }
1596: return j;
1597: }
1598:
1.137 brouard 1599: /* void cutv(char *u,char *v, char*t, char occ) */
1600: /* { */
1601: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1602: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1603: /* gives u="abcdef2ghi" and v="j" *\/ */
1604: /* int i,lg,j,p=0; */
1605: /* i=0; */
1606: /* lg=strlen(t); */
1607: /* for(j=0; j<=lg-1; j++) { */
1608: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1609: /* } */
1.126 brouard 1610:
1.137 brouard 1611: /* for(j=0; j<p; j++) { */
1612: /* (u[j] = t[j]); */
1613: /* } */
1614: /* u[p]='\0'; */
1.126 brouard 1615:
1.137 brouard 1616: /* for(j=0; j<= lg; j++) { */
1617: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1618: /* } */
1619: /* } */
1.126 brouard 1620:
1.160 brouard 1621: #ifdef _WIN32
1622: char * strsep(char **pp, const char *delim)
1623: {
1624: char *p, *q;
1625:
1626: if ((p = *pp) == NULL)
1627: return 0;
1628: if ((q = strpbrk (p, delim)) != NULL)
1629: {
1630: *pp = q + 1;
1631: *q = '\0';
1632: }
1633: else
1634: *pp = 0;
1635: return p;
1636: }
1637: #endif
1638:
1.126 brouard 1639: /********************** nrerror ********************/
1640:
1641: void nrerror(char error_text[])
1642: {
1643: fprintf(stderr,"ERREUR ...\n");
1644: fprintf(stderr,"%s\n",error_text);
1645: exit(EXIT_FAILURE);
1646: }
1647: /*********************** vector *******************/
1648: double *vector(int nl, int nh)
1649: {
1650: double *v;
1651: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1652: if (!v) nrerror("allocation failure in vector");
1653: return v-nl+NR_END;
1654: }
1655:
1656: /************************ free vector ******************/
1657: void free_vector(double*v, int nl, int nh)
1658: {
1659: free((FREE_ARG)(v+nl-NR_END));
1660: }
1661:
1662: /************************ivector *******************************/
1663: int *ivector(long nl,long nh)
1664: {
1665: int *v;
1666: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1667: if (!v) nrerror("allocation failure in ivector");
1668: return v-nl+NR_END;
1669: }
1670:
1671: /******************free ivector **************************/
1672: void free_ivector(int *v, long nl, long nh)
1673: {
1674: free((FREE_ARG)(v+nl-NR_END));
1675: }
1676:
1677: /************************lvector *******************************/
1678: long *lvector(long nl,long nh)
1679: {
1680: long *v;
1681: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1682: if (!v) nrerror("allocation failure in ivector");
1683: return v-nl+NR_END;
1684: }
1685:
1686: /******************free lvector **************************/
1687: void free_lvector(long *v, long nl, long nh)
1688: {
1689: free((FREE_ARG)(v+nl-NR_END));
1690: }
1691:
1692: /******************* imatrix *******************************/
1693: int **imatrix(long nrl, long nrh, long ncl, long nch)
1694: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1695: {
1696: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1697: int **m;
1698:
1699: /* allocate pointers to rows */
1700: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1701: if (!m) nrerror("allocation failure 1 in matrix()");
1702: m += NR_END;
1703: m -= nrl;
1704:
1705:
1706: /* allocate rows and set pointers to them */
1707: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1708: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1709: m[nrl] += NR_END;
1710: m[nrl] -= ncl;
1711:
1712: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1713:
1714: /* return pointer to array of pointers to rows */
1715: return m;
1716: }
1717:
1718: /****************** free_imatrix *************************/
1719: void free_imatrix(m,nrl,nrh,ncl,nch)
1720: int **m;
1721: long nch,ncl,nrh,nrl;
1722: /* free an int matrix allocated by imatrix() */
1723: {
1724: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1725: free((FREE_ARG) (m+nrl-NR_END));
1726: }
1727:
1728: /******************* matrix *******************************/
1729: double **matrix(long nrl, long nrh, long ncl, long nch)
1730: {
1731: long i, nrow=nrh-nrl+1, ncol=nch-ncl+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: return m;
1.145 brouard 1746: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1747: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1748: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1749: */
1750: }
1751:
1752: /*************************free matrix ************************/
1753: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1754: {
1755: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1756: free((FREE_ARG)(m+nrl-NR_END));
1757: }
1758:
1759: /******************* ma3x *******************************/
1760: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1761: {
1762: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1763: double ***m;
1764:
1765: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1766: if (!m) nrerror("allocation failure 1 in matrix()");
1767: m += NR_END;
1768: m -= nrl;
1769:
1770: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1771: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1772: m[nrl] += NR_END;
1773: m[nrl] -= ncl;
1774:
1775: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1776:
1777: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1778: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1779: m[nrl][ncl] += NR_END;
1780: m[nrl][ncl] -= nll;
1781: for (j=ncl+1; j<=nch; j++)
1782: m[nrl][j]=m[nrl][j-1]+nlay;
1783:
1784: for (i=nrl+1; i<=nrh; i++) {
1785: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1786: for (j=ncl+1; j<=nch; j++)
1787: m[i][j]=m[i][j-1]+nlay;
1788: }
1789: return m;
1790: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1791: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1792: */
1793: }
1794:
1795: /*************************free ma3x ************************/
1796: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1797: {
1798: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1799: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1800: free((FREE_ARG)(m+nrl-NR_END));
1801: }
1802:
1803: /*************** function subdirf ***********/
1804: char *subdirf(char fileres[])
1805: {
1806: /* Caution optionfilefiname is hidden */
1807: strcpy(tmpout,optionfilefiname);
1808: strcat(tmpout,"/"); /* Add to the right */
1809: strcat(tmpout,fileres);
1810: return tmpout;
1811: }
1812:
1813: /*************** function subdirf2 ***********/
1814: char *subdirf2(char fileres[], char *preop)
1815: {
1816:
1817: /* Caution optionfilefiname is hidden */
1818: strcpy(tmpout,optionfilefiname);
1819: strcat(tmpout,"/");
1820: strcat(tmpout,preop);
1821: strcat(tmpout,fileres);
1822: return tmpout;
1823: }
1824:
1825: /*************** function subdirf3 ***********/
1826: char *subdirf3(char fileres[], char *preop, char *preop2)
1827: {
1828:
1829: /* Caution optionfilefiname is hidden */
1830: strcpy(tmpout,optionfilefiname);
1831: strcat(tmpout,"/");
1832: strcat(tmpout,preop);
1833: strcat(tmpout,preop2);
1834: strcat(tmpout,fileres);
1835: return tmpout;
1836: }
1.213 brouard 1837:
1838: /*************** function subdirfext ***********/
1839: char *subdirfext(char fileres[], char *preop, char *postop)
1840: {
1841:
1842: strcpy(tmpout,preop);
1843: strcat(tmpout,fileres);
1844: strcat(tmpout,postop);
1845: return tmpout;
1846: }
1.126 brouard 1847:
1.213 brouard 1848: /*************** function subdirfext3 ***********/
1849: char *subdirfext3(char fileres[], char *preop, char *postop)
1850: {
1851:
1852: /* Caution optionfilefiname is hidden */
1853: strcpy(tmpout,optionfilefiname);
1854: strcat(tmpout,"/");
1855: strcat(tmpout,preop);
1856: strcat(tmpout,fileres);
1857: strcat(tmpout,postop);
1858: return tmpout;
1859: }
1860:
1.162 brouard 1861: char *asc_diff_time(long time_sec, char ascdiff[])
1862: {
1863: long sec_left, days, hours, minutes;
1864: days = (time_sec) / (60*60*24);
1865: sec_left = (time_sec) % (60*60*24);
1866: hours = (sec_left) / (60*60) ;
1867: sec_left = (sec_left) %(60*60);
1868: minutes = (sec_left) /60;
1869: sec_left = (sec_left) % (60);
1870: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1871: return ascdiff;
1872: }
1873:
1.126 brouard 1874: /***************** f1dim *************************/
1875: extern int ncom;
1876: extern double *pcom,*xicom;
1877: extern double (*nrfunc)(double []);
1878:
1879: double f1dim(double x)
1880: {
1881: int j;
1882: double f;
1883: double *xt;
1884:
1885: xt=vector(1,ncom);
1886: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1887: f=(*nrfunc)(xt);
1888: free_vector(xt,1,ncom);
1889: return f;
1890: }
1891:
1892: /*****************brent *************************/
1893: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1894: {
1895: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1896: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1897: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1898: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1899: * returned function value.
1900: */
1.126 brouard 1901: int iter;
1902: double a,b,d,etemp;
1.159 brouard 1903: double fu=0,fv,fw,fx;
1.164 brouard 1904: double ftemp=0.;
1.126 brouard 1905: double p,q,r,tol1,tol2,u,v,w,x,xm;
1906: double e=0.0;
1907:
1908: a=(ax < cx ? ax : cx);
1909: b=(ax > cx ? ax : cx);
1910: x=w=v=bx;
1911: fw=fv=fx=(*f)(x);
1912: for (iter=1;iter<=ITMAX;iter++) {
1913: xm=0.5*(a+b);
1914: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1915: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1916: printf(".");fflush(stdout);
1917: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1918: #ifdef DEBUGBRENT
1.126 brouard 1919: 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);
1920: 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);
1921: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1922: #endif
1923: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1924: *xmin=x;
1925: return fx;
1926: }
1927: ftemp=fu;
1928: if (fabs(e) > tol1) {
1929: r=(x-w)*(fx-fv);
1930: q=(x-v)*(fx-fw);
1931: p=(x-v)*q-(x-w)*r;
1932: q=2.0*(q-r);
1933: if (q > 0.0) p = -p;
1934: q=fabs(q);
1935: etemp=e;
1936: e=d;
1937: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1938: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1939: else {
1.224 brouard 1940: d=p/q;
1941: u=x+d;
1942: if (u-a < tol2 || b-u < tol2)
1943: d=SIGN(tol1,xm-x);
1.126 brouard 1944: }
1945: } else {
1946: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1947: }
1948: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1949: fu=(*f)(u);
1950: if (fu <= fx) {
1951: if (u >= x) a=x; else b=x;
1952: SHFT(v,w,x,u)
1.183 brouard 1953: SHFT(fv,fw,fx,fu)
1954: } else {
1955: if (u < x) a=u; else b=u;
1956: if (fu <= fw || w == x) {
1.224 brouard 1957: v=w;
1958: w=u;
1959: fv=fw;
1960: fw=fu;
1.183 brouard 1961: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1962: v=u;
1963: fv=fu;
1.183 brouard 1964: }
1965: }
1.126 brouard 1966: }
1967: nrerror("Too many iterations in brent");
1968: *xmin=x;
1969: return fx;
1970: }
1971:
1972: /****************** mnbrak ***********************/
1973:
1974: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1975: double (*func)(double))
1.183 brouard 1976: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1977: the downhill direction (defined by the function as evaluated at the initial points) and returns
1978: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1979: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1980: */
1.126 brouard 1981: double ulim,u,r,q, dum;
1982: double fu;
1.187 brouard 1983:
1984: double scale=10.;
1985: int iterscale=0;
1986:
1987: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1988: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1989:
1990:
1991: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1992: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1993: /* *bx = *ax - (*ax - *bx)/scale; */
1994: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1995: /* } */
1996:
1.126 brouard 1997: if (*fb > *fa) {
1998: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1999: SHFT(dum,*fb,*fa,dum)
2000: }
1.126 brouard 2001: *cx=(*bx)+GOLD*(*bx-*ax);
2002: *fc=(*func)(*cx);
1.183 brouard 2003: #ifdef DEBUG
1.224 brouard 2004: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2005: 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 2006: #endif
1.224 brouard 2007: 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 2008: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2009: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2010: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2011: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2012: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2013: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2014: fu=(*func)(u);
1.163 brouard 2015: #ifdef DEBUG
2016: /* f(x)=A(x-u)**2+f(u) */
2017: double A, fparabu;
2018: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2019: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2020: 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);
2021: 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 2022: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2023: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2024: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2025: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2026: #endif
1.184 brouard 2027: #ifdef MNBRAKORIGINAL
1.183 brouard 2028: #else
1.191 brouard 2029: /* if (fu > *fc) { */
2030: /* #ifdef DEBUG */
2031: /* printf("mnbrak4 fu > fc \n"); */
2032: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2033: /* #endif */
2034: /* /\* 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 *\\/ *\/ */
2035: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2036: /* dum=u; /\* Shifting c and u *\/ */
2037: /* u = *cx; */
2038: /* *cx = dum; */
2039: /* dum = fu; */
2040: /* fu = *fc; */
2041: /* *fc =dum; */
2042: /* } else { /\* end *\/ */
2043: /* #ifdef DEBUG */
2044: /* printf("mnbrak3 fu < fc \n"); */
2045: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2046: /* #endif */
2047: /* dum=u; /\* Shifting c and u *\/ */
2048: /* u = *cx; */
2049: /* *cx = dum; */
2050: /* dum = fu; */
2051: /* fu = *fc; */
2052: /* *fc =dum; */
2053: /* } */
1.224 brouard 2054: #ifdef DEBUGMNBRAK
2055: double A, fparabu;
2056: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2057: fparabu= *fa - A*(*ax-u)*(*ax-u);
2058: 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);
2059: 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 2060: #endif
1.191 brouard 2061: dum=u; /* Shifting c and u */
2062: u = *cx;
2063: *cx = dum;
2064: dum = fu;
2065: fu = *fc;
2066: *fc =dum;
1.183 brouard 2067: #endif
1.162 brouard 2068: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2069: #ifdef DEBUG
1.224 brouard 2070: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2071: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2072: #endif
1.126 brouard 2073: fu=(*func)(u);
2074: if (fu < *fc) {
1.183 brouard 2075: #ifdef DEBUG
1.224 brouard 2076: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2077: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2078: #endif
2079: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2080: SHFT(*fb,*fc,fu,(*func)(u))
2081: #ifdef DEBUG
2082: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2083: #endif
2084: }
1.162 brouard 2085: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2086: #ifdef DEBUG
1.224 brouard 2087: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2088: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2089: #endif
1.126 brouard 2090: u=ulim;
2091: fu=(*func)(u);
1.183 brouard 2092: } else { /* u could be left to b (if r > q parabola has a maximum) */
2093: #ifdef DEBUG
1.224 brouard 2094: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2095: 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 2096: #endif
1.126 brouard 2097: u=(*cx)+GOLD*(*cx-*bx);
2098: fu=(*func)(u);
1.224 brouard 2099: #ifdef DEBUG
2100: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2101: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2102: #endif
1.183 brouard 2103: } /* end tests */
1.126 brouard 2104: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2105: SHFT(*fa,*fb,*fc,fu)
2106: #ifdef DEBUG
1.224 brouard 2107: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2108: 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 2109: #endif
2110: } /* 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 2111: }
2112:
2113: /*************** linmin ************************/
1.162 brouard 2114: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2115: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2116: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2117: the value of func at the returned location p . This is actually all accomplished by calling the
2118: routines mnbrak and brent .*/
1.126 brouard 2119: int ncom;
2120: double *pcom,*xicom;
2121: double (*nrfunc)(double []);
2122:
1.224 brouard 2123: #ifdef LINMINORIGINAL
1.126 brouard 2124: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2125: #else
2126: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2127: #endif
1.126 brouard 2128: {
2129: double brent(double ax, double bx, double cx,
2130: double (*f)(double), double tol, double *xmin);
2131: double f1dim(double x);
2132: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2133: double *fc, double (*func)(double));
2134: int j;
2135: double xx,xmin,bx,ax;
2136: double fx,fb,fa;
1.187 brouard 2137:
1.203 brouard 2138: #ifdef LINMINORIGINAL
2139: #else
2140: double scale=10., axs, xxs; /* Scale added for infinity */
2141: #endif
2142:
1.126 brouard 2143: ncom=n;
2144: pcom=vector(1,n);
2145: xicom=vector(1,n);
2146: nrfunc=func;
2147: for (j=1;j<=n;j++) {
2148: pcom[j]=p[j];
1.202 brouard 2149: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2150: }
1.187 brouard 2151:
1.203 brouard 2152: #ifdef LINMINORIGINAL
2153: xx=1.;
2154: #else
2155: axs=0.0;
2156: xxs=1.;
2157: do{
2158: xx= xxs;
2159: #endif
1.187 brouard 2160: ax=0.;
2161: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2162: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2163: /* 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)) */
2164: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2165: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2166: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2167: /* 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 2168: #ifdef LINMINORIGINAL
2169: #else
2170: if (fx != fx){
1.224 brouard 2171: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2172: printf("|");
2173: fprintf(ficlog,"|");
1.203 brouard 2174: #ifdef DEBUGLINMIN
1.224 brouard 2175: 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 2176: #endif
2177: }
1.224 brouard 2178: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2179: #endif
2180:
1.191 brouard 2181: #ifdef DEBUGLINMIN
2182: 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 2183: 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 2184: #endif
1.224 brouard 2185: #ifdef LINMINORIGINAL
2186: #else
2187: if(fb == fx){ /* Flat function in the direction */
2188: xmin=xx;
2189: *flat=1;
2190: }else{
2191: *flat=0;
2192: #endif
2193: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2194: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2195: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2196: /* fmin = f(p[j] + xmin * xi[j]) */
2197: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2198: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2199: #ifdef DEBUG
1.224 brouard 2200: 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);
2201: 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);
2202: #endif
2203: #ifdef LINMINORIGINAL
2204: #else
2205: }
1.126 brouard 2206: #endif
1.191 brouard 2207: #ifdef DEBUGLINMIN
2208: printf("linmin end ");
1.202 brouard 2209: fprintf(ficlog,"linmin end ");
1.191 brouard 2210: #endif
1.126 brouard 2211: for (j=1;j<=n;j++) {
1.203 brouard 2212: #ifdef LINMINORIGINAL
2213: xi[j] *= xmin;
2214: #else
2215: #ifdef DEBUGLINMIN
2216: if(xxs <1.0)
2217: printf(" before xi[%d]=%12.8f", j,xi[j]);
2218: #endif
2219: 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) */
2220: #ifdef DEBUGLINMIN
2221: if(xxs <1.0)
2222: 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 );
2223: #endif
2224: #endif
1.187 brouard 2225: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2226: }
1.191 brouard 2227: #ifdef DEBUGLINMIN
1.203 brouard 2228: printf("\n");
1.191 brouard 2229: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2230: 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 2231: for (j=1;j<=n;j++) {
1.202 brouard 2232: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2233: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2234: if(j % ncovmodel == 0){
1.191 brouard 2235: printf("\n");
1.202 brouard 2236: fprintf(ficlog,"\n");
2237: }
1.191 brouard 2238: }
1.203 brouard 2239: #else
1.191 brouard 2240: #endif
1.126 brouard 2241: free_vector(xicom,1,n);
2242: free_vector(pcom,1,n);
2243: }
2244:
2245:
2246: /*************** powell ************************/
1.162 brouard 2247: /*
2248: Minimization of a function func of n variables. Input consists of an initial starting point
2249: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2250: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2251: such that failure to decrease by more than this amount on one iteration signals doneness. On
2252: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2253: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2254: */
1.224 brouard 2255: #ifdef LINMINORIGINAL
2256: #else
2257: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2258: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2259: #endif
1.126 brouard 2260: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2261: double (*func)(double []))
2262: {
1.224 brouard 2263: #ifdef LINMINORIGINAL
2264: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2265: double (*func)(double []));
1.224 brouard 2266: #else
1.241 brouard 2267: void linmin(double p[], double xi[], int n, double *fret,
2268: double (*func)(double []),int *flat);
1.224 brouard 2269: #endif
1.239 brouard 2270: int i,ibig,j,jk,k;
1.126 brouard 2271: double del,t,*pt,*ptt,*xit;
1.181 brouard 2272: double directest;
1.126 brouard 2273: double fp,fptt;
2274: double *xits;
2275: int niterf, itmp;
1.224 brouard 2276: #ifdef LINMINORIGINAL
2277: #else
2278:
2279: flatdir=ivector(1,n);
2280: for (j=1;j<=n;j++) flatdir[j]=0;
2281: #endif
1.126 brouard 2282:
2283: pt=vector(1,n);
2284: ptt=vector(1,n);
2285: xit=vector(1,n);
2286: xits=vector(1,n);
2287: *fret=(*func)(p);
2288: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2289: rcurr_time = time(NULL);
1.126 brouard 2290: for (*iter=1;;++(*iter)) {
1.187 brouard 2291: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2292: ibig=0;
2293: del=0.0;
1.157 brouard 2294: rlast_time=rcurr_time;
2295: /* (void) gettimeofday(&curr_time,&tzp); */
2296: rcurr_time = time(NULL);
2297: curr_time = *localtime(&rcurr_time);
2298: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2299: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2300: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2301: for (i=1;i<=n;i++) {
1.126 brouard 2302: fprintf(ficrespow," %.12lf", p[i]);
2303: }
1.239 brouard 2304: fprintf(ficrespow,"\n");fflush(ficrespow);
2305: printf("\n#model= 1 + age ");
2306: fprintf(ficlog,"\n#model= 1 + age ");
2307: if(nagesqr==1){
1.241 brouard 2308: printf(" + age*age ");
2309: fprintf(ficlog," + age*age ");
1.239 brouard 2310: }
2311: for(j=1;j <=ncovmodel-2;j++){
2312: if(Typevar[j]==0) {
2313: printf(" + V%d ",Tvar[j]);
2314: fprintf(ficlog," + V%d ",Tvar[j]);
2315: }else if(Typevar[j]==1) {
2316: printf(" + V%d*age ",Tvar[j]);
2317: fprintf(ficlog," + V%d*age ",Tvar[j]);
2318: }else if(Typevar[j]==2) {
2319: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2320: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2321: }
2322: }
1.126 brouard 2323: printf("\n");
1.239 brouard 2324: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2325: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2326: fprintf(ficlog,"\n");
1.239 brouard 2327: for(i=1,jk=1; i <=nlstate; i++){
2328: for(k=1; k <=(nlstate+ndeath); k++){
2329: if (k != i) {
2330: printf("%d%d ",i,k);
2331: fprintf(ficlog,"%d%d ",i,k);
2332: for(j=1; j <=ncovmodel; j++){
2333: printf("%12.7f ",p[jk]);
2334: fprintf(ficlog,"%12.7f ",p[jk]);
2335: jk++;
2336: }
2337: printf("\n");
2338: fprintf(ficlog,"\n");
2339: }
2340: }
2341: }
1.241 brouard 2342: if(*iter <=3 && *iter >1){
1.157 brouard 2343: tml = *localtime(&rcurr_time);
2344: strcpy(strcurr,asctime(&tml));
2345: rforecast_time=rcurr_time;
1.126 brouard 2346: itmp = strlen(strcurr);
2347: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2348: strcurr[itmp-1]='\0';
1.162 brouard 2349: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2350: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2351: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2352: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2353: forecast_time = *localtime(&rforecast_time);
2354: strcpy(strfor,asctime(&forecast_time));
2355: itmp = strlen(strfor);
2356: if(strfor[itmp-1]=='\n')
2357: strfor[itmp-1]='\0';
2358: 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);
2359: 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 2360: }
2361: }
1.187 brouard 2362: for (i=1;i<=n;i++) { /* For each direction i */
2363: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2364: fptt=(*fret);
2365: #ifdef DEBUG
1.203 brouard 2366: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2367: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2368: #endif
1.203 brouard 2369: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2370: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2371: #ifdef LINMINORIGINAL
1.188 brouard 2372: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2373: #else
2374: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2375: flatdir[i]=flat; /* Function is vanishing in that direction i */
2376: #endif
2377: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2378: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2379: /* because that direction will be replaced unless the gain del is small */
2380: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2381: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2382: /* with the new direction. */
2383: del=fabs(fptt-(*fret));
2384: ibig=i;
1.126 brouard 2385: }
2386: #ifdef DEBUG
2387: printf("%d %.12e",i,(*fret));
2388: fprintf(ficlog,"%d %.12e",i,(*fret));
2389: for (j=1;j<=n;j++) {
1.224 brouard 2390: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2391: printf(" x(%d)=%.12e",j,xit[j]);
2392: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2393: }
2394: for(j=1;j<=n;j++) {
1.225 brouard 2395: printf(" p(%d)=%.12e",j,p[j]);
2396: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2397: }
2398: printf("\n");
2399: fprintf(ficlog,"\n");
2400: #endif
1.187 brouard 2401: } /* end loop on each direction i */
2402: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2403: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2404: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2405: for(j=1;j<=n;j++) {
1.225 brouard 2406: if(flatdir[j] >0){
2407: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2408: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2409: }
2410: /* printf("\n"); */
2411: /* fprintf(ficlog,"\n"); */
2412: }
1.243 brouard 2413: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2414: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2415: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2416: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2417: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2418: /* decreased of more than 3.84 */
2419: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2420: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2421: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2422:
1.188 brouard 2423: /* Starting the program with initial values given by a former maximization will simply change */
2424: /* the scales of the directions and the directions, because the are reset to canonical directions */
2425: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2426: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2427: #ifdef DEBUG
2428: int k[2],l;
2429: k[0]=1;
2430: k[1]=-1;
2431: printf("Max: %.12e",(*func)(p));
2432: fprintf(ficlog,"Max: %.12e",(*func)(p));
2433: for (j=1;j<=n;j++) {
2434: printf(" %.12e",p[j]);
2435: fprintf(ficlog," %.12e",p[j]);
2436: }
2437: printf("\n");
2438: fprintf(ficlog,"\n");
2439: for(l=0;l<=1;l++) {
2440: for (j=1;j<=n;j++) {
2441: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2442: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2443: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2444: }
2445: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2446: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2447: }
2448: #endif
2449:
1.224 brouard 2450: #ifdef LINMINORIGINAL
2451: #else
2452: free_ivector(flatdir,1,n);
2453: #endif
1.126 brouard 2454: free_vector(xit,1,n);
2455: free_vector(xits,1,n);
2456: free_vector(ptt,1,n);
2457: free_vector(pt,1,n);
2458: return;
1.192 brouard 2459: } /* enough precision */
1.240 brouard 2460: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2461: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2462: ptt[j]=2.0*p[j]-pt[j];
2463: xit[j]=p[j]-pt[j];
2464: pt[j]=p[j];
2465: }
1.181 brouard 2466: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2467: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2468: if (*iter <=4) {
1.225 brouard 2469: #else
2470: #endif
1.224 brouard 2471: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2472: #else
1.161 brouard 2473: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2474: #endif
1.162 brouard 2475: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2476: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2477: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2478: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2479: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2480: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2481: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2482: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2483: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2484: /* Even if f3 <f1, directest can be negative and t >0 */
2485: /* mu² and del² are equal when f3=f1 */
2486: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2487: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2488: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2489: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2490: #ifdef NRCORIGINAL
2491: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2492: #else
2493: 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 2494: t= t- del*SQR(fp-fptt);
1.183 brouard 2495: #endif
1.202 brouard 2496: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2497: #ifdef DEBUG
1.181 brouard 2498: 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);
2499: 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 2500: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2501: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2502: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2503: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2504: 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);
2505: 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);
2506: #endif
1.183 brouard 2507: #ifdef POWELLORIGINAL
2508: if (t < 0.0) { /* Then we use it for new direction */
2509: #else
1.182 brouard 2510: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2511: 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 2512: 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 2513: 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 2514: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2515: }
1.181 brouard 2516: if (directest < 0.0) { /* Then we use it for new direction */
2517: #endif
1.191 brouard 2518: #ifdef DEBUGLINMIN
1.234 brouard 2519: printf("Before linmin in direction P%d-P0\n",n);
2520: for (j=1;j<=n;j++) {
2521: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2522: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2523: if(j % ncovmodel == 0){
2524: printf("\n");
2525: fprintf(ficlog,"\n");
2526: }
2527: }
1.224 brouard 2528: #endif
2529: #ifdef LINMINORIGINAL
1.234 brouard 2530: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2531: #else
1.234 brouard 2532: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2533: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2534: #endif
1.234 brouard 2535:
1.191 brouard 2536: #ifdef DEBUGLINMIN
1.234 brouard 2537: for (j=1;j<=n;j++) {
2538: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2539: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2540: if(j % ncovmodel == 0){
2541: printf("\n");
2542: fprintf(ficlog,"\n");
2543: }
2544: }
1.224 brouard 2545: #endif
1.234 brouard 2546: for (j=1;j<=n;j++) {
2547: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2548: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2549: }
1.224 brouard 2550: #ifdef LINMINORIGINAL
2551: #else
1.234 brouard 2552: for (j=1, flatd=0;j<=n;j++) {
2553: if(flatdir[j]>0)
2554: flatd++;
2555: }
2556: if(flatd >0){
1.255 brouard 2557: printf("%d flat directions: ",flatd);
2558: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2559: for (j=1;j<=n;j++) {
2560: if(flatdir[j]>0){
2561: printf("%d ",j);
2562: fprintf(ficlog,"%d ",j);
2563: }
2564: }
2565: printf("\n");
2566: fprintf(ficlog,"\n");
2567: }
1.191 brouard 2568: #endif
1.234 brouard 2569: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2570: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2571:
1.126 brouard 2572: #ifdef DEBUG
1.234 brouard 2573: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2574: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2575: for(j=1;j<=n;j++){
2576: printf(" %lf",xit[j]);
2577: fprintf(ficlog," %lf",xit[j]);
2578: }
2579: printf("\n");
2580: fprintf(ficlog,"\n");
1.126 brouard 2581: #endif
1.192 brouard 2582: } /* end of t or directest negative */
1.224 brouard 2583: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2584: #else
1.234 brouard 2585: } /* end if (fptt < fp) */
1.192 brouard 2586: #endif
1.225 brouard 2587: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2588: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2589: #else
1.224 brouard 2590: #endif
1.234 brouard 2591: } /* loop iteration */
1.126 brouard 2592: }
1.234 brouard 2593:
1.126 brouard 2594: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2595:
1.235 brouard 2596: 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 2597: {
1.279 brouard 2598: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2599: * (and selected quantitative values in nres)
2600: * by left multiplying the unit
2601: * matrix by transitions matrix until convergence is reached with precision ftolpl
2602: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2603: * Wx is row vector: population in state 1, population in state 2, population dead
2604: * or prevalence in state 1, prevalence in state 2, 0
2605: * newm is the matrix after multiplications, its rows are identical at a factor.
2606: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2607: * Output is prlim.
2608: * Initial matrix pimij
2609: */
1.206 brouard 2610: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2611: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2612: /* 0, 0 , 1} */
2613: /*
2614: * and after some iteration: */
2615: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2616: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2617: /* 0, 0 , 1} */
2618: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2619: /* {0.51571254859325999, 0.4842874514067399, */
2620: /* 0.51326036147820708, 0.48673963852179264} */
2621: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2622:
1.126 brouard 2623: int i, ii,j,k;
1.209 brouard 2624: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2625: /* double **matprod2(); */ /* test */
1.218 brouard 2626: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2627: double **newm;
1.209 brouard 2628: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2629: int ncvloop=0;
1.288 brouard 2630: int first=0;
1.169 brouard 2631:
1.209 brouard 2632: min=vector(1,nlstate);
2633: max=vector(1,nlstate);
2634: meandiff=vector(1,nlstate);
2635:
1.218 brouard 2636: /* Starting with matrix unity */
1.126 brouard 2637: for (ii=1;ii<=nlstate+ndeath;ii++)
2638: for (j=1;j<=nlstate+ndeath;j++){
2639: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2640: }
1.169 brouard 2641:
2642: cov[1]=1.;
2643:
2644: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2645: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2646: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2647: ncvloop++;
1.126 brouard 2648: newm=savm;
2649: /* Covariates have to be included here again */
1.138 brouard 2650: cov[2]=agefin;
1.187 brouard 2651: if(nagesqr==1)
2652: cov[3]= agefin*agefin;;
1.234 brouard 2653: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2654: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2655: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2656: /* 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 2657: }
2658: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2659: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2660: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2661: /* 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 2662: }
1.237 brouard 2663: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2664: if(Dummy[Tvar[Tage[k]]]){
2665: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2666: } else{
1.235 brouard 2667: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2668: }
1.235 brouard 2669: /* 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 2670: }
1.237 brouard 2671: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2672: /* 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 2673: if(Dummy[Tvard[k][1]==0]){
2674: if(Dummy[Tvard[k][2]==0]){
2675: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2676: }else{
2677: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2678: }
2679: }else{
2680: if(Dummy[Tvard[k][2]==0]){
2681: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2682: }else{
2683: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2684: }
2685: }
1.234 brouard 2686: }
1.138 brouard 2687: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2688: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2689: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2690: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2691: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2692: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2693: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2694:
1.126 brouard 2695: savm=oldm;
2696: oldm=newm;
1.209 brouard 2697:
2698: for(j=1; j<=nlstate; j++){
2699: max[j]=0.;
2700: min[j]=1.;
2701: }
2702: for(i=1;i<=nlstate;i++){
2703: sumnew=0;
2704: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2705: for(j=1; j<=nlstate; j++){
2706: prlim[i][j]= newm[i][j]/(1-sumnew);
2707: max[j]=FMAX(max[j],prlim[i][j]);
2708: min[j]=FMIN(min[j],prlim[i][j]);
2709: }
2710: }
2711:
1.126 brouard 2712: maxmax=0.;
1.209 brouard 2713: for(j=1; j<=nlstate; j++){
2714: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2715: maxmax=FMAX(maxmax,meandiff[j]);
2716: /* 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 2717: } /* j loop */
1.203 brouard 2718: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2719: /* 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 2720: if(maxmax < ftolpl){
1.209 brouard 2721: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2722: free_vector(min,1,nlstate);
2723: free_vector(max,1,nlstate);
2724: free_vector(meandiff,1,nlstate);
1.126 brouard 2725: return prlim;
2726: }
1.288 brouard 2727: } /* agefin loop */
1.208 brouard 2728: /* After some age loop it doesn't converge */
1.288 brouard 2729: if(!first){
2730: first=1;
2731: 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);
2732: }
2733: 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);
2734:
1.209 brouard 2735: /* 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); */
2736: free_vector(min,1,nlstate);
2737: free_vector(max,1,nlstate);
2738: free_vector(meandiff,1,nlstate);
1.208 brouard 2739:
1.169 brouard 2740: return prlim; /* should not reach here */
1.126 brouard 2741: }
2742:
1.217 brouard 2743:
2744: /**** Back Prevalence limit (stable or period prevalence) ****************/
2745:
1.218 brouard 2746: /* 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) */
2747: /* 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 2748: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2749: {
1.264 brouard 2750: /* 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 2751: matrix by transitions matrix until convergence is reached with precision ftolpl */
2752: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2753: /* Wx is row vector: population in state 1, population in state 2, population dead */
2754: /* or prevalence in state 1, prevalence in state 2, 0 */
2755: /* newm is the matrix after multiplications, its rows are identical at a factor */
2756: /* Initial matrix pimij */
2757: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2758: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2759: /* 0, 0 , 1} */
2760: /*
2761: * and after some iteration: */
2762: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2763: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2764: /* 0, 0 , 1} */
2765: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2766: /* {0.51571254859325999, 0.4842874514067399, */
2767: /* 0.51326036147820708, 0.48673963852179264} */
2768: /* If we start from prlim again, prlim tends to a constant matrix */
2769:
2770: int i, ii,j,k;
1.247 brouard 2771: int first=0;
1.217 brouard 2772: double *min, *max, *meandiff, maxmax,sumnew=0.;
2773: /* double **matprod2(); */ /* test */
2774: double **out, cov[NCOVMAX+1], **bmij();
2775: double **newm;
1.218 brouard 2776: double **dnewm, **doldm, **dsavm; /* for use */
2777: double **oldm, **savm; /* for use */
2778:
1.217 brouard 2779: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2780: int ncvloop=0;
2781:
2782: min=vector(1,nlstate);
2783: max=vector(1,nlstate);
2784: meandiff=vector(1,nlstate);
2785:
1.266 brouard 2786: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2787: oldm=oldms; savm=savms;
2788:
2789: /* Starting with matrix unity */
2790: for (ii=1;ii<=nlstate+ndeath;ii++)
2791: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2792: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2793: }
2794:
2795: cov[1]=1.;
2796:
2797: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2798: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2799: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2800: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2801: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2802: ncvloop++;
1.218 brouard 2803: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2804: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2805: /* Covariates have to be included here again */
2806: cov[2]=agefin;
2807: if(nagesqr==1)
2808: cov[3]= agefin*agefin;;
1.242 brouard 2809: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2810: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2811: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2812: /* 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 2813: }
2814: /* for (k=1; k<=cptcovn;k++) { */
2815: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2816: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2817: /* /\* 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])]); *\/ */
2818: /* } */
2819: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2820: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2821: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2822: /* 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]); */
2823: }
2824: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2825: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2826: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2827: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2828: for (k=1; k<=cptcovage;k++){ /* For product with age */
2829: if(Dummy[Tvar[Tage[k]]]){
2830: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2831: } else{
2832: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2833: }
2834: /* 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]); */
2835: }
2836: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2837: /* 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]); */
2838: if(Dummy[Tvard[k][1]==0]){
2839: if(Dummy[Tvard[k][2]==0]){
2840: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2841: }else{
2842: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2843: }
2844: }else{
2845: if(Dummy[Tvard[k][2]==0]){
2846: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2847: }else{
2848: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2849: }
2850: }
1.217 brouard 2851: }
2852:
2853: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2854: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2855: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2856: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2857: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2858: /* ij should be linked to the correct index of cov */
2859: /* age and covariate values ij are in 'cov', but we need to pass
2860: * ij for the observed prevalence at age and status and covariate
2861: * number: prevacurrent[(int)agefin][ii][ij]
2862: */
2863: /* 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 *\/ */
2864: /* 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 *\/ */
2865: 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 2866: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2867: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2868: /* for(i=1; i<=nlstate+ndeath; i++) { */
2869: /* printf("%d newm= ",i); */
2870: /* for(j=1;j<=nlstate+ndeath;j++) { */
2871: /* printf("%f ",newm[i][j]); */
2872: /* } */
2873: /* printf("oldm * "); */
2874: /* for(j=1;j<=nlstate+ndeath;j++) { */
2875: /* printf("%f ",oldm[i][j]); */
2876: /* } */
1.268 brouard 2877: /* printf(" bmmij "); */
1.266 brouard 2878: /* for(j=1;j<=nlstate+ndeath;j++) { */
2879: /* printf("%f ",pmmij[i][j]); */
2880: /* } */
2881: /* printf("\n"); */
2882: /* } */
2883: /* } */
1.217 brouard 2884: savm=oldm;
2885: oldm=newm;
1.266 brouard 2886:
1.217 brouard 2887: for(j=1; j<=nlstate; j++){
2888: max[j]=0.;
2889: min[j]=1.;
2890: }
2891: for(j=1; j<=nlstate; j++){
2892: for(i=1;i<=nlstate;i++){
1.234 brouard 2893: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2894: bprlim[i][j]= newm[i][j];
2895: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2896: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2897: }
2898: }
1.218 brouard 2899:
1.217 brouard 2900: maxmax=0.;
2901: for(i=1; i<=nlstate; i++){
2902: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2903: maxmax=FMAX(maxmax,meandiff[i]);
2904: /* 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 2905: } /* i loop */
1.217 brouard 2906: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2907: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2908: if(maxmax < ftolpl){
1.220 brouard 2909: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2910: free_vector(min,1,nlstate);
2911: free_vector(max,1,nlstate);
2912: free_vector(meandiff,1,nlstate);
2913: return bprlim;
2914: }
1.288 brouard 2915: } /* agefin loop */
1.217 brouard 2916: /* After some age loop it doesn't converge */
1.288 brouard 2917: if(!first){
1.247 brouard 2918: first=1;
2919: 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\
2920: 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);
2921: }
2922: 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 2923: 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);
2924: /* 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); */
2925: free_vector(min,1,nlstate);
2926: free_vector(max,1,nlstate);
2927: free_vector(meandiff,1,nlstate);
2928:
2929: return bprlim; /* should not reach here */
2930: }
2931:
1.126 brouard 2932: /*************** transition probabilities ***************/
2933:
2934: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2935: {
1.138 brouard 2936: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2937: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2938: model to the ncovmodel covariates (including constant and age).
2939: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2940: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2941: ncth covariate in the global vector x is given by the formula:
2942: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2943: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2944: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2945: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2946: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2947: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2948: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2949: */
2950: double s1, lnpijopii;
1.126 brouard 2951: /*double t34;*/
1.164 brouard 2952: int i,j, nc, ii, jj;
1.126 brouard 2953:
1.223 brouard 2954: for(i=1; i<= nlstate; i++){
2955: for(j=1; j<i;j++){
2956: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2957: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2958: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2959: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2960: }
2961: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2962: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2963: }
2964: for(j=i+1; j<=nlstate+ndeath;j++){
2965: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2966: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2967: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2968: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2969: }
2970: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2971: }
2972: }
1.218 brouard 2973:
1.223 brouard 2974: for(i=1; i<= nlstate; i++){
2975: s1=0;
2976: for(j=1; j<i; j++){
2977: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2978: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2979: }
2980: for(j=i+1; j<=nlstate+ndeath; j++){
2981: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2982: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2983: }
2984: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2985: ps[i][i]=1./(s1+1.);
2986: /* Computing other pijs */
2987: for(j=1; j<i; j++)
2988: ps[i][j]= exp(ps[i][j])*ps[i][i];
2989: for(j=i+1; j<=nlstate+ndeath; j++)
2990: ps[i][j]= exp(ps[i][j])*ps[i][i];
2991: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2992: } /* end i */
1.218 brouard 2993:
1.223 brouard 2994: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2995: for(jj=1; jj<= nlstate+ndeath; jj++){
2996: ps[ii][jj]=0;
2997: ps[ii][ii]=1;
2998: }
2999: }
1.294 brouard 3000:
3001:
1.223 brouard 3002: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3003: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3004: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3005: /* } */
3006: /* printf("\n "); */
3007: /* } */
3008: /* printf("\n ");printf("%lf ",cov[2]);*/
3009: /*
3010: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3011: goto end;*/
1.266 brouard 3012: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3013: }
3014:
1.218 brouard 3015: /*************** backward transition probabilities ***************/
3016:
3017: /* 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 ) */
3018: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3019: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3020: {
1.266 brouard 3021: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
3022: * 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 3023: */
1.218 brouard 3024: int i, ii, j,k;
1.222 brouard 3025:
3026: double **out, **pmij();
3027: double sumnew=0.;
1.218 brouard 3028: double agefin;
1.292 brouard 3029: 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 3030: double **dnewm, **dsavm, **doldm;
3031: double **bbmij;
3032:
1.218 brouard 3033: doldm=ddoldms; /* global pointers */
1.222 brouard 3034: dnewm=ddnewms;
3035: dsavm=ddsavms;
3036:
3037: agefin=cov[2];
1.268 brouard 3038: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3039: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3040: the observed prevalence (with this covariate ij) at beginning of transition */
3041: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3042:
3043: /* P_x */
1.266 brouard 3044: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3045: /* outputs pmmij which is a stochastic matrix in row */
3046:
3047: /* Diag(w_x) */
1.292 brouard 3048: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3049: sumnew=0.;
1.269 brouard 3050: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3051: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3052: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3053: sumnew+=prevacurrent[(int)agefin][ii][ij];
3054: }
3055: if(sumnew >0.01){ /* At least some value in the prevalence */
3056: for (ii=1;ii<=nlstate+ndeath;ii++){
3057: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3058: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3059: }
3060: }else{
3061: for (ii=1;ii<=nlstate+ndeath;ii++){
3062: for (j=1;j<=nlstate+ndeath;j++)
3063: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3064: }
3065: /* if(sumnew <0.9){ */
3066: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3067: /* } */
3068: }
3069: k3=0.0; /* We put the last diagonal to 0 */
3070: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3071: doldm[ii][ii]= k3;
3072: }
3073: /* End doldm, At the end doldm is diag[(w_i)] */
3074:
1.292 brouard 3075: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3076: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3077:
1.292 brouard 3078: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3079: /* 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 3080: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3081: sumnew=0.;
1.222 brouard 3082: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3083: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3084: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3085: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3086: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3087: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3088: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3089: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3090: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3091: /* }else */
1.268 brouard 3092: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3093: } /*End ii */
3094: } /* 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 */
3095:
1.292 brouard 3096: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3097: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3098: /* end bmij */
1.266 brouard 3099: return ps; /*pointer is unchanged */
1.218 brouard 3100: }
1.217 brouard 3101: /*************** transition probabilities ***************/
3102:
1.218 brouard 3103: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3104: {
3105: /* According to parameters values stored in x and the covariate's values stored in cov,
3106: computes the probability to be observed in state j being in state i by appying the
3107: model to the ncovmodel covariates (including constant and age).
3108: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3109: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3110: ncth covariate in the global vector x is given by the formula:
3111: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3112: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3113: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3114: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3115: Outputs ps[i][j] the probability to be observed in j being in j according to
3116: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3117: */
3118: double s1, lnpijopii;
3119: /*double t34;*/
3120: int i,j, nc, ii, jj;
3121:
1.234 brouard 3122: for(i=1; i<= nlstate; i++){
3123: for(j=1; j<i;j++){
3124: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3125: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3126: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3127: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3128: }
3129: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3130: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3131: }
3132: for(j=i+1; j<=nlstate+ndeath;j++){
3133: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3134: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3135: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3136: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3137: }
3138: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3139: }
3140: }
3141:
3142: for(i=1; i<= nlstate; i++){
3143: s1=0;
3144: for(j=1; j<i; j++){
3145: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3146: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3147: }
3148: for(j=i+1; j<=nlstate+ndeath; j++){
3149: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3150: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3151: }
3152: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3153: ps[i][i]=1./(s1+1.);
3154: /* Computing other pijs */
3155: for(j=1; j<i; j++)
3156: ps[i][j]= exp(ps[i][j])*ps[i][i];
3157: for(j=i+1; j<=nlstate+ndeath; j++)
3158: ps[i][j]= exp(ps[i][j])*ps[i][i];
3159: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3160: } /* end i */
3161:
3162: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3163: for(jj=1; jj<= nlstate+ndeath; jj++){
3164: ps[ii][jj]=0;
3165: ps[ii][ii]=1;
3166: }
3167: }
1.296 brouard 3168: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3169: for(jj=1; jj<= nlstate+ndeath; jj++){
3170: s1=0.;
3171: for(ii=1; ii<= nlstate+ndeath; ii++){
3172: s1+=ps[ii][jj];
3173: }
3174: for(ii=1; ii<= nlstate; ii++){
3175: ps[ii][jj]=ps[ii][jj]/s1;
3176: }
3177: }
3178: /* Transposition */
3179: for(jj=1; jj<= nlstate+ndeath; jj++){
3180: for(ii=jj; ii<= nlstate+ndeath; ii++){
3181: s1=ps[ii][jj];
3182: ps[ii][jj]=ps[jj][ii];
3183: ps[jj][ii]=s1;
3184: }
3185: }
3186: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3187: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3188: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3189: /* } */
3190: /* printf("\n "); */
3191: /* } */
3192: /* printf("\n ");printf("%lf ",cov[2]);*/
3193: /*
3194: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3195: goto end;*/
3196: return ps;
1.217 brouard 3197: }
3198:
3199:
1.126 brouard 3200: /**************** Product of 2 matrices ******************/
3201:
1.145 brouard 3202: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3203: {
3204: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3205: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3206: /* in, b, out are matrice of pointers which should have been initialized
3207: before: only the contents of out is modified. The function returns
3208: a pointer to pointers identical to out */
1.145 brouard 3209: int i, j, k;
1.126 brouard 3210: for(i=nrl; i<= nrh; i++)
1.145 brouard 3211: for(k=ncolol; k<=ncoloh; k++){
3212: out[i][k]=0.;
3213: for(j=ncl; j<=nch; j++)
3214: out[i][k] +=in[i][j]*b[j][k];
3215: }
1.126 brouard 3216: return out;
3217: }
3218:
3219:
3220: /************* Higher Matrix Product ***************/
3221:
1.235 brouard 3222: 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 3223: {
1.218 brouard 3224: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3225: 'nhstepm*hstepm*stepm' months (i.e. until
3226: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3227: nhstepm*hstepm matrices.
3228: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3229: (typically every 2 years instead of every month which is too big
3230: for the memory).
3231: Model is determined by parameters x and covariates have to be
3232: included manually here.
3233:
3234: */
3235:
3236: int i, j, d, h, k;
1.131 brouard 3237: double **out, cov[NCOVMAX+1];
1.126 brouard 3238: double **newm;
1.187 brouard 3239: double agexact;
1.214 brouard 3240: double agebegin, ageend;
1.126 brouard 3241:
3242: /* Hstepm could be zero and should return the unit matrix */
3243: for (i=1;i<=nlstate+ndeath;i++)
3244: for (j=1;j<=nlstate+ndeath;j++){
3245: oldm[i][j]=(i==j ? 1.0 : 0.0);
3246: po[i][j][0]=(i==j ? 1.0 : 0.0);
3247: }
3248: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3249: for(h=1; h <=nhstepm; h++){
3250: for(d=1; d <=hstepm; d++){
3251: newm=savm;
3252: /* Covariates have to be included here again */
3253: cov[1]=1.;
1.214 brouard 3254: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3255: cov[2]=agexact;
3256: if(nagesqr==1)
1.227 brouard 3257: cov[3]= agexact*agexact;
1.235 brouard 3258: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3259: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3260: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3261: /* 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)); */
3262: }
3263: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3264: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3265: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3266: /* 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]); */
3267: }
3268: for (k=1; k<=cptcovage;k++){
3269: if(Dummy[Tvar[Tage[k]]]){
3270: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3271: } else{
3272: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3273: }
3274: /* 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]); */
3275: }
3276: for (k=1; k<=cptcovprod;k++){ /* */
3277: /* 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]); */
3278: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3279: }
3280: /* for (k=1; k<=cptcovn;k++) */
3281: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3282: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3283: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3284: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3285: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3286:
3287:
1.126 brouard 3288: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3289: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3290: /* right multiplication of oldm by the current matrix */
1.126 brouard 3291: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3292: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3293: /* if((int)age == 70){ */
3294: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3295: /* for(i=1; i<=nlstate+ndeath; i++) { */
3296: /* printf("%d pmmij ",i); */
3297: /* for(j=1;j<=nlstate+ndeath;j++) { */
3298: /* printf("%f ",pmmij[i][j]); */
3299: /* } */
3300: /* printf(" oldm "); */
3301: /* for(j=1;j<=nlstate+ndeath;j++) { */
3302: /* printf("%f ",oldm[i][j]); */
3303: /* } */
3304: /* printf("\n"); */
3305: /* } */
3306: /* } */
1.126 brouard 3307: savm=oldm;
3308: oldm=newm;
3309: }
3310: for(i=1; i<=nlstate+ndeath; i++)
3311: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3312: po[i][j][h]=newm[i][j];
3313: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3314: }
1.128 brouard 3315: /*printf("h=%d ",h);*/
1.126 brouard 3316: } /* end h */
1.267 brouard 3317: /* printf("\n H=%d \n",h); */
1.126 brouard 3318: return po;
3319: }
3320:
1.217 brouard 3321: /************* Higher Back Matrix Product ***************/
1.218 brouard 3322: /* 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 3323: 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 3324: {
1.266 brouard 3325: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3326: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3327: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3328: nhstepm*hstepm matrices.
3329: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3330: (typically every 2 years instead of every month which is too big
1.217 brouard 3331: for the memory).
1.218 brouard 3332: Model is determined by parameters x and covariates have to be
1.266 brouard 3333: included manually here. Then we use a call to bmij(x and cov)
3334: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3335: */
1.217 brouard 3336:
3337: int i, j, d, h, k;
1.266 brouard 3338: double **out, cov[NCOVMAX+1], **bmij();
3339: double **newm, ***newmm;
1.217 brouard 3340: double agexact;
3341: double agebegin, ageend;
1.222 brouard 3342: double **oldm, **savm;
1.217 brouard 3343:
1.266 brouard 3344: newmm=po; /* To be saved */
3345: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3346: /* Hstepm could be zero and should return the unit matrix */
3347: for (i=1;i<=nlstate+ndeath;i++)
3348: for (j=1;j<=nlstate+ndeath;j++){
3349: oldm[i][j]=(i==j ? 1.0 : 0.0);
3350: po[i][j][0]=(i==j ? 1.0 : 0.0);
3351: }
3352: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3353: for(h=1; h <=nhstepm; h++){
3354: for(d=1; d <=hstepm; d++){
3355: newm=savm;
3356: /* Covariates have to be included here again */
3357: cov[1]=1.;
1.271 brouard 3358: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3359: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3360: cov[2]=agexact;
3361: if(nagesqr==1)
1.222 brouard 3362: cov[3]= agexact*agexact;
1.266 brouard 3363: for (k=1; k<=cptcovn;k++){
3364: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3365: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3366: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3367: /* 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)); */
3368: }
1.267 brouard 3369: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3370: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3371: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3372: /* 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]); */
3373: }
3374: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3375: if(Dummy[Tvar[Tage[k]]]){
3376: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3377: } else{
3378: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3379: }
3380: /* 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]); */
3381: }
3382: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3383: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3384: }
1.217 brouard 3385: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3386: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3387:
1.218 brouard 3388: /* Careful transposed matrix */
1.266 brouard 3389: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3390: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3391: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3392: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3393: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3394: /* if((int)age == 70){ */
3395: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3396: /* for(i=1; i<=nlstate+ndeath; i++) { */
3397: /* printf("%d pmmij ",i); */
3398: /* for(j=1;j<=nlstate+ndeath;j++) { */
3399: /* printf("%f ",pmmij[i][j]); */
3400: /* } */
3401: /* printf(" oldm "); */
3402: /* for(j=1;j<=nlstate+ndeath;j++) { */
3403: /* printf("%f ",oldm[i][j]); */
3404: /* } */
3405: /* printf("\n"); */
3406: /* } */
3407: /* } */
3408: savm=oldm;
3409: oldm=newm;
3410: }
3411: for(i=1; i<=nlstate+ndeath; i++)
3412: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3413: po[i][j][h]=newm[i][j];
1.268 brouard 3414: /* if(h==nhstepm) */
3415: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3416: }
1.268 brouard 3417: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3418: } /* end h */
1.268 brouard 3419: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3420: return po;
3421: }
3422:
3423:
1.162 brouard 3424: #ifdef NLOPT
3425: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3426: double fret;
3427: double *xt;
3428: int j;
3429: myfunc_data *d2 = (myfunc_data *) pd;
3430: /* xt = (p1-1); */
3431: xt=vector(1,n);
3432: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3433:
3434: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3435: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3436: printf("Function = %.12lf ",fret);
3437: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3438: printf("\n");
3439: free_vector(xt,1,n);
3440: return fret;
3441: }
3442: #endif
1.126 brouard 3443:
3444: /*************** log-likelihood *************/
3445: double func( double *x)
3446: {
1.226 brouard 3447: int i, ii, j, k, mi, d, kk;
3448: int ioffset=0;
3449: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3450: double **out;
3451: double lli; /* Individual log likelihood */
3452: int s1, s2;
1.228 brouard 3453: 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 3454: double bbh, survp;
3455: long ipmx;
3456: double agexact;
3457: /*extern weight */
3458: /* We are differentiating ll according to initial status */
3459: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3460: /*for(i=1;i<imx;i++)
3461: printf(" %d\n",s[4][i]);
3462: */
1.162 brouard 3463:
1.226 brouard 3464: ++countcallfunc;
1.162 brouard 3465:
1.226 brouard 3466: cov[1]=1.;
1.126 brouard 3467:
1.226 brouard 3468: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3469: ioffset=0;
1.226 brouard 3470: if(mle==1){
3471: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3472: /* Computes the values of the ncovmodel covariates of the model
3473: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3474: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3475: to be observed in j being in i according to the model.
3476: */
1.243 brouard 3477: ioffset=2+nagesqr ;
1.233 brouard 3478: /* Fixed */
1.234 brouard 3479: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3480: 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)*/
3481: }
1.226 brouard 3482: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3483: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3484: has been calculated etc */
3485: /* For an individual i, wav[i] gives the number of effective waves */
3486: /* We compute the contribution to Likelihood of each effective transition
3487: mw[mi][i] is real wave of the mi th effectve wave */
3488: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3489: s2=s[mw[mi+1][i]][i];
3490: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3491: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3492: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3493: */
3494: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3495: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3496: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3497: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3498: }
3499: for (ii=1;ii<=nlstate+ndeath;ii++)
3500: for (j=1;j<=nlstate+ndeath;j++){
3501: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3502: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3503: }
3504: for(d=0; d<dh[mi][i]; d++){
3505: newm=savm;
3506: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3507: cov[2]=agexact;
3508: if(nagesqr==1)
3509: cov[3]= agexact*agexact; /* Should be changed here */
3510: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3511: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3512: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3513: else
3514: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3515: }
3516: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3517: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3518: savm=oldm;
3519: oldm=newm;
3520: } /* end mult */
3521:
3522: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3523: /* But now since version 0.9 we anticipate for bias at large stepm.
3524: * If stepm is larger than one month (smallest stepm) and if the exact delay
3525: * (in months) between two waves is not a multiple of stepm, we rounded to
3526: * the nearest (and in case of equal distance, to the lowest) interval but now
3527: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3528: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3529: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3530: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3531: * -stepm/2 to stepm/2 .
3532: * For stepm=1 the results are the same as for previous versions of Imach.
3533: * For stepm > 1 the results are less biased than in previous versions.
3534: */
1.234 brouard 3535: s1=s[mw[mi][i]][i];
3536: s2=s[mw[mi+1][i]][i];
3537: bbh=(double)bh[mi][i]/(double)stepm;
3538: /* bias bh is positive if real duration
3539: * is higher than the multiple of stepm and negative otherwise.
3540: */
3541: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3542: if( s2 > nlstate){
3543: /* i.e. if s2 is a death state and if the date of death is known
3544: then the contribution to the likelihood is the probability to
3545: die between last step unit time and current step unit time,
3546: which is also equal to probability to die before dh
3547: minus probability to die before dh-stepm .
3548: In version up to 0.92 likelihood was computed
3549: as if date of death was unknown. Death was treated as any other
3550: health state: the date of the interview describes the actual state
3551: and not the date of a change in health state. The former idea was
3552: to consider that at each interview the state was recorded
3553: (healthy, disable or death) and IMaCh was corrected; but when we
3554: introduced the exact date of death then we should have modified
3555: the contribution of an exact death to the likelihood. This new
3556: contribution is smaller and very dependent of the step unit
3557: stepm. It is no more the probability to die between last interview
3558: and month of death but the probability to survive from last
3559: interview up to one month before death multiplied by the
3560: probability to die within a month. Thanks to Chris
3561: Jackson for correcting this bug. Former versions increased
3562: mortality artificially. The bad side is that we add another loop
3563: which slows down the processing. The difference can be up to 10%
3564: lower mortality.
3565: */
3566: /* If, at the beginning of the maximization mostly, the
3567: cumulative probability or probability to be dead is
3568: constant (ie = 1) over time d, the difference is equal to
3569: 0. out[s1][3] = savm[s1][3]: probability, being at state
3570: s1 at precedent wave, to be dead a month before current
3571: wave is equal to probability, being at state s1 at
3572: precedent wave, to be dead at mont of the current
3573: wave. Then the observed probability (that this person died)
3574: is null according to current estimated parameter. In fact,
3575: it should be very low but not zero otherwise the log go to
3576: infinity.
3577: */
1.183 brouard 3578: /* #ifdef INFINITYORIGINAL */
3579: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3580: /* #else */
3581: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3582: /* lli=log(mytinydouble); */
3583: /* else */
3584: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3585: /* #endif */
1.226 brouard 3586: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3587:
1.226 brouard 3588: } else if ( s2==-1 ) { /* alive */
3589: for (j=1,survp=0. ; j<=nlstate; j++)
3590: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3591: /*survp += out[s1][j]; */
3592: lli= log(survp);
3593: }
3594: else if (s2==-4) {
3595: for (j=3,survp=0. ; j<=nlstate; j++)
3596: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3597: lli= log(survp);
3598: }
3599: else if (s2==-5) {
3600: for (j=1,survp=0. ; j<=2; j++)
3601: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3602: lli= log(survp);
3603: }
3604: else{
3605: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3606: /* 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 */
3607: }
3608: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3609: /*if(lli ==000.0)*/
3610: /*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); */
3611: ipmx +=1;
3612: sw += weight[i];
3613: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3614: /* if (lli < log(mytinydouble)){ */
3615: /* 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); */
3616: /* 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]); */
3617: /* } */
3618: } /* end of wave */
3619: } /* end of individual */
3620: } else if(mle==2){
3621: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3622: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3623: for(mi=1; mi<= wav[i]-1; mi++){
3624: for (ii=1;ii<=nlstate+ndeath;ii++)
3625: for (j=1;j<=nlstate+ndeath;j++){
3626: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3627: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3628: }
3629: for(d=0; d<=dh[mi][i]; d++){
3630: newm=savm;
3631: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3632: cov[2]=agexact;
3633: if(nagesqr==1)
3634: cov[3]= agexact*agexact;
3635: for (kk=1; kk<=cptcovage;kk++) {
3636: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3637: }
3638: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3639: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3640: savm=oldm;
3641: oldm=newm;
3642: } /* end mult */
3643:
3644: s1=s[mw[mi][i]][i];
3645: s2=s[mw[mi+1][i]][i];
3646: bbh=(double)bh[mi][i]/(double)stepm;
3647: 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 */
3648: ipmx +=1;
3649: sw += weight[i];
3650: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3651: } /* end of wave */
3652: } /* end of individual */
3653: } else if(mle==3){ /* exponential inter-extrapolation */
3654: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3655: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3656: for(mi=1; mi<= wav[i]-1; mi++){
3657: for (ii=1;ii<=nlstate+ndeath;ii++)
3658: for (j=1;j<=nlstate+ndeath;j++){
3659: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3660: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3661: }
3662: for(d=0; d<dh[mi][i]; d++){
3663: newm=savm;
3664: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3665: cov[2]=agexact;
3666: if(nagesqr==1)
3667: cov[3]= agexact*agexact;
3668: for (kk=1; kk<=cptcovage;kk++) {
3669: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3670: }
3671: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3672: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3673: savm=oldm;
3674: oldm=newm;
3675: } /* end mult */
3676:
3677: s1=s[mw[mi][i]][i];
3678: s2=s[mw[mi+1][i]][i];
3679: bbh=(double)bh[mi][i]/(double)stepm;
3680: 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 */
3681: ipmx +=1;
3682: sw += weight[i];
3683: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3684: } /* end of wave */
3685: } /* end of individual */
3686: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3687: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3688: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3689: for(mi=1; mi<= wav[i]-1; mi++){
3690: for (ii=1;ii<=nlstate+ndeath;ii++)
3691: for (j=1;j<=nlstate+ndeath;j++){
3692: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3693: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3694: }
3695: for(d=0; d<dh[mi][i]; d++){
3696: newm=savm;
3697: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3698: cov[2]=agexact;
3699: if(nagesqr==1)
3700: cov[3]= agexact*agexact;
3701: for (kk=1; kk<=cptcovage;kk++) {
3702: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3703: }
1.126 brouard 3704:
1.226 brouard 3705: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3706: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3707: savm=oldm;
3708: oldm=newm;
3709: } /* end mult */
3710:
3711: s1=s[mw[mi][i]][i];
3712: s2=s[mw[mi+1][i]][i];
3713: if( s2 > nlstate){
3714: lli=log(out[s1][s2] - savm[s1][s2]);
3715: } else if ( s2==-1 ) { /* alive */
3716: for (j=1,survp=0. ; j<=nlstate; j++)
3717: survp += out[s1][j];
3718: lli= log(survp);
3719: }else{
3720: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3721: }
3722: ipmx +=1;
3723: sw += weight[i];
3724: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3725: /* 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 3726: } /* end of wave */
3727: } /* end of individual */
3728: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3729: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3730: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3731: for(mi=1; mi<= wav[i]-1; mi++){
3732: for (ii=1;ii<=nlstate+ndeath;ii++)
3733: for (j=1;j<=nlstate+ndeath;j++){
3734: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3735: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3736: }
3737: for(d=0; d<dh[mi][i]; d++){
3738: newm=savm;
3739: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3740: cov[2]=agexact;
3741: if(nagesqr==1)
3742: cov[3]= agexact*agexact;
3743: for (kk=1; kk<=cptcovage;kk++) {
3744: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3745: }
1.126 brouard 3746:
1.226 brouard 3747: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3748: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3749: savm=oldm;
3750: oldm=newm;
3751: } /* end mult */
3752:
3753: s1=s[mw[mi][i]][i];
3754: s2=s[mw[mi+1][i]][i];
3755: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3756: ipmx +=1;
3757: sw += weight[i];
3758: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3759: /*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]);*/
3760: } /* end of wave */
3761: } /* end of individual */
3762: } /* End of if */
3763: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3764: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3765: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3766: return -l;
1.126 brouard 3767: }
3768:
3769: /*************** log-likelihood *************/
3770: double funcone( double *x)
3771: {
1.228 brouard 3772: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3773: int i, ii, j, k, mi, d, kk;
1.228 brouard 3774: int ioffset=0;
1.131 brouard 3775: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3776: double **out;
3777: double lli; /* Individual log likelihood */
3778: double llt;
3779: int s1, s2;
1.228 brouard 3780: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3781:
1.126 brouard 3782: double bbh, survp;
1.187 brouard 3783: double agexact;
1.214 brouard 3784: double agebegin, ageend;
1.126 brouard 3785: /*extern weight */
3786: /* We are differentiating ll according to initial status */
3787: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3788: /*for(i=1;i<imx;i++)
3789: printf(" %d\n",s[4][i]);
3790: */
3791: cov[1]=1.;
3792:
3793: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3794: ioffset=0;
3795: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3796: /* ioffset=2+nagesqr+cptcovage; */
3797: ioffset=2+nagesqr;
1.232 brouard 3798: /* Fixed */
1.224 brouard 3799: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3800: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3801: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3802: 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)*/
3803: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3804: /* cov[2+6]=covar[Tvar[6]][i]; */
3805: /* cov[2+6]=covar[2][i]; V2 */
3806: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3807: /* cov[2+7]=covar[Tvar[7]][i]; */
3808: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3809: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3810: /* cov[2+9]=covar[Tvar[9]][i]; */
3811: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3812: }
1.232 brouard 3813: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3814: /* 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?)*\/ */
3815: /* } */
1.231 brouard 3816: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3817: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3818: /* } */
1.225 brouard 3819:
1.233 brouard 3820:
3821: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3822: /* Wave varying (but not age varying) */
3823: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3824: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3825: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3826: }
1.232 brouard 3827: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3828: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3829: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3830: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3831: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3832: /* 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 3833: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3834: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3835: /* /\* 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]); *\/ */
3836: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3837: /* } */
1.126 brouard 3838: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3839: for (j=1;j<=nlstate+ndeath;j++){
3840: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3841: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3842: }
1.214 brouard 3843:
3844: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3845: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3846: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3847: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3848: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3849: and mw[mi+1][i]. dh depends on stepm.*/
3850: newm=savm;
1.247 brouard 3851: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3852: cov[2]=agexact;
3853: if(nagesqr==1)
3854: cov[3]= agexact*agexact;
3855: for (kk=1; kk<=cptcovage;kk++) {
3856: if(!FixedV[Tvar[Tage[kk]]])
3857: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3858: else
3859: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3860: }
3861: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3862: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3863: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3864: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3865: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3866: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3867: savm=oldm;
3868: oldm=newm;
1.126 brouard 3869: } /* end mult */
3870:
3871: s1=s[mw[mi][i]][i];
3872: s2=s[mw[mi+1][i]][i];
1.217 brouard 3873: /* if(s2==-1){ */
1.268 brouard 3874: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3875: /* /\* exit(1); *\/ */
3876: /* } */
1.126 brouard 3877: bbh=(double)bh[mi][i]/(double)stepm;
3878: /* bias is positive if real duration
3879: * is higher than the multiple of stepm and negative otherwise.
3880: */
3881: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3882: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3883: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3884: for (j=1,survp=0. ; j<=nlstate; j++)
3885: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3886: lli= log(survp);
1.126 brouard 3887: }else if (mle==1){
1.242 brouard 3888: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3889: } else if(mle==2){
1.242 brouard 3890: 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 3891: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3892: 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 3893: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3894: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3895: } else{ /* mle=0 back to 1 */
1.242 brouard 3896: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3897: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3898: } /* End of if */
3899: ipmx +=1;
3900: sw += weight[i];
3901: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3902: /*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 3903: if(globpr){
1.246 brouard 3904: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3905: %11.6f %11.6f %11.6f ", \
1.242 brouard 3906: 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 3907: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3908: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3909: llt +=ll[k]*gipmx/gsw;
3910: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3911: }
3912: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3913: }
1.232 brouard 3914: } /* end of wave */
3915: } /* end of individual */
3916: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3917: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3918: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3919: if(globpr==0){ /* First time we count the contributions and weights */
3920: gipmx=ipmx;
3921: gsw=sw;
3922: }
3923: return -l;
1.126 brouard 3924: }
3925:
3926:
3927: /*************** function likelione ***********/
1.292 brouard 3928: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 3929: {
3930: /* This routine should help understanding what is done with
3931: the selection of individuals/waves and
3932: to check the exact contribution to the likelihood.
3933: Plotting could be done.
3934: */
3935: int k;
3936:
3937: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3938: strcpy(fileresilk,"ILK_");
1.202 brouard 3939: strcat(fileresilk,fileresu);
1.126 brouard 3940: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3941: printf("Problem with resultfile: %s\n", fileresilk);
3942: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3943: }
1.214 brouard 3944: 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");
3945: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3946: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3947: for(k=1; k<=nlstate; k++)
3948: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3949: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3950: }
3951:
1.292 brouard 3952: *fretone=(*func)(p);
1.126 brouard 3953: if(*globpri !=0){
3954: fclose(ficresilk);
1.205 brouard 3955: if (mle ==0)
3956: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3957: else if(mle >=1)
3958: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3959: 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 3960: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3961:
3962: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3963: 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 3964: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3965: }
1.207 brouard 3966: 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 3967: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3968: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3969: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3970: fflush(fichtm);
1.205 brouard 3971: }
1.126 brouard 3972: return;
3973: }
3974:
3975:
3976: /*********** Maximum Likelihood Estimation ***************/
3977:
3978: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3979: {
1.165 brouard 3980: int i,j, iter=0;
1.126 brouard 3981: double **xi;
3982: double fret;
3983: double fretone; /* Only one call to likelihood */
3984: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3985:
3986: #ifdef NLOPT
3987: int creturn;
3988: nlopt_opt opt;
3989: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3990: double *lb;
3991: double minf; /* the minimum objective value, upon return */
3992: double * p1; /* Shifted parameters from 0 instead of 1 */
3993: myfunc_data dinst, *d = &dinst;
3994: #endif
3995:
3996:
1.126 brouard 3997: xi=matrix(1,npar,1,npar);
3998: for (i=1;i<=npar;i++)
3999: for (j=1;j<=npar;j++)
4000: xi[i][j]=(i==j ? 1.0 : 0.0);
4001: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4002: strcpy(filerespow,"POW_");
1.126 brouard 4003: strcat(filerespow,fileres);
4004: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4005: printf("Problem with resultfile: %s\n", filerespow);
4006: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4007: }
4008: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4009: for (i=1;i<=nlstate;i++)
4010: for(j=1;j<=nlstate+ndeath;j++)
4011: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4012: fprintf(ficrespow,"\n");
1.162 brouard 4013: #ifdef POWELL
1.126 brouard 4014: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4015: #endif
1.126 brouard 4016:
1.162 brouard 4017: #ifdef NLOPT
4018: #ifdef NEWUOA
4019: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4020: #else
4021: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4022: #endif
4023: lb=vector(0,npar-1);
4024: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4025: nlopt_set_lower_bounds(opt, lb);
4026: nlopt_set_initial_step1(opt, 0.1);
4027:
4028: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4029: d->function = func;
4030: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4031: nlopt_set_min_objective(opt, myfunc, d);
4032: nlopt_set_xtol_rel(opt, ftol);
4033: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4034: printf("nlopt failed! %d\n",creturn);
4035: }
4036: else {
4037: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4038: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4039: iter=1; /* not equal */
4040: }
4041: nlopt_destroy(opt);
4042: #endif
1.126 brouard 4043: free_matrix(xi,1,npar,1,npar);
4044: fclose(ficrespow);
1.203 brouard 4045: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4046: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4047: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4048:
4049: }
4050:
4051: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4052: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4053: {
4054: double **a,**y,*x,pd;
1.203 brouard 4055: /* double **hess; */
1.164 brouard 4056: int i, j;
1.126 brouard 4057: int *indx;
4058:
4059: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4060: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4061: void lubksb(double **a, int npar, int *indx, double b[]) ;
4062: void ludcmp(double **a, int npar, int *indx, double *d) ;
4063: double gompertz(double p[]);
1.203 brouard 4064: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4065:
4066: printf("\nCalculation of the hessian matrix. Wait...\n");
4067: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4068: for (i=1;i<=npar;i++){
1.203 brouard 4069: printf("%d-",i);fflush(stdout);
4070: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4071:
4072: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4073:
4074: /* printf(" %f ",p[i]);
4075: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4076: }
4077:
4078: for (i=1;i<=npar;i++) {
4079: for (j=1;j<=npar;j++) {
4080: if (j>i) {
1.203 brouard 4081: printf(".%d-%d",i,j);fflush(stdout);
4082: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4083: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4084:
4085: hess[j][i]=hess[i][j];
4086: /*printf(" %lf ",hess[i][j]);*/
4087: }
4088: }
4089: }
4090: printf("\n");
4091: fprintf(ficlog,"\n");
4092:
4093: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4094: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4095:
4096: a=matrix(1,npar,1,npar);
4097: y=matrix(1,npar,1,npar);
4098: x=vector(1,npar);
4099: indx=ivector(1,npar);
4100: for (i=1;i<=npar;i++)
4101: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4102: ludcmp(a,npar,indx,&pd);
4103:
4104: for (j=1;j<=npar;j++) {
4105: for (i=1;i<=npar;i++) x[i]=0;
4106: x[j]=1;
4107: lubksb(a,npar,indx,x);
4108: for (i=1;i<=npar;i++){
4109: matcov[i][j]=x[i];
4110: }
4111: }
4112:
4113: printf("\n#Hessian matrix#\n");
4114: fprintf(ficlog,"\n#Hessian matrix#\n");
4115: for (i=1;i<=npar;i++) {
4116: for (j=1;j<=npar;j++) {
1.203 brouard 4117: printf("%.6e ",hess[i][j]);
4118: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4119: }
4120: printf("\n");
4121: fprintf(ficlog,"\n");
4122: }
4123:
1.203 brouard 4124: /* printf("\n#Covariance matrix#\n"); */
4125: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4126: /* for (i=1;i<=npar;i++) { */
4127: /* for (j=1;j<=npar;j++) { */
4128: /* printf("%.6e ",matcov[i][j]); */
4129: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4130: /* } */
4131: /* printf("\n"); */
4132: /* fprintf(ficlog,"\n"); */
4133: /* } */
4134:
1.126 brouard 4135: /* Recompute Inverse */
1.203 brouard 4136: /* for (i=1;i<=npar;i++) */
4137: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4138: /* ludcmp(a,npar,indx,&pd); */
4139:
4140: /* printf("\n#Hessian matrix recomputed#\n"); */
4141:
4142: /* for (j=1;j<=npar;j++) { */
4143: /* for (i=1;i<=npar;i++) x[i]=0; */
4144: /* x[j]=1; */
4145: /* lubksb(a,npar,indx,x); */
4146: /* for (i=1;i<=npar;i++){ */
4147: /* y[i][j]=x[i]; */
4148: /* printf("%.3e ",y[i][j]); */
4149: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4150: /* } */
4151: /* printf("\n"); */
4152: /* fprintf(ficlog,"\n"); */
4153: /* } */
4154:
4155: /* Verifying the inverse matrix */
4156: #ifdef DEBUGHESS
4157: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4158:
1.203 brouard 4159: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4160: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4161:
4162: for (j=1;j<=npar;j++) {
4163: for (i=1;i<=npar;i++){
1.203 brouard 4164: printf("%.2f ",y[i][j]);
4165: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4166: }
4167: printf("\n");
4168: fprintf(ficlog,"\n");
4169: }
1.203 brouard 4170: #endif
1.126 brouard 4171:
4172: free_matrix(a,1,npar,1,npar);
4173: free_matrix(y,1,npar,1,npar);
4174: free_vector(x,1,npar);
4175: free_ivector(indx,1,npar);
1.203 brouard 4176: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4177:
4178:
4179: }
4180:
4181: /*************** hessian matrix ****************/
4182: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4183: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4184: int i;
4185: int l=1, lmax=20;
1.203 brouard 4186: double k1,k2, res, fx;
1.132 brouard 4187: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4188: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4189: int k=0,kmax=10;
4190: double l1;
4191:
4192: fx=func(x);
4193: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4194: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4195: l1=pow(10,l);
4196: delts=delt;
4197: for(k=1 ; k <kmax; k=k+1){
4198: delt = delta*(l1*k);
4199: p2[theta]=x[theta] +delt;
1.145 brouard 4200: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4201: p2[theta]=x[theta]-delt;
4202: k2=func(p2)-fx;
4203: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4204: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4205:
1.203 brouard 4206: #ifdef DEBUGHESSII
1.126 brouard 4207: 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);
4208: 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);
4209: #endif
4210: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4211: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4212: k=kmax;
4213: }
4214: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4215: k=kmax; l=lmax*10;
1.126 brouard 4216: }
4217: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4218: delts=delt;
4219: }
1.203 brouard 4220: } /* End loop k */
1.126 brouard 4221: }
4222: delti[theta]=delts;
4223: return res;
4224:
4225: }
4226:
1.203 brouard 4227: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4228: {
4229: int i;
1.164 brouard 4230: int l=1, lmax=20;
1.126 brouard 4231: double k1,k2,k3,k4,res,fx;
1.132 brouard 4232: double p2[MAXPARM+1];
1.203 brouard 4233: int k, kmax=1;
4234: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4235:
4236: int firstime=0;
1.203 brouard 4237:
1.126 brouard 4238: fx=func(x);
1.203 brouard 4239: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4240: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4241: p2[thetai]=x[thetai]+delti[thetai]*k;
4242: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4243: k1=func(p2)-fx;
4244:
1.203 brouard 4245: p2[thetai]=x[thetai]+delti[thetai]*k;
4246: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4247: k2=func(p2)-fx;
4248:
1.203 brouard 4249: p2[thetai]=x[thetai]-delti[thetai]*k;
4250: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4251: k3=func(p2)-fx;
4252:
1.203 brouard 4253: p2[thetai]=x[thetai]-delti[thetai]*k;
4254: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4255: k4=func(p2)-fx;
1.203 brouard 4256: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4257: if(k1*k2*k3*k4 <0.){
1.208 brouard 4258: firstime=1;
1.203 brouard 4259: kmax=kmax+10;
1.208 brouard 4260: }
4261: if(kmax >=10 || firstime ==1){
1.246 brouard 4262: 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);
4263: 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 4264: 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);
4265: 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);
4266: }
4267: #ifdef DEBUGHESSIJ
4268: v1=hess[thetai][thetai];
4269: v2=hess[thetaj][thetaj];
4270: cv12=res;
4271: /* Computing eigen value of Hessian matrix */
4272: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4273: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4274: if ((lc2 <0) || (lc1 <0) ){
4275: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4276: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4277: 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);
4278: 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);
4279: }
1.126 brouard 4280: #endif
4281: }
4282: return res;
4283: }
4284:
1.203 brouard 4285: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4286: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4287: /* { */
4288: /* int i; */
4289: /* int l=1, lmax=20; */
4290: /* double k1,k2,k3,k4,res,fx; */
4291: /* double p2[MAXPARM+1]; */
4292: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4293: /* int k=0,kmax=10; */
4294: /* double l1; */
4295:
4296: /* fx=func(x); */
4297: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4298: /* l1=pow(10,l); */
4299: /* delts=delt; */
4300: /* for(k=1 ; k <kmax; k=k+1){ */
4301: /* delt = delti*(l1*k); */
4302: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4303: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4304: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4305: /* k1=func(p2)-fx; */
4306:
4307: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4308: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4309: /* k2=func(p2)-fx; */
4310:
4311: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4312: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4313: /* k3=func(p2)-fx; */
4314:
4315: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4316: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4317: /* k4=func(p2)-fx; */
4318: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4319: /* #ifdef DEBUGHESSIJ */
4320: /* 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); */
4321: /* 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); */
4322: /* #endif */
4323: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4324: /* k=kmax; */
4325: /* } */
4326: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4327: /* k=kmax; l=lmax*10; */
4328: /* } */
4329: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4330: /* delts=delt; */
4331: /* } */
4332: /* } /\* End loop k *\/ */
4333: /* } */
4334: /* delti[theta]=delts; */
4335: /* return res; */
4336: /* } */
4337:
4338:
1.126 brouard 4339: /************** Inverse of matrix **************/
4340: void ludcmp(double **a, int n, int *indx, double *d)
4341: {
4342: int i,imax,j,k;
4343: double big,dum,sum,temp;
4344: double *vv;
4345:
4346: vv=vector(1,n);
4347: *d=1.0;
4348: for (i=1;i<=n;i++) {
4349: big=0.0;
4350: for (j=1;j<=n;j++)
4351: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4352: if (big == 0.0){
4353: printf(" Singular Hessian matrix at row %d:\n",i);
4354: for (j=1;j<=n;j++) {
4355: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4356: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4357: }
4358: fflush(ficlog);
4359: fclose(ficlog);
4360: nrerror("Singular matrix in routine ludcmp");
4361: }
1.126 brouard 4362: vv[i]=1.0/big;
4363: }
4364: for (j=1;j<=n;j++) {
4365: for (i=1;i<j;i++) {
4366: sum=a[i][j];
4367: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4368: a[i][j]=sum;
4369: }
4370: big=0.0;
4371: for (i=j;i<=n;i++) {
4372: sum=a[i][j];
4373: for (k=1;k<j;k++)
4374: sum -= a[i][k]*a[k][j];
4375: a[i][j]=sum;
4376: if ( (dum=vv[i]*fabs(sum)) >= big) {
4377: big=dum;
4378: imax=i;
4379: }
4380: }
4381: if (j != imax) {
4382: for (k=1;k<=n;k++) {
4383: dum=a[imax][k];
4384: a[imax][k]=a[j][k];
4385: a[j][k]=dum;
4386: }
4387: *d = -(*d);
4388: vv[imax]=vv[j];
4389: }
4390: indx[j]=imax;
4391: if (a[j][j] == 0.0) a[j][j]=TINY;
4392: if (j != n) {
4393: dum=1.0/(a[j][j]);
4394: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4395: }
4396: }
4397: free_vector(vv,1,n); /* Doesn't work */
4398: ;
4399: }
4400:
4401: void lubksb(double **a, int n, int *indx, double b[])
4402: {
4403: int i,ii=0,ip,j;
4404: double sum;
4405:
4406: for (i=1;i<=n;i++) {
4407: ip=indx[i];
4408: sum=b[ip];
4409: b[ip]=b[i];
4410: if (ii)
4411: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4412: else if (sum) ii=i;
4413: b[i]=sum;
4414: }
4415: for (i=n;i>=1;i--) {
4416: sum=b[i];
4417: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4418: b[i]=sum/a[i][i];
4419: }
4420: }
4421:
4422: void pstamp(FILE *fichier)
4423: {
1.196 brouard 4424: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4425: }
4426:
1.297 brouard 4427: void date2dmy(double date,double *day, double *month, double *year){
4428: double yp=0., yp1=0., yp2=0.;
4429:
4430: yp1=modf(date,&yp);/* extracts integral of date in yp and
4431: fractional in yp1 */
4432: *year=yp;
4433: yp2=modf((yp1*12),&yp);
4434: *month=yp;
4435: yp1=modf((yp2*30.5),&yp);
4436: *day=yp;
4437: if(*day==0) *day=1;
4438: if(*month==0) *month=1;
4439: }
4440:
1.253 brouard 4441:
4442:
1.126 brouard 4443: /************ Frequencies ********************/
1.251 brouard 4444: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4445: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4446: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4447: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4448:
1.265 brouard 4449: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4450: int iind=0, iage=0;
4451: int mi; /* Effective wave */
4452: int first;
4453: double ***freq; /* Frequencies */
1.268 brouard 4454: 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 */
4455: 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 4456: double *meanq, *stdq, *idq;
1.226 brouard 4457: double **meanqt;
4458: double *pp, **prop, *posprop, *pospropt;
4459: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4460: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4461: double agebegin, ageend;
4462:
4463: pp=vector(1,nlstate);
1.251 brouard 4464: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4465: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4466: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4467: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4468: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4469: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4470: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4471: meanqt=matrix(1,lastpass,1,nqtveff);
4472: strcpy(fileresp,"P_");
4473: strcat(fileresp,fileresu);
4474: /*strcat(fileresphtm,fileresu);*/
4475: if((ficresp=fopen(fileresp,"w"))==NULL) {
4476: printf("Problem with prevalence resultfile: %s\n", fileresp);
4477: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4478: exit(0);
4479: }
1.240 brouard 4480:
1.226 brouard 4481: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4482: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4483: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4484: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4485: fflush(ficlog);
4486: exit(70);
4487: }
4488: else{
4489: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4490: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4491: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4492: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4493: }
1.237 brouard 4494: 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 4495:
1.226 brouard 4496: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4497: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4498: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4499: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4500: fflush(ficlog);
4501: exit(70);
1.240 brouard 4502: } else{
1.226 brouard 4503: 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 4504: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4505: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4506: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4507: }
1.240 brouard 4508: 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);
4509:
1.253 brouard 4510: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4511: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4512: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4513: j1=0;
1.126 brouard 4514:
1.227 brouard 4515: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4516: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4517: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4518:
4519:
1.226 brouard 4520: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4521: reference=low_education V1=0,V2=0
4522: med_educ V1=1 V2=0,
4523: high_educ V1=0 V2=1
4524: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4525: */
1.249 brouard 4526: dateintsum=0;
4527: k2cpt=0;
4528:
1.253 brouard 4529: if(cptcoveff == 0 )
1.265 brouard 4530: nl=1; /* Constant and age model only */
1.253 brouard 4531: else
4532: nl=2;
1.265 brouard 4533:
4534: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4535: /* Loop on nj=1 or 2 if dummy covariates j!=0
4536: * Loop on j1(1 to 2**cptcoveff) covariate combination
4537: * freq[s1][s2][iage] =0.
4538: * Loop on iind
4539: * ++freq[s1][s2][iage] weighted
4540: * end iind
4541: * if covariate and j!0
4542: * headers Variable on one line
4543: * endif cov j!=0
4544: * header of frequency table by age
4545: * Loop on age
4546: * pp[s1]+=freq[s1][s2][iage] weighted
4547: * pos+=freq[s1][s2][iage] weighted
4548: * Loop on s1 initial state
4549: * fprintf(ficresp
4550: * end s1
4551: * end age
4552: * if j!=0 computes starting values
4553: * end compute starting values
4554: * end j1
4555: * end nl
4556: */
1.253 brouard 4557: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4558: if(nj==1)
4559: j=0; /* First pass for the constant */
1.265 brouard 4560: else{
1.253 brouard 4561: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4562: }
1.251 brouard 4563: first=1;
1.265 brouard 4564: 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 4565: posproptt=0.;
4566: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4567: scanf("%d", i);*/
4568: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4569: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4570: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4571: freq[i][s2][m]=0;
1.251 brouard 4572:
4573: for (i=1; i<=nlstate; i++) {
1.240 brouard 4574: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4575: prop[i][m]=0;
4576: posprop[i]=0;
4577: pospropt[i]=0;
4578: }
1.283 brouard 4579: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4580: idq[z1]=0.;
4581: meanq[z1]=0.;
4582: stdq[z1]=0.;
1.283 brouard 4583: }
4584: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4585: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4586: /* meanqt[m][z1]=0.; */
4587: /* } */
4588: /* } */
1.251 brouard 4589: /* dateintsum=0; */
4590: /* k2cpt=0; */
4591:
1.265 brouard 4592: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4593: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4594: bool=1;
4595: if(j !=0){
4596: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4597: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4598: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4599: /* if(Tvaraff[z1] ==-20){ */
4600: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4601: /* }else if(Tvaraff[z1] ==-10){ */
4602: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4603: /* }else */
4604: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4605: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4606: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4607: /* 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",
4608: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4609: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4610: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4611: } /* Onlyf fixed */
4612: } /* end z1 */
4613: } /* cptcovn > 0 */
4614: } /* end any */
4615: }/* end j==0 */
1.265 brouard 4616: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4617: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4618: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4619: m=mw[mi][iind];
4620: if(j!=0){
4621: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4622: for (z1=1; z1<=cptcoveff; z1++) {
4623: if( Fixed[Tmodelind[z1]]==1){
4624: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4625: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4626: value is -1, we don't select. It differs from the
4627: constant and age model which counts them. */
4628: bool=0; /* not selected */
4629: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4630: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4631: bool=0;
4632: }
4633: }
4634: }
4635: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4636: } /* end j==0 */
4637: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4638: if(bool==1){ /*Selected */
1.251 brouard 4639: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4640: and mw[mi+1][iind]. dh depends on stepm. */
4641: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4642: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4643: if(m >=firstpass && m <=lastpass){
4644: k2=anint[m][iind]+(mint[m][iind]/12.);
4645: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4646: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4647: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4648: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4649: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4650: if (m<lastpass) {
4651: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4652: /* 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]); */
4653: if(s[m][iind]==-1)
4654: 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.));
4655: 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 4656: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean */
4657: idq[z1]=idq[z1]+weight[iind];
4658: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4659: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4660: }
1.251 brouard 4661: /* if((int)agev[m][iind] == 55) */
4662: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4663: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4664: 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 4665: }
1.251 brouard 4666: } /* end if between passes */
4667: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4668: dateintsum=dateintsum+k2; /* on all covariates ?*/
4669: k2cpt++;
4670: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4671: }
1.251 brouard 4672: }else{
4673: bool=1;
4674: }/* end bool 2 */
4675: } /* end m */
1.284 brouard 4676: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4677: /* idq[z1]=idq[z1]+weight[iind]; */
4678: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4679: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4680: /* } */
1.251 brouard 4681: } /* end bool */
4682: } /* end iind = 1 to imx */
4683: /* prop[s][age] is feeded for any initial and valid live state as well as
4684: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4685:
4686:
4687: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4688: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4689: pstamp(ficresp);
1.251 brouard 4690: if (cptcoveff>0 && j!=0){
1.265 brouard 4691: pstamp(ficresp);
1.251 brouard 4692: printf( "\n#********** Variable ");
4693: fprintf(ficresp, "\n#********** Variable ");
4694: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4695: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4696: fprintf(ficlog, "\n#********** Variable ");
4697: for (z1=1; z1<=cptcoveff; z1++){
4698: if(!FixedV[Tvaraff[z1]]){
4699: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4700: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4701: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4702: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4703: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4704: }else{
1.251 brouard 4705: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4706: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4707: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4708: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4709: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4710: }
4711: }
4712: printf( "**********\n#");
4713: fprintf(ficresp, "**********\n#");
4714: fprintf(ficresphtm, "**********</h3>\n");
4715: fprintf(ficresphtmfr, "**********</h3>\n");
4716: fprintf(ficlog, "**********\n");
4717: }
1.284 brouard 4718: /*
4719: Printing means of quantitative variables if any
4720: */
4721: for (z1=1; z1<= nqfveff; z1++) {
1.285 brouard 4722: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.0f individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.284 brouard 4723: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
4724: if(weightopt==1){
4725: printf(" Weighted mean and standard deviation of");
4726: fprintf(ficlog," Weighted mean and standard deviation of");
4727: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4728: }
1.285 brouard 4729: 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]));
4730: 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]));
4731: 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 4732: }
4733: /* for (z1=1; z1<= nqtveff; z1++) { */
4734: /* for(m=1;m<=lastpass;m++){ */
4735: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4736: /* } */
4737: /* } */
1.283 brouard 4738:
1.251 brouard 4739: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4740: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4741: fprintf(ficresp, " Age");
4742: 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 4743: for(i=1; i<=nlstate;i++) {
1.265 brouard 4744: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4745: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4746: }
1.265 brouard 4747: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4748: fprintf(ficresphtm, "\n");
4749:
4750: /* Header of frequency table by age */
4751: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4752: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4753: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4754: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4755: if(s2!=0 && m!=0)
4756: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4757: }
1.226 brouard 4758: }
1.251 brouard 4759: fprintf(ficresphtmfr, "\n");
4760:
4761: /* For each age */
4762: for(iage=iagemin; iage <= iagemax+3; iage++){
4763: fprintf(ficresphtm,"<tr>");
4764: if(iage==iagemax+1){
4765: fprintf(ficlog,"1");
4766: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4767: }else if(iage==iagemax+2){
4768: fprintf(ficlog,"0");
4769: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4770: }else if(iage==iagemax+3){
4771: fprintf(ficlog,"Total");
4772: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4773: }else{
1.240 brouard 4774: if(first==1){
1.251 brouard 4775: first=0;
4776: printf("See log file for details...\n");
4777: }
4778: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4779: fprintf(ficlog,"Age %d", iage);
4780: }
1.265 brouard 4781: for(s1=1; s1 <=nlstate ; s1++){
4782: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4783: pp[s1] += freq[s1][m][iage];
1.251 brouard 4784: }
1.265 brouard 4785: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4786: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4787: pos += freq[s1][m][iage];
4788: if(pp[s1]>=1.e-10){
1.251 brouard 4789: if(first==1){
1.265 brouard 4790: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4791: }
1.265 brouard 4792: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4793: }else{
4794: if(first==1)
1.265 brouard 4795: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4796: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4797: }
4798: }
4799:
1.265 brouard 4800: for(s1=1; s1 <=nlstate ; s1++){
4801: /* posprop[s1]=0; */
4802: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4803: pp[s1] += freq[s1][m][iage];
4804: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4805:
4806: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4807: pos += pp[s1]; /* pos is the total number of transitions until this age */
4808: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4809: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4810: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4811: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4812: }
4813:
4814: /* Writing ficresp */
4815: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4816: if( iage <= iagemax){
4817: fprintf(ficresp," %d",iage);
4818: }
4819: }else if( nj==2){
4820: if( iage <= iagemax){
4821: fprintf(ficresp," %d",iage);
4822: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4823: }
1.240 brouard 4824: }
1.265 brouard 4825: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4826: if(pos>=1.e-5){
1.251 brouard 4827: if(first==1)
1.265 brouard 4828: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4829: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4830: }else{
4831: if(first==1)
1.265 brouard 4832: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4833: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4834: }
4835: if( iage <= iagemax){
4836: if(pos>=1.e-5){
1.265 brouard 4837: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4838: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4839: }else if( nj==2){
4840: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4841: }
4842: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4843: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4844: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4845: } else{
4846: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4847: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4848: }
1.240 brouard 4849: }
1.265 brouard 4850: pospropt[s1] +=posprop[s1];
4851: } /* end loop s1 */
1.251 brouard 4852: /* pospropt=0.; */
1.265 brouard 4853: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4854: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4855: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4856: if(first==1){
1.265 brouard 4857: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4858: }
1.265 brouard 4859: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4860: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4861: }
1.265 brouard 4862: if(s1!=0 && m!=0)
4863: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4864: }
1.265 brouard 4865: } /* end loop s1 */
1.251 brouard 4866: posproptt=0.;
1.265 brouard 4867: for(s1=1; s1 <=nlstate; s1++){
4868: posproptt += pospropt[s1];
1.251 brouard 4869: }
4870: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4871: fprintf(ficresphtm,"</tr>\n");
4872: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4873: if(iage <= iagemax)
4874: fprintf(ficresp,"\n");
1.240 brouard 4875: }
1.251 brouard 4876: if(first==1)
4877: printf("Others in log...\n");
4878: fprintf(ficlog,"\n");
4879: } /* end loop age iage */
1.265 brouard 4880:
1.251 brouard 4881: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4882: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4883: if(posproptt < 1.e-5){
1.265 brouard 4884: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4885: }else{
1.265 brouard 4886: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4887: }
1.226 brouard 4888: }
1.251 brouard 4889: fprintf(ficresphtm,"</tr>\n");
4890: fprintf(ficresphtm,"</table>\n");
4891: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4892: if(posproptt < 1.e-5){
1.251 brouard 4893: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4894: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4895: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4896: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4897: invalidvarcomb[j1]=1;
1.226 brouard 4898: }else{
1.251 brouard 4899: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4900: invalidvarcomb[j1]=0;
1.226 brouard 4901: }
1.251 brouard 4902: fprintf(ficresphtmfr,"</table>\n");
4903: fprintf(ficlog,"\n");
4904: if(j!=0){
4905: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4906: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4907: for(k=1; k <=(nlstate+ndeath); k++){
4908: if (k != i) {
1.265 brouard 4909: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4910: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4911: if(j1==1){ /* All dummy covariates to zero */
4912: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4913: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4914: printf("%d%d ",i,k);
4915: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4916: 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]));
4917: 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]));
4918: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4919: }
1.253 brouard 4920: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4921: for(iage=iagemin; iage <= iagemax+3; iage++){
4922: x[iage]= (double)iage;
4923: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4924: /* 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 4925: }
1.268 brouard 4926: /* Some are not finite, but linreg will ignore these ages */
4927: no=0;
1.253 brouard 4928: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4929: pstart[s1]=b;
4930: pstart[s1-1]=a;
1.252 brouard 4931: }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 */
4932: 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]);
4933: 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 4934: 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 4935: printf("%d%d ",i,k);
4936: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4937: 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 4938: }else{ /* Other cases, like quantitative fixed or varying covariates */
4939: ;
4940: }
4941: /* printf("%12.7f )", param[i][jj][k]); */
4942: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4943: s1++;
1.251 brouard 4944: } /* end jj */
4945: } /* end k!= i */
4946: } /* end k */
1.265 brouard 4947: } /* end i, s1 */
1.251 brouard 4948: } /* end j !=0 */
4949: } /* end selected combination of covariate j1 */
4950: if(j==0){ /* We can estimate starting values from the occurences in each case */
4951: printf("#Freqsummary: Starting values for the constants:\n");
4952: fprintf(ficlog,"\n");
1.265 brouard 4953: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4954: for(k=1; k <=(nlstate+ndeath); k++){
4955: if (k != i) {
4956: printf("%d%d ",i,k);
4957: fprintf(ficlog,"%d%d ",i,k);
4958: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4959: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4960: if(jj==1){ /* Age has to be done */
1.265 brouard 4961: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4962: 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]));
4963: 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 4964: }
4965: /* printf("%12.7f )", param[i][jj][k]); */
4966: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4967: s1++;
1.250 brouard 4968: }
1.251 brouard 4969: printf("\n");
4970: fprintf(ficlog,"\n");
1.250 brouard 4971: }
4972: }
1.284 brouard 4973: } /* end of state i */
1.251 brouard 4974: printf("#Freqsummary\n");
4975: fprintf(ficlog,"\n");
1.265 brouard 4976: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4977: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4978: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4979: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4980: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4981: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4982: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4983: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4984: /* } */
4985: }
1.265 brouard 4986: } /* end loop s1 */
1.251 brouard 4987:
4988: printf("\n");
4989: fprintf(ficlog,"\n");
4990: } /* end j=0 */
1.249 brouard 4991: } /* end j */
1.252 brouard 4992:
1.253 brouard 4993: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4994: for(i=1, jk=1; i <=nlstate; i++){
4995: for(j=1; j <=nlstate+ndeath; j++){
4996: if(j!=i){
4997: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4998: printf("%1d%1d",i,j);
4999: fprintf(ficparo,"%1d%1d",i,j);
5000: for(k=1; k<=ncovmodel;k++){
5001: /* printf(" %lf",param[i][j][k]); */
5002: /* fprintf(ficparo," %lf",param[i][j][k]); */
5003: p[jk]=pstart[jk];
5004: printf(" %f ",pstart[jk]);
5005: fprintf(ficparo," %f ",pstart[jk]);
5006: jk++;
5007: }
5008: printf("\n");
5009: fprintf(ficparo,"\n");
5010: }
5011: }
5012: }
5013: } /* end mle=-2 */
1.226 brouard 5014: dateintmean=dateintsum/k2cpt;
1.296 brouard 5015: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5016:
1.226 brouard 5017: fclose(ficresp);
5018: fclose(ficresphtm);
5019: fclose(ficresphtmfr);
1.283 brouard 5020: free_vector(idq,1,nqfveff);
1.226 brouard 5021: free_vector(meanq,1,nqfveff);
1.284 brouard 5022: free_vector(stdq,1,nqfveff);
1.226 brouard 5023: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5024: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5025: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5026: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5027: free_vector(pospropt,1,nlstate);
5028: free_vector(posprop,1,nlstate);
1.251 brouard 5029: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5030: free_vector(pp,1,nlstate);
5031: /* End of freqsummary */
5032: }
1.126 brouard 5033:
1.268 brouard 5034: /* Simple linear regression */
5035: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5036:
5037: /* y=a+bx regression */
5038: double sumx = 0.0; /* sum of x */
5039: double sumx2 = 0.0; /* sum of x**2 */
5040: double sumxy = 0.0; /* sum of x * y */
5041: double sumy = 0.0; /* sum of y */
5042: double sumy2 = 0.0; /* sum of y**2 */
5043: double sume2 = 0.0; /* sum of square or residuals */
5044: double yhat;
5045:
5046: double denom=0;
5047: int i;
5048: int ne=*no;
5049:
5050: for ( i=ifi, ne=0;i<=ila;i++) {
5051: if(!isfinite(x[i]) || !isfinite(y[i])){
5052: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5053: continue;
5054: }
5055: ne=ne+1;
5056: sumx += x[i];
5057: sumx2 += x[i]*x[i];
5058: sumxy += x[i] * y[i];
5059: sumy += y[i];
5060: sumy2 += y[i]*y[i];
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:
5065: denom = (ne * sumx2 - sumx*sumx);
5066: if (denom == 0) {
5067: // vertical, slope m is infinity
5068: *b = INFINITY;
5069: *a = 0;
5070: if (r) *r = 0;
5071: return 1;
5072: }
5073:
5074: *b = (ne * sumxy - sumx * sumy) / denom;
5075: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5076: if (r!=NULL) {
5077: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5078: sqrt((sumx2 - sumx*sumx/ne) *
5079: (sumy2 - sumy*sumy/ne));
5080: }
5081: *no=ne;
5082: for ( i=ifi, ne=0;i<=ila;i++) {
5083: if(!isfinite(x[i]) || !isfinite(y[i])){
5084: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5085: continue;
5086: }
5087: ne=ne+1;
5088: yhat = y[i] - *a -*b* x[i];
5089: sume2 += yhat * yhat ;
5090:
5091: denom = (ne * sumx2 - sumx*sumx);
5092: /* 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); */
5093: }
5094: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5095: *sa= *sb * sqrt(sumx2/ne);
5096:
5097: return 0;
5098: }
5099:
1.126 brouard 5100: /************ Prevalence ********************/
1.227 brouard 5101: 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)
5102: {
5103: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5104: in each health status at the date of interview (if between dateprev1 and dateprev2).
5105: We still use firstpass and lastpass as another selection.
5106: */
1.126 brouard 5107:
1.227 brouard 5108: int i, m, jk, j1, bool, z1,j, iv;
5109: int mi; /* Effective wave */
5110: int iage;
5111: double agebegin, ageend;
5112:
5113: double **prop;
5114: double posprop;
5115: double y2; /* in fractional years */
5116: int iagemin, iagemax;
5117: int first; /** to stop verbosity which is redirected to log file */
5118:
5119: iagemin= (int) agemin;
5120: iagemax= (int) agemax;
5121: /*pp=vector(1,nlstate);*/
1.251 brouard 5122: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5123: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5124: j1=0;
1.222 brouard 5125:
1.227 brouard 5126: /*j=cptcoveff;*/
5127: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5128:
1.288 brouard 5129: first=0;
1.227 brouard 5130: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5131: for (i=1; i<=nlstate; i++)
1.251 brouard 5132: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5133: prop[i][iage]=0.0;
5134: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5135: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5136: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5137:
5138: for (i=1; i<=imx; i++) { /* Each individual */
5139: bool=1;
5140: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5141: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5142: m=mw[mi][i];
5143: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5144: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5145: for (z1=1; z1<=cptcoveff; z1++){
5146: if( Fixed[Tmodelind[z1]]==1){
5147: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5148: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5149: bool=0;
5150: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5151: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5152: bool=0;
5153: }
5154: }
5155: if(bool==1){ /* Otherwise we skip that wave/person */
5156: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5157: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5158: if(m >=firstpass && m <=lastpass){
5159: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5160: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5161: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5162: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5163: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5164: 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);
5165: exit(1);
5166: }
5167: if (s[m][i]>0 && s[m][i]<=nlstate) {
5168: /*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]]);*/
5169: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5170: prop[s[m][i]][iagemax+3] += weight[i];
5171: } /* end valid statuses */
5172: } /* end selection of dates */
5173: } /* end selection of waves */
5174: } /* end bool */
5175: } /* end wave */
5176: } /* end individual */
5177: for(i=iagemin; i <= iagemax+3; i++){
5178: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5179: posprop += prop[jk][i];
5180: }
5181:
5182: for(jk=1; jk <=nlstate ; jk++){
5183: if( i <= iagemax){
5184: if(posprop>=1.e-5){
5185: probs[i][jk][j1]= prop[jk][i]/posprop;
5186: } else{
1.288 brouard 5187: if(!first){
5188: first=1;
1.266 brouard 5189: 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]);
5190: }else{
1.288 brouard 5191: 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 5192: }
5193: }
5194: }
5195: }/* end jk */
5196: }/* end i */
1.222 brouard 5197: /*} *//* end i1 */
1.227 brouard 5198: } /* end j1 */
1.222 brouard 5199:
1.227 brouard 5200: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5201: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5202: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5203: } /* End of prevalence */
1.126 brouard 5204:
5205: /************* Waves Concatenation ***************/
5206:
5207: 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)
5208: {
1.298 ! brouard 5209: /* Concatenates waves: wav[i] is the number of effective (useful waves in the sense that a non interview is useless) of individual i.
1.126 brouard 5210: Death is a valid wave (if date is known).
5211: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5212: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 ! brouard 5213: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5214: */
1.126 brouard 5215:
1.224 brouard 5216: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5217: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5218: double sum=0., jmean=0.;*/
1.224 brouard 5219: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5220: int j, k=0,jk, ju, jl;
5221: double sum=0.;
5222: first=0;
1.214 brouard 5223: firstwo=0;
1.217 brouard 5224: firsthree=0;
1.218 brouard 5225: firstfour=0;
1.164 brouard 5226: jmin=100000;
1.126 brouard 5227: jmax=-1;
5228: jmean=0.;
1.224 brouard 5229:
5230: /* Treating live states */
1.214 brouard 5231: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5232: mi=0; /* First valid wave */
1.227 brouard 5233: mli=0; /* Last valid wave */
1.126 brouard 5234: m=firstpass;
1.214 brouard 5235: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5236: 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 */
5237: mli=m-1;/* mw[++mi][i]=m-1; */
5238: }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 */
5239: mw[++mi][i]=m;
5240: mli=m;
1.224 brouard 5241: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5242: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5243: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5244: }
1.227 brouard 5245: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5246: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5247: break;
1.224 brouard 5248: #else
1.227 brouard 5249: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5250: if(firsthree == 0){
1.262 brouard 5251: 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 5252: firsthree=1;
5253: }
1.262 brouard 5254: 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 5255: mw[++mi][i]=m;
5256: mli=m;
5257: }
5258: if(s[m][i]==-2){ /* Vital status is really unknown */
5259: nbwarn++;
5260: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5261: 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);
5262: 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);
5263: }
5264: break;
5265: }
5266: break;
1.224 brouard 5267: #endif
1.227 brouard 5268: }/* End m >= lastpass */
1.126 brouard 5269: }/* end while */
1.224 brouard 5270:
1.227 brouard 5271: /* 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 5272: /* After last pass */
1.224 brouard 5273: /* Treating death states */
1.214 brouard 5274: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5275: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5276: /* } */
1.126 brouard 5277: mi++; /* Death is another wave */
5278: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5279: /* Only death is a correct wave */
1.126 brouard 5280: mw[mi][i]=m;
1.257 brouard 5281: } /* else not in a death state */
1.224 brouard 5282: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5283: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5284: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5285: 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 */
5286: nbwarn++;
5287: if(firstfiv==0){
5288: 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 );
5289: firstfiv=1;
5290: }else{
5291: 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 );
5292: }
5293: }else{ /* Death occured afer last wave potential bias */
5294: nberr++;
5295: if(firstwo==0){
1.257 brouard 5296: 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 5297: firstwo=1;
5298: }
1.257 brouard 5299: 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 5300: }
1.257 brouard 5301: }else{ /* if date of interview is unknown */
1.227 brouard 5302: /* death is known but not confirmed by death status at any wave */
5303: if(firstfour==0){
5304: 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 );
5305: firstfour=1;
5306: }
5307: 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 5308: }
1.224 brouard 5309: } /* end if date of death is known */
5310: #endif
5311: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5312: /* wav[i]=mw[mi][i]; */
1.126 brouard 5313: if(mi==0){
5314: nbwarn++;
5315: if(first==0){
1.227 brouard 5316: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5317: first=1;
1.126 brouard 5318: }
5319: if(first==1){
1.227 brouard 5320: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5321: }
5322: } /* end mi==0 */
5323: } /* End individuals */
1.214 brouard 5324: /* wav and mw are no more changed */
1.223 brouard 5325:
1.214 brouard 5326:
1.126 brouard 5327: for(i=1; i<=imx; i++){
5328: for(mi=1; mi<wav[i];mi++){
5329: if (stepm <=0)
1.227 brouard 5330: dh[mi][i]=1;
1.126 brouard 5331: else{
1.260 brouard 5332: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5333: if (agedc[i] < 2*AGESUP) {
5334: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5335: if(j==0) j=1; /* Survives at least one month after exam */
5336: else if(j<0){
5337: nberr++;
5338: 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]);
5339: j=1; /* Temporary Dangerous patch */
5340: 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);
5341: 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]);
5342: 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);
5343: }
5344: k=k+1;
5345: if (j >= jmax){
5346: jmax=j;
5347: ijmax=i;
5348: }
5349: if (j <= jmin){
5350: jmin=j;
5351: ijmin=i;
5352: }
5353: sum=sum+j;
5354: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5355: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5356: }
5357: }
5358: else{
5359: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5360: /* 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 5361:
1.227 brouard 5362: k=k+1;
5363: if (j >= jmax) {
5364: jmax=j;
5365: ijmax=i;
5366: }
5367: else if (j <= jmin){
5368: jmin=j;
5369: ijmin=i;
5370: }
5371: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5372: /*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]);*/
5373: if(j<0){
5374: nberr++;
5375: 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]);
5376: 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]);
5377: }
5378: sum=sum+j;
5379: }
5380: jk= j/stepm;
5381: jl= j -jk*stepm;
5382: ju= j -(jk+1)*stepm;
5383: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5384: if(jl==0){
5385: dh[mi][i]=jk;
5386: bh[mi][i]=0;
5387: }else{ /* We want a negative bias in order to only have interpolation ie
5388: * to avoid the price of an extra matrix product in likelihood */
5389: dh[mi][i]=jk+1;
5390: bh[mi][i]=ju;
5391: }
5392: }else{
5393: if(jl <= -ju){
5394: dh[mi][i]=jk;
5395: bh[mi][i]=jl; /* bias is positive if real duration
5396: * is higher than the multiple of stepm and negative otherwise.
5397: */
5398: }
5399: else{
5400: dh[mi][i]=jk+1;
5401: bh[mi][i]=ju;
5402: }
5403: if(dh[mi][i]==0){
5404: dh[mi][i]=1; /* At least one step */
5405: bh[mi][i]=ju; /* At least one step */
5406: /* 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);*/
5407: }
5408: } /* end if mle */
1.126 brouard 5409: }
5410: } /* end wave */
5411: }
5412: jmean=sum/k;
5413: 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 5414: 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 5415: }
1.126 brouard 5416:
5417: /*********** Tricode ****************************/
1.220 brouard 5418: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5419: {
5420: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5421: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5422: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5423: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5424: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5425: */
1.130 brouard 5426:
1.242 brouard 5427: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5428: int modmaxcovj=0; /* Modality max of covariates j */
5429: int cptcode=0; /* Modality max of covariates j */
5430: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5431:
5432:
1.242 brouard 5433: /* cptcoveff=0; */
5434: /* *cptcov=0; */
1.126 brouard 5435:
1.242 brouard 5436: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5437: for (k=1; k <= maxncov; k++)
5438: for(j=1; j<=2; j++)
5439: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5440:
1.242 brouard 5441: /* Loop on covariates without age and products and no quantitative variable */
5442: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5443: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5444: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5445: switch(Fixed[k]) {
5446: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5447: 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*/
5448: ij=(int)(covar[Tvar[k]][i]);
5449: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5450: * If product of Vn*Vm, still boolean *:
5451: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5452: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5453: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5454: modality of the nth covariate of individual i. */
5455: if (ij > modmaxcovj)
5456: modmaxcovj=ij;
5457: else if (ij < modmincovj)
5458: modmincovj=ij;
1.287 brouard 5459: if (ij <0 || ij >1 ){
5460: printf("Information, IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5461: fprintf(ficlog,"Information, currently IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5462: }
5463: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5464: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5465: exit(1);
5466: }else
5467: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5468: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5469: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5470: /* getting the maximum value of the modality of the covariate
5471: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5472: female ies 1, then modmaxcovj=1.
5473: */
5474: } /* end for loop on individuals i */
5475: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5476: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5477: cptcode=modmaxcovj;
5478: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5479: /*for (i=0; i<=cptcode; i++) {*/
5480: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5481: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5482: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5483: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5484: if( j != -1){
5485: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5486: covariate for which somebody answered excluding
5487: undefined. Usually 2: 0 and 1. */
5488: }
5489: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5490: covariate for which somebody answered including
5491: undefined. Usually 3: -1, 0 and 1. */
5492: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5493: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5494: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5495:
1.242 brouard 5496: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5497: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5498: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5499: /* modmincovj=3; modmaxcovj = 7; */
5500: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5501: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5502: /* defining two dummy variables: variables V1_1 and V1_2.*/
5503: /* nbcode[Tvar[j]][ij]=k; */
5504: /* nbcode[Tvar[j]][1]=0; */
5505: /* nbcode[Tvar[j]][2]=1; */
5506: /* nbcode[Tvar[j]][3]=2; */
5507: /* To be continued (not working yet). */
5508: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5509:
5510: /* 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*/
5511: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5512: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5513: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5514: /*, could be restored in the future */
5515: 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 5516: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5517: break;
5518: }
5519: ij++;
1.287 brouard 5520: 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 5521: cptcode = ij; /* New max modality for covar j */
5522: } /* end of loop on modality i=-1 to 1 or more */
5523: break;
5524: case 1: /* Testing on varying covariate, could be simple and
5525: * should look at waves or product of fixed *
5526: * varying. No time to test -1, assuming 0 and 1 only */
5527: ij=0;
5528: for(i=0; i<=1;i++){
5529: nbcode[Tvar[k]][++ij]=i;
5530: }
5531: break;
5532: default:
5533: break;
5534: } /* end switch */
5535: } /* end dummy test */
1.287 brouard 5536: } /* 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 5537:
5538: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5539: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5540: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5541: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5542: 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 */
5543: 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 */
5544: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5545: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5546:
5547: ij=0;
5548: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5549: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5550: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5551: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5552: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5553: /* If product not in single variable we don't print results */
5554: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5555: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5556: 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*/
5557: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5558: 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 */
5559: if(Fixed[k]!=0)
5560: anyvaryingduminmodel=1;
5561: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5562: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5563: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5564: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5565: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5566: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5567: }
5568: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5569: /* ij--; */
5570: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5571: *cptcov=ij; /*Number of total real effective covariates: effective
5572: * because they can be excluded from the model and real
5573: * if in the model but excluded because missing values, but how to get k from ij?*/
5574: for(j=ij+1; j<= cptcovt; j++){
5575: Tvaraff[j]=0;
5576: Tmodelind[j]=0;
5577: }
5578: for(j=ntveff+1; j<= cptcovt; j++){
5579: TmodelInvind[j]=0;
5580: }
5581: /* To be sorted */
5582: ;
5583: }
1.126 brouard 5584:
1.145 brouard 5585:
1.126 brouard 5586: /*********** Health Expectancies ****************/
5587:
1.235 brouard 5588: 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 5589:
5590: {
5591: /* Health expectancies, no variances */
1.164 brouard 5592: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5593: int nhstepma, nstepma; /* Decreasing with age */
5594: double age, agelim, hf;
5595: double ***p3mat;
5596: double eip;
5597:
1.238 brouard 5598: /* pstamp(ficreseij); */
1.126 brouard 5599: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5600: fprintf(ficreseij,"# Age");
5601: for(i=1; i<=nlstate;i++){
5602: for(j=1; j<=nlstate;j++){
5603: fprintf(ficreseij," e%1d%1d ",i,j);
5604: }
5605: fprintf(ficreseij," e%1d. ",i);
5606: }
5607: fprintf(ficreseij,"\n");
5608:
5609:
5610: if(estepm < stepm){
5611: printf ("Problem %d lower than %d\n",estepm, stepm);
5612: }
5613: else hstepm=estepm;
5614: /* We compute the life expectancy from trapezoids spaced every estepm months
5615: * This is mainly to measure the difference between two models: for example
5616: * if stepm=24 months pijx are given only every 2 years and by summing them
5617: * we are calculating an estimate of the Life Expectancy assuming a linear
5618: * progression in between and thus overestimating or underestimating according
5619: * to the curvature of the survival function. If, for the same date, we
5620: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5621: * to compare the new estimate of Life expectancy with the same linear
5622: * hypothesis. A more precise result, taking into account a more precise
5623: * curvature will be obtained if estepm is as small as stepm. */
5624:
5625: /* For example we decided to compute the life expectancy with the smallest unit */
5626: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5627: nhstepm is the number of hstepm from age to agelim
5628: nstepm is the number of stepm from age to agelin.
1.270 brouard 5629: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5630: and note for a fixed period like estepm months */
5631: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5632: survival function given by stepm (the optimization length). Unfortunately it
5633: means that if the survival funtion is printed only each two years of age and if
5634: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5635: results. So we changed our mind and took the option of the best precision.
5636: */
5637: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5638:
5639: agelim=AGESUP;
5640: /* If stepm=6 months */
5641: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5642: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5643:
5644: /* nhstepm age range expressed in number of stepm */
5645: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5646: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5647: /* if (stepm >= YEARM) hstepm=1;*/
5648: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5649: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5650:
5651: for (age=bage; age<=fage; age ++){
5652: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5653: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5654: /* if (stepm >= YEARM) hstepm=1;*/
5655: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5656:
5657: /* If stepm=6 months */
5658: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5659: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5660:
1.235 brouard 5661: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5662:
5663: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5664:
5665: printf("%d|",(int)age);fflush(stdout);
5666: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5667:
5668: /* Computing expectancies */
5669: for(i=1; i<=nlstate;i++)
5670: for(j=1; j<=nlstate;j++)
5671: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5672: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5673:
5674: /* 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]);*/
5675:
5676: }
5677:
5678: fprintf(ficreseij,"%3.0f",age );
5679: for(i=1; i<=nlstate;i++){
5680: eip=0;
5681: for(j=1; j<=nlstate;j++){
5682: eip +=eij[i][j][(int)age];
5683: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5684: }
5685: fprintf(ficreseij,"%9.4f", eip );
5686: }
5687: fprintf(ficreseij,"\n");
5688:
5689: }
5690: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5691: printf("\n");
5692: fprintf(ficlog,"\n");
5693:
5694: }
5695:
1.235 brouard 5696: 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 5697:
5698: {
5699: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5700: to initial status i, ei. .
1.126 brouard 5701: */
5702: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5703: int nhstepma, nstepma; /* Decreasing with age */
5704: double age, agelim, hf;
5705: double ***p3matp, ***p3matm, ***varhe;
5706: double **dnewm,**doldm;
5707: double *xp, *xm;
5708: double **gp, **gm;
5709: double ***gradg, ***trgradg;
5710: int theta;
5711:
5712: double eip, vip;
5713:
5714: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5715: xp=vector(1,npar);
5716: xm=vector(1,npar);
5717: dnewm=matrix(1,nlstate*nlstate,1,npar);
5718: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5719:
5720: pstamp(ficresstdeij);
5721: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5722: fprintf(ficresstdeij,"# Age");
5723: for(i=1; i<=nlstate;i++){
5724: for(j=1; j<=nlstate;j++)
5725: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5726: fprintf(ficresstdeij," e%1d. ",i);
5727: }
5728: fprintf(ficresstdeij,"\n");
5729:
5730: pstamp(ficrescveij);
5731: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5732: fprintf(ficrescveij,"# Age");
5733: for(i=1; i<=nlstate;i++)
5734: for(j=1; j<=nlstate;j++){
5735: cptj= (j-1)*nlstate+i;
5736: for(i2=1; i2<=nlstate;i2++)
5737: for(j2=1; j2<=nlstate;j2++){
5738: cptj2= (j2-1)*nlstate+i2;
5739: if(cptj2 <= cptj)
5740: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5741: }
5742: }
5743: fprintf(ficrescveij,"\n");
5744:
5745: if(estepm < stepm){
5746: printf ("Problem %d lower than %d\n",estepm, stepm);
5747: }
5748: else hstepm=estepm;
5749: /* We compute the life expectancy from trapezoids spaced every estepm months
5750: * This is mainly to measure the difference between two models: for example
5751: * if stepm=24 months pijx are given only every 2 years and by summing them
5752: * we are calculating an estimate of the Life Expectancy assuming a linear
5753: * progression in between and thus overestimating or underestimating according
5754: * to the curvature of the survival function. If, for the same date, we
5755: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5756: * to compare the new estimate of Life expectancy with the same linear
5757: * hypothesis. A more precise result, taking into account a more precise
5758: * curvature will be obtained if estepm is as small as stepm. */
5759:
5760: /* For example we decided to compute the life expectancy with the smallest unit */
5761: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5762: nhstepm is the number of hstepm from age to agelim
5763: nstepm is the number of stepm from age to agelin.
5764: Look at hpijx to understand the reason of that which relies in memory size
5765: and note for a fixed period like estepm months */
5766: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5767: survival function given by stepm (the optimization length). Unfortunately it
5768: means that if the survival funtion is printed only each two years of age and if
5769: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5770: results. So we changed our mind and took the option of the best precision.
5771: */
5772: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5773:
5774: /* If stepm=6 months */
5775: /* nhstepm age range expressed in number of stepm */
5776: agelim=AGESUP;
5777: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5778: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5779: /* if (stepm >= YEARM) hstepm=1;*/
5780: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5781:
5782: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5783: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5784: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5785: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5786: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5787: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5788:
5789: for (age=bage; age<=fage; age ++){
5790: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5791: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5792: /* if (stepm >= YEARM) hstepm=1;*/
5793: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5794:
1.126 brouard 5795: /* If stepm=6 months */
5796: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5797: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5798:
5799: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5800:
1.126 brouard 5801: /* Computing Variances of health expectancies */
5802: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5803: decrease memory allocation */
5804: for(theta=1; theta <=npar; theta++){
5805: for(i=1; i<=npar; i++){
1.222 brouard 5806: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5807: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5808: }
1.235 brouard 5809: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5810: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5811:
1.126 brouard 5812: for(j=1; j<= nlstate; j++){
1.222 brouard 5813: for(i=1; i<=nlstate; i++){
5814: for(h=0; h<=nhstepm-1; h++){
5815: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5816: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5817: }
5818: }
1.126 brouard 5819: }
1.218 brouard 5820:
1.126 brouard 5821: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5822: for(h=0; h<=nhstepm-1; h++){
5823: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5824: }
1.126 brouard 5825: }/* End theta */
5826:
5827:
5828: for(h=0; h<=nhstepm-1; h++)
5829: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5830: for(theta=1; theta <=npar; theta++)
5831: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5832:
1.218 brouard 5833:
1.222 brouard 5834: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5835: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5836: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5837:
1.222 brouard 5838: printf("%d|",(int)age);fflush(stdout);
5839: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5840: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5841: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5842: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5843: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5844: for(ij=1;ij<=nlstate*nlstate;ij++)
5845: for(ji=1;ji<=nlstate*nlstate;ji++)
5846: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5847: }
5848: }
1.218 brouard 5849:
1.126 brouard 5850: /* Computing expectancies */
1.235 brouard 5851: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5852: for(i=1; i<=nlstate;i++)
5853: for(j=1; j<=nlstate;j++)
1.222 brouard 5854: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5855: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5856:
1.222 brouard 5857: /* 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 5858:
1.222 brouard 5859: }
1.269 brouard 5860:
5861: /* Standard deviation of expectancies ij */
1.126 brouard 5862: fprintf(ficresstdeij,"%3.0f",age );
5863: for(i=1; i<=nlstate;i++){
5864: eip=0.;
5865: vip=0.;
5866: for(j=1; j<=nlstate;j++){
1.222 brouard 5867: eip += eij[i][j][(int)age];
5868: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5869: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5870: 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 5871: }
5872: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5873: }
5874: fprintf(ficresstdeij,"\n");
1.218 brouard 5875:
1.269 brouard 5876: /* Variance of expectancies ij */
1.126 brouard 5877: fprintf(ficrescveij,"%3.0f",age );
5878: for(i=1; i<=nlstate;i++)
5879: for(j=1; j<=nlstate;j++){
1.222 brouard 5880: cptj= (j-1)*nlstate+i;
5881: for(i2=1; i2<=nlstate;i2++)
5882: for(j2=1; j2<=nlstate;j2++){
5883: cptj2= (j2-1)*nlstate+i2;
5884: if(cptj2 <= cptj)
5885: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5886: }
1.126 brouard 5887: }
5888: fprintf(ficrescveij,"\n");
1.218 brouard 5889:
1.126 brouard 5890: }
5891: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5892: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5893: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5894: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5895: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5896: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5897: printf("\n");
5898: fprintf(ficlog,"\n");
1.218 brouard 5899:
1.126 brouard 5900: free_vector(xm,1,npar);
5901: free_vector(xp,1,npar);
5902: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5903: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5904: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5905: }
1.218 brouard 5906:
1.126 brouard 5907: /************ Variance ******************/
1.235 brouard 5908: 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 5909: {
1.279 brouard 5910: /** Variance of health expectancies
5911: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5912: * double **newm;
5913: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5914: */
1.218 brouard 5915:
5916: /* int movingaverage(); */
5917: double **dnewm,**doldm;
5918: double **dnewmp,**doldmp;
5919: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 5920: int first=0;
1.218 brouard 5921: int k;
5922: double *xp;
1.279 brouard 5923: double **gp, **gm; /**< for var eij */
5924: double ***gradg, ***trgradg; /**< for var eij */
5925: double **gradgp, **trgradgp; /**< for var p point j */
5926: double *gpp, *gmp; /**< for var p point j */
5927: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5928: double ***p3mat;
5929: double age,agelim, hf;
5930: /* double ***mobaverage; */
5931: int theta;
5932: char digit[4];
5933: char digitp[25];
5934:
5935: char fileresprobmorprev[FILENAMELENGTH];
5936:
5937: if(popbased==1){
5938: if(mobilav!=0)
5939: strcpy(digitp,"-POPULBASED-MOBILAV_");
5940: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5941: }
5942: else
5943: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5944:
1.218 brouard 5945: /* if (mobilav!=0) { */
5946: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5947: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5948: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5949: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5950: /* } */
5951: /* } */
5952:
5953: strcpy(fileresprobmorprev,"PRMORPREV-");
5954: sprintf(digit,"%-d",ij);
5955: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5956: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5957: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5958: strcat(fileresprobmorprev,fileresu);
5959: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5960: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5961: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5962: }
5963: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5964: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5965: pstamp(ficresprobmorprev);
5966: 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 5967: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5968: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5969: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5970: }
5971: for(j=1;j<=cptcoveff;j++)
5972: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5973: fprintf(ficresprobmorprev,"\n");
5974:
1.218 brouard 5975: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5976: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5977: fprintf(ficresprobmorprev," p.%-d SE",j);
5978: for(i=1; i<=nlstate;i++)
5979: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5980: }
5981: fprintf(ficresprobmorprev,"\n");
5982:
5983: fprintf(ficgp,"\n# Routine varevsij");
5984: fprintf(ficgp,"\nunset title \n");
5985: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5986: 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");
5987: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 5988:
1.218 brouard 5989: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5990: pstamp(ficresvij);
5991: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5992: if(popbased==1)
5993: 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);
5994: else
5995: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5996: fprintf(ficresvij,"# Age");
5997: for(i=1; i<=nlstate;i++)
5998: for(j=1; j<=nlstate;j++)
5999: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6000: fprintf(ficresvij,"\n");
6001:
6002: xp=vector(1,npar);
6003: dnewm=matrix(1,nlstate,1,npar);
6004: doldm=matrix(1,nlstate,1,nlstate);
6005: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6006: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6007:
6008: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6009: gpp=vector(nlstate+1,nlstate+ndeath);
6010: gmp=vector(nlstate+1,nlstate+ndeath);
6011: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6012:
1.218 brouard 6013: if(estepm < stepm){
6014: printf ("Problem %d lower than %d\n",estepm, stepm);
6015: }
6016: else hstepm=estepm;
6017: /* For example we decided to compute the life expectancy with the smallest unit */
6018: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6019: nhstepm is the number of hstepm from age to agelim
6020: nstepm is the number of stepm from age to agelim.
6021: Look at function hpijx to understand why because of memory size limitations,
6022: we decided (b) to get a life expectancy respecting the most precise curvature of the
6023: survival function given by stepm (the optimization length). Unfortunately it
6024: means that if the survival funtion is printed every two years of age and if
6025: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6026: results. So we changed our mind and took the option of the best precision.
6027: */
6028: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6029: agelim = AGESUP;
6030: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6031: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6032: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6033: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6034: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6035: gp=matrix(0,nhstepm,1,nlstate);
6036: gm=matrix(0,nhstepm,1,nlstate);
6037:
6038:
6039: for(theta=1; theta <=npar; theta++){
6040: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6041: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6042: }
1.279 brouard 6043: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6044: * returns into prlim .
1.288 brouard 6045: */
1.242 brouard 6046: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6047:
6048: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6049: if (popbased==1) {
6050: if(mobilav ==0){
6051: for(i=1; i<=nlstate;i++)
6052: prlim[i][i]=probs[(int)age][i][ij];
6053: }else{ /* mobilav */
6054: for(i=1; i<=nlstate;i++)
6055: prlim[i][i]=mobaverage[(int)age][i][ij];
6056: }
6057: }
1.295 brouard 6058: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6059: */
6060: 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 6061: /**< 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 6062: * at horizon h in state j including mortality.
6063: */
1.218 brouard 6064: for(j=1; j<= nlstate; j++){
6065: for(h=0; h<=nhstepm; h++){
6066: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6067: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6068: }
6069: }
1.279 brouard 6070: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6071: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6072: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6073: */
6074: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6075: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6076: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6077: }
6078:
6079: /* Again with minus shift */
1.218 brouard 6080:
6081: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6082: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6083:
1.242 brouard 6084: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6085:
6086: if (popbased==1) {
6087: if(mobilav ==0){
6088: for(i=1; i<=nlstate;i++)
6089: prlim[i][i]=probs[(int)age][i][ij];
6090: }else{ /* mobilav */
6091: for(i=1; i<=nlstate;i++)
6092: prlim[i][i]=mobaverage[(int)age][i][ij];
6093: }
6094: }
6095:
1.235 brouard 6096: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6097:
6098: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6099: for(h=0; h<=nhstepm; h++){
6100: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6101: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6102: }
6103: }
6104: /* This for computing probability of death (h=1 means
6105: computed over hstepm matrices product = hstepm*stepm months)
6106: as a weighted average of prlim.
6107: */
6108: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6109: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6110: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6111: }
1.279 brouard 6112: /* end shifting computations */
6113:
6114: /**< Computing gradient matrix at horizon h
6115: */
1.218 brouard 6116: for(j=1; j<= nlstate; j++) /* vareij */
6117: for(h=0; h<=nhstepm; h++){
6118: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6119: }
1.279 brouard 6120: /**< Gradient of overall mortality p.3 (or p.j)
6121: */
6122: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6123: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6124: }
6125:
6126: } /* End theta */
1.279 brouard 6127:
6128: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6129: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6130:
6131: for(h=0; h<=nhstepm; h++) /* veij */
6132: for(j=1; j<=nlstate;j++)
6133: for(theta=1; theta <=npar; theta++)
6134: trgradg[h][j][theta]=gradg[h][theta][j];
6135:
6136: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6137: for(theta=1; theta <=npar; theta++)
6138: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6139: /**< as well as its transposed matrix
6140: */
1.218 brouard 6141:
6142: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6143: for(i=1;i<=nlstate;i++)
6144: for(j=1;j<=nlstate;j++)
6145: vareij[i][j][(int)age] =0.;
1.279 brouard 6146:
6147: /* Computing trgradg by matcov by gradg at age and summing over h
6148: * and k (nhstepm) formula 15 of article
6149: * Lievre-Brouard-Heathcote
6150: */
6151:
1.218 brouard 6152: for(h=0;h<=nhstepm;h++){
6153: for(k=0;k<=nhstepm;k++){
6154: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6155: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6156: for(i=1;i<=nlstate;i++)
6157: for(j=1;j<=nlstate;j++)
6158: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6159: }
6160: }
6161:
1.279 brouard 6162: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6163: * p.j overall mortality formula 49 but computed directly because
6164: * we compute the grad (wix pijx) instead of grad (pijx),even if
6165: * wix is independent of theta.
6166: */
1.218 brouard 6167: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6168: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6169: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6170: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6171: varppt[j][i]=doldmp[j][i];
6172: /* end ppptj */
6173: /* x centered again */
6174:
1.242 brouard 6175: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6176:
6177: if (popbased==1) {
6178: if(mobilav ==0){
6179: for(i=1; i<=nlstate;i++)
6180: prlim[i][i]=probs[(int)age][i][ij];
6181: }else{ /* mobilav */
6182: for(i=1; i<=nlstate;i++)
6183: prlim[i][i]=mobaverage[(int)age][i][ij];
6184: }
6185: }
6186:
6187: /* This for computing probability of death (h=1 means
6188: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6189: as a weighted average of prlim.
6190: */
1.235 brouard 6191: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6192: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6193: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6194: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6195: }
6196: /* end probability of death */
6197:
6198: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6199: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6200: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6201: for(i=1; i<=nlstate;i++){
6202: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6203: }
6204: }
6205: fprintf(ficresprobmorprev,"\n");
6206:
6207: fprintf(ficresvij,"%.0f ",age );
6208: for(i=1; i<=nlstate;i++)
6209: for(j=1; j<=nlstate;j++){
6210: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6211: }
6212: fprintf(ficresvij,"\n");
6213: free_matrix(gp,0,nhstepm,1,nlstate);
6214: free_matrix(gm,0,nhstepm,1,nlstate);
6215: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6216: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6217: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6218: } /* End age */
6219: free_vector(gpp,nlstate+1,nlstate+ndeath);
6220: free_vector(gmp,nlstate+1,nlstate+ndeath);
6221: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6222: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6223: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6224: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6225: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6226: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6227: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6228: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6229: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6230: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6231: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6232: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6233: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6234: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6235: 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);
6236: /* 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 6237: */
1.218 brouard 6238: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6239: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6240:
1.218 brouard 6241: free_vector(xp,1,npar);
6242: free_matrix(doldm,1,nlstate,1,nlstate);
6243: free_matrix(dnewm,1,nlstate,1,npar);
6244: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6245: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6246: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6247: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6248: fclose(ficresprobmorprev);
6249: fflush(ficgp);
6250: fflush(fichtm);
6251: } /* end varevsij */
1.126 brouard 6252:
6253: /************ Variance of prevlim ******************/
1.269 brouard 6254: 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 6255: {
1.205 brouard 6256: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6257: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6258:
1.268 brouard 6259: double **dnewmpar,**doldm;
1.126 brouard 6260: int i, j, nhstepm, hstepm;
6261: double *xp;
6262: double *gp, *gm;
6263: double **gradg, **trgradg;
1.208 brouard 6264: double **mgm, **mgp;
1.126 brouard 6265: double age,agelim;
6266: int theta;
6267:
6268: pstamp(ficresvpl);
1.288 brouard 6269: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6270: fprintf(ficresvpl,"# Age ");
6271: if(nresult >=1)
6272: fprintf(ficresvpl," Result# ");
1.126 brouard 6273: for(i=1; i<=nlstate;i++)
6274: fprintf(ficresvpl," %1d-%1d",i,i);
6275: fprintf(ficresvpl,"\n");
6276:
6277: xp=vector(1,npar);
1.268 brouard 6278: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6279: doldm=matrix(1,nlstate,1,nlstate);
6280:
6281: hstepm=1*YEARM; /* Every year of age */
6282: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6283: agelim = AGESUP;
6284: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6285: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6286: if (stepm >= YEARM) hstepm=1;
6287: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6288: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6289: mgp=matrix(1,npar,1,nlstate);
6290: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6291: gp=vector(1,nlstate);
6292: gm=vector(1,nlstate);
6293:
6294: for(theta=1; theta <=npar; theta++){
6295: for(i=1; i<=npar; i++){ /* Computes gradient */
6296: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6297: }
1.288 brouard 6298: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6299: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6300: /* else */
6301: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6302: for(i=1;i<=nlstate;i++){
1.126 brouard 6303: gp[i] = prlim[i][i];
1.208 brouard 6304: mgp[theta][i] = prlim[i][i];
6305: }
1.126 brouard 6306: for(i=1; i<=npar; i++) /* Computes gradient */
6307: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6308: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6309: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6310: /* else */
6311: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6312: for(i=1;i<=nlstate;i++){
1.126 brouard 6313: gm[i] = prlim[i][i];
1.208 brouard 6314: mgm[theta][i] = prlim[i][i];
6315: }
1.126 brouard 6316: for(i=1;i<=nlstate;i++)
6317: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6318: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6319: } /* End theta */
6320:
6321: trgradg =matrix(1,nlstate,1,npar);
6322:
6323: for(j=1; j<=nlstate;j++)
6324: for(theta=1; theta <=npar; theta++)
6325: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6326: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6327: /* printf("\nmgm mgp %d ",(int)age); */
6328: /* for(j=1; j<=nlstate;j++){ */
6329: /* printf(" %d ",j); */
6330: /* for(theta=1; theta <=npar; theta++) */
6331: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6332: /* printf("\n "); */
6333: /* } */
6334: /* } */
6335: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6336: /* printf("\n gradg %d ",(int)age); */
6337: /* for(j=1; j<=nlstate;j++){ */
6338: /* printf("%d ",j); */
6339: /* for(theta=1; theta <=npar; theta++) */
6340: /* printf("%d %lf ",theta,gradg[theta][j]); */
6341: /* printf("\n "); */
6342: /* } */
6343: /* } */
1.126 brouard 6344:
6345: for(i=1;i<=nlstate;i++)
6346: varpl[i][(int)age] =0.;
1.209 brouard 6347: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6348: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6349: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6350: }else{
1.268 brouard 6351: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6352: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6353: }
1.126 brouard 6354: for(i=1;i<=nlstate;i++)
6355: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6356:
6357: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6358: if(nresult >=1)
6359: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6360: for(i=1; i<=nlstate;i++){
1.126 brouard 6361: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6362: /* for(j=1;j<=nlstate;j++) */
6363: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6364: }
1.126 brouard 6365: fprintf(ficresvpl,"\n");
6366: free_vector(gp,1,nlstate);
6367: free_vector(gm,1,nlstate);
1.208 brouard 6368: free_matrix(mgm,1,npar,1,nlstate);
6369: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6370: free_matrix(gradg,1,npar,1,nlstate);
6371: free_matrix(trgradg,1,nlstate,1,npar);
6372: } /* End age */
6373:
6374: free_vector(xp,1,npar);
6375: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6376: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6377:
6378: }
6379:
6380:
6381: /************ Variance of backprevalence limit ******************/
1.269 brouard 6382: 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 6383: {
6384: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6385: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6386:
6387: double **dnewmpar,**doldm;
6388: int i, j, nhstepm, hstepm;
6389: double *xp;
6390: double *gp, *gm;
6391: double **gradg, **trgradg;
6392: double **mgm, **mgp;
6393: double age,agelim;
6394: int theta;
6395:
6396: pstamp(ficresvbl);
6397: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6398: fprintf(ficresvbl,"# Age ");
6399: if(nresult >=1)
6400: fprintf(ficresvbl," Result# ");
6401: for(i=1; i<=nlstate;i++)
6402: fprintf(ficresvbl," %1d-%1d",i,i);
6403: fprintf(ficresvbl,"\n");
6404:
6405: xp=vector(1,npar);
6406: dnewmpar=matrix(1,nlstate,1,npar);
6407: doldm=matrix(1,nlstate,1,nlstate);
6408:
6409: hstepm=1*YEARM; /* Every year of age */
6410: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6411: agelim = AGEINF;
6412: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6413: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6414: if (stepm >= YEARM) hstepm=1;
6415: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6416: gradg=matrix(1,npar,1,nlstate);
6417: mgp=matrix(1,npar,1,nlstate);
6418: mgm=matrix(1,npar,1,nlstate);
6419: gp=vector(1,nlstate);
6420: gm=vector(1,nlstate);
6421:
6422: for(theta=1; theta <=npar; theta++){
6423: for(i=1; i<=npar; i++){ /* Computes gradient */
6424: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6425: }
6426: if(mobilavproj > 0 )
6427: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6428: else
6429: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6430: for(i=1;i<=nlstate;i++){
6431: gp[i] = bprlim[i][i];
6432: mgp[theta][i] = bprlim[i][i];
6433: }
6434: for(i=1; i<=npar; i++) /* Computes gradient */
6435: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6436: if(mobilavproj > 0 )
6437: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6438: else
6439: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6440: for(i=1;i<=nlstate;i++){
6441: gm[i] = bprlim[i][i];
6442: mgm[theta][i] = bprlim[i][i];
6443: }
6444: for(i=1;i<=nlstate;i++)
6445: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6446: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6447: } /* End theta */
6448:
6449: trgradg =matrix(1,nlstate,1,npar);
6450:
6451: for(j=1; j<=nlstate;j++)
6452: for(theta=1; theta <=npar; theta++)
6453: trgradg[j][theta]=gradg[theta][j];
6454: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6455: /* printf("\nmgm mgp %d ",(int)age); */
6456: /* for(j=1; j<=nlstate;j++){ */
6457: /* printf(" %d ",j); */
6458: /* for(theta=1; theta <=npar; theta++) */
6459: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6460: /* printf("\n "); */
6461: /* } */
6462: /* } */
6463: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6464: /* printf("\n gradg %d ",(int)age); */
6465: /* for(j=1; j<=nlstate;j++){ */
6466: /* printf("%d ",j); */
6467: /* for(theta=1; theta <=npar; theta++) */
6468: /* printf("%d %lf ",theta,gradg[theta][j]); */
6469: /* printf("\n "); */
6470: /* } */
6471: /* } */
6472:
6473: for(i=1;i<=nlstate;i++)
6474: varbpl[i][(int)age] =0.;
6475: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6476: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6477: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6478: }else{
6479: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6480: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6481: }
6482: for(i=1;i<=nlstate;i++)
6483: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6484:
6485: fprintf(ficresvbl,"%.0f ",age );
6486: if(nresult >=1)
6487: fprintf(ficresvbl,"%d ",nres );
6488: for(i=1; i<=nlstate;i++)
6489: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6490: fprintf(ficresvbl,"\n");
6491: free_vector(gp,1,nlstate);
6492: free_vector(gm,1,nlstate);
6493: free_matrix(mgm,1,npar,1,nlstate);
6494: free_matrix(mgp,1,npar,1,nlstate);
6495: free_matrix(gradg,1,npar,1,nlstate);
6496: free_matrix(trgradg,1,nlstate,1,npar);
6497: } /* End age */
6498:
6499: free_vector(xp,1,npar);
6500: free_matrix(doldm,1,nlstate,1,npar);
6501: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6502:
6503: }
6504:
6505: /************ Variance of one-step probabilities ******************/
6506: 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 6507: {
6508: int i, j=0, k1, l1, tj;
6509: int k2, l2, j1, z1;
6510: int k=0, l;
6511: int first=1, first1, first2;
6512: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6513: double **dnewm,**doldm;
6514: double *xp;
6515: double *gp, *gm;
6516: double **gradg, **trgradg;
6517: double **mu;
6518: double age, cov[NCOVMAX+1];
6519: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6520: int theta;
6521: char fileresprob[FILENAMELENGTH];
6522: char fileresprobcov[FILENAMELENGTH];
6523: char fileresprobcor[FILENAMELENGTH];
6524: double ***varpij;
6525:
6526: strcpy(fileresprob,"PROB_");
6527: strcat(fileresprob,fileres);
6528: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6529: printf("Problem with resultfile: %s\n", fileresprob);
6530: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6531: }
6532: strcpy(fileresprobcov,"PROBCOV_");
6533: strcat(fileresprobcov,fileresu);
6534: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6535: printf("Problem with resultfile: %s\n", fileresprobcov);
6536: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6537: }
6538: strcpy(fileresprobcor,"PROBCOR_");
6539: strcat(fileresprobcor,fileresu);
6540: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6541: printf("Problem with resultfile: %s\n", fileresprobcor);
6542: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6543: }
6544: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6545: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6546: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6547: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6548: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6549: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6550: pstamp(ficresprob);
6551: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6552: fprintf(ficresprob,"# Age");
6553: pstamp(ficresprobcov);
6554: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6555: fprintf(ficresprobcov,"# Age");
6556: pstamp(ficresprobcor);
6557: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6558: fprintf(ficresprobcor,"# Age");
1.126 brouard 6559:
6560:
1.222 brouard 6561: for(i=1; i<=nlstate;i++)
6562: for(j=1; j<=(nlstate+ndeath);j++){
6563: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6564: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6565: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6566: }
6567: /* fprintf(ficresprob,"\n");
6568: fprintf(ficresprobcov,"\n");
6569: fprintf(ficresprobcor,"\n");
6570: */
6571: xp=vector(1,npar);
6572: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6573: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6574: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6575: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6576: first=1;
6577: fprintf(ficgp,"\n# Routine varprob");
6578: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6579: fprintf(fichtm,"\n");
6580:
1.288 brouard 6581: 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 6582: 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);
6583: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6584: and drawn. It helps understanding how is the covariance between two incidences.\
6585: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6586: 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 6587: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6588: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6589: standard deviations wide on each axis. <br>\
6590: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6591: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6592: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6593:
1.222 brouard 6594: cov[1]=1;
6595: /* tj=cptcoveff; */
1.225 brouard 6596: tj = (int) pow(2,cptcoveff);
1.222 brouard 6597: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6598: j1=0;
1.224 brouard 6599: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6600: if (cptcovn>0) {
6601: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6602: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6603: fprintf(ficresprob, "**********\n#\n");
6604: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6605: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6606: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6607:
1.222 brouard 6608: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6609: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6610: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6611:
6612:
1.222 brouard 6613: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6614: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6615: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6616:
1.222 brouard 6617: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6618: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6619: fprintf(ficresprobcor, "**********\n#");
6620: if(invalidvarcomb[j1]){
6621: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6622: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6623: continue;
6624: }
6625: }
6626: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6627: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6628: gp=vector(1,(nlstate)*(nlstate+ndeath));
6629: gm=vector(1,(nlstate)*(nlstate+ndeath));
6630: for (age=bage; age<=fage; age ++){
6631: cov[2]=age;
6632: if(nagesqr==1)
6633: cov[3]= age*age;
6634: for (k=1; k<=cptcovn;k++) {
6635: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6636: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6637: * 1 1 1 1 1
6638: * 2 2 1 1 1
6639: * 3 1 2 1 1
6640: */
6641: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6642: }
6643: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6644: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6645: for (k=1; k<=cptcovprod;k++)
6646: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6647:
6648:
1.222 brouard 6649: for(theta=1; theta <=npar; theta++){
6650: for(i=1; i<=npar; i++)
6651: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6652:
1.222 brouard 6653: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6654:
1.222 brouard 6655: k=0;
6656: for(i=1; i<= (nlstate); i++){
6657: for(j=1; j<=(nlstate+ndeath);j++){
6658: k=k+1;
6659: gp[k]=pmmij[i][j];
6660: }
6661: }
1.220 brouard 6662:
1.222 brouard 6663: for(i=1; i<=npar; i++)
6664: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6665:
1.222 brouard 6666: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6667: k=0;
6668: for(i=1; i<=(nlstate); i++){
6669: for(j=1; j<=(nlstate+ndeath);j++){
6670: k=k+1;
6671: gm[k]=pmmij[i][j];
6672: }
6673: }
1.220 brouard 6674:
1.222 brouard 6675: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6676: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6677: }
1.126 brouard 6678:
1.222 brouard 6679: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6680: for(theta=1; theta <=npar; theta++)
6681: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6682:
1.222 brouard 6683: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6684: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6685:
1.222 brouard 6686: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6687:
1.222 brouard 6688: k=0;
6689: for(i=1; i<=(nlstate); i++){
6690: for(j=1; j<=(nlstate+ndeath);j++){
6691: k=k+1;
6692: mu[k][(int) age]=pmmij[i][j];
6693: }
6694: }
6695: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6696: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6697: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6698:
1.222 brouard 6699: /*printf("\n%d ",(int)age);
6700: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6701: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6702: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6703: }*/
1.220 brouard 6704:
1.222 brouard 6705: fprintf(ficresprob,"\n%d ",(int)age);
6706: fprintf(ficresprobcov,"\n%d ",(int)age);
6707: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6708:
1.222 brouard 6709: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6710: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6711: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6712: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6713: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6714: }
6715: i=0;
6716: for (k=1; k<=(nlstate);k++){
6717: for (l=1; l<=(nlstate+ndeath);l++){
6718: i++;
6719: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6720: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6721: for (j=1; j<=i;j++){
6722: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6723: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6724: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6725: }
6726: }
6727: }/* end of loop for state */
6728: } /* end of loop for age */
6729: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6730: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6731: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6732: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6733:
6734: /* Confidence intervalle of pij */
6735: /*
6736: fprintf(ficgp,"\nunset parametric;unset label");
6737: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6738: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6739: 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);
6740: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6741: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6742: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6743: */
6744:
6745: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6746: first1=1;first2=2;
6747: for (k2=1; k2<=(nlstate);k2++){
6748: for (l2=1; l2<=(nlstate+ndeath);l2++){
6749: if(l2==k2) continue;
6750: j=(k2-1)*(nlstate+ndeath)+l2;
6751: for (k1=1; k1<=(nlstate);k1++){
6752: for (l1=1; l1<=(nlstate+ndeath);l1++){
6753: if(l1==k1) continue;
6754: i=(k1-1)*(nlstate+ndeath)+l1;
6755: if(i<=j) continue;
6756: for (age=bage; age<=fage; age ++){
6757: if ((int)age %5==0){
6758: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6759: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6760: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6761: mu1=mu[i][(int) age]/stepm*YEARM ;
6762: mu2=mu[j][(int) age]/stepm*YEARM;
6763: c12=cv12/sqrt(v1*v2);
6764: /* Computing eigen value of matrix of covariance */
6765: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6766: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6767: if ((lc2 <0) || (lc1 <0) ){
6768: if(first2==1){
6769: first1=0;
6770: 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);
6771: }
6772: 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);
6773: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6774: /* lc2=fabs(lc2); */
6775: }
1.220 brouard 6776:
1.222 brouard 6777: /* Eigen vectors */
1.280 brouard 6778: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6779: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6780: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6781: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6782: }else
6783: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6784: /*v21=sqrt(1.-v11*v11); *//* error */
6785: v21=(lc1-v1)/cv12*v11;
6786: v12=-v21;
6787: v22=v11;
6788: tnalp=v21/v11;
6789: if(first1==1){
6790: first1=0;
6791: 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);
6792: }
6793: 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);
6794: /*printf(fignu*/
6795: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6796: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6797: if(first==1){
6798: first=0;
6799: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6800: fprintf(ficgp,"\nset parametric;unset label");
6801: 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);
6802: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6803: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6804: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6805: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6806: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6807: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6808: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6809: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6810: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6811: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6812: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6813: 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 6814: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6815: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6816: }else{
6817: first=0;
6818: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6819: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6820: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6821: 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 6822: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6823: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6824: }/* if first */
6825: } /* age mod 5 */
6826: } /* end loop age */
6827: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6828: first=1;
6829: } /*l12 */
6830: } /* k12 */
6831: } /*l1 */
6832: }/* k1 */
6833: } /* loop on combination of covariates j1 */
6834: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6835: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6836: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6837: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6838: free_vector(xp,1,npar);
6839: fclose(ficresprob);
6840: fclose(ficresprobcov);
6841: fclose(ficresprobcor);
6842: fflush(ficgp);
6843: fflush(fichtmcov);
6844: }
1.126 brouard 6845:
6846:
6847: /******************* Printing html file ***********/
1.201 brouard 6848: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6849: int lastpass, int stepm, int weightopt, char model[],\
6850: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 6851: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
6852: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
6853: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 6854: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6855:
6856: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6857: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6858: </ul>");
1.237 brouard 6859: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6860: </ul>", model);
1.214 brouard 6861: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6862: 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",
6863: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6864: 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 6865: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6866: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6867: fprintf(fichtm,"\
6868: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6869: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6870: fprintf(fichtm,"\
1.217 brouard 6871: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6872: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6873: fprintf(fichtm,"\
1.288 brouard 6874: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6875: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6876: fprintf(fichtm,"\
1.288 brouard 6877: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6878: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6879: fprintf(fichtm,"\
1.211 brouard 6880: - (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 6881: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6882: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6883: if(prevfcast==1){
6884: fprintf(fichtm,"\
6885: - Prevalence projections by age and states: \
1.201 brouard 6886: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6887: }
1.126 brouard 6888:
6889:
1.225 brouard 6890: m=pow(2,cptcoveff);
1.222 brouard 6891: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6892:
1.264 brouard 6893: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6894:
6895: jj1=0;
6896:
6897: fprintf(fichtm," \n<ul>");
6898: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6899: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6900: if(m != 1 && TKresult[nres]!= k1)
6901: continue;
6902: jj1++;
6903: if (cptcovn > 0) {
6904: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6905: for (cpt=1; cpt<=cptcoveff;cpt++){
6906: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6907: }
6908: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6909: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6910: }
6911: fprintf(fichtm,"\">");
6912:
6913: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6914: fprintf(fichtm,"************ Results for covariates");
6915: for (cpt=1; cpt<=cptcoveff;cpt++){
6916: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6917: }
6918: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6919: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6920: }
6921: if(invalidvarcomb[k1]){
6922: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6923: continue;
6924: }
6925: fprintf(fichtm,"</a></li>");
6926: } /* cptcovn >0 */
6927: }
6928: fprintf(fichtm," \n</ul>");
6929:
1.222 brouard 6930: jj1=0;
1.237 brouard 6931:
6932: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6933: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6934: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6935: continue;
1.220 brouard 6936:
1.222 brouard 6937: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6938: jj1++;
6939: if (cptcovn > 0) {
1.264 brouard 6940: fprintf(fichtm,"\n<p><a name=\"rescov");
6941: for (cpt=1; cpt<=cptcoveff;cpt++){
6942: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6943: }
6944: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6945: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6946: }
6947: fprintf(fichtm,"\"</a>");
6948:
1.222 brouard 6949: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6950: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6951: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6952: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6953: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6954: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6955: }
1.237 brouard 6956: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6957: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6958: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6959: }
6960:
1.230 brouard 6961: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6962: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6963: if(invalidvarcomb[k1]){
6964: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6965: printf("\nCombination (%d) ignored because no cases \n",k1);
6966: continue;
6967: }
6968: }
6969: /* aij, bij */
1.259 brouard 6970: 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 6971: <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 6972: /* Pij */
1.241 brouard 6973: 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> \
6974: <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 6975: /* Quasi-incidences */
6976: 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 6977: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6978: 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 6979: 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> \
6980: <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 6981: /* Survival functions (period) in state j */
6982: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6983: 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 6984: <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 6985: }
6986: /* State specific survival functions (period) */
6987: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6988: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
6989: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 6990: <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 6991: }
1.288 brouard 6992: /* Period (forward stable) prevalence in each health state */
1.222 brouard 6993: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6994: 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> \
6995: <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 6996: }
1.296 brouard 6997: if(prevbcast==1){
1.288 brouard 6998: /* Backward prevalence in each health state */
1.222 brouard 6999: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7000: 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 7001: <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 7002: }
1.217 brouard 7003: }
1.222 brouard 7004: if(prevfcast==1){
1.288 brouard 7005: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7006: for(cpt=1; cpt<=nlstate;cpt++){
1.288 brouard 7007: 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> \
1.296 brouard 7008: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateprojd, dateprojf, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7009: }
7010: }
1.296 brouard 7011: if(prevbcast==1){
1.268 brouard 7012: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7013: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7014: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7015: 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 \
7016: 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) \
7017: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
7018: <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 7019: }
7020: }
1.220 brouard 7021:
1.222 brouard 7022: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 7023: 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> \
7024: <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 7025: }
7026: /* } /\* end i1 *\/ */
7027: }/* End k1 */
7028: fprintf(fichtm,"</ul>");
1.126 brouard 7029:
1.222 brouard 7030: fprintf(fichtm,"\
1.126 brouard 7031: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7032: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7033: - 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 7034: But because parameters are usually highly correlated (a higher incidence of disability \
7035: and a higher incidence of recovery can give very close observed transition) it might \
7036: be very useful to look not only at linear confidence intervals estimated from the \
7037: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7038: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7039: covariance matrix of the one-step probabilities. \
7040: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7041:
1.222 brouard 7042: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7043: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7044: fprintf(fichtm,"\
1.126 brouard 7045: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7046: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7047:
1.222 brouard 7048: fprintf(fichtm,"\
1.126 brouard 7049: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7050: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7051: fprintf(fichtm,"\
1.126 brouard 7052: - 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): \
7053: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7054: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7055: fprintf(fichtm,"\
1.126 brouard 7056: - (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): \
7057: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7058: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7059: fprintf(fichtm,"\
1.288 brouard 7060: - 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 7061: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7062: fprintf(fichtm,"\
1.128 brouard 7063: - 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 7064: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7065: fprintf(fichtm,"\
1.288 brouard 7066: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7067: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7068:
7069: /* if(popforecast==1) fprintf(fichtm,"\n */
7070: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7071: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7072: /* <br>",fileres,fileres,fileres,fileres); */
7073: /* else */
7074: /* 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 7075: fflush(fichtm);
7076: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7077:
1.225 brouard 7078: m=pow(2,cptcoveff);
1.222 brouard 7079: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7080:
1.222 brouard 7081: jj1=0;
1.237 brouard 7082:
1.241 brouard 7083: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7084: for(k1=1; k1<=m;k1++){
1.253 brouard 7085: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7086: continue;
1.222 brouard 7087: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7088: jj1++;
1.126 brouard 7089: if (cptcovn > 0) {
7090: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7091: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7092: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7093: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7094: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7095: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7096: }
7097:
1.126 brouard 7098: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7099:
1.222 brouard 7100: if(invalidvarcomb[k1]){
7101: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7102: continue;
7103: }
1.126 brouard 7104: }
7105: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7106: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 7107: 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 7108: <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 7109: }
7110: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 7111: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
7112: true period expectancies (those weighted with period prevalences are also\
7113: drawn in addition to the population based expectancies computed using\
1.241 brouard 7114: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
7115: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7116: /* } /\* end i1 *\/ */
7117: }/* End k1 */
1.241 brouard 7118: }/* End nres */
1.222 brouard 7119: fprintf(fichtm,"</ul>");
7120: fflush(fichtm);
1.126 brouard 7121: }
7122:
7123: /******************* Gnuplot file **************/
1.296 brouard 7124: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double bage, double fage , int prevfcast, int prevbcast, char pathc[], double p[], int offyear, int offbyear){
1.126 brouard 7125:
7126: char dirfileres[132],optfileres[132];
1.264 brouard 7127: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7128: 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 7129: int lv=0, vlv=0, kl=0;
1.130 brouard 7130: int ng=0;
1.201 brouard 7131: int vpopbased;
1.223 brouard 7132: int ioffset; /* variable offset for columns */
1.270 brouard 7133: int iyearc=1; /* variable column for year of projection */
7134: int iagec=1; /* variable column for age of projection */
1.235 brouard 7135: int nres=0; /* Index of resultline */
1.266 brouard 7136: int istart=1; /* For starting graphs in projections */
1.219 brouard 7137:
1.126 brouard 7138: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7139: /* printf("Problem with file %s",optionfilegnuplot); */
7140: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7141: /* } */
7142:
7143: /*#ifdef windows */
7144: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7145: /*#endif */
1.225 brouard 7146: m=pow(2,cptcoveff);
1.126 brouard 7147:
1.274 brouard 7148: /* diagram of the model */
7149: fprintf(ficgp,"\n#Diagram of the model \n");
7150: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7151: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7152: 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);
7153:
7154: 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);
7155: fprintf(ficgp,"\n#show arrow\nunset label\n");
7156: 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);
7157: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7158: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7159: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7160: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7161:
1.202 brouard 7162: /* Contribution to likelihood */
7163: /* Plot the probability implied in the likelihood */
1.223 brouard 7164: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7165: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7166: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7167: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7168: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7169: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7170: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7171: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7172: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7173: 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));
7174: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7175: 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));
7176: for (i=1; i<= nlstate ; i ++) {
7177: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7178: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7179: 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);
7180: for (j=2; j<= nlstate+ndeath ; j ++) {
7181: 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);
7182: }
7183: fprintf(ficgp,";\nset out; unset ylabel;\n");
7184: }
7185: /* 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 */
7186: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7187: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7188: fprintf(ficgp,"\nset out;unset log\n");
7189: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7190:
1.126 brouard 7191: strcpy(dirfileres,optionfilefiname);
7192: strcpy(optfileres,"vpl");
1.223 brouard 7193: /* 1eme*/
1.238 brouard 7194: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7195: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7196: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7197: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7198: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7199: continue;
7200: /* We are interested in selected combination by the resultline */
1.246 brouard 7201: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7202: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7203: strcpy(gplotlabel,"(");
1.238 brouard 7204: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7205: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7206: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7207: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7208: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7209: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7210: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7211: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7212: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7213: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7214: }
7215: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7216: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7217: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7218: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7219: }
7220: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7221: /* printf("\n#\n"); */
1.238 brouard 7222: fprintf(ficgp,"\n#\n");
7223: if(invalidvarcomb[k1]){
1.260 brouard 7224: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7225: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7226: continue;
7227: }
1.235 brouard 7228:
1.241 brouard 7229: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7230: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7231: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7232: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7233: 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);
7234: /* 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); */
7235: /* k1-1 error should be nres-1*/
1.238 brouard 7236: for (i=1; i<= nlstate ; i ++) {
7237: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7238: else fprintf(ficgp," %%*lf (%%*lf)");
7239: }
1.288 brouard 7240: 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 7241: for (i=1; i<= nlstate ; i ++) {
7242: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7243: else fprintf(ficgp," %%*lf (%%*lf)");
7244: }
1.260 brouard 7245: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7246: for (i=1; i<= nlstate ; i ++) {
7247: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7248: else fprintf(ficgp," %%*lf (%%*lf)");
7249: }
1.265 brouard 7250: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7251:
7252: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7253: if(cptcoveff ==0){
1.271 brouard 7254: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7255: }else{
7256: kl=0;
7257: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7258: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7259: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7260: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7261: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7262: vlv= nbcode[Tvaraff[k]][lv];
7263: kl++;
7264: /* 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 *\/ */
7265: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7266: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7267: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7268: if(k==cptcoveff){
7269: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7270: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7271: }else{
7272: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7273: kl++;
7274: }
7275: } /* end covariate */
7276: } /* end if no covariate */
7277:
1.296 brouard 7278: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7279: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7280: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7281: if(cptcoveff ==0){
1.245 brouard 7282: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7283: }else{
7284: kl=0;
7285: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7286: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7287: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7288: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7289: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7290: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7291: kl++;
1.238 brouard 7292: /* 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 *\/ */
7293: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7294: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7295: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7296: if(k==cptcoveff){
1.245 brouard 7297: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7298: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7299: }else{
7300: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7301: kl++;
7302: }
7303: } /* end covariate */
7304: } /* end if no covariate */
1.296 brouard 7305: if(prevbcast == 1){
1.268 brouard 7306: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7307: /* k1-1 error should be nres-1*/
7308: for (i=1; i<= nlstate ; i ++) {
7309: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7310: else fprintf(ficgp," %%*lf (%%*lf)");
7311: }
1.271 brouard 7312: 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 7313: for (i=1; i<= nlstate ; i ++) {
7314: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7315: else fprintf(ficgp," %%*lf (%%*lf)");
7316: }
1.276 brouard 7317: 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 7318: for (i=1; i<= nlstate ; i ++) {
7319: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7320: else fprintf(ficgp," %%*lf (%%*lf)");
7321: }
1.274 brouard 7322: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7323: } /* end if backprojcast */
1.296 brouard 7324: } /* end if prevbcast */
1.276 brouard 7325: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7326: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7327: } /* nres */
1.201 brouard 7328: } /* k1 */
7329: } /* cpt */
1.235 brouard 7330:
7331:
1.126 brouard 7332: /*2 eme*/
1.238 brouard 7333: for (k1=1; k1<= m ; k1 ++){
7334: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7335: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7336: continue;
7337: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7338: strcpy(gplotlabel,"(");
1.238 brouard 7339: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7340: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7341: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7342: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7343: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7344: vlv= nbcode[Tvaraff[k]][lv];
7345: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7346: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7347: }
1.237 brouard 7348: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7349: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7350: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7351: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7352: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7353: }
1.264 brouard 7354: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7355: fprintf(ficgp,"\n#\n");
1.223 brouard 7356: if(invalidvarcomb[k1]){
7357: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7358: continue;
7359: }
1.219 brouard 7360:
1.241 brouard 7361: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7362: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7363: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7364: if(vpopbased==0){
1.238 brouard 7365: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7366: }else
1.238 brouard 7367: fprintf(ficgp,"\nreplot ");
7368: for (i=1; i<= nlstate+1 ; i ++) {
7369: k=2*i;
1.261 brouard 7370: 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 7371: for (j=1; j<= nlstate+1 ; j ++) {
7372: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7373: else fprintf(ficgp," %%*lf (%%*lf)");
7374: }
7375: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7376: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7377: 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 7378: for (j=1; j<= nlstate+1 ; j ++) {
7379: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7380: else fprintf(ficgp," %%*lf (%%*lf)");
7381: }
7382: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7383: 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 7384: for (j=1; j<= nlstate+1 ; j ++) {
7385: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7386: else fprintf(ficgp," %%*lf (%%*lf)");
7387: }
7388: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7389: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7390: } /* state */
7391: } /* vpopbased */
1.264 brouard 7392: 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 7393: } /* end nres */
7394: } /* k1 end 2 eme*/
7395:
7396:
7397: /*3eme*/
7398: for (k1=1; k1<= m ; k1 ++){
7399: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7400: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7401: continue;
7402:
7403: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7404: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7405: strcpy(gplotlabel,"(");
1.238 brouard 7406: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7407: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7408: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7409: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7410: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7411: vlv= nbcode[Tvaraff[k]][lv];
7412: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7413: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7414: }
7415: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7416: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7417: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7418: }
1.264 brouard 7419: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7420: fprintf(ficgp,"\n#\n");
7421: if(invalidvarcomb[k1]){
7422: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7423: continue;
7424: }
7425:
7426: /* k=2+nlstate*(2*cpt-2); */
7427: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7428: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7429: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7430: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7431: 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 7432: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7433: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7434: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7435: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7436: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7437: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7438:
1.238 brouard 7439: */
7440: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7441: 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 7442: /* 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 7443:
1.238 brouard 7444: }
1.261 brouard 7445: 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 7446: }
1.264 brouard 7447: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7448: } /* end nres */
7449: } /* end kl 3eme */
1.126 brouard 7450:
1.223 brouard 7451: /* 4eme */
1.201 brouard 7452: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7453: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7454: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7455: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7456: continue;
1.238 brouard 7457: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7458: strcpy(gplotlabel,"(");
1.238 brouard 7459: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7460: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7461: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7462: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7463: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7464: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7465: vlv= nbcode[Tvaraff[k]][lv];
7466: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7467: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7468: }
7469: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7470: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7471: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7472: }
1.264 brouard 7473: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7474: fprintf(ficgp,"\n#\n");
7475: if(invalidvarcomb[k1]){
7476: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7477: continue;
1.223 brouard 7478: }
1.238 brouard 7479:
1.241 brouard 7480: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7481: 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 7482: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7483: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7484: k=3;
7485: for (i=1; i<= nlstate ; i ++){
7486: if(i==1){
7487: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7488: }else{
7489: fprintf(ficgp,", '' ");
7490: }
7491: l=(nlstate+ndeath)*(i-1)+1;
7492: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7493: for (j=2; j<= nlstate+ndeath ; j ++)
7494: fprintf(ficgp,"+$%d",k+l+j-1);
7495: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7496: } /* nlstate */
1.264 brouard 7497: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7498: } /* end cpt state*/
7499: } /* end nres */
7500: } /* end covariate k1 */
7501:
1.220 brouard 7502: /* 5eme */
1.201 brouard 7503: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7504: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7505: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7506: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7507: continue;
1.238 brouard 7508: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7509: strcpy(gplotlabel,"(");
1.238 brouard 7510: 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);
7511: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7512: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7513: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7514: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7515: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7516: vlv= nbcode[Tvaraff[k]][lv];
7517: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7518: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7519: }
7520: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7521: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7522: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7523: }
1.264 brouard 7524: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7525: fprintf(ficgp,"\n#\n");
7526: if(invalidvarcomb[k1]){
7527: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7528: continue;
7529: }
1.227 brouard 7530:
1.241 brouard 7531: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7532: 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 7533: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7534: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7535: k=3;
7536: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7537: if(j==1)
7538: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7539: else
7540: fprintf(ficgp,", '' ");
7541: l=(nlstate+ndeath)*(cpt-1) +j;
7542: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7543: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7544: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7545: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7546: } /* nlstate */
7547: fprintf(ficgp,", '' ");
7548: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7549: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7550: l=(nlstate+ndeath)*(cpt-1) +j;
7551: if(j < nlstate)
7552: fprintf(ficgp,"$%d +",k+l);
7553: else
7554: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7555: }
1.264 brouard 7556: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7557: } /* end cpt state*/
7558: } /* end covariate */
7559: } /* end nres */
1.227 brouard 7560:
1.220 brouard 7561: /* 6eme */
1.202 brouard 7562: /* CV preval stable (period) for each covariate */
1.237 brouard 7563: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7564: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7565: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7566: continue;
1.255 brouard 7567: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7568: strcpy(gplotlabel,"(");
1.288 brouard 7569: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7570: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7571: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7572: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7573: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7574: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7575: vlv= nbcode[Tvaraff[k]][lv];
7576: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7577: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7578: }
1.237 brouard 7579: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7580: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7581: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7582: }
1.264 brouard 7583: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7584: fprintf(ficgp,"\n#\n");
1.223 brouard 7585: if(invalidvarcomb[k1]){
1.227 brouard 7586: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7587: continue;
1.223 brouard 7588: }
1.227 brouard 7589:
1.241 brouard 7590: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7591: 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 7592: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7593: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7594: k=3; /* Offset */
1.255 brouard 7595: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7596: if(i==1)
7597: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7598: else
7599: fprintf(ficgp,", '' ");
1.255 brouard 7600: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7601: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7602: for (j=2; j<= nlstate ; j ++)
7603: fprintf(ficgp,"+$%d",k+l+j-1);
7604: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7605: } /* nlstate */
1.264 brouard 7606: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7607: } /* end cpt state*/
7608: } /* end covariate */
1.227 brouard 7609:
7610:
1.220 brouard 7611: /* 7eme */
1.296 brouard 7612: if(prevbcast == 1){
1.288 brouard 7613: /* CV backward prevalence for each covariate */
1.237 brouard 7614: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7615: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7616: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7617: continue;
1.268 brouard 7618: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7619: strcpy(gplotlabel,"(");
1.288 brouard 7620: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7621: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7622: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7623: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7624: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7625: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7626: vlv= nbcode[Tvaraff[k]][lv];
7627: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7628: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7629: }
1.237 brouard 7630: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7631: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7632: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7633: }
1.264 brouard 7634: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7635: fprintf(ficgp,"\n#\n");
7636: if(invalidvarcomb[k1]){
7637: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7638: continue;
7639: }
7640:
1.241 brouard 7641: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7642: 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 7643: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7644: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7645: k=3; /* Offset */
1.268 brouard 7646: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7647: if(i==1)
7648: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7649: else
7650: fprintf(ficgp,", '' ");
7651: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7652: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7653: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7654: /* 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 7655: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7656: /* for (j=2; j<= nlstate ; j ++) */
7657: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7658: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7659: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7660: } /* nlstate */
1.264 brouard 7661: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7662: } /* end cpt state*/
7663: } /* end covariate */
1.296 brouard 7664: } /* End if prevbcast */
1.218 brouard 7665:
1.223 brouard 7666: /* 8eme */
1.218 brouard 7667: if(prevfcast==1){
1.288 brouard 7668: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7669:
1.237 brouard 7670: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7671: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7672: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7673: continue;
1.211 brouard 7674: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7675: strcpy(gplotlabel,"(");
1.288 brouard 7676: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7677: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7678: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7679: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7680: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7681: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7682: vlv= nbcode[Tvaraff[k]][lv];
7683: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7684: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7685: }
1.237 brouard 7686: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7687: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7688: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7689: }
1.264 brouard 7690: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7691: fprintf(ficgp,"\n#\n");
7692: if(invalidvarcomb[k1]){
7693: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7694: continue;
7695: }
7696:
7697: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7698: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7699: 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 7700: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7701: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7702:
7703: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7704: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7705: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7706: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7707: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7708: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7709: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7710: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7711: if(i==istart){
1.227 brouard 7712: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7713: }else{
7714: fprintf(ficgp,",\\\n '' ");
7715: }
7716: if(cptcoveff ==0){ /* No covariate */
7717: ioffset=2; /* Age is in 2 */
7718: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7719: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7720: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7721: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7722: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7723: if(i==nlstate+1){
1.270 brouard 7724: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7725: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7726: fprintf(ficgp,",\\\n '' ");
7727: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7728: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7729: offyear, \
1.268 brouard 7730: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7731: }else
1.227 brouard 7732: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7733: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7734: }else{ /* more than 2 covariates */
1.270 brouard 7735: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7736: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7737: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7738: iyearc=ioffset-1;
7739: iagec=ioffset;
1.227 brouard 7740: fprintf(ficgp," u %d:(",ioffset);
7741: kl=0;
7742: strcpy(gplotcondition,"(");
7743: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7744: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7745: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7746: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7747: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7748: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7749: kl++;
7750: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7751: kl++;
7752: if(k <cptcoveff && cptcoveff>1)
7753: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7754: }
7755: strcpy(gplotcondition+strlen(gplotcondition),")");
7756: /* 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 *\/ */
7757: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7758: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7759: /* '' 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*/
7760: if(i==nlstate+1){
1.270 brouard 7761: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7762: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7763: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7764: fprintf(ficgp," u %d:(",iagec);
7765: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7766: iyearc, iagec, offyear, \
7767: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7768: /* '' 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 7769: }else{
7770: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7771: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7772: }
7773: } /* end if covariate */
7774: } /* nlstate */
1.264 brouard 7775: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7776: } /* end cpt state*/
7777: } /* end covariate */
7778: } /* End if prevfcast */
1.227 brouard 7779:
1.296 brouard 7780: if(prevbcast==1){
1.268 brouard 7781: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7782:
7783: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7784: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7785: if(m != 1 && TKresult[nres]!= k1)
7786: continue;
7787: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7788: strcpy(gplotlabel,"(");
7789: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7790: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7791: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7792: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7793: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7794: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7795: vlv= nbcode[Tvaraff[k]][lv];
7796: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7797: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7798: }
7799: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7800: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7801: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7802: }
7803: strcpy(gplotlabel+strlen(gplotlabel),")");
7804: fprintf(ficgp,"\n#\n");
7805: if(invalidvarcomb[k1]){
7806: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7807: continue;
7808: }
7809:
7810: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7811: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7812: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7813: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7814: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7815:
7816: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7817: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7818: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7819: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7820: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7821: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7822: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7823: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7824: if(i==istart){
7825: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7826: }else{
7827: fprintf(ficgp,",\\\n '' ");
7828: }
7829: if(cptcoveff ==0){ /* No covariate */
7830: ioffset=2; /* Age is in 2 */
7831: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7832: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7833: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7834: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7835: fprintf(ficgp," u %d:(", ioffset);
7836: if(i==nlstate+1){
1.270 brouard 7837: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7838: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7839: fprintf(ficgp,",\\\n '' ");
7840: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7841: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7842: offbyear, \
7843: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7844: }else
7845: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7846: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7847: }else{ /* more than 2 covariates */
1.270 brouard 7848: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7849: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7850: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7851: iyearc=ioffset-1;
7852: iagec=ioffset;
1.268 brouard 7853: fprintf(ficgp," u %d:(",ioffset);
7854: kl=0;
7855: strcpy(gplotcondition,"(");
7856: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7857: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7858: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7859: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7860: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7861: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7862: kl++;
7863: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7864: kl++;
7865: if(k <cptcoveff && cptcoveff>1)
7866: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7867: }
7868: strcpy(gplotcondition+strlen(gplotcondition),")");
7869: /* 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 *\/ */
7870: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7871: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7872: /* '' 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*/
7873: if(i==nlstate+1){
1.270 brouard 7874: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7875: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7876: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7877: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7878: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7879: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7880: iyearc,iagec,offbyear, \
7881: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7882: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7883: }else{
7884: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7885: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7886: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7887: }
7888: } /* end if covariate */
7889: } /* nlstate */
7890: fprintf(ficgp,"\nset out; unset label;\n");
7891: } /* end cpt state*/
7892: } /* end covariate */
1.296 brouard 7893: } /* End if prevbcast */
1.268 brouard 7894:
1.227 brouard 7895:
1.238 brouard 7896: /* 9eme writing MLE parameters */
7897: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7898: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7899: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7900: for(k=1; k <=(nlstate+ndeath); k++){
7901: if (k != i) {
1.227 brouard 7902: fprintf(ficgp,"# current state %d\n",k);
7903: for(j=1; j <=ncovmodel; j++){
7904: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7905: jk++;
7906: }
7907: fprintf(ficgp,"\n");
1.126 brouard 7908: }
7909: }
1.223 brouard 7910: }
1.187 brouard 7911: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7912:
1.145 brouard 7913: /*goto avoid;*/
1.238 brouard 7914: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7915: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7916: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7917: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7918: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7919: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7920: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7921: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7922: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7923: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7924: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7925: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7926: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7927: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7928: fprintf(ficgp,"#\n");
1.223 brouard 7929: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7930: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7931: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7932: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7933: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7934: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7935: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7936: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7937: continue;
1.264 brouard 7938: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7939: strcpy(gplotlabel,"(");
1.276 brouard 7940: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7941: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7942: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7943: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7944: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7945: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7946: vlv= nbcode[Tvaraff[k]][lv];
7947: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7948: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7949: }
1.237 brouard 7950: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7951: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7952: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7953: }
1.264 brouard 7954: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7955: fprintf(ficgp,"\n#\n");
1.264 brouard 7956: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7957: fprintf(ficgp,"\nset key outside ");
7958: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7959: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7960: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7961: if (ng==1){
7962: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7963: fprintf(ficgp,"\nunset log y");
7964: }else if (ng==2){
7965: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7966: fprintf(ficgp,"\nset log y");
7967: }else if (ng==3){
7968: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7969: fprintf(ficgp,"\nset log y");
7970: }else
7971: fprintf(ficgp,"\nunset title ");
7972: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7973: i=1;
7974: for(k2=1; k2<=nlstate; k2++) {
7975: k3=i;
7976: for(k=1; k<=(nlstate+ndeath); k++) {
7977: if (k != k2){
7978: switch( ng) {
7979: case 1:
7980: if(nagesqr==0)
7981: fprintf(ficgp," p%d+p%d*x",i,i+1);
7982: else /* nagesqr =1 */
7983: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7984: break;
7985: case 2: /* ng=2 */
7986: if(nagesqr==0)
7987: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7988: else /* nagesqr =1 */
7989: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7990: break;
7991: case 3:
7992: if(nagesqr==0)
7993: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7994: else /* nagesqr =1 */
7995: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7996: break;
7997: }
7998: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7999: ijp=1; /* product no age */
8000: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8001: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8002: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8003: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8004: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8005: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8006: if(DummyV[j]==0){
8007: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8008: }else{ /* quantitative */
8009: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8010: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8011: }
8012: ij++;
1.237 brouard 8013: }
1.268 brouard 8014: }
8015: }else if(cptcovprod >0){
8016: if(j==Tprod[ijp]) { /* */
8017: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8018: if(ijp <=cptcovprod) { /* Product */
8019: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8020: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8021: /* 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)]); */
8022: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8023: }else{ /* Vn is dummy and Vm is quanti */
8024: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8025: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8026: }
8027: }else{ /* Vn*Vm Vn is quanti */
8028: if(DummyV[Tvard[ijp][2]]==0){
8029: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8030: }else{ /* Both quanti */
8031: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8032: }
1.237 brouard 8033: }
1.268 brouard 8034: ijp++;
1.237 brouard 8035: }
1.268 brouard 8036: } /* end Tprod */
1.237 brouard 8037: } else{ /* simple covariate */
1.264 brouard 8038: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8039: if(Dummy[j]==0){
8040: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8041: }else{ /* quantitative */
8042: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8043: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8044: }
1.237 brouard 8045: } /* end simple */
8046: } /* end j */
1.223 brouard 8047: }else{
8048: i=i-ncovmodel;
8049: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8050: fprintf(ficgp," (1.");
8051: }
1.227 brouard 8052:
1.223 brouard 8053: if(ng != 1){
8054: fprintf(ficgp,")/(1");
1.227 brouard 8055:
1.264 brouard 8056: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8057: if(nagesqr==0)
1.264 brouard 8058: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8059: else /* nagesqr =1 */
1.264 brouard 8060: 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 8061:
1.223 brouard 8062: ij=1;
8063: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8064: if(cptcovage >0){
8065: if((j-2)==Tage[ij]) { /* Bug valgrind */
8066: if(ij <=cptcovage) { /* Bug valgrind */
8067: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8068: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8069: ij++;
8070: }
8071: }
8072: }else
8073: 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 8074: }
8075: fprintf(ficgp,")");
8076: }
8077: fprintf(ficgp,")");
8078: if(ng ==2)
1.276 brouard 8079: 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 8080: else /* ng= 3 */
1.276 brouard 8081: 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 8082: }else{ /* end ng <> 1 */
8083: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8084: 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 8085: }
8086: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8087: fprintf(ficgp,",");
8088: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8089: fprintf(ficgp,",");
8090: i=i+ncovmodel;
8091: } /* end k */
8092: } /* end k2 */
1.276 brouard 8093: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8094: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8095: } /* end k1 */
1.223 brouard 8096: } /* end ng */
8097: /* avoid: */
8098: fflush(ficgp);
1.126 brouard 8099: } /* end gnuplot */
8100:
8101:
8102: /*************** Moving average **************/
1.219 brouard 8103: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8104: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8105:
1.222 brouard 8106: int i, cpt, cptcod;
8107: int modcovmax =1;
8108: int mobilavrange, mob;
8109: int iage=0;
1.288 brouard 8110: int firstA1=0, firstA2=0;
1.222 brouard 8111:
1.266 brouard 8112: double sum=0., sumr=0.;
1.222 brouard 8113: double age;
1.266 brouard 8114: double *sumnewp, *sumnewm, *sumnewmr;
8115: double *agemingood, *agemaxgood;
8116: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8117:
8118:
1.278 brouard 8119: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8120: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8121:
8122: sumnewp = vector(1,ncovcombmax);
8123: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8124: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8125: agemingood = vector(1,ncovcombmax);
1.266 brouard 8126: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8127: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8128: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8129:
8130: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8131: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8132: sumnewp[cptcod]=0.;
1.266 brouard 8133: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8134: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8135: }
8136: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8137:
1.266 brouard 8138: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8139: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8140: else mobilavrange=mobilav;
8141: for (age=bage; age<=fage; age++)
8142: for (i=1; i<=nlstate;i++)
8143: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8144: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8145: /* We keep the original values on the extreme ages bage, fage and for
8146: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8147: we use a 5 terms etc. until the borders are no more concerned.
8148: */
8149: for (mob=3;mob <=mobilavrange;mob=mob+2){
8150: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8151: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8152: sumnewm[cptcod]=0.;
8153: for (i=1; i<=nlstate;i++){
1.222 brouard 8154: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8155: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8156: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8157: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8158: }
8159: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8160: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8161: } /* end i */
8162: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8163: } /* end cptcod */
1.222 brouard 8164: }/* end age */
8165: }/* end mob */
1.266 brouard 8166: }else{
8167: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8168: return -1;
1.266 brouard 8169: }
8170:
8171: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8172: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8173: if(invalidvarcomb[cptcod]){
8174: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8175: continue;
8176: }
1.219 brouard 8177:
1.266 brouard 8178: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8179: sumnewm[cptcod]=0.;
8180: sumnewmr[cptcod]=0.;
8181: for (i=1; i<=nlstate;i++){
8182: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8183: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8184: }
8185: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8186: agemingoodr[cptcod]=age;
8187: }
8188: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8189: agemingood[cptcod]=age;
8190: }
8191: } /* age */
8192: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8193: sumnewm[cptcod]=0.;
1.266 brouard 8194: sumnewmr[cptcod]=0.;
1.222 brouard 8195: for (i=1; i<=nlstate;i++){
8196: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8197: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8198: }
8199: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8200: agemaxgoodr[cptcod]=age;
1.222 brouard 8201: }
8202: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8203: agemaxgood[cptcod]=age;
8204: }
8205: } /* age */
8206: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8207: /* but they will change */
1.288 brouard 8208: firstA1=0;firstA2=0;
1.266 brouard 8209: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8210: sumnewm[cptcod]=0.;
8211: sumnewmr[cptcod]=0.;
8212: for (i=1; i<=nlstate;i++){
8213: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8214: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8215: }
8216: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8217: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8218: agemaxgoodr[cptcod]=age; /* age min */
8219: for (i=1; i<=nlstate;i++)
8220: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8221: }else{ /* bad we change the value with the values of good ages */
8222: for (i=1; i<=nlstate;i++){
8223: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8224: } /* i */
8225: } /* end bad */
8226: }else{
8227: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8228: agemaxgood[cptcod]=age;
8229: }else{ /* bad we change the value with the values of good ages */
8230: for (i=1; i<=nlstate;i++){
8231: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8232: } /* i */
8233: } /* end bad */
8234: }/* end else */
8235: sum=0.;sumr=0.;
8236: for (i=1; i<=nlstate;i++){
8237: sum+=mobaverage[(int)age][i][cptcod];
8238: sumr+=probs[(int)age][i][cptcod];
8239: }
8240: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8241: if(!firstA1){
8242: firstA1=1;
8243: 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);
8244: }
8245: 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 8246: } /* end bad */
8247: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8248: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8249: if(!firstA2){
8250: firstA2=1;
8251: 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);
8252: }
8253: 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 8254: } /* end bad */
8255: }/* age */
1.266 brouard 8256:
8257: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8258: sumnewm[cptcod]=0.;
1.266 brouard 8259: sumnewmr[cptcod]=0.;
1.222 brouard 8260: for (i=1; i<=nlstate;i++){
8261: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8262: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8263: }
8264: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8265: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8266: agemingoodr[cptcod]=age;
8267: for (i=1; i<=nlstate;i++)
8268: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8269: }else{ /* bad we change the value with the values of good ages */
8270: for (i=1; i<=nlstate;i++){
8271: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8272: } /* i */
8273: } /* end bad */
8274: }else{
8275: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8276: agemingood[cptcod]=age;
8277: }else{ /* bad */
8278: for (i=1; i<=nlstate;i++){
8279: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8280: } /* i */
8281: } /* end bad */
8282: }/* end else */
8283: sum=0.;sumr=0.;
8284: for (i=1; i<=nlstate;i++){
8285: sum+=mobaverage[(int)age][i][cptcod];
8286: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8287: }
1.266 brouard 8288: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8289: 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 8290: } /* end bad */
8291: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8292: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8293: 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 8294: } /* end bad */
8295: }/* age */
1.266 brouard 8296:
1.222 brouard 8297:
8298: for (age=bage; age<=fage; age++){
1.235 brouard 8299: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8300: sumnewp[cptcod]=0.;
8301: sumnewm[cptcod]=0.;
8302: for (i=1; i<=nlstate;i++){
8303: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8304: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8305: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8306: }
8307: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8308: }
8309: /* printf("\n"); */
8310: /* } */
1.266 brouard 8311:
1.222 brouard 8312: /* brutal averaging */
1.266 brouard 8313: /* for (i=1; i<=nlstate;i++){ */
8314: /* for (age=1; age<=bage; age++){ */
8315: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8316: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8317: /* } */
8318: /* for (age=fage; age<=AGESUP; age++){ */
8319: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8320: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8321: /* } */
8322: /* } /\* end i status *\/ */
8323: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8324: /* for (age=1; age<=AGESUP; age++){ */
8325: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8326: /* mobaverage[(int)age][i][cptcod]=0.; */
8327: /* } */
8328: /* } */
1.222 brouard 8329: }/* end cptcod */
1.266 brouard 8330: free_vector(agemaxgoodr,1, ncovcombmax);
8331: free_vector(agemaxgood,1, ncovcombmax);
8332: free_vector(agemingood,1, ncovcombmax);
8333: free_vector(agemingoodr,1, ncovcombmax);
8334: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8335: free_vector(sumnewm,1, ncovcombmax);
8336: free_vector(sumnewp,1, ncovcombmax);
8337: return 0;
8338: }/* End movingaverage */
1.218 brouard 8339:
1.126 brouard 8340:
1.296 brouard 8341:
1.126 brouard 8342: /************** Forecasting ******************/
1.296 brouard 8343: /* void prevforecast(char fileres[], double dateintmean, double anprojd, double mprojd, double jprojd, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double anprojf, double p[], int cptcoveff)*/
8344: void prevforecast(char fileres[], double dateintmean, double dateprojd, double dateprojf, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double p[], int cptcoveff){
8345: /* dateintemean, mean date of interviews
8346: dateprojd, year, month, day of starting projection
8347: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8348: agemin, agemax range of age
8349: dateprev1 dateprev2 range of dates during which prevalence is computed
8350: */
1.296 brouard 8351: /* double anprojd, mprojd, jprojd; */
8352: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8353: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8354: double agec; /* generic age */
1.296 brouard 8355: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8356: double *popeffectif,*popcount;
8357: double ***p3mat;
1.218 brouard 8358: /* double ***mobaverage; */
1.126 brouard 8359: char fileresf[FILENAMELENGTH];
8360:
8361: agelim=AGESUP;
1.211 brouard 8362: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8363: in each health status at the date of interview (if between dateprev1 and dateprev2).
8364: We still use firstpass and lastpass as another selection.
8365: */
1.214 brouard 8366: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8367: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8368:
1.201 brouard 8369: strcpy(fileresf,"F_");
8370: strcat(fileresf,fileresu);
1.126 brouard 8371: if((ficresf=fopen(fileresf,"w"))==NULL) {
8372: printf("Problem with forecast resultfile: %s\n", fileresf);
8373: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8374: }
1.235 brouard 8375: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8376: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8377:
1.225 brouard 8378: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8379:
8380:
8381: stepsize=(int) (stepm+YEARM-1)/YEARM;
8382: if (stepm<=12) stepsize=1;
8383: if(estepm < stepm){
8384: printf ("Problem %d lower than %d\n",estepm, stepm);
8385: }
1.270 brouard 8386: else{
8387: hstepm=estepm;
8388: }
8389: if(estepm > stepm){ /* Yes every two year */
8390: stepsize=2;
8391: }
1.296 brouard 8392: hstepm=hstepm/stepm;
1.126 brouard 8393:
1.296 brouard 8394:
8395: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8396: /* fractional in yp1 *\/ */
8397: /* aintmean=yp; */
8398: /* yp2=modf((yp1*12),&yp); */
8399: /* mintmean=yp; */
8400: /* yp1=modf((yp2*30.5),&yp); */
8401: /* jintmean=yp; */
8402: /* if(jintmean==0) jintmean=1; */
8403: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8404:
1.296 brouard 8405:
8406: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8407: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8408: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8409: i1=pow(2,cptcoveff);
1.126 brouard 8410: if (cptcovn < 1){i1=1;}
8411:
1.296 brouard 8412: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8413:
8414: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8415:
1.126 brouard 8416: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8417: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8418: for(k=1; k<=i1;k++){
1.253 brouard 8419: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8420: continue;
1.227 brouard 8421: if(invalidvarcomb[k]){
8422: printf("\nCombination (%d) projection ignored because no cases \n",k);
8423: continue;
8424: }
8425: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8426: for(j=1;j<=cptcoveff;j++) {
8427: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8428: }
1.235 brouard 8429: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8430: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8431: }
1.227 brouard 8432: fprintf(ficresf," yearproj age");
8433: for(j=1; j<=nlstate+ndeath;j++){
8434: for(i=1; i<=nlstate;i++)
8435: fprintf(ficresf," p%d%d",i,j);
8436: fprintf(ficresf," wp.%d",j);
8437: }
1.296 brouard 8438: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8439: fprintf(ficresf,"\n");
1.296 brouard 8440: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8441: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8442: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8443: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8444: nhstepm = nhstepm/hstepm;
8445: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8446: oldm=oldms;savm=savms;
1.268 brouard 8447: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8448: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8449: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8450: for (h=0; h<=nhstepm; h++){
8451: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8452: break;
8453: }
8454: }
8455: fprintf(ficresf,"\n");
8456: for(j=1;j<=cptcoveff;j++)
8457: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8458: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8459:
8460: for(j=1; j<=nlstate+ndeath;j++) {
8461: ppij=0.;
8462: for(i=1; i<=nlstate;i++) {
1.278 brouard 8463: if (mobilav>=1)
8464: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8465: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8466: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8467: }
1.268 brouard 8468: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8469: } /* end i */
8470: fprintf(ficresf," %.3f", ppij);
8471: }/* end j */
1.227 brouard 8472: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8473: } /* end agec */
1.266 brouard 8474: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8475: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8476: } /* end yearp */
8477: } /* end k */
1.219 brouard 8478:
1.126 brouard 8479: fclose(ficresf);
1.215 brouard 8480: printf("End of Computing forecasting \n");
8481: fprintf(ficlog,"End of Computing forecasting\n");
8482:
1.126 brouard 8483: }
8484:
1.269 brouard 8485: /************** Back Forecasting ******************/
1.296 brouard 8486: /* 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){ */
8487: void prevbackforecast(char fileres[], double ***prevacurrent, double dateintmean, double dateprojd, double dateprojf, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double p[], int cptcoveff){
8488: /* back1, year, month, day of starting backprojection
1.267 brouard 8489: agemin, agemax range of age
8490: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8491: anback2 year of end of backprojection (same day and month as back1).
8492: prevacurrent and prev are prevalences.
1.267 brouard 8493: */
8494: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8495: double agec; /* generic age */
1.296 brouard 8496: double agelim, ppij, ppi, yp,yp1,yp2,jintmean,mintmean,aintmean;
1.267 brouard 8497: double *popeffectif,*popcount;
8498: double ***p3mat;
8499: /* double ***mobaverage; */
8500: char fileresfb[FILENAMELENGTH];
8501:
1.268 brouard 8502: agelim=AGEINF;
1.267 brouard 8503: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8504: in each health status at the date of interview (if between dateprev1 and dateprev2).
8505: We still use firstpass and lastpass as another selection.
8506: */
8507: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8508: /* firstpass, lastpass, stepm, weightopt, model); */
8509:
8510: /*Do we need to compute prevalence again?*/
8511:
8512: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8513:
8514: strcpy(fileresfb,"FB_");
8515: strcat(fileresfb,fileresu);
8516: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8517: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8518: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8519: }
8520: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8521: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8522:
8523: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8524:
8525:
8526: stepsize=(int) (stepm+YEARM-1)/YEARM;
8527: if (stepm<=12) stepsize=1;
8528: if(estepm < stepm){
8529: printf ("Problem %d lower than %d\n",estepm, stepm);
8530: }
1.270 brouard 8531: else{
8532: hstepm=estepm;
8533: }
8534: if(estepm >= stepm){ /* Yes every two year */
8535: stepsize=2;
8536: }
1.267 brouard 8537:
8538: hstepm=hstepm/stepm;
1.296 brouard 8539: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8540: /* fractional in yp1 *\/ */
8541: /* aintmean=yp; */
8542: /* yp2=modf((yp1*12),&yp); */
8543: /* mintmean=yp; */
8544: /* yp1=modf((yp2*30.5),&yp); */
8545: /* jintmean=yp; */
8546: /* if(jintmean==0) jintmean=1; */
8547: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8548:
8549: i1=pow(2,cptcoveff);
8550: if (cptcovn < 1){i1=1;}
8551:
1.296 brouard 8552: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8553: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8554:
8555: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8556:
8557: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8558: for(k=1; k<=i1;k++){
8559: if(i1 != 1 && TKresult[nres]!= k)
8560: continue;
8561: if(invalidvarcomb[k]){
8562: printf("\nCombination (%d) projection ignored because no cases \n",k);
8563: continue;
8564: }
1.268 brouard 8565: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8566: for(j=1;j<=cptcoveff;j++) {
8567: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8568: }
8569: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8570: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8571: }
8572: fprintf(ficresfb," yearbproj age");
8573: for(j=1; j<=nlstate+ndeath;j++){
8574: for(i=1; i<=nlstate;i++)
1.268 brouard 8575: fprintf(ficresfb," b%d%d",i,j);
8576: fprintf(ficresfb," b.%d",j);
1.267 brouard 8577: }
1.296 brouard 8578: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8579: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8580: fprintf(ficresfb,"\n");
1.296 brouard 8581: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8582: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8583: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8584: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8585: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8586: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8587: nhstepm = nhstepm/hstepm;
8588: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8589: oldm=oldms;savm=savms;
1.268 brouard 8590: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8591: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8592: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8593: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8594: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8595: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8596: for (h=0; h<=nhstepm; h++){
1.268 brouard 8597: if (h*hstepm/YEARM*stepm ==-yearp) {
8598: break;
8599: }
8600: }
8601: fprintf(ficresfb,"\n");
8602: for(j=1;j<=cptcoveff;j++)
8603: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8604: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8605: for(i=1; i<=nlstate+ndeath;i++) {
8606: ppij=0.;ppi=0.;
8607: for(j=1; j<=nlstate;j++) {
8608: /* if (mobilav==1) */
1.269 brouard 8609: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8610: ppi=ppi+prevacurrent[(int)agec][j][k];
8611: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8612: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8613: /* else { */
8614: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8615: /* } */
1.268 brouard 8616: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8617: } /* end j */
8618: if(ppi <0.99){
8619: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8620: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8621: }
8622: fprintf(ficresfb," %.3f", ppij);
8623: }/* end j */
1.267 brouard 8624: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8625: } /* end agec */
8626: } /* end yearp */
8627: } /* end k */
1.217 brouard 8628:
1.267 brouard 8629: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8630:
1.267 brouard 8631: fclose(ficresfb);
8632: printf("End of Computing Back forecasting \n");
8633: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8634:
1.267 brouard 8635: }
1.217 brouard 8636:
1.269 brouard 8637: /* Variance of prevalence limit: varprlim */
8638: 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 8639: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8640:
8641: char fileresvpl[FILENAMELENGTH];
8642: FILE *ficresvpl;
8643: double **oldm, **savm;
8644: double **varpl; /* Variances of prevalence limits by age */
8645: int i1, k, nres, j ;
8646:
8647: strcpy(fileresvpl,"VPL_");
8648: strcat(fileresvpl,fileresu);
8649: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8650: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8651: exit(0);
8652: }
1.288 brouard 8653: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8654: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8655:
8656: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8657: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8658:
8659: i1=pow(2,cptcoveff);
8660: if (cptcovn < 1){i1=1;}
8661:
8662: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8663: for(k=1; k<=i1;k++){
8664: if(i1 != 1 && TKresult[nres]!= k)
8665: continue;
8666: fprintf(ficresvpl,"\n#****** ");
8667: printf("\n#****** ");
8668: fprintf(ficlog,"\n#****** ");
8669: for(j=1;j<=cptcoveff;j++) {
8670: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8671: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8672: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8673: }
8674: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8675: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8676: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8677: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8678: }
8679: fprintf(ficresvpl,"******\n");
8680: printf("******\n");
8681: fprintf(ficlog,"******\n");
8682:
8683: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8684: oldm=oldms;savm=savms;
8685: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8686: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8687: /*}*/
8688: }
8689:
8690: fclose(ficresvpl);
1.288 brouard 8691: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8692: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8693:
8694: }
8695: /* Variance of back prevalence: varbprlim */
8696: 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){
8697: /*------- Variance of back (stable) prevalence------*/
8698:
8699: char fileresvbl[FILENAMELENGTH];
8700: FILE *ficresvbl;
8701:
8702: double **oldm, **savm;
8703: double **varbpl; /* Variances of back prevalence limits by age */
8704: int i1, k, nres, j ;
8705:
8706: strcpy(fileresvbl,"VBL_");
8707: strcat(fileresvbl,fileresu);
8708: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8709: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8710: exit(0);
8711: }
8712: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8713: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8714:
8715:
8716: i1=pow(2,cptcoveff);
8717: if (cptcovn < 1){i1=1;}
8718:
8719: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8720: for(k=1; k<=i1;k++){
8721: if(i1 != 1 && TKresult[nres]!= k)
8722: continue;
8723: fprintf(ficresvbl,"\n#****** ");
8724: printf("\n#****** ");
8725: fprintf(ficlog,"\n#****** ");
8726: for(j=1;j<=cptcoveff;j++) {
8727: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8728: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8729: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8730: }
8731: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8732: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8733: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8734: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8735: }
8736: fprintf(ficresvbl,"******\n");
8737: printf("******\n");
8738: fprintf(ficlog,"******\n");
8739:
8740: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8741: oldm=oldms;savm=savms;
8742:
8743: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8744: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8745: /*}*/
8746: }
8747:
8748: fclose(ficresvbl);
8749: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8750: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8751:
8752: } /* End of varbprlim */
8753:
1.126 brouard 8754: /************** Forecasting *****not tested NB*************/
1.227 brouard 8755: /* 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 8756:
1.227 brouard 8757: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8758: /* int *popage; */
8759: /* double calagedatem, agelim, kk1, kk2; */
8760: /* double *popeffectif,*popcount; */
8761: /* double ***p3mat,***tabpop,***tabpopprev; */
8762: /* /\* double ***mobaverage; *\/ */
8763: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8764:
1.227 brouard 8765: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8766: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8767: /* agelim=AGESUP; */
8768: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8769:
1.227 brouard 8770: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8771:
8772:
1.227 brouard 8773: /* strcpy(filerespop,"POP_"); */
8774: /* strcat(filerespop,fileresu); */
8775: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8776: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8777: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8778: /* } */
8779: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8780: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8781:
1.227 brouard 8782: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8783:
1.227 brouard 8784: /* /\* if (mobilav!=0) { *\/ */
8785: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8786: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8787: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8788: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8789: /* /\* } *\/ */
8790: /* /\* } *\/ */
1.126 brouard 8791:
1.227 brouard 8792: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8793: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8794:
1.227 brouard 8795: /* agelim=AGESUP; */
1.126 brouard 8796:
1.227 brouard 8797: /* hstepm=1; */
8798: /* hstepm=hstepm/stepm; */
1.218 brouard 8799:
1.227 brouard 8800: /* if (popforecast==1) { */
8801: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8802: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8803: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8804: /* } */
8805: /* popage=ivector(0,AGESUP); */
8806: /* popeffectif=vector(0,AGESUP); */
8807: /* popcount=vector(0,AGESUP); */
1.126 brouard 8808:
1.227 brouard 8809: /* i=1; */
8810: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8811:
1.227 brouard 8812: /* imx=i; */
8813: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8814: /* } */
1.218 brouard 8815:
1.227 brouard 8816: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8817: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8818: /* k=k+1; */
8819: /* fprintf(ficrespop,"\n#******"); */
8820: /* for(j=1;j<=cptcoveff;j++) { */
8821: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8822: /* } */
8823: /* fprintf(ficrespop,"******\n"); */
8824: /* fprintf(ficrespop,"# Age"); */
8825: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8826: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8827:
1.227 brouard 8828: /* for (cpt=0; cpt<=0;cpt++) { */
8829: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8830:
1.227 brouard 8831: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8832: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8833: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8834:
1.227 brouard 8835: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8836: /* oldm=oldms;savm=savms; */
8837: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8838:
1.227 brouard 8839: /* for (h=0; h<=nhstepm; h++){ */
8840: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8841: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8842: /* } */
8843: /* for(j=1; j<=nlstate+ndeath;j++) { */
8844: /* kk1=0.;kk2=0; */
8845: /* for(i=1; i<=nlstate;i++) { */
8846: /* if (mobilav==1) */
8847: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8848: /* else { */
8849: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8850: /* } */
8851: /* } */
8852: /* if (h==(int)(calagedatem+12*cpt)){ */
8853: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8854: /* /\*fprintf(ficrespop," %.3f", kk1); */
8855: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8856: /* } */
8857: /* } */
8858: /* for(i=1; i<=nlstate;i++){ */
8859: /* kk1=0.; */
8860: /* for(j=1; j<=nlstate;j++){ */
8861: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8862: /* } */
8863: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8864: /* } */
1.218 brouard 8865:
1.227 brouard 8866: /* if (h==(int)(calagedatem+12*cpt)) */
8867: /* for(j=1; j<=nlstate;j++) */
8868: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8869: /* } */
8870: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8871: /* } */
8872: /* } */
1.218 brouard 8873:
1.227 brouard 8874: /* /\******\/ */
1.218 brouard 8875:
1.227 brouard 8876: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8877: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8878: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8879: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8880: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8881:
1.227 brouard 8882: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8883: /* oldm=oldms;savm=savms; */
8884: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8885: /* for (h=0; h<=nhstepm; h++){ */
8886: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8887: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8888: /* } */
8889: /* for(j=1; j<=nlstate+ndeath;j++) { */
8890: /* kk1=0.;kk2=0; */
8891: /* for(i=1; i<=nlstate;i++) { */
8892: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8893: /* } */
8894: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8895: /* } */
8896: /* } */
8897: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8898: /* } */
8899: /* } */
8900: /* } */
8901: /* } */
1.218 brouard 8902:
1.227 brouard 8903: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8904:
1.227 brouard 8905: /* if (popforecast==1) { */
8906: /* free_ivector(popage,0,AGESUP); */
8907: /* free_vector(popeffectif,0,AGESUP); */
8908: /* free_vector(popcount,0,AGESUP); */
8909: /* } */
8910: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8911: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8912: /* fclose(ficrespop); */
8913: /* } /\* End of popforecast *\/ */
1.218 brouard 8914:
1.126 brouard 8915: int fileappend(FILE *fichier, char *optionfich)
8916: {
8917: if((fichier=fopen(optionfich,"a"))==NULL) {
8918: printf("Problem with file: %s\n", optionfich);
8919: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8920: return (0);
8921: }
8922: fflush(fichier);
8923: return (1);
8924: }
8925:
8926:
8927: /**************** function prwizard **********************/
8928: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8929: {
8930:
8931: /* Wizard to print covariance matrix template */
8932:
1.164 brouard 8933: char ca[32], cb[32];
8934: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8935: int numlinepar;
8936:
8937: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8938: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8939: for(i=1; i <=nlstate; i++){
8940: jj=0;
8941: for(j=1; j <=nlstate+ndeath; j++){
8942: if(j==i) continue;
8943: jj++;
8944: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8945: printf("%1d%1d",i,j);
8946: fprintf(ficparo,"%1d%1d",i,j);
8947: for(k=1; k<=ncovmodel;k++){
8948: /* printf(" %lf",param[i][j][k]); */
8949: /* fprintf(ficparo," %lf",param[i][j][k]); */
8950: printf(" 0.");
8951: fprintf(ficparo," 0.");
8952: }
8953: printf("\n");
8954: fprintf(ficparo,"\n");
8955: }
8956: }
8957: printf("# Scales (for hessian or gradient estimation)\n");
8958: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8959: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8960: for(i=1; i <=nlstate; i++){
8961: jj=0;
8962: for(j=1; j <=nlstate+ndeath; j++){
8963: if(j==i) continue;
8964: jj++;
8965: fprintf(ficparo,"%1d%1d",i,j);
8966: printf("%1d%1d",i,j);
8967: fflush(stdout);
8968: for(k=1; k<=ncovmodel;k++){
8969: /* printf(" %le",delti3[i][j][k]); */
8970: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8971: printf(" 0.");
8972: fprintf(ficparo," 0.");
8973: }
8974: numlinepar++;
8975: printf("\n");
8976: fprintf(ficparo,"\n");
8977: }
8978: }
8979: printf("# Covariance matrix\n");
8980: /* # 121 Var(a12)\n\ */
8981: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8982: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8983: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8984: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8985: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8986: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8987: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8988: fflush(stdout);
8989: fprintf(ficparo,"# Covariance matrix\n");
8990: /* # 121 Var(a12)\n\ */
8991: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8992: /* # ...\n\ */
8993: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8994:
8995: for(itimes=1;itimes<=2;itimes++){
8996: jj=0;
8997: for(i=1; i <=nlstate; i++){
8998: for(j=1; j <=nlstate+ndeath; j++){
8999: if(j==i) continue;
9000: for(k=1; k<=ncovmodel;k++){
9001: jj++;
9002: ca[0]= k+'a'-1;ca[1]='\0';
9003: if(itimes==1){
9004: printf("#%1d%1d%d",i,j,k);
9005: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9006: }else{
9007: printf("%1d%1d%d",i,j,k);
9008: fprintf(ficparo,"%1d%1d%d",i,j,k);
9009: /* printf(" %.5le",matcov[i][j]); */
9010: }
9011: ll=0;
9012: for(li=1;li <=nlstate; li++){
9013: for(lj=1;lj <=nlstate+ndeath; lj++){
9014: if(lj==li) continue;
9015: for(lk=1;lk<=ncovmodel;lk++){
9016: ll++;
9017: if(ll<=jj){
9018: cb[0]= lk +'a'-1;cb[1]='\0';
9019: if(ll<jj){
9020: if(itimes==1){
9021: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9022: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9023: }else{
9024: printf(" 0.");
9025: fprintf(ficparo," 0.");
9026: }
9027: }else{
9028: if(itimes==1){
9029: printf(" Var(%s%1d%1d)",ca,i,j);
9030: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9031: }else{
9032: printf(" 0.");
9033: fprintf(ficparo," 0.");
9034: }
9035: }
9036: }
9037: } /* end lk */
9038: } /* end lj */
9039: } /* end li */
9040: printf("\n");
9041: fprintf(ficparo,"\n");
9042: numlinepar++;
9043: } /* end k*/
9044: } /*end j */
9045: } /* end i */
9046: } /* end itimes */
9047:
9048: } /* end of prwizard */
9049: /******************* Gompertz Likelihood ******************************/
9050: double gompertz(double x[])
9051: {
9052: double A,B,L=0.0,sump=0.,num=0.;
9053: int i,n=0; /* n is the size of the sample */
9054:
1.220 brouard 9055: for (i=1;i<=imx ; i++) {
1.126 brouard 9056: sump=sump+weight[i];
9057: /* sump=sump+1;*/
9058: num=num+1;
9059: }
9060:
9061:
9062: /* for (i=0; i<=imx; i++)
9063: 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]);*/
9064:
9065: for (i=1;i<=imx ; i++)
9066: {
9067: if (cens[i] == 1 && wav[i]>1)
9068: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9069:
9070: if (cens[i] == 0 && wav[i]>1)
9071: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
9072: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
9073:
9074: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9075: if (wav[i] > 1 ) { /* ??? */
9076: L=L+A*weight[i];
9077: /* 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]);*/
9078: }
9079: }
9080:
9081: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9082:
9083: return -2*L*num/sump;
9084: }
9085:
1.136 brouard 9086: #ifdef GSL
9087: /******************* Gompertz_f Likelihood ******************************/
9088: double gompertz_f(const gsl_vector *v, void *params)
9089: {
9090: double A,B,LL=0.0,sump=0.,num=0.;
9091: double *x= (double *) v->data;
9092: int i,n=0; /* n is the size of the sample */
9093:
9094: for (i=0;i<=imx-1 ; i++) {
9095: sump=sump+weight[i];
9096: /* sump=sump+1;*/
9097: num=num+1;
9098: }
9099:
9100:
9101: /* for (i=0; i<=imx; i++)
9102: 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]);*/
9103: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9104: for (i=1;i<=imx ; i++)
9105: {
9106: if (cens[i] == 1 && wav[i]>1)
9107: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9108:
9109: if (cens[i] == 0 && wav[i]>1)
9110: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9111: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9112:
9113: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9114: if (wav[i] > 1 ) { /* ??? */
9115: LL=LL+A*weight[i];
9116: /* 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]);*/
9117: }
9118: }
9119:
9120: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9121: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9122:
9123: return -2*LL*num/sump;
9124: }
9125: #endif
9126:
1.126 brouard 9127: /******************* Printing html file ***********/
1.201 brouard 9128: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9129: int lastpass, int stepm, int weightopt, char model[],\
9130: int imx, double p[],double **matcov,double agemortsup){
9131: int i,k;
9132:
9133: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9134: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9135: for (i=1;i<=2;i++)
9136: 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 9137: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9138: fprintf(fichtm,"</ul>");
9139:
9140: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9141:
9142: 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>");
9143:
9144: for (k=agegomp;k<(agemortsup-2);k++)
9145: 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]);
9146:
9147:
9148: fflush(fichtm);
9149: }
9150:
9151: /******************* Gnuplot file **************/
1.201 brouard 9152: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9153:
9154: char dirfileres[132],optfileres[132];
1.164 brouard 9155:
1.126 brouard 9156: int ng;
9157:
9158:
9159: /*#ifdef windows */
9160: fprintf(ficgp,"cd \"%s\" \n",pathc);
9161: /*#endif */
9162:
9163:
9164: strcpy(dirfileres,optionfilefiname);
9165: strcpy(optfileres,"vpl");
1.199 brouard 9166: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9167: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9168: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9169: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9170: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9171:
9172: }
9173:
1.136 brouard 9174: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9175: {
1.126 brouard 9176:
1.136 brouard 9177: /*-------- data file ----------*/
9178: FILE *fic;
9179: char dummy[]=" ";
1.240 brouard 9180: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9181: int lstra;
1.136 brouard 9182: int linei, month, year,iout;
9183: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9184: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9185: char *stratrunc;
1.223 brouard 9186:
1.240 brouard 9187: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9188: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9189:
1.240 brouard 9190: for(v=1; v <=ncovcol;v++){
9191: DummyV[v]=0;
9192: FixedV[v]=0;
9193: }
9194: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9195: DummyV[v]=1;
9196: FixedV[v]=0;
9197: }
9198: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9199: DummyV[v]=0;
9200: FixedV[v]=1;
9201: }
9202: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9203: DummyV[v]=1;
9204: FixedV[v]=1;
9205: }
9206: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9207: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9208: 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]);
9209: }
1.126 brouard 9210:
1.136 brouard 9211: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9212: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9213: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9214: }
1.126 brouard 9215:
1.136 brouard 9216: i=1;
9217: linei=0;
9218: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9219: linei=linei+1;
9220: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9221: if(line[j] == '\t')
9222: line[j] = ' ';
9223: }
9224: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9225: ;
9226: };
9227: line[j+1]=0; /* Trims blanks at end of line */
9228: if(line[0]=='#'){
9229: fprintf(ficlog,"Comment line\n%s\n",line);
9230: printf("Comment line\n%s\n",line);
9231: continue;
9232: }
9233: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9234: strcpy(line, linetmp);
1.223 brouard 9235:
9236: /* Loops on waves */
9237: for (j=maxwav;j>=1;j--){
9238: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9239: cutv(stra, strb, line, ' ');
9240: if(strb[0]=='.') { /* Missing value */
9241: lval=-1;
9242: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9243: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9244: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9245: 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);
9246: 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);
9247: return 1;
9248: }
9249: }else{
9250: errno=0;
9251: /* what_kind_of_number(strb); */
9252: dval=strtod(strb,&endptr);
9253: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9254: /* if(strb != endptr && *endptr == '\0') */
9255: /* dval=dlval; */
9256: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9257: if( strb[0]=='\0' || (*endptr != '\0')){
9258: 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);
9259: 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);
9260: return 1;
9261: }
9262: cotqvar[j][iv][i]=dval;
9263: cotvar[j][ntv+iv][i]=dval;
9264: }
9265: strcpy(line,stra);
1.223 brouard 9266: }/* end loop ntqv */
1.225 brouard 9267:
1.223 brouard 9268: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9269: cutv(stra, strb, line, ' ');
9270: if(strb[0]=='.') { /* Missing value */
9271: lval=-1;
9272: }else{
9273: errno=0;
9274: lval=strtol(strb,&endptr,10);
9275: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9276: if( strb[0]=='\0' || (*endptr != '\0')){
9277: 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);
9278: 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);
9279: return 1;
9280: }
9281: }
9282: if(lval <-1 || lval >1){
9283: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9284: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9285: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9286: For example, for multinomial values like 1, 2 and 3,\n \
9287: build V1=0 V2=0 for the reference value (1),\n \
9288: V1=1 V2=0 for (2) \n \
1.223 brouard 9289: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9290: output of IMaCh is often meaningless.\n \
1.223 brouard 9291: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9292: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9293: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9294: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9295: For example, for multinomial values like 1, 2 and 3,\n \
9296: build V1=0 V2=0 for the reference value (1),\n \
9297: V1=1 V2=0 for (2) \n \
1.223 brouard 9298: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9299: output of IMaCh is often meaningless.\n \
1.223 brouard 9300: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9301: return 1;
9302: }
9303: cotvar[j][iv][i]=(double)(lval);
9304: strcpy(line,stra);
1.223 brouard 9305: }/* end loop ntv */
1.225 brouard 9306:
1.223 brouard 9307: /* Statuses at wave */
1.137 brouard 9308: cutv(stra, strb, line, ' ');
1.223 brouard 9309: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9310: lval=-1;
1.136 brouard 9311: }else{
1.238 brouard 9312: errno=0;
9313: lval=strtol(strb,&endptr,10);
9314: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9315: if( strb[0]=='\0' || (*endptr != '\0')){
9316: 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);
9317: 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);
9318: return 1;
9319: }
1.136 brouard 9320: }
1.225 brouard 9321:
1.136 brouard 9322: s[j][i]=lval;
1.225 brouard 9323:
1.223 brouard 9324: /* Date of Interview */
1.136 brouard 9325: strcpy(line,stra);
9326: cutv(stra, strb,line,' ');
1.169 brouard 9327: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9328: }
1.169 brouard 9329: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9330: month=99;
9331: year=9999;
1.136 brouard 9332: }else{
1.225 brouard 9333: 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);
9334: 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);
9335: return 1;
1.136 brouard 9336: }
9337: anint[j][i]= (double) year;
9338: mint[j][i]= (double)month;
9339: strcpy(line,stra);
1.223 brouard 9340: } /* End loop on waves */
1.225 brouard 9341:
1.223 brouard 9342: /* Date of death */
1.136 brouard 9343: cutv(stra, strb,line,' ');
1.169 brouard 9344: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9345: }
1.169 brouard 9346: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9347: month=99;
9348: year=9999;
9349: }else{
1.141 brouard 9350: 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 9351: 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);
9352: return 1;
1.136 brouard 9353: }
9354: andc[i]=(double) year;
9355: moisdc[i]=(double) month;
9356: strcpy(line,stra);
9357:
1.223 brouard 9358: /* Date of birth */
1.136 brouard 9359: cutv(stra, strb,line,' ');
1.169 brouard 9360: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9361: }
1.169 brouard 9362: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9363: month=99;
9364: year=9999;
9365: }else{
1.141 brouard 9366: 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);
9367: 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 9368: return 1;
1.136 brouard 9369: }
9370: if (year==9999) {
1.141 brouard 9371: 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);
9372: 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 9373: return 1;
9374:
1.136 brouard 9375: }
9376: annais[i]=(double)(year);
9377: moisnais[i]=(double)(month);
9378: strcpy(line,stra);
1.225 brouard 9379:
1.223 brouard 9380: /* Sample weight */
1.136 brouard 9381: cutv(stra, strb,line,' ');
9382: errno=0;
9383: dval=strtod(strb,&endptr);
9384: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9385: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9386: 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 9387: fflush(ficlog);
9388: return 1;
9389: }
9390: weight[i]=dval;
9391: strcpy(line,stra);
1.225 brouard 9392:
1.223 brouard 9393: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9394: cutv(stra, strb, line, ' ');
9395: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9396: lval=-1;
1.223 brouard 9397: }else{
1.225 brouard 9398: errno=0;
9399: /* what_kind_of_number(strb); */
9400: dval=strtod(strb,&endptr);
9401: /* if(strb != endptr && *endptr == '\0') */
9402: /* dval=dlval; */
9403: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9404: if( strb[0]=='\0' || (*endptr != '\0')){
9405: 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);
9406: 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);
9407: return 1;
9408: }
9409: coqvar[iv][i]=dval;
1.226 brouard 9410: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9411: }
9412: strcpy(line,stra);
9413: }/* end loop nqv */
1.136 brouard 9414:
1.223 brouard 9415: /* Covariate values */
1.136 brouard 9416: for (j=ncovcol;j>=1;j--){
9417: cutv(stra, strb,line,' ');
1.223 brouard 9418: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9419: lval=-1;
1.136 brouard 9420: }else{
1.225 brouard 9421: errno=0;
9422: lval=strtol(strb,&endptr,10);
9423: if( strb[0]=='\0' || (*endptr != '\0')){
9424: 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);
9425: 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);
9426: return 1;
9427: }
1.136 brouard 9428: }
9429: if(lval <-1 || lval >1){
1.225 brouard 9430: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9431: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9432: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9433: For example, for multinomial values like 1, 2 and 3,\n \
9434: build V1=0 V2=0 for the reference value (1),\n \
9435: V1=1 V2=0 for (2) \n \
1.136 brouard 9436: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9437: output of IMaCh is often meaningless.\n \
1.136 brouard 9438: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9439: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9440: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9441: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9442: For example, for multinomial values like 1, 2 and 3,\n \
9443: build V1=0 V2=0 for the reference value (1),\n \
9444: V1=1 V2=0 for (2) \n \
1.136 brouard 9445: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9446: output of IMaCh is often meaningless.\n \
1.136 brouard 9447: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9448: return 1;
1.136 brouard 9449: }
9450: covar[j][i]=(double)(lval);
9451: strcpy(line,stra);
9452: }
9453: lstra=strlen(stra);
1.225 brouard 9454:
1.136 brouard 9455: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9456: stratrunc = &(stra[lstra-9]);
9457: num[i]=atol(stratrunc);
9458: }
9459: else
9460: num[i]=atol(stra);
9461: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9462: 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;}*/
9463:
9464: i=i+1;
9465: } /* End loop reading data */
1.225 brouard 9466:
1.136 brouard 9467: *imax=i-1; /* Number of individuals */
9468: fclose(fic);
1.225 brouard 9469:
1.136 brouard 9470: return (0);
1.164 brouard 9471: /* endread: */
1.225 brouard 9472: printf("Exiting readdata: ");
9473: fclose(fic);
9474: return (1);
1.223 brouard 9475: }
1.126 brouard 9476:
1.234 brouard 9477: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9478: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9479: while (*p2 == ' ')
1.234 brouard 9480: p2++;
9481: /* while ((*p1++ = *p2++) !=0) */
9482: /* ; */
9483: /* do */
9484: /* while (*p2 == ' ') */
9485: /* p2++; */
9486: /* while (*p1++ == *p2++); */
9487: *stri=p2;
1.145 brouard 9488: }
9489:
1.235 brouard 9490: int decoderesult ( char resultline[], int nres)
1.230 brouard 9491: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9492: {
1.235 brouard 9493: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9494: char resultsav[MAXLINE];
1.234 brouard 9495: int resultmodel[MAXLINE];
9496: int modelresult[MAXLINE];
1.230 brouard 9497: char stra[80], strb[80], strc[80], strd[80],stre[80];
9498:
1.234 brouard 9499: removefirstspace(&resultline);
1.233 brouard 9500: printf("decoderesult:%s\n",resultline);
1.230 brouard 9501:
9502: if (strstr(resultline,"v") !=0){
9503: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9504: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9505: return 1;
9506: }
9507: trimbb(resultsav, resultline);
9508: if (strlen(resultsav) >1){
9509: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9510: }
1.253 brouard 9511: if(j == 0){ /* Resultline but no = */
9512: TKresult[nres]=0; /* Combination for the nresult and the model */
9513: return (0);
9514: }
9515:
1.234 brouard 9516: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9517: 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);
9518: 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);
9519: }
9520: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9521: if(nbocc(resultsav,'=') >1){
9522: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9523: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9524: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9525: }else
9526: cutl(strc,strd,resultsav,'=');
1.230 brouard 9527: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9528:
1.230 brouard 9529: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9530: Tvarsel[k]=atoi(strc);
9531: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9532: /* cptcovsel++; */
9533: if (nbocc(stra,'=') >0)
9534: strcpy(resultsav,stra); /* and analyzes it */
9535: }
1.235 brouard 9536: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9537: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9538: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9539: match=0;
1.236 brouard 9540: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9541: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9542: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9543: match=1;
9544: break;
9545: }
9546: }
9547: if(match == 0){
9548: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9549: }
9550: }
9551: }
1.235 brouard 9552: /* Checking for missing or useless values in comparison of current model needs */
9553: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9554: match=0;
1.235 brouard 9555: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9556: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9557: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9558: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9559: ++match;
9560: }
9561: }
9562: }
9563: if(match == 0){
9564: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9565: }else if(match > 1){
9566: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9567: }
9568: }
1.235 brouard 9569:
1.234 brouard 9570: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9571: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9572: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9573: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9574: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9575: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9576: /* 1 0 0 0 */
9577: /* 2 1 0 0 */
9578: /* 3 0 1 0 */
9579: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9580: /* 5 0 0 1 */
9581: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9582: /* 7 0 1 1 */
9583: /* 8 1 1 1 */
1.237 brouard 9584: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9585: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9586: /* V5*age V5 known which value for nres? */
9587: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9588: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9589: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9590: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9591: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9592: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9593: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9594: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9595: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9596: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9597: k4++;;
9598: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9599: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9600: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9601: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9602: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9603: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9604: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9605: k4q++;;
9606: }
9607: }
1.234 brouard 9608:
1.235 brouard 9609: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9610: return (0);
9611: }
1.235 brouard 9612:
1.230 brouard 9613: int decodemodel( char model[], int lastobs)
9614: /**< This routine decodes the model and returns:
1.224 brouard 9615: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9616: * - nagesqr = 1 if age*age in the model, otherwise 0.
9617: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9618: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9619: * - cptcovage number of covariates with age*products =2
9620: * - cptcovs number of simple covariates
9621: * - 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
9622: * which is a new column after the 9 (ncovcol) variables.
9623: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9624: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9625: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9626: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9627: */
1.136 brouard 9628: {
1.238 brouard 9629: int i, j, k, ks, v;
1.227 brouard 9630: int j1, k1, k2, k3, k4;
1.136 brouard 9631: char modelsav[80];
1.145 brouard 9632: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9633: char *strpt;
1.136 brouard 9634:
1.145 brouard 9635: /*removespace(model);*/
1.136 brouard 9636: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9637: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9638: if (strstr(model,"AGE") !=0){
1.192 brouard 9639: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9640: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9641: return 1;
9642: }
1.141 brouard 9643: if (strstr(model,"v") !=0){
9644: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9645: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9646: return 1;
9647: }
1.187 brouard 9648: strcpy(modelsav,model);
9649: if ((strpt=strstr(model,"age*age")) !=0){
9650: printf(" strpt=%s, model=%s\n",strpt, model);
9651: if(strpt != model){
1.234 brouard 9652: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9653: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9654: corresponding column of parameters.\n",model);
1.234 brouard 9655: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9656: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9657: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9658: return 1;
1.225 brouard 9659: }
1.187 brouard 9660: nagesqr=1;
9661: if (strstr(model,"+age*age") !=0)
1.234 brouard 9662: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9663: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9664: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9665: else
1.234 brouard 9666: substrchaine(modelsav, model, "age*age");
1.187 brouard 9667: }else
9668: nagesqr=0;
9669: if (strlen(modelsav) >1){
9670: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9671: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9672: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9673: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9674: * cst, age and age*age
9675: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9676: /* including age products which are counted in cptcovage.
9677: * but the covariates which are products must be treated
9678: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9679: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9680: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9681:
9682:
1.187 brouard 9683: /* Design
9684: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9685: * < ncovcol=8 >
9686: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9687: * k= 1 2 3 4 5 6 7 8
9688: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9689: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9690: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9691: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9692: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9693: * Tage[++cptcovage]=k
9694: * if products, new covar are created after ncovcol with k1
9695: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9696: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9697: * 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
9698: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9699: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9700: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9701: * < ncovcol=8 >
9702: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9703: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9704: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9705: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9706: * p Tprod[1]@2={ 6, 5}
9707: *p Tvard[1][1]@4= {7, 8, 5, 6}
9708: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9709: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9710: *How to reorganize?
9711: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9712: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9713: * {2, 1, 4, 8, 5, 6, 3, 7}
9714: * Struct []
9715: */
1.225 brouard 9716:
1.187 brouard 9717: /* This loop fills the array Tvar from the string 'model'.*/
9718: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9719: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9720: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9721: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9722: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9723: /* k=1 Tvar[1]=2 (from V2) */
9724: /* k=5 Tvar[5] */
9725: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9726: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9727: /* } */
1.198 brouard 9728: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9729: /*
9730: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9731: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9732: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9733: }
1.187 brouard 9734: cptcovage=0;
9735: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9736: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9737: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9738: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9739: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9740: /*scanf("%d",i);*/
9741: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9742: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9743: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9744: /* covar is not filled and then is empty */
9745: cptcovprod--;
9746: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9747: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9748: Typevar[k]=1; /* 1 for age product */
9749: cptcovage++; /* Sums the number of covariates which include age as a product */
9750: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9751: /*printf("stre=%s ", stre);*/
9752: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9753: cptcovprod--;
9754: cutl(stre,strb,strc,'V');
9755: Tvar[k]=atoi(stre);
9756: Typevar[k]=1; /* 1 for age product */
9757: cptcovage++;
9758: Tage[cptcovage]=k;
9759: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9760: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9761: cptcovn++;
9762: cptcovprodnoage++;k1++;
9763: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9764: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9765: because this model-covariate is a construction we invent a new column
9766: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9767: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9768: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9769: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9770: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9771: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9772: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9773: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9774: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9775: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9776: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9777: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9778: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9779: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9780: for (i=1; i<=lastobs;i++){
9781: /* Computes the new covariate which is a product of
9782: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9783: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9784: }
9785: } /* End age is not in the model */
9786: } /* End if model includes a product */
9787: else { /* no more sum */
9788: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9789: /* scanf("%d",i);*/
9790: cutl(strd,strc,strb,'V');
9791: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9792: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9793: Tvar[k]=atoi(strd);
9794: Typevar[k]=0; /* 0 for simple covariates */
9795: }
9796: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9797: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9798: scanf("%d",i);*/
1.187 brouard 9799: } /* end of loop + on total covariates */
9800: } /* end if strlen(modelsave == 0) age*age might exist */
9801: } /* end if strlen(model == 0) */
1.136 brouard 9802:
9803: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9804: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9805:
1.136 brouard 9806: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9807: printf("cptcovprod=%d ", cptcovprod);
9808: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9809: scanf("%d ",i);*/
9810:
9811:
1.230 brouard 9812: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9813: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9814: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9815: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9816: k = 1 2 3 4 5 6 7 8 9
9817: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9818: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9819: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9820: Dummy[k] 1 0 0 0 3 1 1 2 3
9821: Tmodelind[combination of covar]=k;
1.225 brouard 9822: */
9823: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9824: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9825: /* 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 9826: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9827: printf("Model=%s\n\
9828: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9829: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9830: 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);
9831: fprintf(ficlog,"Model=%s\n\
9832: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9833: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9834: 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 9835: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9836: 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 */
9837: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9838: Fixed[k]= 0;
9839: Dummy[k]= 0;
1.225 brouard 9840: ncoveff++;
1.232 brouard 9841: ncovf++;
1.234 brouard 9842: nsd++;
9843: modell[k].maintype= FTYPE;
9844: TvarsD[nsd]=Tvar[k];
9845: TvarsDind[nsd]=k;
9846: TvarF[ncovf]=Tvar[k];
9847: TvarFind[ncovf]=k;
9848: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9849: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9850: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9851: Fixed[k]= 0;
9852: Dummy[k]= 0;
9853: ncoveff++;
9854: ncovf++;
9855: modell[k].maintype= FTYPE;
9856: TvarF[ncovf]=Tvar[k];
9857: TvarFind[ncovf]=k;
1.230 brouard 9858: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9859: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9860: }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 9861: Fixed[k]= 0;
9862: Dummy[k]= 1;
1.230 brouard 9863: nqfveff++;
1.234 brouard 9864: modell[k].maintype= FTYPE;
9865: modell[k].subtype= FQ;
9866: nsq++;
9867: TvarsQ[nsq]=Tvar[k];
9868: TvarsQind[nsq]=k;
1.232 brouard 9869: ncovf++;
1.234 brouard 9870: TvarF[ncovf]=Tvar[k];
9871: TvarFind[ncovf]=k;
1.231 brouard 9872: 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 9873: 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 9874: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9875: Fixed[k]= 1;
9876: Dummy[k]= 0;
1.225 brouard 9877: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9878: modell[k].maintype= VTYPE;
9879: modell[k].subtype= VD;
9880: nsd++;
9881: TvarsD[nsd]=Tvar[k];
9882: TvarsDind[nsd]=k;
9883: ncovv++; /* Only simple time varying variables */
9884: TvarV[ncovv]=Tvar[k];
1.242 brouard 9885: 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 9886: 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 */
9887: 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 9888: 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);
9889: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9890: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9891: Fixed[k]= 1;
9892: Dummy[k]= 1;
9893: nqtveff++;
9894: modell[k].maintype= VTYPE;
9895: modell[k].subtype= VQ;
9896: ncovv++; /* Only simple time varying variables */
9897: nsq++;
9898: TvarsQ[nsq]=Tvar[k];
9899: TvarsQind[nsq]=k;
9900: TvarV[ncovv]=Tvar[k];
1.242 brouard 9901: 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 9902: 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 */
9903: 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 9904: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9905: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9906: 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 9907: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9908: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9909: ncova++;
9910: TvarA[ncova]=Tvar[k];
9911: TvarAind[ncova]=k;
1.231 brouard 9912: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9913: Fixed[k]= 2;
9914: Dummy[k]= 2;
9915: modell[k].maintype= ATYPE;
9916: modell[k].subtype= APFD;
9917: /* ncoveff++; */
1.227 brouard 9918: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9919: Fixed[k]= 2;
9920: Dummy[k]= 3;
9921: modell[k].maintype= ATYPE;
9922: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9923: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9924: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9925: Fixed[k]= 3;
9926: Dummy[k]= 2;
9927: modell[k].maintype= ATYPE;
9928: modell[k].subtype= APVD; /* Product age * varying dummy */
9929: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9930: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9931: Fixed[k]= 3;
9932: Dummy[k]= 3;
9933: modell[k].maintype= ATYPE;
9934: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9935: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9936: }
9937: }else if (Typevar[k] == 2) { /* product without age */
9938: k1=Tposprod[k];
9939: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9940: if(Tvard[k1][2] <=ncovcol){
9941: Fixed[k]= 1;
9942: Dummy[k]= 0;
9943: modell[k].maintype= FTYPE;
9944: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9945: ncovf++; /* Fixed variables without age */
9946: TvarF[ncovf]=Tvar[k];
9947: TvarFind[ncovf]=k;
9948: }else if(Tvard[k1][2] <=ncovcol+nqv){
9949: Fixed[k]= 0; /* or 2 ?*/
9950: Dummy[k]= 1;
9951: modell[k].maintype= FTYPE;
9952: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9953: ncovf++; /* Varying variables without age */
9954: TvarF[ncovf]=Tvar[k];
9955: TvarFind[ncovf]=k;
9956: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9957: Fixed[k]= 1;
9958: Dummy[k]= 0;
9959: modell[k].maintype= VTYPE;
9960: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9961: ncovv++; /* Varying variables without age */
9962: TvarV[ncovv]=Tvar[k];
9963: TvarVind[ncovv]=k;
9964: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9965: Fixed[k]= 1;
9966: Dummy[k]= 1;
9967: modell[k].maintype= VTYPE;
9968: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9969: ncovv++; /* Varying variables without age */
9970: TvarV[ncovv]=Tvar[k];
9971: TvarVind[ncovv]=k;
9972: }
1.227 brouard 9973: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9974: if(Tvard[k1][2] <=ncovcol){
9975: Fixed[k]= 0; /* or 2 ?*/
9976: Dummy[k]= 1;
9977: modell[k].maintype= FTYPE;
9978: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9979: ncovf++; /* Fixed variables without age */
9980: TvarF[ncovf]=Tvar[k];
9981: TvarFind[ncovf]=k;
9982: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9983: Fixed[k]= 1;
9984: Dummy[k]= 1;
9985: modell[k].maintype= VTYPE;
9986: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9987: ncovv++; /* Varying variables without age */
9988: TvarV[ncovv]=Tvar[k];
9989: TvarVind[ncovv]=k;
9990: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9991: Fixed[k]= 1;
9992: Dummy[k]= 1;
9993: modell[k].maintype= VTYPE;
9994: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9995: ncovv++; /* Varying variables without age */
9996: TvarV[ncovv]=Tvar[k];
9997: TvarVind[ncovv]=k;
9998: ncovv++; /* Varying variables without age */
9999: TvarV[ncovv]=Tvar[k];
10000: TvarVind[ncovv]=k;
10001: }
1.227 brouard 10002: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10003: if(Tvard[k1][2] <=ncovcol){
10004: Fixed[k]= 1;
10005: Dummy[k]= 1;
10006: modell[k].maintype= VTYPE;
10007: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10008: ncovv++; /* Varying variables without age */
10009: TvarV[ncovv]=Tvar[k];
10010: TvarVind[ncovv]=k;
10011: }else if(Tvard[k1][2] <=ncovcol+nqv){
10012: Fixed[k]= 1;
10013: Dummy[k]= 1;
10014: modell[k].maintype= VTYPE;
10015: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10016: ncovv++; /* Varying variables without age */
10017: TvarV[ncovv]=Tvar[k];
10018: TvarVind[ncovv]=k;
10019: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10020: Fixed[k]= 1;
10021: Dummy[k]= 0;
10022: modell[k].maintype= VTYPE;
10023: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10024: ncovv++; /* Varying variables without age */
10025: TvarV[ncovv]=Tvar[k];
10026: TvarVind[ncovv]=k;
10027: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10028: Fixed[k]= 1;
10029: Dummy[k]= 1;
10030: modell[k].maintype= VTYPE;
10031: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10032: ncovv++; /* Varying variables without age */
10033: TvarV[ncovv]=Tvar[k];
10034: TvarVind[ncovv]=k;
10035: }
1.227 brouard 10036: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10037: if(Tvard[k1][2] <=ncovcol){
10038: Fixed[k]= 1;
10039: Dummy[k]= 1;
10040: modell[k].maintype= VTYPE;
10041: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10042: ncovv++; /* Varying variables without age */
10043: TvarV[ncovv]=Tvar[k];
10044: TvarVind[ncovv]=k;
10045: }else if(Tvard[k1][2] <=ncovcol+nqv){
10046: Fixed[k]= 1;
10047: Dummy[k]= 1;
10048: modell[k].maintype= VTYPE;
10049: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10050: ncovv++; /* Varying variables without age */
10051: TvarV[ncovv]=Tvar[k];
10052: TvarVind[ncovv]=k;
10053: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10054: Fixed[k]= 1;
10055: Dummy[k]= 1;
10056: modell[k].maintype= VTYPE;
10057: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10058: ncovv++; /* Varying variables without age */
10059: TvarV[ncovv]=Tvar[k];
10060: TvarVind[ncovv]=k;
10061: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10062: Fixed[k]= 1;
10063: Dummy[k]= 1;
10064: modell[k].maintype= VTYPE;
10065: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10066: ncovv++; /* Varying variables without age */
10067: TvarV[ncovv]=Tvar[k];
10068: TvarVind[ncovv]=k;
10069: }
1.227 brouard 10070: }else{
1.240 brouard 10071: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10072: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10073: } /*end k1*/
1.225 brouard 10074: }else{
1.226 brouard 10075: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10076: 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 10077: }
1.227 brouard 10078: 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 10079: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10080: 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]);
10081: }
10082: /* Searching for doublons in the model */
10083: for(k1=1; k1<= cptcovt;k1++){
10084: for(k2=1; k2 <k1;k2++){
1.285 brouard 10085: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10086: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10087: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10088: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10089: 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]);
10090: 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 10091: return(1);
10092: }
10093: }else if (Typevar[k1] ==2){
10094: k3=Tposprod[k1];
10095: k4=Tposprod[k2];
10096: 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])) ){
10097: 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]]);
10098: 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);
10099: return(1);
10100: }
10101: }
1.227 brouard 10102: }
10103: }
1.225 brouard 10104: }
10105: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10106: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10107: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10108: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10109: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10110: /*endread:*/
1.225 brouard 10111: printf("Exiting decodemodel: ");
10112: return (1);
1.136 brouard 10113: }
10114:
1.169 brouard 10115: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10116: {/* Check ages at death */
1.136 brouard 10117: int i, m;
1.218 brouard 10118: int firstone=0;
10119:
1.136 brouard 10120: for (i=1; i<=imx; i++) {
10121: for(m=2; (m<= maxwav); m++) {
10122: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10123: anint[m][i]=9999;
1.216 brouard 10124: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10125: s[m][i]=-1;
1.136 brouard 10126: }
10127: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10128: *nberr = *nberr + 1;
1.218 brouard 10129: if(firstone == 0){
10130: firstone=1;
1.260 brouard 10131: 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 10132: }
1.262 brouard 10133: 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 10134: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10135: }
10136: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10137: (*nberr)++;
1.259 brouard 10138: 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 10139: 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 10140: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10141: }
10142: }
10143: }
10144:
10145: for (i=1; i<=imx; i++) {
10146: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10147: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10148: 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 10149: if (s[m][i] >= nlstate+1) {
1.169 brouard 10150: if(agedc[i]>0){
10151: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10152: agev[m][i]=agedc[i];
1.214 brouard 10153: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10154: }else {
1.136 brouard 10155: if ((int)andc[i]!=9999){
10156: nbwarn++;
10157: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10158: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10159: agev[m][i]=-1;
10160: }
10161: }
1.169 brouard 10162: } /* agedc > 0 */
1.214 brouard 10163: } /* end if */
1.136 brouard 10164: else if(s[m][i] !=9){ /* Standard case, age in fractional
10165: years but with the precision of a month */
10166: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10167: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10168: agev[m][i]=1;
10169: else if(agev[m][i] < *agemin){
10170: *agemin=agev[m][i];
10171: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10172: }
10173: else if(agev[m][i] >*agemax){
10174: *agemax=agev[m][i];
1.156 brouard 10175: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10176: }
10177: /*agev[m][i]=anint[m][i]-annais[i];*/
10178: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10179: } /* en if 9*/
1.136 brouard 10180: else { /* =9 */
1.214 brouard 10181: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10182: agev[m][i]=1;
10183: s[m][i]=-1;
10184: }
10185: }
1.214 brouard 10186: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10187: agev[m][i]=1;
1.214 brouard 10188: else{
10189: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10190: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10191: agev[m][i]=0;
10192: }
10193: } /* End for lastpass */
10194: }
1.136 brouard 10195:
10196: for (i=1; i<=imx; i++) {
10197: for(m=firstpass; (m<=lastpass); m++){
10198: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10199: (*nberr)++;
1.136 brouard 10200: 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);
10201: 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);
10202: return 1;
10203: }
10204: }
10205: }
10206:
10207: /*for (i=1; i<=imx; i++){
10208: for (m=firstpass; (m<lastpass); m++){
10209: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10210: }
10211:
10212: }*/
10213:
10214:
1.139 brouard 10215: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10216: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10217:
10218: return (0);
1.164 brouard 10219: /* endread:*/
1.136 brouard 10220: printf("Exiting calandcheckages: ");
10221: return (1);
10222: }
10223:
1.172 brouard 10224: #if defined(_MSC_VER)
10225: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10226: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10227: //#include "stdafx.h"
10228: //#include <stdio.h>
10229: //#include <tchar.h>
10230: //#include <windows.h>
10231: //#include <iostream>
10232: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10233:
10234: LPFN_ISWOW64PROCESS fnIsWow64Process;
10235:
10236: BOOL IsWow64()
10237: {
10238: BOOL bIsWow64 = FALSE;
10239:
10240: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10241: // (HANDLE, PBOOL);
10242:
10243: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10244:
10245: HMODULE module = GetModuleHandle(_T("kernel32"));
10246: const char funcName[] = "IsWow64Process";
10247: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10248: GetProcAddress(module, funcName);
10249:
10250: if (NULL != fnIsWow64Process)
10251: {
10252: if (!fnIsWow64Process(GetCurrentProcess(),
10253: &bIsWow64))
10254: //throw std::exception("Unknown error");
10255: printf("Unknown error\n");
10256: }
10257: return bIsWow64 != FALSE;
10258: }
10259: #endif
1.177 brouard 10260:
1.191 brouard 10261: void syscompilerinfo(int logged)
1.292 brouard 10262: {
10263: #include <stdint.h>
10264:
10265: /* #include "syscompilerinfo.h"*/
1.185 brouard 10266: /* command line Intel compiler 32bit windows, XP compatible:*/
10267: /* /GS /W3 /Gy
10268: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10269: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10270: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10271: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10272: */
10273: /* 64 bits */
1.185 brouard 10274: /*
10275: /GS /W3 /Gy
10276: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10277: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10278: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10279: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10280: /* Optimization are useless and O3 is slower than O2 */
10281: /*
10282: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10283: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10284: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10285: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10286: */
1.186 brouard 10287: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10288: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10289: /PDB:"visual studio
10290: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10291: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10292: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10293: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10294: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10295: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10296: uiAccess='false'"
10297: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10298: /NOLOGO /TLBID:1
10299: */
1.292 brouard 10300:
10301:
1.177 brouard 10302: #if defined __INTEL_COMPILER
1.178 brouard 10303: #if defined(__GNUC__)
10304: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10305: #endif
1.177 brouard 10306: #elif defined(__GNUC__)
1.179 brouard 10307: #ifndef __APPLE__
1.174 brouard 10308: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10309: #endif
1.177 brouard 10310: struct utsname sysInfo;
1.178 brouard 10311: int cross = CROSS;
10312: if (cross){
10313: printf("Cross-");
1.191 brouard 10314: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10315: }
1.174 brouard 10316: #endif
10317:
1.191 brouard 10318: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10319: #if defined(__clang__)
1.191 brouard 10320: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10321: #endif
10322: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10323: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10324: #endif
10325: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10326: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10327: #endif
10328: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10329: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10330: #endif
10331: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10332: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10333: #endif
10334: #if defined(_MSC_VER)
1.191 brouard 10335: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10336: #endif
10337: #if defined(__PGI)
1.191 brouard 10338: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10339: #endif
10340: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10341: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10342: #endif
1.191 brouard 10343: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10344:
1.167 brouard 10345: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10346: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10347: // Windows (x64 and x86)
1.191 brouard 10348: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10349: #elif __unix__ // all unices, not all compilers
10350: // Unix
1.191 brouard 10351: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10352: #elif __linux__
10353: // linux
1.191 brouard 10354: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10355: #elif __APPLE__
1.174 brouard 10356: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10357: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10358: #endif
10359:
10360: /* __MINGW32__ */
10361: /* __CYGWIN__ */
10362: /* __MINGW64__ */
10363: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10364: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10365: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10366: /* _WIN64 // Defined for applications for Win64. */
10367: /* _M_X64 // Defined for compilations that target x64 processors. */
10368: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10369:
1.167 brouard 10370: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10371: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10372: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10373: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10374: #else
1.191 brouard 10375: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10376: #endif
10377:
1.169 brouard 10378: #if defined(__GNUC__)
10379: # if defined(__GNUC_PATCHLEVEL__)
10380: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10381: + __GNUC_MINOR__ * 100 \
10382: + __GNUC_PATCHLEVEL__)
10383: # else
10384: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10385: + __GNUC_MINOR__ * 100)
10386: # endif
1.174 brouard 10387: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10388: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10389:
10390: if (uname(&sysInfo) != -1) {
10391: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10392: 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 10393: }
10394: else
10395: perror("uname() error");
1.179 brouard 10396: //#ifndef __INTEL_COMPILER
10397: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10398: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10399: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10400: #endif
1.169 brouard 10401: #endif
1.172 brouard 10402:
1.286 brouard 10403: // void main ()
1.172 brouard 10404: // {
1.169 brouard 10405: #if defined(_MSC_VER)
1.174 brouard 10406: if (IsWow64()){
1.191 brouard 10407: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10408: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10409: }
10410: else{
1.191 brouard 10411: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10412: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10413: }
1.172 brouard 10414: // printf("\nPress Enter to continue...");
10415: // getchar();
10416: // }
10417:
1.169 brouard 10418: #endif
10419:
1.167 brouard 10420:
1.219 brouard 10421: }
1.136 brouard 10422:
1.219 brouard 10423: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10424: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10425: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10426: /* double ftolpl = 1.e-10; */
1.180 brouard 10427: double age, agebase, agelim;
1.203 brouard 10428: double tot;
1.180 brouard 10429:
1.202 brouard 10430: strcpy(filerespl,"PL_");
10431: strcat(filerespl,fileresu);
10432: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10433: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10434: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10435: }
1.288 brouard 10436: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10437: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10438: pstamp(ficrespl);
1.288 brouard 10439: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10440: fprintf(ficrespl,"#Age ");
10441: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10442: fprintf(ficrespl,"\n");
1.180 brouard 10443:
1.219 brouard 10444: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10445:
1.219 brouard 10446: agebase=ageminpar;
10447: agelim=agemaxpar;
1.180 brouard 10448:
1.227 brouard 10449: /* i1=pow(2,ncoveff); */
1.234 brouard 10450: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10451: if (cptcovn < 1){i1=1;}
1.180 brouard 10452:
1.238 brouard 10453: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10454: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10455: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10456: continue;
1.235 brouard 10457:
1.238 brouard 10458: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10459: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10460: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10461: /* k=k+1; */
10462: /* to clean */
10463: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10464: fprintf(ficrespl,"#******");
10465: printf("#******");
10466: fprintf(ficlog,"#******");
10467: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10468: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10469: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10470: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10471: }
10472: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10473: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10474: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10475: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10476: }
10477: fprintf(ficrespl,"******\n");
10478: printf("******\n");
10479: fprintf(ficlog,"******\n");
10480: if(invalidvarcomb[k]){
10481: printf("\nCombination (%d) ignored because no case \n",k);
10482: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10483: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10484: continue;
10485: }
1.219 brouard 10486:
1.238 brouard 10487: fprintf(ficrespl,"#Age ");
10488: for(j=1;j<=cptcoveff;j++) {
10489: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10490: }
10491: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10492: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10493:
1.238 brouard 10494: for (age=agebase; age<=agelim; age++){
10495: /* for (age=agebase; age<=agebase; age++){ */
10496: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10497: fprintf(ficrespl,"%.0f ",age );
10498: for(j=1;j<=cptcoveff;j++)
10499: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10500: tot=0.;
10501: for(i=1; i<=nlstate;i++){
10502: tot += prlim[i][i];
10503: fprintf(ficrespl," %.5f", prlim[i][i]);
10504: }
10505: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10506: } /* Age */
10507: /* was end of cptcod */
10508: } /* cptcov */
10509: } /* nres */
1.219 brouard 10510: return 0;
1.180 brouard 10511: }
10512:
1.218 brouard 10513: 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 10514: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10515:
10516: /* Computes the back prevalence limit for any combination of covariate values
10517: * at any age between ageminpar and agemaxpar
10518: */
1.235 brouard 10519: int i, j, k, i1, nres=0 ;
1.217 brouard 10520: /* double ftolpl = 1.e-10; */
10521: double age, agebase, agelim;
10522: double tot;
1.218 brouard 10523: /* double ***mobaverage; */
10524: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10525:
10526: strcpy(fileresplb,"PLB_");
10527: strcat(fileresplb,fileresu);
10528: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10529: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10530: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10531: }
1.288 brouard 10532: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10533: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10534: pstamp(ficresplb);
1.288 brouard 10535: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10536: fprintf(ficresplb,"#Age ");
10537: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10538: fprintf(ficresplb,"\n");
10539:
1.218 brouard 10540:
10541: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10542:
10543: agebase=ageminpar;
10544: agelim=agemaxpar;
10545:
10546:
1.227 brouard 10547: i1=pow(2,cptcoveff);
1.218 brouard 10548: if (cptcovn < 1){i1=1;}
1.227 brouard 10549:
1.238 brouard 10550: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10551: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10552: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10553: continue;
10554: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10555: fprintf(ficresplb,"#******");
10556: printf("#******");
10557: fprintf(ficlog,"#******");
10558: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10559: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10560: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10561: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10562: }
10563: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10564: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10565: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10566: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10567: }
10568: fprintf(ficresplb,"******\n");
10569: printf("******\n");
10570: fprintf(ficlog,"******\n");
10571: if(invalidvarcomb[k]){
10572: printf("\nCombination (%d) ignored because no cases \n",k);
10573: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10574: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10575: continue;
10576: }
1.218 brouard 10577:
1.238 brouard 10578: fprintf(ficresplb,"#Age ");
10579: for(j=1;j<=cptcoveff;j++) {
10580: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10581: }
10582: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10583: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10584:
10585:
1.238 brouard 10586: for (age=agebase; age<=agelim; age++){
10587: /* for (age=agebase; age<=agebase; age++){ */
10588: if(mobilavproj > 0){
10589: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10590: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10591: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10592: }else if (mobilavproj == 0){
10593: 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);
10594: 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);
10595: exit(1);
10596: }else{
10597: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10598: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10599: /* printf("TOTOT\n"); */
10600: /* exit(1); */
1.238 brouard 10601: }
10602: fprintf(ficresplb,"%.0f ",age );
10603: for(j=1;j<=cptcoveff;j++)
10604: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10605: tot=0.;
10606: for(i=1; i<=nlstate;i++){
10607: tot += bprlim[i][i];
10608: fprintf(ficresplb," %.5f", bprlim[i][i]);
10609: }
10610: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10611: } /* Age */
10612: /* was end of cptcod */
1.255 brouard 10613: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10614: } /* end of any combination */
10615: } /* end of nres */
1.218 brouard 10616: /* hBijx(p, bage, fage); */
10617: /* fclose(ficrespijb); */
10618:
10619: return 0;
1.217 brouard 10620: }
1.218 brouard 10621:
1.180 brouard 10622: int hPijx(double *p, int bage, int fage){
10623: /*------------- h Pij x at various ages ------------*/
10624:
10625: int stepsize;
10626: int agelim;
10627: int hstepm;
10628: int nhstepm;
1.235 brouard 10629: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10630:
10631: double agedeb;
10632: double ***p3mat;
10633:
1.201 brouard 10634: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10635: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10636: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10637: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10638: }
10639: printf("Computing pij: result on file '%s' \n", filerespij);
10640: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10641:
10642: stepsize=(int) (stepm+YEARM-1)/YEARM;
10643: /*if (stepm<=24) stepsize=2;*/
10644:
10645: agelim=AGESUP;
10646: hstepm=stepsize*YEARM; /* Every year of age */
10647: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10648:
1.180 brouard 10649: /* hstepm=1; aff par mois*/
10650: pstamp(ficrespij);
10651: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10652: i1= pow(2,cptcoveff);
1.218 brouard 10653: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10654: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10655: /* k=k+1; */
1.235 brouard 10656: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10657: for(k=1; k<=i1;k++){
1.253 brouard 10658: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10659: continue;
1.183 brouard 10660: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10661: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10662: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10663: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10664: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10665: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10666: }
1.183 brouard 10667: fprintf(ficrespij,"******\n");
10668:
10669: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10670: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10671: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10672:
10673: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10674:
1.183 brouard 10675: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10676: oldm=oldms;savm=savms;
1.235 brouard 10677: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10678: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10679: for(i=1; i<=nlstate;i++)
10680: for(j=1; j<=nlstate+ndeath;j++)
10681: fprintf(ficrespij," %1d-%1d",i,j);
10682: fprintf(ficrespij,"\n");
10683: for (h=0; h<=nhstepm; h++){
10684: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10685: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10686: for(i=1; i<=nlstate;i++)
10687: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10688: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10689: fprintf(ficrespij,"\n");
10690: }
1.183 brouard 10691: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10692: fprintf(ficrespij,"\n");
10693: }
1.180 brouard 10694: /*}*/
10695: }
1.218 brouard 10696: return 0;
1.180 brouard 10697: }
1.218 brouard 10698:
10699: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10700: /*------------- h Bij x at various ages ------------*/
10701:
10702: int stepsize;
1.218 brouard 10703: /* int agelim; */
10704: int ageminl;
1.217 brouard 10705: int hstepm;
10706: int nhstepm;
1.238 brouard 10707: int h, i, i1, j, k, nres;
1.218 brouard 10708:
1.217 brouard 10709: double agedeb;
10710: double ***p3mat;
1.218 brouard 10711:
10712: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10713: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10714: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10715: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10716: }
10717: printf("Computing pij back: result on file '%s' \n", filerespijb);
10718: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10719:
10720: stepsize=(int) (stepm+YEARM-1)/YEARM;
10721: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10722:
1.218 brouard 10723: /* agelim=AGESUP; */
1.289 brouard 10724: ageminl=AGEINF; /* was 30 */
1.218 brouard 10725: hstepm=stepsize*YEARM; /* Every year of age */
10726: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10727:
10728: /* hstepm=1; aff par mois*/
10729: pstamp(ficrespijb);
1.255 brouard 10730: 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 10731: i1= pow(2,cptcoveff);
1.218 brouard 10732: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10733: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10734: /* k=k+1; */
1.238 brouard 10735: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10736: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10737: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10738: continue;
10739: fprintf(ficrespijb,"\n#****** ");
10740: for(j=1;j<=cptcoveff;j++)
10741: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10742: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10743: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10744: }
10745: fprintf(ficrespijb,"******\n");
1.264 brouard 10746: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10747: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10748: continue;
10749: }
10750:
10751: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10752: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10753: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 10754: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm+0.1)-1; /* Typically 20 years = 20*12/6=40 or 55*12/24=27.5-1.1=>27 */
10755: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 10756:
10757: /* nhstepm=nhstepm*YEARM; aff par mois*/
10758:
1.266 brouard 10759: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10760: /* and memory limitations if stepm is small */
10761:
1.238 brouard 10762: /* oldm=oldms;savm=savms; */
10763: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10764: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10765: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10766: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10767: for(i=1; i<=nlstate;i++)
10768: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10769: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10770: fprintf(ficrespijb,"\n");
1.238 brouard 10771: for (h=0; h<=nhstepm; h++){
10772: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10773: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10774: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10775: for(i=1; i<=nlstate;i++)
10776: for(j=1; j<=nlstate+ndeath;j++)
10777: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10778: fprintf(ficrespijb,"\n");
10779: }
10780: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10781: fprintf(ficrespijb,"\n");
10782: } /* end age deb */
10783: } /* end combination */
10784: } /* end nres */
1.218 brouard 10785: return 0;
10786: } /* hBijx */
1.217 brouard 10787:
1.180 brouard 10788:
1.136 brouard 10789: /***********************************************/
10790: /**************** Main Program *****************/
10791: /***********************************************/
10792:
10793: int main(int argc, char *argv[])
10794: {
10795: #ifdef GSL
10796: const gsl_multimin_fminimizer_type *T;
10797: size_t iteri = 0, it;
10798: int rval = GSL_CONTINUE;
10799: int status = GSL_SUCCESS;
10800: double ssval;
10801: #endif
10802: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 10803: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
10804: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 10805: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10806: int jj, ll, li, lj, lk;
1.136 brouard 10807: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10808: int num_filled;
1.136 brouard 10809: int itimes;
10810: int NDIM=2;
10811: int vpopbased=0;
1.235 brouard 10812: int nres=0;
1.258 brouard 10813: int endishere=0;
1.277 brouard 10814: int noffset=0;
1.274 brouard 10815: int ncurrv=0; /* Temporary variable */
10816:
1.164 brouard 10817: char ca[32], cb[32];
1.136 brouard 10818: /* FILE *fichtm; *//* Html File */
10819: /* FILE *ficgp;*/ /*Gnuplot File */
10820: struct stat info;
1.191 brouard 10821: double agedeb=0.;
1.194 brouard 10822:
10823: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10824: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10825:
1.165 brouard 10826: double fret;
1.191 brouard 10827: double dum=0.; /* Dummy variable */
1.136 brouard 10828: double ***p3mat;
1.218 brouard 10829: /* double ***mobaverage; */
1.164 brouard 10830:
10831: char line[MAXLINE];
1.197 brouard 10832: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10833:
1.234 brouard 10834: char modeltemp[MAXLINE];
1.230 brouard 10835: char resultline[MAXLINE];
10836:
1.136 brouard 10837: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10838: char *tok, *val; /* pathtot */
1.290 brouard 10839: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 10840: int c, h , cpt, c2;
1.191 brouard 10841: int jl=0;
10842: int i1, j1, jk, stepsize=0;
1.194 brouard 10843: int count=0;
10844:
1.164 brouard 10845: int *tab;
1.136 brouard 10846: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 10847: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
10848: /* double anprojf, mprojf, jprojf; */
10849: /* double jintmean,mintmean,aintmean; */
10850: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10851: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10852: double yrfproj= 10.0; /* Number of years of forward projections */
10853: double yrbproj= 10.0; /* Number of years of backward projections */
10854: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 10855: int mobilav=0,popforecast=0;
1.191 brouard 10856: int hstepm=0, nhstepm=0;
1.136 brouard 10857: int agemortsup;
10858: float sumlpop=0.;
10859: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10860: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10861:
1.191 brouard 10862: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10863: double ftolpl=FTOL;
10864: double **prlim;
1.217 brouard 10865: double **bprlim;
1.136 brouard 10866: double ***param; /* Matrix of parameters */
1.251 brouard 10867: double ***paramstart; /* Matrix of starting parameter values */
10868: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10869: double **matcov; /* Matrix of covariance */
1.203 brouard 10870: double **hess; /* Hessian matrix */
1.136 brouard 10871: double ***delti3; /* Scale */
10872: double *delti; /* Scale */
10873: double ***eij, ***vareij;
10874: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10875:
1.136 brouard 10876: double *epj, vepp;
1.164 brouard 10877:
1.273 brouard 10878: double dateprev1, dateprev2;
1.296 brouard 10879: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
10880: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
10881:
1.217 brouard 10882:
1.136 brouard 10883: double **ximort;
1.145 brouard 10884: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10885: int *dcwave;
10886:
1.164 brouard 10887: char z[1]="c";
1.136 brouard 10888:
10889: /*char *strt;*/
10890: char strtend[80];
1.126 brouard 10891:
1.164 brouard 10892:
1.126 brouard 10893: /* setlocale (LC_ALL, ""); */
10894: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10895: /* textdomain (PACKAGE); */
10896: /* setlocale (LC_CTYPE, ""); */
10897: /* setlocale (LC_MESSAGES, ""); */
10898:
10899: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10900: rstart_time = time(NULL);
10901: /* (void) gettimeofday(&start_time,&tzp);*/
10902: start_time = *localtime(&rstart_time);
1.126 brouard 10903: curr_time=start_time;
1.157 brouard 10904: /*tml = *localtime(&start_time.tm_sec);*/
10905: /* strcpy(strstart,asctime(&tml)); */
10906: strcpy(strstart,asctime(&start_time));
1.126 brouard 10907:
10908: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10909: /* tp.tm_sec = tp.tm_sec +86400; */
10910: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10911: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10912: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10913: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10914: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10915: /* strt=asctime(&tmg); */
10916: /* printf("Time(after) =%s",strstart); */
10917: /* (void) time (&time_value);
10918: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10919: * tm = *localtime(&time_value);
10920: * strstart=asctime(&tm);
10921: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10922: */
10923:
10924: nberr=0; /* Number of errors and warnings */
10925: nbwarn=0;
1.184 brouard 10926: #ifdef WIN32
10927: _getcwd(pathcd, size);
10928: #else
1.126 brouard 10929: getcwd(pathcd, size);
1.184 brouard 10930: #endif
1.191 brouard 10931: syscompilerinfo(0);
1.196 brouard 10932: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10933: if(argc <=1){
10934: printf("\nEnter the parameter file name: ");
1.205 brouard 10935: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10936: printf("ERROR Empty parameter file name\n");
10937: goto end;
10938: }
1.126 brouard 10939: i=strlen(pathr);
10940: if(pathr[i-1]=='\n')
10941: pathr[i-1]='\0';
1.156 brouard 10942: i=strlen(pathr);
1.205 brouard 10943: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10944: pathr[i-1]='\0';
1.205 brouard 10945: }
10946: i=strlen(pathr);
10947: if( i==0 ){
10948: printf("ERROR Empty parameter file name\n");
10949: goto end;
10950: }
10951: for (tok = pathr; tok != NULL; ){
1.126 brouard 10952: printf("Pathr |%s|\n",pathr);
10953: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10954: printf("val= |%s| pathr=%s\n",val,pathr);
10955: strcpy (pathtot, val);
10956: if(pathr[0] == '\0') break; /* Dirty */
10957: }
10958: }
1.281 brouard 10959: else if (argc<=2){
10960: strcpy(pathtot,argv[1]);
10961: }
1.126 brouard 10962: else{
10963: strcpy(pathtot,argv[1]);
1.281 brouard 10964: strcpy(z,argv[2]);
10965: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 10966: }
10967: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10968: /*cygwin_split_path(pathtot,path,optionfile);
10969: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10970: /* cutv(path,optionfile,pathtot,'\\');*/
10971:
10972: /* Split argv[0], imach program to get pathimach */
10973: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10974: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10975: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10976: /* strcpy(pathimach,argv[0]); */
10977: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10978: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10979: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10980: #ifdef WIN32
10981: _chdir(path); /* Can be a relative path */
10982: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10983: #else
1.126 brouard 10984: chdir(path); /* Can be a relative path */
1.184 brouard 10985: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10986: #endif
10987: printf("Current directory %s!\n",pathcd);
1.126 brouard 10988: strcpy(command,"mkdir ");
10989: strcat(command,optionfilefiname);
10990: if((outcmd=system(command)) != 0){
1.169 brouard 10991: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10992: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10993: /* fclose(ficlog); */
10994: /* exit(1); */
10995: }
10996: /* if((imk=mkdir(optionfilefiname))<0){ */
10997: /* perror("mkdir"); */
10998: /* } */
10999:
11000: /*-------- arguments in the command line --------*/
11001:
1.186 brouard 11002: /* Main Log file */
1.126 brouard 11003: strcat(filelog, optionfilefiname);
11004: strcat(filelog,".log"); /* */
11005: if((ficlog=fopen(filelog,"w"))==NULL) {
11006: printf("Problem with logfile %s\n",filelog);
11007: goto end;
11008: }
11009: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11010: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11011: fprintf(ficlog,"\nEnter the parameter file name: \n");
11012: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11013: path=%s \n\
11014: optionfile=%s\n\
11015: optionfilext=%s\n\
1.156 brouard 11016: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11017:
1.197 brouard 11018: syscompilerinfo(1);
1.167 brouard 11019:
1.126 brouard 11020: printf("Local time (at start):%s",strstart);
11021: fprintf(ficlog,"Local time (at start): %s",strstart);
11022: fflush(ficlog);
11023: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11024: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11025:
11026: /* */
11027: strcpy(fileres,"r");
11028: strcat(fileres, optionfilefiname);
1.201 brouard 11029: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11030: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11031: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11032:
1.186 brouard 11033: /* Main ---------arguments file --------*/
1.126 brouard 11034:
11035: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11036: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11037: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11038: fflush(ficlog);
1.149 brouard 11039: /* goto end; */
11040: exit(70);
1.126 brouard 11041: }
11042:
11043: strcpy(filereso,"o");
1.201 brouard 11044: strcat(filereso,fileresu);
1.126 brouard 11045: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11046: printf("Problem with Output resultfile: %s\n", filereso);
11047: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11048: fflush(ficlog);
11049: goto end;
11050: }
1.278 brouard 11051: /*-------- Rewriting parameter file ----------*/
11052: strcpy(rfileres,"r"); /* "Rparameterfile */
11053: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11054: strcat(rfileres,"."); /* */
11055: strcat(rfileres,optionfilext); /* Other files have txt extension */
11056: if((ficres =fopen(rfileres,"w"))==NULL) {
11057: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11058: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11059: fflush(ficlog);
11060: goto end;
11061: }
11062: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11063:
1.278 brouard 11064:
1.126 brouard 11065: /* Reads comments: lines beginning with '#' */
11066: numlinepar=0;
1.277 brouard 11067: /* Is it a BOM UTF-8 Windows file? */
11068: /* First parameter line */
1.197 brouard 11069: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11070: noffset=0;
11071: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11072: {
11073: noffset=noffset+3;
11074: printf("# File is an UTF8 Bom.\n"); // 0xBF
11075: }
11076: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
11077: {
11078: noffset=noffset+2;
11079: printf("# File is an UTF16BE BOM file\n");
11080: }
11081: else if( line[0] == 0 && line[1] == 0)
11082: {
11083: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11084: noffset=noffset+4;
11085: printf("# File is an UTF16BE BOM file\n");
11086: }
11087: } else{
11088: ;/*printf(" Not a BOM file\n");*/
11089: }
11090:
1.197 brouard 11091: /* If line starts with a # it is a comment */
1.277 brouard 11092: if (line[noffset] == '#') {
1.197 brouard 11093: numlinepar++;
11094: fputs(line,stdout);
11095: fputs(line,ficparo);
1.278 brouard 11096: fputs(line,ficres);
1.197 brouard 11097: fputs(line,ficlog);
11098: continue;
11099: }else
11100: break;
11101: }
11102: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11103: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11104: if (num_filled != 5) {
11105: printf("Should be 5 parameters\n");
1.283 brouard 11106: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11107: }
1.126 brouard 11108: numlinepar++;
1.197 brouard 11109: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11110: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11111: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11112: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11113: }
11114: /* Second parameter line */
11115: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11116: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11117: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11118: if (line[0] == '#') {
11119: numlinepar++;
1.283 brouard 11120: printf("%s",line);
11121: fprintf(ficres,"%s",line);
11122: fprintf(ficparo,"%s",line);
11123: fprintf(ficlog,"%s",line);
1.197 brouard 11124: continue;
11125: }else
11126: break;
11127: }
1.223 brouard 11128: 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", \
11129: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11130: if (num_filled != 11) {
11131: 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 11132: printf("but line=%s\n",line);
1.283 brouard 11133: 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");
11134: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11135: }
1.286 brouard 11136: if( lastpass > maxwav){
11137: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11138: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11139: fflush(ficlog);
11140: goto end;
11141: }
11142: 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 11143: 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 11144: 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 11145: 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 11146: }
1.203 brouard 11147: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11148: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11149: /* Third parameter line */
11150: while(fgets(line, MAXLINE, ficpar)) {
11151: /* If line starts with a # it is a comment */
11152: if (line[0] == '#') {
11153: numlinepar++;
1.283 brouard 11154: printf("%s",line);
11155: fprintf(ficres,"%s",line);
11156: fprintf(ficparo,"%s",line);
11157: fprintf(ficlog,"%s",line);
1.197 brouard 11158: continue;
11159: }else
11160: break;
11161: }
1.201 brouard 11162: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11163: if (num_filled != 1){
11164: printf("ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
11165: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
1.197 brouard 11166: model[0]='\0';
11167: goto end;
11168: }
11169: else{
11170: if (model[0]=='+'){
11171: for(i=1; i<=strlen(model);i++)
11172: modeltemp[i-1]=model[i];
1.201 brouard 11173: strcpy(model,modeltemp);
1.197 brouard 11174: }
11175: }
1.199 brouard 11176: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11177: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11178: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11179: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11180: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11181: }
11182: /* 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); */
11183: /* numlinepar=numlinepar+3; /\* In general *\/ */
11184: /* 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 11185: /* 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); */
11186: /* 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 11187: fflush(ficlog);
1.190 brouard 11188: /* if(model[0]=='#'|| model[0]== '\0'){ */
11189: if(model[0]=='#'){
1.279 brouard 11190: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11191: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11192: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11193: if(mle != -1){
1.279 brouard 11194: 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 11195: exit(1);
11196: }
11197: }
1.126 brouard 11198: while((c=getc(ficpar))=='#' && c!= EOF){
11199: ungetc(c,ficpar);
11200: fgets(line, MAXLINE, ficpar);
11201: numlinepar++;
1.195 brouard 11202: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11203: z[0]=line[1];
11204: }
11205: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11206: fputs(line, stdout);
11207: //puts(line);
1.126 brouard 11208: fputs(line,ficparo);
11209: fputs(line,ficlog);
11210: }
11211: ungetc(c,ficpar);
11212:
11213:
1.290 brouard 11214: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11215: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11216: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11217: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11218: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11219: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11220: v1+v2*age+v2*v3 makes cptcovn = 3
11221: */
11222: if (strlen(model)>1)
1.187 brouard 11223: 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 11224: else
1.187 brouard 11225: ncovmodel=2; /* Constant and age */
1.133 brouard 11226: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11227: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11228: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11229: 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);
11230: 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);
11231: fflush(stdout);
11232: fclose (ficlog);
11233: goto end;
11234: }
1.126 brouard 11235: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11236: delti=delti3[1][1];
11237: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11238: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11239: /* We could also provide initial parameters values giving by simple logistic regression
11240: * only one way, that is without matrix product. We will have nlstate maximizations */
11241: /* for(i=1;i<nlstate;i++){ */
11242: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11243: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11244: /* } */
1.126 brouard 11245: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11246: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11247: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11248: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11249: fclose (ficparo);
11250: fclose (ficlog);
11251: goto end;
11252: exit(0);
1.220 brouard 11253: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11254: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11255: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11256: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11257: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11258: matcov=matrix(1,npar,1,npar);
1.203 brouard 11259: hess=matrix(1,npar,1,npar);
1.220 brouard 11260: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11261: /* Read guessed parameters */
1.126 brouard 11262: /* Reads comments: lines beginning with '#' */
11263: while((c=getc(ficpar))=='#' && c!= EOF){
11264: ungetc(c,ficpar);
11265: fgets(line, MAXLINE, ficpar);
11266: numlinepar++;
1.141 brouard 11267: fputs(line,stdout);
1.126 brouard 11268: fputs(line,ficparo);
11269: fputs(line,ficlog);
11270: }
11271: ungetc(c,ficpar);
11272:
11273: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11274: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11275: for(i=1; i <=nlstate; i++){
1.234 brouard 11276: j=0;
1.126 brouard 11277: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11278: if(jj==i) continue;
11279: j++;
1.292 brouard 11280: while((c=getc(ficpar))=='#' && c!= EOF){
11281: ungetc(c,ficpar);
11282: fgets(line, MAXLINE, ficpar);
11283: numlinepar++;
11284: fputs(line,stdout);
11285: fputs(line,ficparo);
11286: fputs(line,ficlog);
11287: }
11288: ungetc(c,ficpar);
1.234 brouard 11289: fscanf(ficpar,"%1d%1d",&i1,&j1);
11290: if ((i1 != i) || (j1 != jj)){
11291: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11292: It might be a problem of design; if ncovcol and the model are correct\n \
11293: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11294: exit(1);
11295: }
11296: fprintf(ficparo,"%1d%1d",i1,j1);
11297: if(mle==1)
11298: printf("%1d%1d",i,jj);
11299: fprintf(ficlog,"%1d%1d",i,jj);
11300: for(k=1; k<=ncovmodel;k++){
11301: fscanf(ficpar," %lf",¶m[i][j][k]);
11302: if(mle==1){
11303: printf(" %lf",param[i][j][k]);
11304: fprintf(ficlog," %lf",param[i][j][k]);
11305: }
11306: else
11307: fprintf(ficlog," %lf",param[i][j][k]);
11308: fprintf(ficparo," %lf",param[i][j][k]);
11309: }
11310: fscanf(ficpar,"\n");
11311: numlinepar++;
11312: if(mle==1)
11313: printf("\n");
11314: fprintf(ficlog,"\n");
11315: fprintf(ficparo,"\n");
1.126 brouard 11316: }
11317: }
11318: fflush(ficlog);
1.234 brouard 11319:
1.251 brouard 11320: /* Reads parameters values */
1.126 brouard 11321: p=param[1][1];
1.251 brouard 11322: pstart=paramstart[1][1];
1.126 brouard 11323:
11324: /* Reads comments: lines beginning with '#' */
11325: while((c=getc(ficpar))=='#' && c!= EOF){
11326: ungetc(c,ficpar);
11327: fgets(line, MAXLINE, ficpar);
11328: numlinepar++;
1.141 brouard 11329: fputs(line,stdout);
1.126 brouard 11330: fputs(line,ficparo);
11331: fputs(line,ficlog);
11332: }
11333: ungetc(c,ficpar);
11334:
11335: for(i=1; i <=nlstate; i++){
11336: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11337: fscanf(ficpar,"%1d%1d",&i1,&j1);
11338: if ( (i1-i) * (j1-j) != 0){
11339: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11340: exit(1);
11341: }
11342: printf("%1d%1d",i,j);
11343: fprintf(ficparo,"%1d%1d",i1,j1);
11344: fprintf(ficlog,"%1d%1d",i1,j1);
11345: for(k=1; k<=ncovmodel;k++){
11346: fscanf(ficpar,"%le",&delti3[i][j][k]);
11347: printf(" %le",delti3[i][j][k]);
11348: fprintf(ficparo," %le",delti3[i][j][k]);
11349: fprintf(ficlog," %le",delti3[i][j][k]);
11350: }
11351: fscanf(ficpar,"\n");
11352: numlinepar++;
11353: printf("\n");
11354: fprintf(ficparo,"\n");
11355: fprintf(ficlog,"\n");
1.126 brouard 11356: }
11357: }
11358: fflush(ficlog);
1.234 brouard 11359:
1.145 brouard 11360: /* Reads covariance matrix */
1.126 brouard 11361: delti=delti3[1][1];
1.220 brouard 11362:
11363:
1.126 brouard 11364: /* 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 11365:
1.126 brouard 11366: /* Reads comments: lines beginning with '#' */
11367: while((c=getc(ficpar))=='#' && c!= EOF){
11368: ungetc(c,ficpar);
11369: fgets(line, MAXLINE, ficpar);
11370: numlinepar++;
1.141 brouard 11371: fputs(line,stdout);
1.126 brouard 11372: fputs(line,ficparo);
11373: fputs(line,ficlog);
11374: }
11375: ungetc(c,ficpar);
1.220 brouard 11376:
1.126 brouard 11377: matcov=matrix(1,npar,1,npar);
1.203 brouard 11378: hess=matrix(1,npar,1,npar);
1.131 brouard 11379: for(i=1; i <=npar; i++)
11380: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11381:
1.194 brouard 11382: /* Scans npar lines */
1.126 brouard 11383: for(i=1; i <=npar; i++){
1.226 brouard 11384: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11385: if(count != 3){
1.226 brouard 11386: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11387: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11388: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11389: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11390: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11391: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11392: exit(1);
1.220 brouard 11393: }else{
1.226 brouard 11394: if(mle==1)
11395: printf("%1d%1d%d",i1,j1,jk);
11396: }
11397: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11398: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11399: for(j=1; j <=i; j++){
1.226 brouard 11400: fscanf(ficpar," %le",&matcov[i][j]);
11401: if(mle==1){
11402: printf(" %.5le",matcov[i][j]);
11403: }
11404: fprintf(ficlog," %.5le",matcov[i][j]);
11405: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11406: }
11407: fscanf(ficpar,"\n");
11408: numlinepar++;
11409: if(mle==1)
1.220 brouard 11410: printf("\n");
1.126 brouard 11411: fprintf(ficlog,"\n");
11412: fprintf(ficparo,"\n");
11413: }
1.194 brouard 11414: /* End of read covariance matrix npar lines */
1.126 brouard 11415: for(i=1; i <=npar; i++)
11416: for(j=i+1;j<=npar;j++)
1.226 brouard 11417: matcov[i][j]=matcov[j][i];
1.126 brouard 11418:
11419: if(mle==1)
11420: printf("\n");
11421: fprintf(ficlog,"\n");
11422:
11423: fflush(ficlog);
11424:
11425: } /* End of mle != -3 */
1.218 brouard 11426:
1.186 brouard 11427: /* Main data
11428: */
1.290 brouard 11429: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11430: /* num=lvector(1,n); */
11431: /* moisnais=vector(1,n); */
11432: /* annais=vector(1,n); */
11433: /* moisdc=vector(1,n); */
11434: /* andc=vector(1,n); */
11435: /* weight=vector(1,n); */
11436: /* agedc=vector(1,n); */
11437: /* cod=ivector(1,n); */
11438: /* for(i=1;i<=n;i++){ */
11439: num=lvector(firstobs,lastobs);
11440: moisnais=vector(firstobs,lastobs);
11441: annais=vector(firstobs,lastobs);
11442: moisdc=vector(firstobs,lastobs);
11443: andc=vector(firstobs,lastobs);
11444: weight=vector(firstobs,lastobs);
11445: agedc=vector(firstobs,lastobs);
11446: cod=ivector(firstobs,lastobs);
11447: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11448: num[i]=0;
11449: moisnais[i]=0;
11450: annais[i]=0;
11451: moisdc[i]=0;
11452: andc[i]=0;
11453: agedc[i]=0;
11454: cod[i]=0;
11455: weight[i]=1.0; /* Equal weights, 1 by default */
11456: }
1.290 brouard 11457: mint=matrix(1,maxwav,firstobs,lastobs);
11458: anint=matrix(1,maxwav,firstobs,lastobs);
11459: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11460: tab=ivector(1,NCOVMAX);
1.144 brouard 11461: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11462: 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 11463:
1.136 brouard 11464: /* Reads data from file datafile */
11465: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11466: goto end;
11467:
11468: /* Calculation of the number of parameters from char model */
1.234 brouard 11469: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11470: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11471: k=3 V4 Tvar[k=3]= 4 (from V4)
11472: k=2 V1 Tvar[k=2]= 1 (from V1)
11473: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11474: */
11475:
11476: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11477: TvarsDind=ivector(1,NCOVMAX); /* */
11478: TvarsD=ivector(1,NCOVMAX); /* */
11479: TvarsQind=ivector(1,NCOVMAX); /* */
11480: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11481: TvarF=ivector(1,NCOVMAX); /* */
11482: TvarFind=ivector(1,NCOVMAX); /* */
11483: TvarV=ivector(1,NCOVMAX); /* */
11484: TvarVind=ivector(1,NCOVMAX); /* */
11485: TvarA=ivector(1,NCOVMAX); /* */
11486: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11487: TvarFD=ivector(1,NCOVMAX); /* */
11488: TvarFDind=ivector(1,NCOVMAX); /* */
11489: TvarFQ=ivector(1,NCOVMAX); /* */
11490: TvarFQind=ivector(1,NCOVMAX); /* */
11491: TvarVD=ivector(1,NCOVMAX); /* */
11492: TvarVDind=ivector(1,NCOVMAX); /* */
11493: TvarVQ=ivector(1,NCOVMAX); /* */
11494: TvarVQind=ivector(1,NCOVMAX); /* */
11495:
1.230 brouard 11496: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11497: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11498: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11499: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11500: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11501: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11502: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11503: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11504: */
11505: /* For model-covariate k tells which data-covariate to use but
11506: because this model-covariate is a construction we invent a new column
11507: ncovcol + k1
11508: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11509: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11510: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11511: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11512: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11513: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11514: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11515: */
1.145 brouard 11516: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11517: 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 11518: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11519: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11520: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11521: 4 covariates (3 plus signs)
11522: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11523: */
1.230 brouard 11524: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11525: * individual dummy, fixed or varying:
11526: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11527: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11528: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11529: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11530: * Tmodelind[1]@9={9,0,3,2,}*/
11531: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11532: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11533: * individual quantitative, fixed or varying:
11534: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11535: * 3, 1, 0, 0, 0, 0, 0, 0},
11536: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11537: /* Main decodemodel */
11538:
1.187 brouard 11539:
1.223 brouard 11540: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11541: goto end;
11542:
1.137 brouard 11543: if((double)(lastobs-imx)/(double)imx > 1.10){
11544: nbwarn++;
11545: 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);
11546: 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);
11547: }
1.136 brouard 11548: /* if(mle==1){*/
1.137 brouard 11549: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11550: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11551: }
11552:
11553: /*-calculation of age at interview from date of interview and age at death -*/
11554: agev=matrix(1,maxwav,1,imx);
11555:
11556: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11557: goto end;
11558:
1.126 brouard 11559:
1.136 brouard 11560: agegomp=(int)agemin;
1.290 brouard 11561: free_vector(moisnais,firstobs,lastobs);
11562: free_vector(annais,firstobs,lastobs);
1.126 brouard 11563: /* free_matrix(mint,1,maxwav,1,n);
11564: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11565: /* free_vector(moisdc,1,n); */
11566: /* free_vector(andc,1,n); */
1.145 brouard 11567: /* */
11568:
1.126 brouard 11569: wav=ivector(1,imx);
1.214 brouard 11570: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11571: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11572: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11573: 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.*/
11574: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11575: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11576:
11577: /* Concatenates waves */
1.214 brouard 11578: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11579: Death is a valid wave (if date is known).
11580: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11581: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11582: and mw[mi+1][i]. dh depends on stepm.
11583: */
11584:
1.126 brouard 11585: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11586: /* Concatenates waves */
1.145 brouard 11587:
1.290 brouard 11588: free_vector(moisdc,firstobs,lastobs);
11589: free_vector(andc,firstobs,lastobs);
1.215 brouard 11590:
1.126 brouard 11591: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11592: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11593: ncodemax[1]=1;
1.145 brouard 11594: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11595: cptcoveff=0;
1.220 brouard 11596: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11597: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11598: }
11599:
11600: ncovcombmax=pow(2,cptcoveff);
11601: invalidvarcomb=ivector(1, ncovcombmax);
11602: for(i=1;i<ncovcombmax;i++)
11603: invalidvarcomb[i]=0;
11604:
1.211 brouard 11605: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11606: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11607: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11608:
1.200 brouard 11609: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11610: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11611: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11612: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11613: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11614: * (currently 0 or 1) in the data.
11615: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11616: * corresponding modality (h,j).
11617: */
11618:
1.145 brouard 11619: h=0;
11620: /*if (cptcovn > 0) */
1.126 brouard 11621: m=pow(2,cptcoveff);
11622:
1.144 brouard 11623: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11624: * For k=4 covariates, h goes from 1 to m=2**k
11625: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11626: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11627: * h\k 1 2 3 4
1.143 brouard 11628: *______________________________
11629: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11630: * 2 2 1 1 1
11631: * 3 i=2 1 2 1 1
11632: * 4 2 2 1 1
11633: * 5 i=3 1 i=2 1 2 1
11634: * 6 2 1 2 1
11635: * 7 i=4 1 2 2 1
11636: * 8 2 2 2 1
1.197 brouard 11637: * 9 i=5 1 i=3 1 i=2 1 2
11638: * 10 2 1 1 2
11639: * 11 i=6 1 2 1 2
11640: * 12 2 2 1 2
11641: * 13 i=7 1 i=4 1 2 2
11642: * 14 2 1 2 2
11643: * 15 i=8 1 2 2 2
11644: * 16 2 2 2 2
1.143 brouard 11645: */
1.212 brouard 11646: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11647: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11648: * and the value of each covariate?
11649: * V1=1, V2=1, V3=2, V4=1 ?
11650: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11651: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11652: * In order to get the real value in the data, we use nbcode
11653: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11654: * We are keeping this crazy system in order to be able (in the future?)
11655: * to have more than 2 values (0 or 1) for a covariate.
11656: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11657: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11658: * bbbbbbbb
11659: * 76543210
11660: * h-1 00000101 (6-1=5)
1.219 brouard 11661: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11662: * &
11663: * 1 00000001 (1)
1.219 brouard 11664: * 00000000 = 1 & ((h-1) >> (k-1))
11665: * +1= 00000001 =1
1.211 brouard 11666: *
11667: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11668: * h' 1101 =2^3+2^2+0x2^1+2^0
11669: * >>k' 11
11670: * & 00000001
11671: * = 00000001
11672: * +1 = 00000010=2 = codtabm(14,3)
11673: * Reverse h=6 and m=16?
11674: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11675: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11676: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11677: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11678: * V3=decodtabm(14,3,2**4)=2
11679: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11680: *(h-1) >> (j-1) 0011 =13 >> 2
11681: * &1 000000001
11682: * = 000000001
11683: * +1= 000000010 =2
11684: * 2211
11685: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11686: * V3=2
1.220 brouard 11687: * codtabm and decodtabm are identical
1.211 brouard 11688: */
11689:
1.145 brouard 11690:
11691: free_ivector(Ndum,-1,NCOVMAX);
11692:
11693:
1.126 brouard 11694:
1.186 brouard 11695: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11696: strcpy(optionfilegnuplot,optionfilefiname);
11697: if(mle==-3)
1.201 brouard 11698: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11699: strcat(optionfilegnuplot,".gp");
11700:
11701: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11702: printf("Problem with file %s",optionfilegnuplot);
11703: }
11704: else{
1.204 brouard 11705: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11706: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11707: //fprintf(ficgp,"set missing 'NaNq'\n");
11708: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11709: }
11710: /* fclose(ficgp);*/
1.186 brouard 11711:
11712:
11713: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11714:
11715: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11716: if(mle==-3)
1.201 brouard 11717: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11718: strcat(optionfilehtm,".htm");
11719: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11720: printf("Problem with %s \n",optionfilehtm);
11721: exit(0);
1.126 brouard 11722: }
11723:
11724: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11725: strcat(optionfilehtmcov,"-cov.htm");
11726: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11727: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11728: }
11729: else{
11730: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11731: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11732: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11733: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11734: }
11735:
1.213 brouard 11736: 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 11737: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11738: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11739: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11740: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11741: \n\
11742: <hr size=\"2\" color=\"#EC5E5E\">\
11743: <ul><li><h4>Parameter files</h4>\n\
11744: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11745: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11746: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11747: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11748: - Date and time at start: %s</ul>\n",\
11749: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11750: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11751: fileres,fileres,\
11752: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11753: fflush(fichtm);
11754:
11755: strcpy(pathr,path);
11756: strcat(pathr,optionfilefiname);
1.184 brouard 11757: #ifdef WIN32
11758: _chdir(optionfilefiname); /* Move to directory named optionfile */
11759: #else
1.126 brouard 11760: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11761: #endif
11762:
1.126 brouard 11763:
1.220 brouard 11764: /* Calculates basic frequencies. Computes observed prevalence at single age
11765: and for any valid combination of covariates
1.126 brouard 11766: and prints on file fileres'p'. */
1.251 brouard 11767: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11768: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11769:
11770: fprintf(fichtm,"\n");
1.286 brouard 11771: 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 11772: ftol, stepm);
11773: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11774: ncurrv=1;
11775: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11776: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11777: ncurrv=i;
11778: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11779: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 11780: ncurrv=i;
11781: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11782: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 11783: ncurrv=i;
11784: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11785: 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", \
11786: nlstate, ndeath, maxwav, mle, weightopt);
11787:
11788: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11789: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11790:
11791:
11792: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11793: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11794: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11795: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11796: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11797: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11798: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11799: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11800: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11801:
1.126 brouard 11802: /* For Powell, parameters are in a vector p[] starting at p[1]
11803: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11804: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11805:
11806: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11807: /* For mortality only */
1.126 brouard 11808: if (mle==-3){
1.136 brouard 11809: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11810: for(i=1;i<=NDIM;i++)
11811: for(j=1;j<=NDIM;j++)
11812: ximort[i][j]=0.;
1.186 brouard 11813: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 11814: cens=ivector(firstobs,lastobs);
11815: ageexmed=vector(firstobs,lastobs);
11816: agecens=vector(firstobs,lastobs);
11817: dcwave=ivector(firstobs,lastobs);
1.223 brouard 11818:
1.126 brouard 11819: for (i=1; i<=imx; i++){
11820: dcwave[i]=-1;
11821: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11822: if (s[m][i]>nlstate) {
11823: dcwave[i]=m;
11824: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11825: break;
11826: }
1.126 brouard 11827: }
1.226 brouard 11828:
1.126 brouard 11829: for (i=1; i<=imx; i++) {
11830: if (wav[i]>0){
1.226 brouard 11831: ageexmed[i]=agev[mw[1][i]][i];
11832: j=wav[i];
11833: agecens[i]=1.;
11834:
11835: if (ageexmed[i]> 1 && wav[i] > 0){
11836: agecens[i]=agev[mw[j][i]][i];
11837: cens[i]= 1;
11838: }else if (ageexmed[i]< 1)
11839: cens[i]= -1;
11840: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11841: cens[i]=0 ;
1.126 brouard 11842: }
11843: else cens[i]=-1;
11844: }
11845:
11846: for (i=1;i<=NDIM;i++) {
11847: for (j=1;j<=NDIM;j++)
1.226 brouard 11848: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11849: }
11850:
1.145 brouard 11851: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11852: /*printf("%lf %lf", p[1], p[2]);*/
11853:
11854:
1.136 brouard 11855: #ifdef GSL
11856: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11857: #else
1.126 brouard 11858: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11859: #endif
1.201 brouard 11860: strcpy(filerespow,"POW-MORT_");
11861: strcat(filerespow,fileresu);
1.126 brouard 11862: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11863: printf("Problem with resultfile: %s\n", filerespow);
11864: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11865: }
1.136 brouard 11866: #ifdef GSL
11867: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11868: #else
1.126 brouard 11869: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11870: #endif
1.126 brouard 11871: /* for (i=1;i<=nlstate;i++)
11872: for(j=1;j<=nlstate+ndeath;j++)
11873: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11874: */
11875: fprintf(ficrespow,"\n");
1.136 brouard 11876: #ifdef GSL
11877: /* gsl starts here */
11878: T = gsl_multimin_fminimizer_nmsimplex;
11879: gsl_multimin_fminimizer *sfm = NULL;
11880: gsl_vector *ss, *x;
11881: gsl_multimin_function minex_func;
11882:
11883: /* Initial vertex size vector */
11884: ss = gsl_vector_alloc (NDIM);
11885:
11886: if (ss == NULL){
11887: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11888: }
11889: /* Set all step sizes to 1 */
11890: gsl_vector_set_all (ss, 0.001);
11891:
11892: /* Starting point */
1.126 brouard 11893:
1.136 brouard 11894: x = gsl_vector_alloc (NDIM);
11895:
11896: if (x == NULL){
11897: gsl_vector_free(ss);
11898: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11899: }
11900:
11901: /* Initialize method and iterate */
11902: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11903: /* gsl_vector_set(x, 0, 0.0268); */
11904: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11905: gsl_vector_set(x, 0, p[1]);
11906: gsl_vector_set(x, 1, p[2]);
11907:
11908: minex_func.f = &gompertz_f;
11909: minex_func.n = NDIM;
11910: minex_func.params = (void *)&p; /* ??? */
11911:
11912: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11913: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11914:
11915: printf("Iterations beginning .....\n\n");
11916: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11917:
11918: iteri=0;
11919: while (rval == GSL_CONTINUE){
11920: iteri++;
11921: status = gsl_multimin_fminimizer_iterate(sfm);
11922:
11923: if (status) printf("error: %s\n", gsl_strerror (status));
11924: fflush(0);
11925:
11926: if (status)
11927: break;
11928:
11929: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11930: ssval = gsl_multimin_fminimizer_size (sfm);
11931:
11932: if (rval == GSL_SUCCESS)
11933: printf ("converged to a local maximum at\n");
11934:
11935: printf("%5d ", iteri);
11936: for (it = 0; it < NDIM; it++){
11937: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11938: }
11939: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11940: }
11941:
11942: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11943:
11944: gsl_vector_free(x); /* initial values */
11945: gsl_vector_free(ss); /* inital step size */
11946: for (it=0; it<NDIM; it++){
11947: p[it+1]=gsl_vector_get(sfm->x,it);
11948: fprintf(ficrespow," %.12lf", p[it]);
11949: }
11950: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11951: #endif
11952: #ifdef POWELL
11953: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11954: #endif
1.126 brouard 11955: fclose(ficrespow);
11956:
1.203 brouard 11957: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11958:
11959: for(i=1; i <=NDIM; i++)
11960: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11961: matcov[i][j]=matcov[j][i];
1.126 brouard 11962:
11963: printf("\nCovariance matrix\n ");
1.203 brouard 11964: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11965: for(i=1; i <=NDIM; i++) {
11966: for(j=1;j<=NDIM;j++){
1.220 brouard 11967: printf("%f ",matcov[i][j]);
11968: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11969: }
1.203 brouard 11970: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11971: }
11972:
11973: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11974: for (i=1;i<=NDIM;i++) {
1.126 brouard 11975: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11976: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11977: }
1.126 brouard 11978: lsurv=vector(1,AGESUP);
11979: lpop=vector(1,AGESUP);
11980: tpop=vector(1,AGESUP);
11981: lsurv[agegomp]=100000;
11982:
11983: for (k=agegomp;k<=AGESUP;k++) {
11984: agemortsup=k;
11985: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11986: }
11987:
11988: for (k=agegomp;k<agemortsup;k++)
11989: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11990:
11991: for (k=agegomp;k<agemortsup;k++){
11992: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11993: sumlpop=sumlpop+lpop[k];
11994: }
11995:
11996: tpop[agegomp]=sumlpop;
11997: for (k=agegomp;k<(agemortsup-3);k++){
11998: /* tpop[k+1]=2;*/
11999: tpop[k+1]=tpop[k]-lpop[k];
12000: }
12001:
12002:
12003: printf("\nAge lx qx dx Lx Tx e(x)\n");
12004: for (k=agegomp;k<(agemortsup-2);k++)
12005: 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]);
12006:
12007:
12008: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12009: ageminpar=50;
12010: agemaxpar=100;
1.194 brouard 12011: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12012: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12013: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12014: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12015: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12016: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12017: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12018: }else{
12019: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12020: 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 12021: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12022: }
1.201 brouard 12023: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12024: stepm, weightopt,\
12025: model,imx,p,matcov,agemortsup);
12026:
12027: free_vector(lsurv,1,AGESUP);
12028: free_vector(lpop,1,AGESUP);
12029: free_vector(tpop,1,AGESUP);
1.220 brouard 12030: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12031: free_ivector(dcwave,firstobs,lastobs);
12032: free_vector(agecens,firstobs,lastobs);
12033: free_vector(ageexmed,firstobs,lastobs);
12034: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12035: #ifdef GSL
1.136 brouard 12036: #endif
1.186 brouard 12037: } /* Endof if mle==-3 mortality only */
1.205 brouard 12038: /* Standard */
12039: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12040: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12041: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12042: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12043: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12044: for (k=1; k<=npar;k++)
12045: printf(" %d %8.5f",k,p[k]);
12046: printf("\n");
1.205 brouard 12047: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12048: /* mlikeli uses func not funcone */
1.247 brouard 12049: /* for(i=1;i<nlstate;i++){ */
12050: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12051: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12052: /* } */
1.205 brouard 12053: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12054: }
12055: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12056: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12057: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12058: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12059: }
12060: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12061: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12062: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12063: for (k=1; k<=npar;k++)
12064: printf(" %d %8.5f",k,p[k]);
12065: printf("\n");
12066:
12067: /*--------- results files --------------*/
1.283 brouard 12068: /* 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 12069:
12070:
12071: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12072: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12073: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12074: for(i=1,jk=1; i <=nlstate; i++){
12075: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12076: if (k != i) {
12077: printf("%d%d ",i,k);
12078: fprintf(ficlog,"%d%d ",i,k);
12079: fprintf(ficres,"%1d%1d ",i,k);
12080: for(j=1; j <=ncovmodel; j++){
12081: printf("%12.7f ",p[jk]);
12082: fprintf(ficlog,"%12.7f ",p[jk]);
12083: fprintf(ficres,"%12.7f ",p[jk]);
12084: jk++;
12085: }
12086: printf("\n");
12087: fprintf(ficlog,"\n");
12088: fprintf(ficres,"\n");
12089: }
1.126 brouard 12090: }
12091: }
1.203 brouard 12092: if(mle != 0){
12093: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12094: ftolhess=ftol; /* Usually correct */
1.203 brouard 12095: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12096: 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");
12097: 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");
12098: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12099: for(k=1; k <=(nlstate+ndeath); k++){
12100: if (k != i) {
12101: printf("%d%d ",i,k);
12102: fprintf(ficlog,"%d%d ",i,k);
12103: for(j=1; j <=ncovmodel; j++){
12104: 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]));
12105: 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]));
12106: jk++;
12107: }
12108: printf("\n");
12109: fprintf(ficlog,"\n");
12110: }
12111: }
1.193 brouard 12112: }
1.203 brouard 12113: } /* end of hesscov and Wald tests */
1.225 brouard 12114:
1.203 brouard 12115: /* */
1.126 brouard 12116: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12117: printf("# Scales (for hessian or gradient estimation)\n");
12118: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12119: for(i=1,jk=1; i <=nlstate; i++){
12120: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12121: if (j!=i) {
12122: fprintf(ficres,"%1d%1d",i,j);
12123: printf("%1d%1d",i,j);
12124: fprintf(ficlog,"%1d%1d",i,j);
12125: for(k=1; k<=ncovmodel;k++){
12126: printf(" %.5e",delti[jk]);
12127: fprintf(ficlog," %.5e",delti[jk]);
12128: fprintf(ficres," %.5e",delti[jk]);
12129: jk++;
12130: }
12131: printf("\n");
12132: fprintf(ficlog,"\n");
12133: fprintf(ficres,"\n");
12134: }
1.126 brouard 12135: }
12136: }
12137:
12138: 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 12139: if(mle >= 1) /* To big for the screen */
1.126 brouard 12140: 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");
12141: 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");
12142: /* # 121 Var(a12)\n\ */
12143: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12144: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12145: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12146: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12147: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12148: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12149: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12150:
12151:
12152: /* Just to have a covariance matrix which will be more understandable
12153: even is we still don't want to manage dictionary of variables
12154: */
12155: for(itimes=1;itimes<=2;itimes++){
12156: jj=0;
12157: for(i=1; i <=nlstate; i++){
1.225 brouard 12158: for(j=1; j <=nlstate+ndeath; j++){
12159: if(j==i) continue;
12160: for(k=1; k<=ncovmodel;k++){
12161: jj++;
12162: ca[0]= k+'a'-1;ca[1]='\0';
12163: if(itimes==1){
12164: if(mle>=1)
12165: printf("#%1d%1d%d",i,j,k);
12166: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12167: fprintf(ficres,"#%1d%1d%d",i,j,k);
12168: }else{
12169: if(mle>=1)
12170: printf("%1d%1d%d",i,j,k);
12171: fprintf(ficlog,"%1d%1d%d",i,j,k);
12172: fprintf(ficres,"%1d%1d%d",i,j,k);
12173: }
12174: ll=0;
12175: for(li=1;li <=nlstate; li++){
12176: for(lj=1;lj <=nlstate+ndeath; lj++){
12177: if(lj==li) continue;
12178: for(lk=1;lk<=ncovmodel;lk++){
12179: ll++;
12180: if(ll<=jj){
12181: cb[0]= lk +'a'-1;cb[1]='\0';
12182: if(ll<jj){
12183: if(itimes==1){
12184: if(mle>=1)
12185: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12186: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12187: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12188: }else{
12189: if(mle>=1)
12190: printf(" %.5e",matcov[jj][ll]);
12191: fprintf(ficlog," %.5e",matcov[jj][ll]);
12192: fprintf(ficres," %.5e",matcov[jj][ll]);
12193: }
12194: }else{
12195: if(itimes==1){
12196: if(mle>=1)
12197: printf(" Var(%s%1d%1d)",ca,i,j);
12198: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12199: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12200: }else{
12201: if(mle>=1)
12202: printf(" %.7e",matcov[jj][ll]);
12203: fprintf(ficlog," %.7e",matcov[jj][ll]);
12204: fprintf(ficres," %.7e",matcov[jj][ll]);
12205: }
12206: }
12207: }
12208: } /* end lk */
12209: } /* end lj */
12210: } /* end li */
12211: if(mle>=1)
12212: printf("\n");
12213: fprintf(ficlog,"\n");
12214: fprintf(ficres,"\n");
12215: numlinepar++;
12216: } /* end k*/
12217: } /*end j */
1.126 brouard 12218: } /* end i */
12219: } /* end itimes */
12220:
12221: fflush(ficlog);
12222: fflush(ficres);
1.225 brouard 12223: while(fgets(line, MAXLINE, ficpar)) {
12224: /* If line starts with a # it is a comment */
12225: if (line[0] == '#') {
12226: numlinepar++;
12227: fputs(line,stdout);
12228: fputs(line,ficparo);
12229: fputs(line,ficlog);
12230: continue;
12231: }else
12232: break;
12233: }
12234:
1.209 brouard 12235: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12236: /* ungetc(c,ficpar); */
12237: /* fgets(line, MAXLINE, ficpar); */
12238: /* fputs(line,stdout); */
12239: /* fputs(line,ficparo); */
12240: /* } */
12241: /* ungetc(c,ficpar); */
1.126 brouard 12242:
12243: estepm=0;
1.209 brouard 12244: 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 12245:
12246: if (num_filled != 6) {
12247: 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);
12248: 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);
12249: goto end;
12250: }
12251: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12252: }
12253: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12254: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12255:
1.209 brouard 12256: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12257: if (estepm==0 || estepm < stepm) estepm=stepm;
12258: if (fage <= 2) {
12259: bage = ageminpar;
12260: fage = agemaxpar;
12261: }
12262:
12263: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12264: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12265: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12266:
1.186 brouard 12267: /* Other stuffs, more or less useful */
1.254 brouard 12268: while(fgets(line, MAXLINE, ficpar)) {
12269: /* If line starts with a # it is a comment */
12270: if (line[0] == '#') {
12271: numlinepar++;
12272: fputs(line,stdout);
12273: fputs(line,ficparo);
12274: fputs(line,ficlog);
12275: continue;
12276: }else
12277: break;
12278: }
12279:
12280: 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){
12281:
12282: if (num_filled != 7) {
12283: 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);
12284: 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);
12285: goto end;
12286: }
12287: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12288: 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);
12289: 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);
12290: 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 12291: }
1.254 brouard 12292:
12293: while(fgets(line, MAXLINE, ficpar)) {
12294: /* If line starts with a # it is a comment */
12295: if (line[0] == '#') {
12296: numlinepar++;
12297: fputs(line,stdout);
12298: fputs(line,ficparo);
12299: fputs(line,ficlog);
12300: continue;
12301: }else
12302: break;
1.126 brouard 12303: }
12304:
12305:
12306: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12307: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12308:
1.254 brouard 12309: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12310: if (num_filled != 1) {
12311: 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);
12312: 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);
12313: goto end;
12314: }
12315: printf("pop_based=%d\n",popbased);
12316: fprintf(ficlog,"pop_based=%d\n",popbased);
12317: fprintf(ficparo,"pop_based=%d\n",popbased);
12318: fprintf(ficres,"pop_based=%d\n",popbased);
12319: }
12320:
1.258 brouard 12321: /* Results */
12322: nresult=0;
12323: do{
12324: if(!fgets(line, MAXLINE, ficpar)){
12325: endishere=1;
12326: parameterline=14;
12327: }else if (line[0] == '#') {
12328: /* If line starts with a # it is a comment */
1.254 brouard 12329: numlinepar++;
12330: fputs(line,stdout);
12331: fputs(line,ficparo);
12332: fputs(line,ficlog);
12333: continue;
1.258 brouard 12334: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12335: parameterline=11;
1.296 brouard 12336: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12337: parameterline=12;
12338: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12339: parameterline=13;
12340: else{
12341: parameterline=14;
1.254 brouard 12342: }
1.258 brouard 12343: switch (parameterline){
12344: case 11:
1.296 brouard 12345: if((num_filled=sscanf(line,"prevforecast=%d starting-proj-date=%lf/%lf/%lf final-proj-date=%lf/%lf/%lf mobil_average=%d\n",&prevfcast,&jproj1,&mproj1,&anproj1,&jproj2,&mproj2,&anproj2,&mobilavproj)) !=EOF && (num_filled == 8)){
12346: fprintf(ficparo,"prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
1.258 brouard 12347: 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);
12348: 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);
12349: 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);
12350: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12351: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12352: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12353: prvforecast = 1;
12354: }
12355: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
12356: printf(" Num_filled=%d, yearsfproj=%lf, mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12357: prvforecast = 2;
12358: }
12359: else {
12360: printf("Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevforecast=1 yearsfproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
12361: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevforecast=1 yearproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
12362: goto end;
1.258 brouard 12363: }
1.254 brouard 12364: break;
1.258 brouard 12365: case 12:
1.296 brouard 12366: if((num_filled=sscanf(line,"prevbackcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&prevbcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj)) !=EOF && (num_filled == 8)){
12367: fprintf(ficparo,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12368: printf("prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12369: fprintf(ficlog,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12370: fprintf(ficres,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12371: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12372: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12373: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12374: prvbackcast = 1;
12375: }
12376: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
12377: printf(" Num_filled=%d, yearsbproj=%lf, mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12378: prvbackcast = 2;
12379: }
12380: else {
12381: printf("Error: Not 8 (data)parameters in line but %d, for example:prevbackcast=1 starting-back-date=1/1/1990 final-back-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevbackcast=1 yearsbproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
12382: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevbackcast=1 starting-back-date=1/1/1990 final-back-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevbackcast=1 yearbproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
12383: goto end;
1.258 brouard 12384: }
1.230 brouard 12385: break;
1.296 brouard 12386: /* /\*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);*\/ */
12387: /* 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){ */
12388: /* if (num_filled != 8) { */
12389: /* 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); */
12390: /* 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); */
12391: /* goto end; */
12392: /* } */
12393: /* 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); */
12394: /* 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); */
12395: /* 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); */
12396: /* 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); */
12397: /* /\* day and month of proj2 are not used but only year anproj2.*\/ */
12398: /* dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.; */
12399: /* dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.; */
12400: /* } */
12401: /* break; */
1.258 brouard 12402: case 13:
12403: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12404: if (num_filled == 0){
12405: resultline[0]='\0';
12406: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12407: 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);
12408: break;
12409: } else if (num_filled != 1){
12410: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12411: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12412: }
12413: nresult++; /* Sum of resultlines */
12414: printf("Result %d: result=%s\n",nresult, resultline);
12415: if(nresult > MAXRESULTLINES){
12416: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12417: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12418: goto end;
12419: }
12420: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12421: fprintf(ficparo,"result: %s\n",resultline);
12422: fprintf(ficres,"result: %s\n",resultline);
12423: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12424: break;
1.258 brouard 12425: case 14:
1.259 brouard 12426: if(ncovmodel >2 && nresult==0 ){
12427: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12428: goto end;
12429: }
1.259 brouard 12430: break;
1.258 brouard 12431: default:
12432: nresult=1;
12433: decoderesult(".",nresult ); /* No covariate */
12434: }
12435: } /* End switch parameterline */
12436: }while(endishere==0); /* End do */
1.126 brouard 12437:
1.230 brouard 12438: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12439: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12440:
12441: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12442: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12443: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12444: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12445: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12446: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12447: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12448: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12449: }else{
1.270 brouard 12450: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12451: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12452: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12453: if(prvforecast==1){
12454: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12455: jprojd=jproj1;
12456: mprojd=mproj1;
12457: anprojd=anproj1;
12458: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12459: jprojf=jproj2;
12460: mprojf=mproj2;
12461: anprojf=anproj2;
12462: } else if(prvforecast == 2){
12463: dateprojd=dateintmean;
12464: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12465: dateprojf=dateintmean+yrfproj;
12466: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12467: }
12468: if(prvbackcast==1){
12469: datebackd=(jback1+12*mback1+365*anback1)/365;
12470: jbackd=jback1;
12471: mbackd=mback1;
12472: anbackd=anback1;
12473: datebackf=(jback2+12*mback2+365*anback2)/365;
12474: jbackf=jback2;
12475: mbackf=mback2;
12476: anbackf=anback2;
12477: } else if(prvbackcast == 2){
12478: datebackd=dateintmean;
12479: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12480: datebackf=dateintmean-yrbproj;
12481: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12482: }
12483:
12484: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12485: }
12486: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12487: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12488: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12489:
1.225 brouard 12490: /*------------ free_vector -------------*/
12491: /* chdir(path); */
1.220 brouard 12492:
1.215 brouard 12493: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12494: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12495: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12496: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12497: free_lvector(num,firstobs,lastobs);
12498: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12499: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12500: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12501: fclose(ficparo);
12502: fclose(ficres);
1.220 brouard 12503:
12504:
1.186 brouard 12505: /* Other results (useful)*/
1.220 brouard 12506:
12507:
1.126 brouard 12508: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12509: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12510: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12511: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12512: fclose(ficrespl);
12513:
12514: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12515: /*#include "hpijx.h"*/
12516: hPijx(p, bage, fage);
1.145 brouard 12517: fclose(ficrespij);
1.227 brouard 12518:
1.220 brouard 12519: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12520: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12521: k=1;
1.126 brouard 12522: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12523:
1.269 brouard 12524: /* Prevalence for each covariate combination in probs[age][status][cov] */
12525: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12526: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12527: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12528: for(k=1;k<=ncovcombmax;k++)
12529: probs[i][j][k]=0.;
1.269 brouard 12530: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12531: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12532: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12533: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12534: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12535: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12536: for(k=1;k<=ncovcombmax;k++)
12537: mobaverages[i][j][k]=0.;
1.219 brouard 12538: mobaverage=mobaverages;
12539: if (mobilav!=0) {
1.235 brouard 12540: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12541: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12542: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12543: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12544: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12545: }
1.269 brouard 12546: } else if (mobilavproj !=0) {
1.235 brouard 12547: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12548: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12549: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12550: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12551: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12552: }
1.269 brouard 12553: }else{
12554: printf("Internal error moving average\n");
12555: fflush(stdout);
12556: exit(1);
1.219 brouard 12557: }
12558: }/* end if moving average */
1.227 brouard 12559:
1.126 brouard 12560: /*---------- Forecasting ------------------*/
1.296 brouard 12561: if(prevfcast==1){
12562: /* /\* if(stepm ==1){*\/ */
12563: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12564: /*This done previously after freqsummary.*/
12565: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
12566: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
12567:
12568: /* } else if (prvforecast==2){ */
12569: /* /\* if(stepm ==1){*\/ */
12570: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12571: /* } */
12572: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
12573: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 12574: }
1.269 brouard 12575:
1.296 brouard 12576: /* Prevbcasting */
12577: if(prevbcast==1){
1.219 brouard 12578: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12579: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12580: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12581:
12582: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12583:
12584: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12585:
1.219 brouard 12586: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12587: fclose(ficresplb);
12588:
1.222 brouard 12589: hBijx(p, bage, fage, mobaverage);
12590: fclose(ficrespijb);
1.219 brouard 12591:
1.296 brouard 12592: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
12593: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
12594: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
12595: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
12596: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
12597: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
12598:
12599:
1.269 brouard 12600: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12601:
12602:
1.269 brouard 12603: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12604: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12605: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12606: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 12607: } /* end Prevbcasting */
1.268 brouard 12608:
1.186 brouard 12609:
12610: /* ------ Other prevalence ratios------------ */
1.126 brouard 12611:
1.215 brouard 12612: free_ivector(wav,1,imx);
12613: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12614: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12615: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12616:
12617:
1.127 brouard 12618: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12619:
1.201 brouard 12620: strcpy(filerese,"E_");
12621: strcat(filerese,fileresu);
1.126 brouard 12622: if((ficreseij=fopen(filerese,"w"))==NULL) {
12623: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12624: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12625: }
1.208 brouard 12626: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12627: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12628:
12629: pstamp(ficreseij);
1.219 brouard 12630:
1.235 brouard 12631: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12632: if (cptcovn < 1){i1=1;}
12633:
12634: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12635: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12636: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12637: continue;
1.219 brouard 12638: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12639: printf("\n#****** ");
1.225 brouard 12640: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12641: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12642: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12643: }
12644: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12645: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12646: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12647: }
12648: fprintf(ficreseij,"******\n");
1.235 brouard 12649: printf("******\n");
1.219 brouard 12650:
12651: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12652: oldm=oldms;savm=savms;
1.235 brouard 12653: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12654:
1.219 brouard 12655: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12656: }
12657: fclose(ficreseij);
1.208 brouard 12658: printf("done evsij\n");fflush(stdout);
12659: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12660:
1.218 brouard 12661:
1.227 brouard 12662: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12663:
1.201 brouard 12664: strcpy(filerest,"T_");
12665: strcat(filerest,fileresu);
1.127 brouard 12666: if((ficrest=fopen(filerest,"w"))==NULL) {
12667: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12668: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12669: }
1.208 brouard 12670: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12671: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12672: strcpy(fileresstde,"STDE_");
12673: strcat(fileresstde,fileresu);
1.126 brouard 12674: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12675: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12676: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12677: }
1.227 brouard 12678: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12679: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12680:
1.201 brouard 12681: strcpy(filerescve,"CVE_");
12682: strcat(filerescve,fileresu);
1.126 brouard 12683: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12684: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12685: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12686: }
1.227 brouard 12687: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12688: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12689:
1.201 brouard 12690: strcpy(fileresv,"V_");
12691: strcat(fileresv,fileresu);
1.126 brouard 12692: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12693: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12694: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12695: }
1.227 brouard 12696: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12697: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12698:
1.235 brouard 12699: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12700: if (cptcovn < 1){i1=1;}
12701:
12702: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12703: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12704: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12705: continue;
1.242 brouard 12706: printf("\n#****** Result for:");
12707: fprintf(ficrest,"\n#****** Result for:");
12708: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12709: for(j=1;j<=cptcoveff;j++){
12710: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12711: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12712: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12713: }
1.235 brouard 12714: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12715: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12716: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12717: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12718: }
1.208 brouard 12719: fprintf(ficrest,"******\n");
1.227 brouard 12720: fprintf(ficlog,"******\n");
12721: printf("******\n");
1.208 brouard 12722:
12723: fprintf(ficresstdeij,"\n#****** ");
12724: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12725: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12726: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12727: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12728: }
1.235 brouard 12729: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12730: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12731: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12732: }
1.208 brouard 12733: fprintf(ficresstdeij,"******\n");
12734: fprintf(ficrescveij,"******\n");
12735:
12736: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12737: /* pstamp(ficresvij); */
1.225 brouard 12738: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12739: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12740: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12741: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12742: }
1.208 brouard 12743: fprintf(ficresvij,"******\n");
12744:
12745: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12746: oldm=oldms;savm=savms;
1.235 brouard 12747: printf(" cvevsij ");
12748: fprintf(ficlog, " cvevsij ");
12749: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12750: printf(" end cvevsij \n ");
12751: fprintf(ficlog, " end cvevsij \n ");
12752:
12753: /*
12754: */
12755: /* goto endfree; */
12756:
12757: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12758: pstamp(ficrest);
12759:
1.269 brouard 12760: epj=vector(1,nlstate+1);
1.208 brouard 12761: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12762: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12763: cptcod= 0; /* To be deleted */
12764: printf("varevsij vpopbased=%d \n",vpopbased);
12765: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12766: 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 12767: 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 ");
12768: if(vpopbased==1)
12769: 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);
12770: else
1.288 brouard 12771: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12772: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12773: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12774: fprintf(ficrest,"\n");
12775: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 12776: printf("Computing age specific forward period (stable) prevalences in each health state \n");
12777: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12778: for(age=bage; age <=fage ;age++){
1.235 brouard 12779: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12780: if (vpopbased==1) {
12781: if(mobilav ==0){
12782: for(i=1; i<=nlstate;i++)
12783: prlim[i][i]=probs[(int)age][i][k];
12784: }else{ /* mobilav */
12785: for(i=1; i<=nlstate;i++)
12786: prlim[i][i]=mobaverage[(int)age][i][k];
12787: }
12788: }
1.219 brouard 12789:
1.227 brouard 12790: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12791: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12792: /* printf(" age %4.0f ",age); */
12793: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12794: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12795: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12796: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12797: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12798: }
12799: epj[nlstate+1] +=epj[j];
12800: }
12801: /* printf(" age %4.0f \n",age); */
1.219 brouard 12802:
1.227 brouard 12803: for(i=1, vepp=0.;i <=nlstate;i++)
12804: for(j=1;j <=nlstate;j++)
12805: vepp += vareij[i][j][(int)age];
12806: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12807: for(j=1;j <=nlstate;j++){
12808: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12809: }
12810: fprintf(ficrest,"\n");
12811: }
1.208 brouard 12812: } /* End vpopbased */
1.269 brouard 12813: free_vector(epj,1,nlstate+1);
1.208 brouard 12814: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12815: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12816: printf("done selection\n");fflush(stdout);
12817: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12818:
1.235 brouard 12819: } /* End k selection */
1.227 brouard 12820:
12821: printf("done State-specific expectancies\n");fflush(stdout);
12822: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12823:
1.288 brouard 12824: /* variance-covariance of forward period prevalence*/
1.269 brouard 12825: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12826:
1.227 brouard 12827:
1.290 brouard 12828: free_vector(weight,firstobs,lastobs);
1.227 brouard 12829: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 12830: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
12831: free_matrix(anint,1,maxwav,firstobs,lastobs);
12832: free_matrix(mint,1,maxwav,firstobs,lastobs);
12833: free_ivector(cod,firstobs,lastobs);
1.227 brouard 12834: free_ivector(tab,1,NCOVMAX);
12835: fclose(ficresstdeij);
12836: fclose(ficrescveij);
12837: fclose(ficresvij);
12838: fclose(ficrest);
12839: fclose(ficpar);
12840:
12841:
1.126 brouard 12842: /*---------- End : free ----------------*/
1.219 brouard 12843: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12844: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12845: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12846: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12847: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12848: } /* mle==-3 arrives here for freeing */
1.227 brouard 12849: /* endfree:*/
12850: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12851: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12852: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 12853: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
12854: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
12855: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
12856: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 12857: free_matrix(matcov,1,npar,1,npar);
12858: free_matrix(hess,1,npar,1,npar);
12859: /*free_vector(delti,1,npar);*/
12860: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12861: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12862: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12863: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12864:
12865: free_ivector(ncodemax,1,NCOVMAX);
12866: free_ivector(ncodemaxwundef,1,NCOVMAX);
12867: free_ivector(Dummy,-1,NCOVMAX);
12868: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12869: free_ivector(DummyV,1,NCOVMAX);
12870: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12871: free_ivector(Typevar,-1,NCOVMAX);
12872: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12873: free_ivector(TvarsQ,1,NCOVMAX);
12874: free_ivector(TvarsQind,1,NCOVMAX);
12875: free_ivector(TvarsD,1,NCOVMAX);
12876: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12877: free_ivector(TvarFD,1,NCOVMAX);
12878: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12879: free_ivector(TvarF,1,NCOVMAX);
12880: free_ivector(TvarFind,1,NCOVMAX);
12881: free_ivector(TvarV,1,NCOVMAX);
12882: free_ivector(TvarVind,1,NCOVMAX);
12883: free_ivector(TvarA,1,NCOVMAX);
12884: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12885: free_ivector(TvarFQ,1,NCOVMAX);
12886: free_ivector(TvarFQind,1,NCOVMAX);
12887: free_ivector(TvarVD,1,NCOVMAX);
12888: free_ivector(TvarVDind,1,NCOVMAX);
12889: free_ivector(TvarVQ,1,NCOVMAX);
12890: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12891: free_ivector(Tvarsel,1,NCOVMAX);
12892: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12893: free_ivector(Tposprod,1,NCOVMAX);
12894: free_ivector(Tprod,1,NCOVMAX);
12895: free_ivector(Tvaraff,1,NCOVMAX);
12896: free_ivector(invalidvarcomb,1,ncovcombmax);
12897: free_ivector(Tage,1,NCOVMAX);
12898: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12899: free_ivector(TmodelInvind,1,NCOVMAX);
12900: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12901:
12902: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12903: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12904: fflush(fichtm);
12905: fflush(ficgp);
12906:
1.227 brouard 12907:
1.126 brouard 12908: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12909: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12910: 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 12911: }else{
12912: printf("End of Imach\n");
12913: fprintf(ficlog,"End of Imach\n");
12914: }
12915: printf("See log file on %s\n",filelog);
12916: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12917: /*(void) gettimeofday(&end_time,&tzp);*/
12918: rend_time = time(NULL);
12919: end_time = *localtime(&rend_time);
12920: /* tml = *localtime(&end_time.tm_sec); */
12921: strcpy(strtend,asctime(&end_time));
1.126 brouard 12922: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12923: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12924: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12925:
1.157 brouard 12926: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12927: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12928: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12929: /* printf("Total time was %d uSec.\n", total_usecs);*/
12930: /* if(fileappend(fichtm,optionfilehtm)){ */
12931: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12932: fclose(fichtm);
12933: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12934: fclose(fichtmcov);
12935: fclose(ficgp);
12936: fclose(ficlog);
12937: /*------ End -----------*/
1.227 brouard 12938:
1.281 brouard 12939:
12940: /* Executes gnuplot */
1.227 brouard 12941:
12942: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12943: #ifdef WIN32
1.227 brouard 12944: if (_chdir(pathcd) != 0)
12945: printf("Can't move to directory %s!\n",path);
12946: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12947: #else
1.227 brouard 12948: if(chdir(pathcd) != 0)
12949: printf("Can't move to directory %s!\n", path);
12950: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12951: #endif
1.126 brouard 12952: printf("Current directory %s!\n",pathcd);
12953: /*strcat(plotcmd,CHARSEPARATOR);*/
12954: sprintf(plotcmd,"gnuplot");
1.157 brouard 12955: #ifdef _WIN32
1.126 brouard 12956: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12957: #endif
12958: if(!stat(plotcmd,&info)){
1.158 brouard 12959: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12960: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12961: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12962: }else
12963: strcpy(pplotcmd,plotcmd);
1.157 brouard 12964: #ifdef __unix
1.126 brouard 12965: strcpy(plotcmd,GNUPLOTPROGRAM);
12966: if(!stat(plotcmd,&info)){
1.158 brouard 12967: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12968: }else
12969: strcpy(pplotcmd,plotcmd);
12970: #endif
12971: }else
12972: strcpy(pplotcmd,plotcmd);
12973:
12974: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12975: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 12976: strcpy(pplotcmd,plotcmd);
1.227 brouard 12977:
1.126 brouard 12978: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 12979: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12980: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12981: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 12982: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 12983: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 12984: strcpy(plotcmd,pplotcmd);
12985: }
1.126 brouard 12986: }
1.158 brouard 12987: printf(" Successful, please wait...");
1.126 brouard 12988: while (z[0] != 'q') {
12989: /* chdir(path); */
1.154 brouard 12990: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12991: scanf("%s",z);
12992: /* if (z[0] == 'c') system("./imach"); */
12993: if (z[0] == 'e') {
1.158 brouard 12994: #ifdef __APPLE__
1.152 brouard 12995: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12996: #elif __linux
12997: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12998: #else
1.152 brouard 12999: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13000: #endif
13001: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13002: system(pplotcmd);
1.126 brouard 13003: }
13004: else if (z[0] == 'g') system(plotcmd);
13005: else if (z[0] == 'q') exit(0);
13006: }
1.227 brouard 13007: end:
1.126 brouard 13008: while (z[0] != 'q') {
1.195 brouard 13009: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13010: scanf("%s",z);
13011: }
1.283 brouard 13012: printf("End\n");
1.282 brouard 13013: exit(0);
1.126 brouard 13014: }
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