Annotation of imach/src/imach.c, revision 1.299
1.299 ! brouard 1: /* $Id: imach.c,v 1.298 2019/05/22 18:19:56 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.299 ! brouard 4: Revision 1.298 2019/05/22 18:19:56 brouard
! 5: *** empty log message ***
! 6:
1.298 brouard 7: Revision 1.297 2019/05/22 17:56:10 brouard
8: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
9:
1.297 brouard 10: Revision 1.296 2019/05/20 13:03:18 brouard
11: Summary: Projection syntax simplified
12:
13:
14: We can now start projections, forward or backward, from the mean date
15: of inteviews up to or down to a number of years of projection:
16: prevforecast=1 yearsfproj=15.3 mobil_average=0
17: or
18: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
19: or
20: prevbackcast=1 yearsbproj=12.3 mobil_average=1
21: or
22: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
23:
1.296 brouard 24: Revision 1.295 2019/05/18 09:52:50 brouard
25: Summary: doxygen tex bug
26:
1.295 brouard 27: Revision 1.294 2019/05/16 14:54:33 brouard
28: Summary: There was some wrong lines added
29:
1.294 brouard 30: Revision 1.293 2019/05/09 15:17:34 brouard
31: *** empty log message ***
32:
1.293 brouard 33: Revision 1.292 2019/05/09 14:17:20 brouard
34: Summary: Some updates
35:
1.292 brouard 36: Revision 1.291 2019/05/09 13:44:18 brouard
37: Summary: Before ncovmax
38:
1.291 brouard 39: Revision 1.290 2019/05/09 13:39:37 brouard
40: Summary: 0.99r18 unlimited number of individuals
41:
42: 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.
43:
1.290 brouard 44: Revision 1.289 2018/12/13 09:16:26 brouard
45: Summary: Bug for young ages (<-30) will be in r17
46:
1.289 brouard 47: Revision 1.288 2018/05/02 20:58:27 brouard
48: Summary: Some bugs fixed
49:
1.288 brouard 50: Revision 1.287 2018/05/01 17:57:25 brouard
51: Summary: Bug fixed by providing frequencies only for non missing covariates
52:
1.287 brouard 53: Revision 1.286 2018/04/27 14:27:04 brouard
54: Summary: some minor bugs
55:
1.286 brouard 56: Revision 1.285 2018/04/21 21:02:16 brouard
57: Summary: Some bugs fixed, valgrind tested
58:
1.285 brouard 59: Revision 1.284 2018/04/20 05:22:13 brouard
60: Summary: Computing mean and stdeviation of fixed quantitative variables
61:
1.284 brouard 62: Revision 1.283 2018/04/19 14:49:16 brouard
63: Summary: Some minor bugs fixed
64:
1.283 brouard 65: Revision 1.282 2018/02/27 22:50:02 brouard
66: *** empty log message ***
67:
1.282 brouard 68: Revision 1.281 2018/02/27 19:25:23 brouard
69: Summary: Adding second argument for quitting
70:
1.281 brouard 71: Revision 1.280 2018/02/21 07:58:13 brouard
72: Summary: 0.99r15
73:
74: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
75:
1.280 brouard 76: Revision 1.279 2017/07/20 13:35:01 brouard
77: Summary: temporary working
78:
1.279 brouard 79: Revision 1.278 2017/07/19 14:09:02 brouard
80: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
81:
1.278 brouard 82: Revision 1.277 2017/07/17 08:53:49 brouard
83: Summary: BOM files can be read now
84:
1.277 brouard 85: Revision 1.276 2017/06/30 15:48:31 brouard
86: Summary: Graphs improvements
87:
1.276 brouard 88: Revision 1.275 2017/06/30 13:39:33 brouard
89: Summary: Saito's color
90:
1.275 brouard 91: Revision 1.274 2017/06/29 09:47:08 brouard
92: Summary: Version 0.99r14
93:
1.274 brouard 94: Revision 1.273 2017/06/27 11:06:02 brouard
95: Summary: More documentation on projections
96:
1.273 brouard 97: Revision 1.272 2017/06/27 10:22:40 brouard
98: Summary: Color of backprojection changed from 6 to 5(yellow)
99:
1.272 brouard 100: Revision 1.271 2017/06/27 10:17:50 brouard
101: Summary: Some bug with rint
102:
1.271 brouard 103: Revision 1.270 2017/05/24 05:45:29 brouard
104: *** empty log message ***
105:
1.270 brouard 106: Revision 1.269 2017/05/23 08:39:25 brouard
107: Summary: Code into subroutine, cleanings
108:
1.269 brouard 109: Revision 1.268 2017/05/18 20:09:32 brouard
110: Summary: backprojection and confidence intervals of backprevalence
111:
1.268 brouard 112: Revision 1.267 2017/05/13 10:25:05 brouard
113: Summary: temporary save for backprojection
114:
1.267 brouard 115: Revision 1.266 2017/05/13 07:26:12 brouard
116: Summary: Version 0.99r13 (improvements and bugs fixed)
117:
1.266 brouard 118: Revision 1.265 2017/04/26 16:22:11 brouard
119: Summary: imach 0.99r13 Some bugs fixed
120:
1.265 brouard 121: Revision 1.264 2017/04/26 06:01:29 brouard
122: Summary: Labels in graphs
123:
1.264 brouard 124: Revision 1.263 2017/04/24 15:23:15 brouard
125: Summary: to save
126:
1.263 brouard 127: Revision 1.262 2017/04/18 16:48:12 brouard
128: *** empty log message ***
129:
1.262 brouard 130: Revision 1.261 2017/04/05 10:14:09 brouard
131: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
132:
1.261 brouard 133: Revision 1.260 2017/04/04 17:46:59 brouard
134: Summary: Gnuplot indexations fixed (humm)
135:
1.260 brouard 136: Revision 1.259 2017/04/04 13:01:16 brouard
137: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
138:
1.259 brouard 139: Revision 1.258 2017/04/03 10:17:47 brouard
140: Summary: Version 0.99r12
141:
142: Some cleanings, conformed with updated documentation.
143:
1.258 brouard 144: Revision 1.257 2017/03/29 16:53:30 brouard
145: Summary: Temp
146:
1.257 brouard 147: Revision 1.256 2017/03/27 05:50:23 brouard
148: Summary: Temporary
149:
1.256 brouard 150: Revision 1.255 2017/03/08 16:02:28 brouard
151: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
152:
1.255 brouard 153: Revision 1.254 2017/03/08 07:13:00 brouard
154: Summary: Fixing data parameter line
155:
1.254 brouard 156: Revision 1.253 2016/12/15 11:59:41 brouard
157: Summary: 0.99 in progress
158:
1.253 brouard 159: Revision 1.252 2016/09/15 21:15:37 brouard
160: *** empty log message ***
161:
1.252 brouard 162: Revision 1.251 2016/09/15 15:01:13 brouard
163: Summary: not working
164:
1.251 brouard 165: Revision 1.250 2016/09/08 16:07:27 brouard
166: Summary: continue
167:
1.250 brouard 168: Revision 1.249 2016/09/07 17:14:18 brouard
169: Summary: Starting values from frequencies
170:
1.249 brouard 171: Revision 1.248 2016/09/07 14:10:18 brouard
172: *** empty log message ***
173:
1.248 brouard 174: Revision 1.247 2016/09/02 11:11:21 brouard
175: *** empty log message ***
176:
1.247 brouard 177: Revision 1.246 2016/09/02 08:49:22 brouard
178: *** empty log message ***
179:
1.246 brouard 180: Revision 1.245 2016/09/02 07:25:01 brouard
181: *** empty log message ***
182:
1.245 brouard 183: Revision 1.244 2016/09/02 07:17:34 brouard
184: *** empty log message ***
185:
1.244 brouard 186: Revision 1.243 2016/09/02 06:45:35 brouard
187: *** empty log message ***
188:
1.243 brouard 189: Revision 1.242 2016/08/30 15:01:20 brouard
190: Summary: Fixing a lots
191:
1.242 brouard 192: Revision 1.241 2016/08/29 17:17:25 brouard
193: Summary: gnuplot problem in Back projection to fix
194:
1.241 brouard 195: Revision 1.240 2016/08/29 07:53:18 brouard
196: Summary: Better
197:
1.240 brouard 198: Revision 1.239 2016/08/26 15:51:03 brouard
199: Summary: Improvement in Powell output in order to copy and paste
200:
201: Author:
202:
1.239 brouard 203: Revision 1.238 2016/08/26 14:23:35 brouard
204: Summary: Starting tests of 0.99
205:
1.238 brouard 206: Revision 1.237 2016/08/26 09:20:19 brouard
207: Summary: to valgrind
208:
1.237 brouard 209: Revision 1.236 2016/08/25 10:50:18 brouard
210: *** empty log message ***
211:
1.236 brouard 212: Revision 1.235 2016/08/25 06:59:23 brouard
213: *** empty log message ***
214:
1.235 brouard 215: Revision 1.234 2016/08/23 16:51:20 brouard
216: *** empty log message ***
217:
1.234 brouard 218: Revision 1.233 2016/08/23 07:40:50 brouard
219: Summary: not working
220:
1.233 brouard 221: Revision 1.232 2016/08/22 14:20:21 brouard
222: Summary: not working
223:
1.232 brouard 224: Revision 1.231 2016/08/22 07:17:15 brouard
225: Summary: not working
226:
1.231 brouard 227: Revision 1.230 2016/08/22 06:55:53 brouard
228: Summary: Not working
229:
1.230 brouard 230: Revision 1.229 2016/07/23 09:45:53 brouard
231: Summary: Completing for func too
232:
1.229 brouard 233: Revision 1.228 2016/07/22 17:45:30 brouard
234: Summary: Fixing some arrays, still debugging
235:
1.227 brouard 236: Revision 1.226 2016/07/12 18:42:34 brouard
237: Summary: temp
238:
1.226 brouard 239: Revision 1.225 2016/07/12 08:40:03 brouard
240: Summary: saving but not running
241:
1.225 brouard 242: Revision 1.224 2016/07/01 13:16:01 brouard
243: Summary: Fixes
244:
1.224 brouard 245: Revision 1.223 2016/02/19 09:23:35 brouard
246: Summary: temporary
247:
1.223 brouard 248: Revision 1.222 2016/02/17 08:14:50 brouard
249: Summary: Probably last 0.98 stable version 0.98r6
250:
1.222 brouard 251: Revision 1.221 2016/02/15 23:35:36 brouard
252: Summary: minor bug
253:
1.220 brouard 254: Revision 1.219 2016/02/15 00:48:12 brouard
255: *** empty log message ***
256:
1.219 brouard 257: Revision 1.218 2016/02/12 11:29:23 brouard
258: Summary: 0.99 Back projections
259:
1.218 brouard 260: Revision 1.217 2015/12/23 17:18:31 brouard
261: Summary: Experimental backcast
262:
1.217 brouard 263: Revision 1.216 2015/12/18 17:32:11 brouard
264: Summary: 0.98r4 Warning and status=-2
265:
266: Version 0.98r4 is now:
267: - displaying an error when status is -1, date of interview unknown and date of death known;
268: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
269: Older changes concerning s=-2, dating from 2005 have been supersed.
270:
1.216 brouard 271: Revision 1.215 2015/12/16 08:52:24 brouard
272: Summary: 0.98r4 working
273:
1.215 brouard 274: Revision 1.214 2015/12/16 06:57:54 brouard
275: Summary: temporary not working
276:
1.214 brouard 277: Revision 1.213 2015/12/11 18:22:17 brouard
278: Summary: 0.98r4
279:
1.213 brouard 280: Revision 1.212 2015/11/21 12:47:24 brouard
281: Summary: minor typo
282:
1.212 brouard 283: Revision 1.211 2015/11/21 12:41:11 brouard
284: Summary: 0.98r3 with some graph of projected cross-sectional
285:
286: Author: Nicolas Brouard
287:
1.211 brouard 288: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 289: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 290: Summary: Adding ftolpl parameter
291: Author: N Brouard
292:
293: We had difficulties to get smoothed confidence intervals. It was due
294: to the period prevalence which wasn't computed accurately. The inner
295: parameter ftolpl is now an outer parameter of the .imach parameter
296: file after estepm. If ftolpl is small 1.e-4 and estepm too,
297: computation are long.
298:
1.209 brouard 299: Revision 1.208 2015/11/17 14:31:57 brouard
300: Summary: temporary
301:
1.208 brouard 302: Revision 1.207 2015/10/27 17:36:57 brouard
303: *** empty log message ***
304:
1.207 brouard 305: Revision 1.206 2015/10/24 07:14:11 brouard
306: *** empty log message ***
307:
1.206 brouard 308: Revision 1.205 2015/10/23 15:50:53 brouard
309: Summary: 0.98r3 some clarification for graphs on likelihood contributions
310:
1.205 brouard 311: Revision 1.204 2015/10/01 16:20:26 brouard
312: Summary: Some new graphs of contribution to likelihood
313:
1.204 brouard 314: Revision 1.203 2015/09/30 17:45:14 brouard
315: Summary: looking at better estimation of the hessian
316:
317: Also a better criteria for convergence to the period prevalence And
318: therefore adding the number of years needed to converge. (The
319: prevalence in any alive state shold sum to one
320:
1.203 brouard 321: Revision 1.202 2015/09/22 19:45:16 brouard
322: Summary: Adding some overall graph on contribution to likelihood. Might change
323:
1.202 brouard 324: Revision 1.201 2015/09/15 17:34:58 brouard
325: Summary: 0.98r0
326:
327: - Some new graphs like suvival functions
328: - Some bugs fixed like model=1+age+V2.
329:
1.201 brouard 330: Revision 1.200 2015/09/09 16:53:55 brouard
331: Summary: Big bug thanks to Flavia
332:
333: Even model=1+age+V2. did not work anymore
334:
1.200 brouard 335: Revision 1.199 2015/09/07 14:09:23 brouard
336: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
337:
1.199 brouard 338: Revision 1.198 2015/09/03 07:14:39 brouard
339: Summary: 0.98q5 Flavia
340:
1.198 brouard 341: Revision 1.197 2015/09/01 18:24:39 brouard
342: *** empty log message ***
343:
1.197 brouard 344: Revision 1.196 2015/08/18 23:17:52 brouard
345: Summary: 0.98q5
346:
1.196 brouard 347: Revision 1.195 2015/08/18 16:28:39 brouard
348: Summary: Adding a hack for testing purpose
349:
350: After reading the title, ftol and model lines, if the comment line has
351: a q, starting with #q, the answer at the end of the run is quit. It
352: permits to run test files in batch with ctest. The former workaround was
353: $ echo q | imach foo.imach
354:
1.195 brouard 355: Revision 1.194 2015/08/18 13:32:00 brouard
356: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
357:
1.194 brouard 358: Revision 1.193 2015/08/04 07:17:42 brouard
359: Summary: 0.98q4
360:
1.193 brouard 361: Revision 1.192 2015/07/16 16:49:02 brouard
362: Summary: Fixing some outputs
363:
1.192 brouard 364: Revision 1.191 2015/07/14 10:00:33 brouard
365: Summary: Some fixes
366:
1.191 brouard 367: Revision 1.190 2015/05/05 08:51:13 brouard
368: Summary: Adding digits in output parameters (7 digits instead of 6)
369:
370: Fix 1+age+.
371:
1.190 brouard 372: Revision 1.189 2015/04/30 14:45:16 brouard
373: Summary: 0.98q2
374:
1.189 brouard 375: Revision 1.188 2015/04/30 08:27:53 brouard
376: *** empty log message ***
377:
1.188 brouard 378: Revision 1.187 2015/04/29 09:11:15 brouard
379: *** empty log message ***
380:
1.187 brouard 381: Revision 1.186 2015/04/23 12:01:52 brouard
382: Summary: V1*age is working now, version 0.98q1
383:
384: Some codes had been disabled in order to simplify and Vn*age was
385: working in the optimization phase, ie, giving correct MLE parameters,
386: but, as usual, outputs were not correct and program core dumped.
387:
1.186 brouard 388: Revision 1.185 2015/03/11 13:26:42 brouard
389: Summary: Inclusion of compile and links command line for Intel Compiler
390:
1.185 brouard 391: Revision 1.184 2015/03/11 11:52:39 brouard
392: Summary: Back from Windows 8. Intel Compiler
393:
1.184 brouard 394: Revision 1.183 2015/03/10 20:34:32 brouard
395: Summary: 0.98q0, trying with directest, mnbrak fixed
396:
397: We use directest instead of original Powell test; probably no
398: incidence on the results, but better justifications;
399: We fixed Numerical Recipes mnbrak routine which was wrong and gave
400: wrong results.
401:
1.183 brouard 402: Revision 1.182 2015/02/12 08:19:57 brouard
403: Summary: Trying to keep directest which seems simpler and more general
404: Author: Nicolas Brouard
405:
1.182 brouard 406: Revision 1.181 2015/02/11 23:22:24 brouard
407: Summary: Comments on Powell added
408:
409: Author:
410:
1.181 brouard 411: Revision 1.180 2015/02/11 17:33:45 brouard
412: Summary: Finishing move from main to function (hpijx and prevalence_limit)
413:
1.180 brouard 414: Revision 1.179 2015/01/04 09:57:06 brouard
415: Summary: back to OS/X
416:
1.179 brouard 417: Revision 1.178 2015/01/04 09:35:48 brouard
418: *** empty log message ***
419:
1.178 brouard 420: Revision 1.177 2015/01/03 18:40:56 brouard
421: Summary: Still testing ilc32 on OSX
422:
1.177 brouard 423: Revision 1.176 2015/01/03 16:45:04 brouard
424: *** empty log message ***
425:
1.176 brouard 426: Revision 1.175 2015/01/03 16:33:42 brouard
427: *** empty log message ***
428:
1.175 brouard 429: Revision 1.174 2015/01/03 16:15:49 brouard
430: Summary: Still in cross-compilation
431:
1.174 brouard 432: Revision 1.173 2015/01/03 12:06:26 brouard
433: Summary: trying to detect cross-compilation
434:
1.173 brouard 435: Revision 1.172 2014/12/27 12:07:47 brouard
436: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
437:
1.172 brouard 438: Revision 1.171 2014/12/23 13:26:59 brouard
439: Summary: Back from Visual C
440:
441: Still problem with utsname.h on Windows
442:
1.171 brouard 443: Revision 1.170 2014/12/23 11:17:12 brouard
444: Summary: Cleaning some \%% back to %%
445:
446: The escape was mandatory for a specific compiler (which one?), but too many warnings.
447:
1.170 brouard 448: Revision 1.169 2014/12/22 23:08:31 brouard
449: Summary: 0.98p
450:
451: Outputs some informations on compiler used, OS etc. Testing on different platforms.
452:
1.169 brouard 453: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 454: Summary: update
1.169 brouard 455:
1.168 brouard 456: Revision 1.167 2014/12/22 13:50:56 brouard
457: Summary: Testing uname and compiler version and if compiled 32 or 64
458:
459: Testing on Linux 64
460:
1.167 brouard 461: Revision 1.166 2014/12/22 11:40:47 brouard
462: *** empty log message ***
463:
1.166 brouard 464: Revision 1.165 2014/12/16 11:20:36 brouard
465: Summary: After compiling on Visual C
466:
467: * imach.c (Module): Merging 1.61 to 1.162
468:
1.165 brouard 469: Revision 1.164 2014/12/16 10:52:11 brouard
470: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
471:
472: * imach.c (Module): Merging 1.61 to 1.162
473:
1.164 brouard 474: Revision 1.163 2014/12/16 10:30:11 brouard
475: * imach.c (Module): Merging 1.61 to 1.162
476:
1.163 brouard 477: Revision 1.162 2014/09/25 11:43:39 brouard
478: Summary: temporary backup 0.99!
479:
1.162 brouard 480: Revision 1.1 2014/09/16 11:06:58 brouard
481: Summary: With some code (wrong) for nlopt
482:
483: Author:
484:
485: Revision 1.161 2014/09/15 20:41:41 brouard
486: Summary: Problem with macro SQR on Intel compiler
487:
1.161 brouard 488: Revision 1.160 2014/09/02 09:24:05 brouard
489: *** empty log message ***
490:
1.160 brouard 491: Revision 1.159 2014/09/01 10:34:10 brouard
492: Summary: WIN32
493: Author: Brouard
494:
1.159 brouard 495: Revision 1.158 2014/08/27 17:11:51 brouard
496: *** empty log message ***
497:
1.158 brouard 498: Revision 1.157 2014/08/27 16:26:55 brouard
499: Summary: Preparing windows Visual studio version
500: Author: Brouard
501:
502: In order to compile on Visual studio, time.h is now correct and time_t
503: and tm struct should be used. difftime should be used but sometimes I
504: just make the differences in raw time format (time(&now).
505: Trying to suppress #ifdef LINUX
506: Add xdg-open for __linux in order to open default browser.
507:
1.157 brouard 508: Revision 1.156 2014/08/25 20:10:10 brouard
509: *** empty log message ***
510:
1.156 brouard 511: Revision 1.155 2014/08/25 18:32:34 brouard
512: Summary: New compile, minor changes
513: Author: Brouard
514:
1.155 brouard 515: Revision 1.154 2014/06/20 17:32:08 brouard
516: Summary: Outputs now all graphs of convergence to period prevalence
517:
1.154 brouard 518: Revision 1.153 2014/06/20 16:45:46 brouard
519: Summary: If 3 live state, convergence to period prevalence on same graph
520: Author: Brouard
521:
1.153 brouard 522: Revision 1.152 2014/06/18 17:54:09 brouard
523: Summary: open browser, use gnuplot on same dir than imach if not found in the path
524:
1.152 brouard 525: Revision 1.151 2014/06/18 16:43:30 brouard
526: *** empty log message ***
527:
1.151 brouard 528: Revision 1.150 2014/06/18 16:42:35 brouard
529: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
530: Author: brouard
531:
1.150 brouard 532: Revision 1.149 2014/06/18 15:51:14 brouard
533: Summary: Some fixes in parameter files errors
534: Author: Nicolas Brouard
535:
1.149 brouard 536: Revision 1.148 2014/06/17 17:38:48 brouard
537: Summary: Nothing new
538: Author: Brouard
539:
540: Just a new packaging for OS/X version 0.98nS
541:
1.148 brouard 542: Revision 1.147 2014/06/16 10:33:11 brouard
543: *** empty log message ***
544:
1.147 brouard 545: Revision 1.146 2014/06/16 10:20:28 brouard
546: Summary: Merge
547: Author: Brouard
548:
549: Merge, before building revised version.
550:
1.146 brouard 551: Revision 1.145 2014/06/10 21:23:15 brouard
552: Summary: Debugging with valgrind
553: Author: Nicolas Brouard
554:
555: Lot of changes in order to output the results with some covariates
556: After the Edimburgh REVES conference 2014, it seems mandatory to
557: improve the code.
558: No more memory valgrind error but a lot has to be done in order to
559: continue the work of splitting the code into subroutines.
560: Also, decodemodel has been improved. Tricode is still not
561: optimal. nbcode should be improved. Documentation has been added in
562: the source code.
563:
1.144 brouard 564: Revision 1.143 2014/01/26 09:45:38 brouard
565: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
566:
567: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
568: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
569:
1.143 brouard 570: Revision 1.142 2014/01/26 03:57:36 brouard
571: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
572:
573: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
574:
1.142 brouard 575: Revision 1.141 2014/01/26 02:42:01 brouard
576: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
577:
1.141 brouard 578: Revision 1.140 2011/09/02 10:37:54 brouard
579: Summary: times.h is ok with mingw32 now.
580:
1.140 brouard 581: Revision 1.139 2010/06/14 07:50:17 brouard
582: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
583: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
584:
1.139 brouard 585: Revision 1.138 2010/04/30 18:19:40 brouard
586: *** empty log message ***
587:
1.138 brouard 588: Revision 1.137 2010/04/29 18:11:38 brouard
589: (Module): Checking covariates for more complex models
590: than V1+V2. A lot of change to be done. Unstable.
591:
1.137 brouard 592: Revision 1.136 2010/04/26 20:30:53 brouard
593: (Module): merging some libgsl code. Fixing computation
594: of likelione (using inter/intrapolation if mle = 0) in order to
595: get same likelihood as if mle=1.
596: Some cleaning of code and comments added.
597:
1.136 brouard 598: Revision 1.135 2009/10/29 15:33:14 brouard
599: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
600:
1.135 brouard 601: Revision 1.134 2009/10/29 13:18:53 brouard
602: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
603:
1.134 brouard 604: Revision 1.133 2009/07/06 10:21:25 brouard
605: just nforces
606:
1.133 brouard 607: Revision 1.132 2009/07/06 08:22:05 brouard
608: Many tings
609:
1.132 brouard 610: Revision 1.131 2009/06/20 16:22:47 brouard
611: Some dimensions resccaled
612:
1.131 brouard 613: Revision 1.130 2009/05/26 06:44:34 brouard
614: (Module): Max Covariate is now set to 20 instead of 8. A
615: lot of cleaning with variables initialized to 0. Trying to make
616: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
617:
1.130 brouard 618: Revision 1.129 2007/08/31 13:49:27 lievre
619: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
620:
1.129 lievre 621: Revision 1.128 2006/06/30 13:02:05 brouard
622: (Module): Clarifications on computing e.j
623:
1.128 brouard 624: Revision 1.127 2006/04/28 18:11:50 brouard
625: (Module): Yes the sum of survivors was wrong since
626: imach-114 because nhstepm was no more computed in the age
627: loop. Now we define nhstepma in the age loop.
628: (Module): In order to speed up (in case of numerous covariates) we
629: compute health expectancies (without variances) in a first step
630: and then all the health expectancies with variances or standard
631: deviation (needs data from the Hessian matrices) which slows the
632: computation.
633: In the future we should be able to stop the program is only health
634: expectancies and graph are needed without standard deviations.
635:
1.127 brouard 636: Revision 1.126 2006/04/28 17:23:28 brouard
637: (Module): Yes the sum of survivors was wrong since
638: imach-114 because nhstepm was no more computed in the age
639: loop. Now we define nhstepma in the age loop.
640: Version 0.98h
641:
1.126 brouard 642: Revision 1.125 2006/04/04 15:20:31 lievre
643: Errors in calculation of health expectancies. Age was not initialized.
644: Forecasting file added.
645:
646: Revision 1.124 2006/03/22 17:13:53 lievre
647: Parameters are printed with %lf instead of %f (more numbers after the comma).
648: The log-likelihood is printed in the log file
649:
650: Revision 1.123 2006/03/20 10:52:43 brouard
651: * imach.c (Module): <title> changed, corresponds to .htm file
652: name. <head> headers where missing.
653:
654: * imach.c (Module): Weights can have a decimal point as for
655: English (a comma might work with a correct LC_NUMERIC environment,
656: otherwise the weight is truncated).
657: Modification of warning when the covariates values are not 0 or
658: 1.
659: Version 0.98g
660:
661: Revision 1.122 2006/03/20 09:45:41 brouard
662: (Module): Weights can have a decimal point as for
663: English (a comma might work with a correct LC_NUMERIC environment,
664: otherwise the weight is truncated).
665: Modification of warning when the covariates values are not 0 or
666: 1.
667: Version 0.98g
668:
669: Revision 1.121 2006/03/16 17:45:01 lievre
670: * imach.c (Module): Comments concerning covariates added
671:
672: * imach.c (Module): refinements in the computation of lli if
673: status=-2 in order to have more reliable computation if stepm is
674: not 1 month. Version 0.98f
675:
676: Revision 1.120 2006/03/16 15:10:38 lievre
677: (Module): refinements in the computation of lli if
678: status=-2 in order to have more reliable computation if stepm is
679: not 1 month. Version 0.98f
680:
681: Revision 1.119 2006/03/15 17:42:26 brouard
682: (Module): Bug if status = -2, the loglikelihood was
683: computed as likelihood omitting the logarithm. Version O.98e
684:
685: Revision 1.118 2006/03/14 18:20:07 brouard
686: (Module): varevsij Comments added explaining the second
687: table of variances if popbased=1 .
688: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
689: (Module): Function pstamp added
690: (Module): Version 0.98d
691:
692: Revision 1.117 2006/03/14 17:16:22 brouard
693: (Module): varevsij Comments added explaining the second
694: table of variances if popbased=1 .
695: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
696: (Module): Function pstamp added
697: (Module): Version 0.98d
698:
699: Revision 1.116 2006/03/06 10:29:27 brouard
700: (Module): Variance-covariance wrong links and
701: varian-covariance of ej. is needed (Saito).
702:
703: Revision 1.115 2006/02/27 12:17:45 brouard
704: (Module): One freematrix added in mlikeli! 0.98c
705:
706: Revision 1.114 2006/02/26 12:57:58 brouard
707: (Module): Some improvements in processing parameter
708: filename with strsep.
709:
710: Revision 1.113 2006/02/24 14:20:24 brouard
711: (Module): Memory leaks checks with valgrind and:
712: datafile was not closed, some imatrix were not freed and on matrix
713: allocation too.
714:
715: Revision 1.112 2006/01/30 09:55:26 brouard
716: (Module): Back to gnuplot.exe instead of wgnuplot.exe
717:
718: Revision 1.111 2006/01/25 20:38:18 brouard
719: (Module): Lots of cleaning and bugs added (Gompertz)
720: (Module): Comments can be added in data file. Missing date values
721: can be a simple dot '.'.
722:
723: Revision 1.110 2006/01/25 00:51:50 brouard
724: (Module): Lots of cleaning and bugs added (Gompertz)
725:
726: Revision 1.109 2006/01/24 19:37:15 brouard
727: (Module): Comments (lines starting with a #) are allowed in data.
728:
729: Revision 1.108 2006/01/19 18:05:42 lievre
730: Gnuplot problem appeared...
731: To be fixed
732:
733: Revision 1.107 2006/01/19 16:20:37 brouard
734: Test existence of gnuplot in imach path
735:
736: Revision 1.106 2006/01/19 13:24:36 brouard
737: Some cleaning and links added in html output
738:
739: Revision 1.105 2006/01/05 20:23:19 lievre
740: *** empty log message ***
741:
742: Revision 1.104 2005/09/30 16:11:43 lievre
743: (Module): sump fixed, loop imx fixed, and simplifications.
744: (Module): If the status is missing at the last wave but we know
745: that the person is alive, then we can code his/her status as -2
746: (instead of missing=-1 in earlier versions) and his/her
747: contributions to the likelihood is 1 - Prob of dying from last
748: health status (= 1-p13= p11+p12 in the easiest case of somebody in
749: the healthy state at last known wave). Version is 0.98
750:
751: Revision 1.103 2005/09/30 15:54:49 lievre
752: (Module): sump fixed, loop imx fixed, and simplifications.
753:
754: Revision 1.102 2004/09/15 17:31:30 brouard
755: Add the possibility to read data file including tab characters.
756:
757: Revision 1.101 2004/09/15 10:38:38 brouard
758: Fix on curr_time
759:
760: Revision 1.100 2004/07/12 18:29:06 brouard
761: Add version for Mac OS X. Just define UNIX in Makefile
762:
763: Revision 1.99 2004/06/05 08:57:40 brouard
764: *** empty log message ***
765:
766: Revision 1.98 2004/05/16 15:05:56 brouard
767: New version 0.97 . First attempt to estimate force of mortality
768: directly from the data i.e. without the need of knowing the health
769: state at each age, but using a Gompertz model: log u =a + b*age .
770: This is the basic analysis of mortality and should be done before any
771: other analysis, in order to test if the mortality estimated from the
772: cross-longitudinal survey is different from the mortality estimated
773: from other sources like vital statistic data.
774:
775: The same imach parameter file can be used but the option for mle should be -3.
776:
1.133 brouard 777: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 778: former routines in order to include the new code within the former code.
779:
780: The output is very simple: only an estimate of the intercept and of
781: the slope with 95% confident intervals.
782:
783: Current limitations:
784: A) Even if you enter covariates, i.e. with the
785: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
786: B) There is no computation of Life Expectancy nor Life Table.
787:
788: Revision 1.97 2004/02/20 13:25:42 lievre
789: Version 0.96d. Population forecasting command line is (temporarily)
790: suppressed.
791:
792: Revision 1.96 2003/07/15 15:38:55 brouard
793: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
794: rewritten within the same printf. Workaround: many printfs.
795:
796: Revision 1.95 2003/07/08 07:54:34 brouard
797: * imach.c (Repository):
798: (Repository): Using imachwizard code to output a more meaningful covariance
799: matrix (cov(a12,c31) instead of numbers.
800:
801: Revision 1.94 2003/06/27 13:00:02 brouard
802: Just cleaning
803:
804: Revision 1.93 2003/06/25 16:33:55 brouard
805: (Module): On windows (cygwin) function asctime_r doesn't
806: exist so I changed back to asctime which exists.
807: (Module): Version 0.96b
808:
809: Revision 1.92 2003/06/25 16:30:45 brouard
810: (Module): On windows (cygwin) function asctime_r doesn't
811: exist so I changed back to asctime which exists.
812:
813: Revision 1.91 2003/06/25 15:30:29 brouard
814: * imach.c (Repository): Duplicated warning errors corrected.
815: (Repository): Elapsed time after each iteration is now output. It
816: helps to forecast when convergence will be reached. Elapsed time
817: is stamped in powell. We created a new html file for the graphs
818: concerning matrix of covariance. It has extension -cov.htm.
819:
820: Revision 1.90 2003/06/24 12:34:15 brouard
821: (Module): Some bugs corrected for windows. Also, when
822: mle=-1 a template is output in file "or"mypar.txt with the design
823: of the covariance matrix to be input.
824:
825: Revision 1.89 2003/06/24 12:30:52 brouard
826: (Module): Some bugs corrected for windows. Also, when
827: mle=-1 a template is output in file "or"mypar.txt with the design
828: of the covariance matrix to be input.
829:
830: Revision 1.88 2003/06/23 17:54:56 brouard
831: * 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.
832:
833: Revision 1.87 2003/06/18 12:26:01 brouard
834: Version 0.96
835:
836: Revision 1.86 2003/06/17 20:04:08 brouard
837: (Module): Change position of html and gnuplot routines and added
838: routine fileappend.
839:
840: Revision 1.85 2003/06/17 13:12:43 brouard
841: * imach.c (Repository): Check when date of death was earlier that
842: current date of interview. It may happen when the death was just
843: prior to the death. In this case, dh was negative and likelihood
844: was wrong (infinity). We still send an "Error" but patch by
845: assuming that the date of death was just one stepm after the
846: interview.
847: (Repository): Because some people have very long ID (first column)
848: we changed int to long in num[] and we added a new lvector for
849: memory allocation. But we also truncated to 8 characters (left
850: truncation)
851: (Repository): No more line truncation errors.
852:
853: Revision 1.84 2003/06/13 21:44:43 brouard
854: * imach.c (Repository): Replace "freqsummary" at a correct
855: place. It differs from routine "prevalence" which may be called
856: many times. Probs is memory consuming and must be used with
857: parcimony.
858: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
859:
860: Revision 1.83 2003/06/10 13:39:11 lievre
861: *** empty log message ***
862:
863: Revision 1.82 2003/06/05 15:57:20 brouard
864: Add log in imach.c and fullversion number is now printed.
865:
866: */
867: /*
868: Interpolated Markov Chain
869:
870: Short summary of the programme:
871:
1.227 brouard 872: This program computes Healthy Life Expectancies or State-specific
873: (if states aren't health statuses) Expectancies from
874: cross-longitudinal data. Cross-longitudinal data consist in:
875:
876: -1- a first survey ("cross") where individuals from different ages
877: are interviewed on their health status or degree of disability (in
878: the case of a health survey which is our main interest)
879:
880: -2- at least a second wave of interviews ("longitudinal") which
881: measure each change (if any) in individual health status. Health
882: expectancies are computed from the time spent in each health state
883: according to a model. More health states you consider, more time is
884: necessary to reach the Maximum Likelihood of the parameters involved
885: in the model. The simplest model is the multinomial logistic model
886: where pij is the probability to be observed in state j at the second
887: wave conditional to be observed in state i at the first
888: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
889: etc , where 'age' is age and 'sex' is a covariate. If you want to
890: have a more complex model than "constant and age", you should modify
891: the program where the markup *Covariates have to be included here
892: again* invites you to do it. More covariates you add, slower the
1.126 brouard 893: convergence.
894:
895: The advantage of this computer programme, compared to a simple
896: multinomial logistic model, is clear when the delay between waves is not
897: identical for each individual. Also, if a individual missed an
898: intermediate interview, the information is lost, but taken into
899: account using an interpolation or extrapolation.
900:
901: hPijx is the probability to be observed in state i at age x+h
902: conditional to the observed state i at age x. The delay 'h' can be
903: split into an exact number (nh*stepm) of unobserved intermediate
904: states. This elementary transition (by month, quarter,
905: semester or year) is modelled as a multinomial logistic. The hPx
906: matrix is simply the matrix product of nh*stepm elementary matrices
907: and the contribution of each individual to the likelihood is simply
908: hPijx.
909:
910: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 911: of the life expectancies. It also computes the period (stable) prevalence.
912:
913: Back prevalence and projections:
1.227 brouard 914:
915: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
916: double agemaxpar, double ftolpl, int *ncvyearp, double
917: dateprev1,double dateprev2, int firstpass, int lastpass, int
918: mobilavproj)
919:
920: Computes the back prevalence limit for any combination of
921: covariate values k at any age between ageminpar and agemaxpar and
922: returns it in **bprlim. In the loops,
923:
924: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
925: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
926:
927: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 928: Computes for any combination of covariates k and any age between bage and fage
929: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
930: oldm=oldms;savm=savms;
1.227 brouard 931:
1.267 brouard 932: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 933: Computes the transition matrix starting at age 'age' over
934: 'nhstepm*hstepm*stepm' months (i.e. until
935: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 936: nhstepm*hstepm matrices.
937:
938: Returns p3mat[i][j][h] after calling
939: p3mat[i][j][h]=matprod2(newm,
940: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
941: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
942: oldm);
1.226 brouard 943:
944: Important routines
945:
946: - func (or funcone), computes logit (pij) distinguishing
947: o fixed variables (single or product dummies or quantitative);
948: o varying variables by:
949: (1) wave (single, product dummies, quantitative),
950: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
951: % fixed dummy (treated) or quantitative (not done because time-consuming);
952: % varying dummy (not done) or quantitative (not done);
953: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
954: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
955: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
956: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
957: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 958:
1.226 brouard 959:
960:
1.133 brouard 961: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
962: Institut national d'études démographiques, Paris.
1.126 brouard 963: This software have been partly granted by Euro-REVES, a concerted action
964: from the European Union.
965: It is copyrighted identically to a GNU software product, ie programme and
966: software can be distributed freely for non commercial use. Latest version
967: can be accessed at http://euroreves.ined.fr/imach .
968:
969: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
970: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
971:
972: **********************************************************************/
973: /*
974: main
975: read parameterfile
976: read datafile
977: concatwav
978: freqsummary
979: if (mle >= 1)
980: mlikeli
981: print results files
982: if mle==1
983: computes hessian
984: read end of parameter file: agemin, agemax, bage, fage, estepm
985: begin-prev-date,...
986: open gnuplot file
987: open html file
1.145 brouard 988: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
989: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
990: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
991: freexexit2 possible for memory heap.
992:
993: h Pij x | pij_nom ficrestpij
994: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
995: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
996: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
997:
998: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
999: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1000: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1001: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1002: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1003:
1.126 brouard 1004: forecasting if prevfcast==1 prevforecast call prevalence()
1005: health expectancies
1006: Variance-covariance of DFLE
1007: prevalence()
1008: movingaverage()
1009: varevsij()
1010: if popbased==1 varevsij(,popbased)
1011: total life expectancies
1012: Variance of period (stable) prevalence
1013: end
1014: */
1015:
1.187 brouard 1016: /* #define DEBUG */
1017: /* #define DEBUGBRENT */
1.203 brouard 1018: /* #define DEBUGLINMIN */
1019: /* #define DEBUGHESS */
1020: #define DEBUGHESSIJ
1.224 brouard 1021: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1022: #define POWELL /* Instead of NLOPT */
1.224 brouard 1023: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1024: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1025: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 1026:
1027: #include <math.h>
1028: #include <stdio.h>
1029: #include <stdlib.h>
1030: #include <string.h>
1.226 brouard 1031: #include <ctype.h>
1.159 brouard 1032:
1033: #ifdef _WIN32
1034: #include <io.h>
1.172 brouard 1035: #include <windows.h>
1036: #include <tchar.h>
1.159 brouard 1037: #else
1.126 brouard 1038: #include <unistd.h>
1.159 brouard 1039: #endif
1.126 brouard 1040:
1041: #include <limits.h>
1042: #include <sys/types.h>
1.171 brouard 1043:
1044: #if defined(__GNUC__)
1045: #include <sys/utsname.h> /* Doesn't work on Windows */
1046: #endif
1047:
1.126 brouard 1048: #include <sys/stat.h>
1049: #include <errno.h>
1.159 brouard 1050: /* extern int errno; */
1.126 brouard 1051:
1.157 brouard 1052: /* #ifdef LINUX */
1053: /* #include <time.h> */
1054: /* #include "timeval.h" */
1055: /* #else */
1056: /* #include <sys/time.h> */
1057: /* #endif */
1058:
1.126 brouard 1059: #include <time.h>
1060:
1.136 brouard 1061: #ifdef GSL
1062: #include <gsl/gsl_errno.h>
1063: #include <gsl/gsl_multimin.h>
1064: #endif
1065:
1.167 brouard 1066:
1.162 brouard 1067: #ifdef NLOPT
1068: #include <nlopt.h>
1069: typedef struct {
1070: double (* function)(double [] );
1071: } myfunc_data ;
1072: #endif
1073:
1.126 brouard 1074: /* #include <libintl.h> */
1075: /* #define _(String) gettext (String) */
1076:
1.251 brouard 1077: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1078:
1079: #define GNUPLOTPROGRAM "gnuplot"
1080: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1081: #define FILENAMELENGTH 132
1082:
1083: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1084: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1085:
1.144 brouard 1086: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1087: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1088:
1089: #define NINTERVMAX 8
1.144 brouard 1090: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1091: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1092: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1093: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1094: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1095: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1096: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1097: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1098: /* #define AGESUP 130 */
1.288 brouard 1099: /* #define AGESUP 150 */
1100: #define AGESUP 200
1.268 brouard 1101: #define AGEINF 0
1.218 brouard 1102: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1103: #define AGEBASE 40
1.194 brouard 1104: #define AGEOVERFLOW 1.e20
1.164 brouard 1105: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1106: #ifdef _WIN32
1107: #define DIRSEPARATOR '\\'
1108: #define CHARSEPARATOR "\\"
1109: #define ODIRSEPARATOR '/'
1110: #else
1.126 brouard 1111: #define DIRSEPARATOR '/'
1112: #define CHARSEPARATOR "/"
1113: #define ODIRSEPARATOR '\\'
1114: #endif
1115:
1.299 ! brouard 1116: /* $Id: imach.c,v 1.298 2019/05/22 18:19:56 brouard Exp $ */
1.126 brouard 1117: /* $State: Exp $ */
1.196 brouard 1118: #include "version.h"
1119: char version[]=__IMACH_VERSION__;
1.283 brouard 1120: 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";
1.299 ! brouard 1121: char fullversion[]="$Revision: 1.298 $ $Date: 2019/05/22 18:19:56 $";
1.126 brouard 1122: char strstart[80];
1123: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1124: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1125: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1126: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1127: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1128: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1129: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1130: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1131: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1132: int cptcovprodnoage=0; /**< Number of covariate products without age */
1133: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1134: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1135: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1136: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1137: int nsd=0; /**< Total number of single dummy variables (output) */
1138: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1139: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1140: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1141: int ntveff=0; /**< ntveff number of effective time varying variables */
1142: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1143: int cptcov=0; /* Working variable */
1.290 brouard 1144: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1145: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1146: int npar=NPARMAX;
1147: int nlstate=2; /* Number of live states */
1148: int ndeath=1; /* Number of dead states */
1.130 brouard 1149: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1150: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1151: int popbased=0;
1152:
1153: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1154: int maxwav=0; /* Maxim number of waves */
1155: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1156: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1157: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1158: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1159: int mle=1, weightopt=0;
1.126 brouard 1160: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1161: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1162: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1163: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1164: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1165: int selected(int kvar); /* Is covariate kvar selected for printing results */
1166:
1.130 brouard 1167: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1168: double **matprod2(); /* test */
1.126 brouard 1169: double **oldm, **newm, **savm; /* Working pointers to matrices */
1170: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1171: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1172:
1.136 brouard 1173: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1174: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1175: FILE *ficlog, *ficrespow;
1.130 brouard 1176: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1177: double fretone; /* Only one call to likelihood */
1.130 brouard 1178: long ipmx=0; /* Number of contributions */
1.126 brouard 1179: double sw; /* Sum of weights */
1180: char filerespow[FILENAMELENGTH];
1181: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1182: FILE *ficresilk;
1183: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1184: FILE *ficresprobmorprev;
1185: FILE *fichtm, *fichtmcov; /* Html File */
1186: FILE *ficreseij;
1187: char filerese[FILENAMELENGTH];
1188: FILE *ficresstdeij;
1189: char fileresstde[FILENAMELENGTH];
1190: FILE *ficrescveij;
1191: char filerescve[FILENAMELENGTH];
1192: FILE *ficresvij;
1193: char fileresv[FILENAMELENGTH];
1.269 brouard 1194:
1.126 brouard 1195: char title[MAXLINE];
1.234 brouard 1196: char model[MAXLINE]; /**< The model line */
1.217 brouard 1197: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1198: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1199: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1200: char command[FILENAMELENGTH];
1201: int outcmd=0;
1202:
1.217 brouard 1203: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1204: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1205: char filelog[FILENAMELENGTH]; /* Log file */
1206: char filerest[FILENAMELENGTH];
1207: char fileregp[FILENAMELENGTH];
1208: char popfile[FILENAMELENGTH];
1209:
1210: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1211:
1.157 brouard 1212: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1213: /* struct timezone tzp; */
1214: /* extern int gettimeofday(); */
1215: struct tm tml, *gmtime(), *localtime();
1216:
1217: extern time_t time();
1218:
1219: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1220: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1221: struct tm tm;
1222:
1.126 brouard 1223: char strcurr[80], strfor[80];
1224:
1225: char *endptr;
1226: long lval;
1227: double dval;
1228:
1229: #define NR_END 1
1230: #define FREE_ARG char*
1231: #define FTOL 1.0e-10
1232:
1233: #define NRANSI
1.240 brouard 1234: #define ITMAX 200
1235: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1236:
1237: #define TOL 2.0e-4
1238:
1239: #define CGOLD 0.3819660
1240: #define ZEPS 1.0e-10
1241: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1242:
1243: #define GOLD 1.618034
1244: #define GLIMIT 100.0
1245: #define TINY 1.0e-20
1246:
1247: static double maxarg1,maxarg2;
1248: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1249: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1250:
1251: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1252: #define rint(a) floor(a+0.5)
1.166 brouard 1253: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1254: #define mytinydouble 1.0e-16
1.166 brouard 1255: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1256: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1257: /* static double dsqrarg; */
1258: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1259: static double sqrarg;
1260: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1261: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1262: int agegomp= AGEGOMP;
1263:
1264: int imx;
1265: int stepm=1;
1266: /* Stepm, step in month: minimum step interpolation*/
1267:
1268: int estepm;
1269: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1270:
1271: int m,nb;
1272: long *num;
1.197 brouard 1273: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1274: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1275: covariate for which somebody answered excluding
1276: undefined. Usually 2: 0 and 1. */
1277: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1278: covariate for which somebody answered including
1279: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1280: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1281: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1282: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1283: double *ageexmed,*agecens;
1284: double dateintmean=0;
1.296 brouard 1285: double anprojd, mprojd, jprojd; /* For eventual projections */
1286: double anprojf, mprojf, jprojf;
1.126 brouard 1287:
1.296 brouard 1288: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1289: double anbackf, mbackf, jbackf;
1290: double jintmean,mintmean,aintmean;
1.126 brouard 1291: double *weight;
1292: int **s; /* Status */
1.141 brouard 1293: double *agedc;
1.145 brouard 1294: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1295: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1296: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1297: double **coqvar; /* Fixed quantitative covariate nqv */
1298: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1299: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1300: double idx;
1301: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1302: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1303: /*k 1 2 3 4 5 6 7 8 9 */
1304: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1305: /* Tndvar[k] 1 2 3 4 5 */
1306: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1307: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1308: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1309: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1310: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1311: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1312: /* Tprod[i]=k 4 7 */
1313: /* Tage[i]=k 5 8 */
1314: /* */
1315: /* Type */
1316: /* V 1 2 3 4 5 */
1317: /* F F V V V */
1318: /* D Q D D Q */
1319: /* */
1320: int *TvarsD;
1321: int *TvarsDind;
1322: int *TvarsQ;
1323: int *TvarsQind;
1324:
1.235 brouard 1325: #define MAXRESULTLINES 10
1326: int nresult=0;
1.258 brouard 1327: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1328: int TKresult[MAXRESULTLINES];
1.237 brouard 1329: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1330: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1331: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1332: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1333: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1334: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1335:
1.234 brouard 1336: /* 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 1337: 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 */
1338: 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 */
1339: 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 */
1340: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1341: 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 */
1342: 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 1343: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1344: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1345: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1346: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1347: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1348: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1349: 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 */
1350: 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 */
1351:
1.230 brouard 1352: int *Tvarsel; /**< Selected covariates for output */
1353: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1354: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1355: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1356: 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 1357: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1358: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1359: int *Tage;
1.227 brouard 1360: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1361: 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 1362: 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*/
1363: 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 1364: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1365: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1366: int **Tvard;
1367: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1368: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1369: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1370: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1371: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1372: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1373: double *lsurv, *lpop, *tpop;
1374:
1.231 brouard 1375: #define FD 1; /* Fixed dummy covariate */
1376: #define FQ 2; /* Fixed quantitative covariate */
1377: #define FP 3; /* Fixed product covariate */
1378: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1379: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1380: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1381: #define VD 10; /* Varying dummy covariate */
1382: #define VQ 11; /* Varying quantitative covariate */
1383: #define VP 12; /* Varying product covariate */
1384: #define VPDD 13; /* Varying product dummy*dummy covariate */
1385: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1386: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1387: #define APFD 16; /* Age product * fixed dummy covariate */
1388: #define APFQ 17; /* Age product * fixed quantitative covariate */
1389: #define APVD 18; /* Age product * varying dummy covariate */
1390: #define APVQ 19; /* Age product * varying quantitative covariate */
1391:
1392: #define FTYPE 1; /* Fixed covariate */
1393: #define VTYPE 2; /* Varying covariate (loop in wave) */
1394: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1395:
1396: struct kmodel{
1397: int maintype; /* main type */
1398: int subtype; /* subtype */
1399: };
1400: struct kmodel modell[NCOVMAX];
1401:
1.143 brouard 1402: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1403: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1404:
1405: /**************** split *************************/
1406: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1407: {
1408: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1409: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1410: */
1411: char *ss; /* pointer */
1.186 brouard 1412: int l1=0, l2=0; /* length counters */
1.126 brouard 1413:
1414: l1 = strlen(path ); /* length of path */
1415: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1416: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1417: if ( ss == NULL ) { /* no directory, so determine current directory */
1418: strcpy( name, path ); /* we got the fullname name because no directory */
1419: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1420: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1421: /* get current working directory */
1422: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1423: #ifdef WIN32
1424: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1425: #else
1426: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1427: #endif
1.126 brouard 1428: return( GLOCK_ERROR_GETCWD );
1429: }
1430: /* got dirc from getcwd*/
1431: printf(" DIRC = %s \n",dirc);
1.205 brouard 1432: } else { /* strip directory from path */
1.126 brouard 1433: ss++; /* after this, the filename */
1434: l2 = strlen( ss ); /* length of filename */
1435: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1436: strcpy( name, ss ); /* save file name */
1437: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1438: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1439: printf(" DIRC2 = %s \n",dirc);
1440: }
1441: /* We add a separator at the end of dirc if not exists */
1442: l1 = strlen( dirc ); /* length of directory */
1443: if( dirc[l1-1] != DIRSEPARATOR ){
1444: dirc[l1] = DIRSEPARATOR;
1445: dirc[l1+1] = 0;
1446: printf(" DIRC3 = %s \n",dirc);
1447: }
1448: ss = strrchr( name, '.' ); /* find last / */
1449: if (ss >0){
1450: ss++;
1451: strcpy(ext,ss); /* save extension */
1452: l1= strlen( name);
1453: l2= strlen(ss)+1;
1454: strncpy( finame, name, l1-l2);
1455: finame[l1-l2]= 0;
1456: }
1457:
1458: return( 0 ); /* we're done */
1459: }
1460:
1461:
1462: /******************************************/
1463:
1464: void replace_back_to_slash(char *s, char*t)
1465: {
1466: int i;
1467: int lg=0;
1468: i=0;
1469: lg=strlen(t);
1470: for(i=0; i<= lg; i++) {
1471: (s[i] = t[i]);
1472: if (t[i]== '\\') s[i]='/';
1473: }
1474: }
1475:
1.132 brouard 1476: char *trimbb(char *out, char *in)
1.137 brouard 1477: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1478: char *s;
1479: s=out;
1480: while (*in != '\0'){
1.137 brouard 1481: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1482: in++;
1483: }
1484: *out++ = *in++;
1485: }
1486: *out='\0';
1487: return s;
1488: }
1489:
1.187 brouard 1490: /* char *substrchaine(char *out, char *in, char *chain) */
1491: /* { */
1492: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1493: /* char *s, *t; */
1494: /* t=in;s=out; */
1495: /* while ((*in != *chain) && (*in != '\0')){ */
1496: /* *out++ = *in++; */
1497: /* } */
1498:
1499: /* /\* *in matches *chain *\/ */
1500: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1501: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1502: /* } */
1503: /* in--; chain--; */
1504: /* while ( (*in != '\0')){ */
1505: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1506: /* *out++ = *in++; */
1507: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1508: /* } */
1509: /* *out='\0'; */
1510: /* out=s; */
1511: /* return out; */
1512: /* } */
1513: char *substrchaine(char *out, char *in, char *chain)
1514: {
1515: /* Substract chain 'chain' from 'in', return and output 'out' */
1516: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1517:
1518: char *strloc;
1519:
1520: strcpy (out, in);
1521: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1522: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1523: if(strloc != NULL){
1524: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1525: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1526: /* strcpy (strloc, strloc +strlen(chain));*/
1527: }
1528: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1529: return out;
1530: }
1531:
1532:
1.145 brouard 1533: char *cutl(char *blocc, char *alocc, char *in, char occ)
1534: {
1.187 brouard 1535: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1536: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1537: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1538: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1539: */
1.160 brouard 1540: char *s, *t;
1.145 brouard 1541: t=in;s=in;
1542: while ((*in != occ) && (*in != '\0')){
1543: *alocc++ = *in++;
1544: }
1545: if( *in == occ){
1546: *(alocc)='\0';
1547: s=++in;
1548: }
1549:
1550: if (s == t) {/* occ not found */
1551: *(alocc-(in-s))='\0';
1552: in=s;
1553: }
1554: while ( *in != '\0'){
1555: *blocc++ = *in++;
1556: }
1557:
1558: *blocc='\0';
1559: return t;
1560: }
1.137 brouard 1561: char *cutv(char *blocc, char *alocc, char *in, char occ)
1562: {
1.187 brouard 1563: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1564: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1565: gives blocc="abcdef2ghi" and alocc="j".
1566: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1567: */
1568: char *s, *t;
1569: t=in;s=in;
1570: while (*in != '\0'){
1571: while( *in == occ){
1572: *blocc++ = *in++;
1573: s=in;
1574: }
1575: *blocc++ = *in++;
1576: }
1577: if (s == t) /* occ not found */
1578: *(blocc-(in-s))='\0';
1579: else
1580: *(blocc-(in-s)-1)='\0';
1581: in=s;
1582: while ( *in != '\0'){
1583: *alocc++ = *in++;
1584: }
1585:
1586: *alocc='\0';
1587: return s;
1588: }
1589:
1.126 brouard 1590: int nbocc(char *s, char occ)
1591: {
1592: int i,j=0;
1593: int lg=20;
1594: i=0;
1595: lg=strlen(s);
1596: for(i=0; i<= lg; i++) {
1.234 brouard 1597: if (s[i] == occ ) j++;
1.126 brouard 1598: }
1599: return j;
1600: }
1601:
1.137 brouard 1602: /* void cutv(char *u,char *v, char*t, char occ) */
1603: /* { */
1604: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1605: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1606: /* gives u="abcdef2ghi" and v="j" *\/ */
1607: /* int i,lg,j,p=0; */
1608: /* i=0; */
1609: /* lg=strlen(t); */
1610: /* for(j=0; j<=lg-1; j++) { */
1611: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1612: /* } */
1.126 brouard 1613:
1.137 brouard 1614: /* for(j=0; j<p; j++) { */
1615: /* (u[j] = t[j]); */
1616: /* } */
1617: /* u[p]='\0'; */
1.126 brouard 1618:
1.137 brouard 1619: /* for(j=0; j<= lg; j++) { */
1620: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1621: /* } */
1622: /* } */
1.126 brouard 1623:
1.160 brouard 1624: #ifdef _WIN32
1625: char * strsep(char **pp, const char *delim)
1626: {
1627: char *p, *q;
1628:
1629: if ((p = *pp) == NULL)
1630: return 0;
1631: if ((q = strpbrk (p, delim)) != NULL)
1632: {
1633: *pp = q + 1;
1634: *q = '\0';
1635: }
1636: else
1637: *pp = 0;
1638: return p;
1639: }
1640: #endif
1641:
1.126 brouard 1642: /********************** nrerror ********************/
1643:
1644: void nrerror(char error_text[])
1645: {
1646: fprintf(stderr,"ERREUR ...\n");
1647: fprintf(stderr,"%s\n",error_text);
1648: exit(EXIT_FAILURE);
1649: }
1650: /*********************** vector *******************/
1651: double *vector(int nl, int nh)
1652: {
1653: double *v;
1654: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1655: if (!v) nrerror("allocation failure in vector");
1656: return v-nl+NR_END;
1657: }
1658:
1659: /************************ free vector ******************/
1660: void free_vector(double*v, int nl, int nh)
1661: {
1662: free((FREE_ARG)(v+nl-NR_END));
1663: }
1664:
1665: /************************ivector *******************************/
1666: int *ivector(long nl,long nh)
1667: {
1668: int *v;
1669: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1670: if (!v) nrerror("allocation failure in ivector");
1671: return v-nl+NR_END;
1672: }
1673:
1674: /******************free ivector **************************/
1675: void free_ivector(int *v, long nl, long nh)
1676: {
1677: free((FREE_ARG)(v+nl-NR_END));
1678: }
1679:
1680: /************************lvector *******************************/
1681: long *lvector(long nl,long nh)
1682: {
1683: long *v;
1684: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1685: if (!v) nrerror("allocation failure in ivector");
1686: return v-nl+NR_END;
1687: }
1688:
1689: /******************free lvector **************************/
1690: void free_lvector(long *v, long nl, long nh)
1691: {
1692: free((FREE_ARG)(v+nl-NR_END));
1693: }
1694:
1695: /******************* imatrix *******************************/
1696: int **imatrix(long nrl, long nrh, long ncl, long nch)
1697: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1698: {
1699: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1700: int **m;
1701:
1702: /* allocate pointers to rows */
1703: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1704: if (!m) nrerror("allocation failure 1 in matrix()");
1705: m += NR_END;
1706: m -= nrl;
1707:
1708:
1709: /* allocate rows and set pointers to them */
1710: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1711: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1712: m[nrl] += NR_END;
1713: m[nrl] -= ncl;
1714:
1715: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1716:
1717: /* return pointer to array of pointers to rows */
1718: return m;
1719: }
1720:
1721: /****************** free_imatrix *************************/
1722: void free_imatrix(m,nrl,nrh,ncl,nch)
1723: int **m;
1724: long nch,ncl,nrh,nrl;
1725: /* free an int matrix allocated by imatrix() */
1726: {
1727: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1728: free((FREE_ARG) (m+nrl-NR_END));
1729: }
1730:
1731: /******************* matrix *******************************/
1732: double **matrix(long nrl, long nrh, long ncl, long nch)
1733: {
1734: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1735: double **m;
1736:
1737: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1738: if (!m) nrerror("allocation failure 1 in matrix()");
1739: m += NR_END;
1740: m -= nrl;
1741:
1742: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1743: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1744: m[nrl] += NR_END;
1745: m[nrl] -= ncl;
1746:
1747: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1748: return m;
1.145 brouard 1749: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1750: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1751: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1752: */
1753: }
1754:
1755: /*************************free matrix ************************/
1756: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1757: {
1758: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1759: free((FREE_ARG)(m+nrl-NR_END));
1760: }
1761:
1762: /******************* ma3x *******************************/
1763: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1764: {
1765: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1766: double ***m;
1767:
1768: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1769: if (!m) nrerror("allocation failure 1 in matrix()");
1770: m += NR_END;
1771: m -= nrl;
1772:
1773: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1774: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1775: m[nrl] += NR_END;
1776: m[nrl] -= ncl;
1777:
1778: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1779:
1780: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1781: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1782: m[nrl][ncl] += NR_END;
1783: m[nrl][ncl] -= nll;
1784: for (j=ncl+1; j<=nch; j++)
1785: m[nrl][j]=m[nrl][j-1]+nlay;
1786:
1787: for (i=nrl+1; i<=nrh; i++) {
1788: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1789: for (j=ncl+1; j<=nch; j++)
1790: m[i][j]=m[i][j-1]+nlay;
1791: }
1792: return m;
1793: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1794: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1795: */
1796: }
1797:
1798: /*************************free ma3x ************************/
1799: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1800: {
1801: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1802: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1803: free((FREE_ARG)(m+nrl-NR_END));
1804: }
1805:
1806: /*************** function subdirf ***********/
1807: char *subdirf(char fileres[])
1808: {
1809: /* Caution optionfilefiname is hidden */
1810: strcpy(tmpout,optionfilefiname);
1811: strcat(tmpout,"/"); /* Add to the right */
1812: strcat(tmpout,fileres);
1813: return tmpout;
1814: }
1815:
1816: /*************** function subdirf2 ***********/
1817: char *subdirf2(char fileres[], char *preop)
1818: {
1819:
1820: /* Caution optionfilefiname is hidden */
1821: strcpy(tmpout,optionfilefiname);
1822: strcat(tmpout,"/");
1823: strcat(tmpout,preop);
1824: strcat(tmpout,fileres);
1825: return tmpout;
1826: }
1827:
1828: /*************** function subdirf3 ***********/
1829: char *subdirf3(char fileres[], char *preop, char *preop2)
1830: {
1831:
1832: /* Caution optionfilefiname is hidden */
1833: strcpy(tmpout,optionfilefiname);
1834: strcat(tmpout,"/");
1835: strcat(tmpout,preop);
1836: strcat(tmpout,preop2);
1837: strcat(tmpout,fileres);
1838: return tmpout;
1839: }
1.213 brouard 1840:
1841: /*************** function subdirfext ***********/
1842: char *subdirfext(char fileres[], char *preop, char *postop)
1843: {
1844:
1845: strcpy(tmpout,preop);
1846: strcat(tmpout,fileres);
1847: strcat(tmpout,postop);
1848: return tmpout;
1849: }
1.126 brouard 1850:
1.213 brouard 1851: /*************** function subdirfext3 ***********/
1852: char *subdirfext3(char fileres[], char *preop, char *postop)
1853: {
1854:
1855: /* Caution optionfilefiname is hidden */
1856: strcpy(tmpout,optionfilefiname);
1857: strcat(tmpout,"/");
1858: strcat(tmpout,preop);
1859: strcat(tmpout,fileres);
1860: strcat(tmpout,postop);
1861: return tmpout;
1862: }
1863:
1.162 brouard 1864: char *asc_diff_time(long time_sec, char ascdiff[])
1865: {
1866: long sec_left, days, hours, minutes;
1867: days = (time_sec) / (60*60*24);
1868: sec_left = (time_sec) % (60*60*24);
1869: hours = (sec_left) / (60*60) ;
1870: sec_left = (sec_left) %(60*60);
1871: minutes = (sec_left) /60;
1872: sec_left = (sec_left) % (60);
1873: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1874: return ascdiff;
1875: }
1876:
1.126 brouard 1877: /***************** f1dim *************************/
1878: extern int ncom;
1879: extern double *pcom,*xicom;
1880: extern double (*nrfunc)(double []);
1881:
1882: double f1dim(double x)
1883: {
1884: int j;
1885: double f;
1886: double *xt;
1887:
1888: xt=vector(1,ncom);
1889: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1890: f=(*nrfunc)(xt);
1891: free_vector(xt,1,ncom);
1892: return f;
1893: }
1894:
1895: /*****************brent *************************/
1896: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1897: {
1898: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1899: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1900: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1901: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1902: * returned function value.
1903: */
1.126 brouard 1904: int iter;
1905: double a,b,d,etemp;
1.159 brouard 1906: double fu=0,fv,fw,fx;
1.164 brouard 1907: double ftemp=0.;
1.126 brouard 1908: double p,q,r,tol1,tol2,u,v,w,x,xm;
1909: double e=0.0;
1910:
1911: a=(ax < cx ? ax : cx);
1912: b=(ax > cx ? ax : cx);
1913: x=w=v=bx;
1914: fw=fv=fx=(*f)(x);
1915: for (iter=1;iter<=ITMAX;iter++) {
1916: xm=0.5*(a+b);
1917: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1918: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1919: printf(".");fflush(stdout);
1920: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1921: #ifdef DEBUGBRENT
1.126 brouard 1922: 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);
1923: 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);
1924: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1925: #endif
1926: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1927: *xmin=x;
1928: return fx;
1929: }
1930: ftemp=fu;
1931: if (fabs(e) > tol1) {
1932: r=(x-w)*(fx-fv);
1933: q=(x-v)*(fx-fw);
1934: p=(x-v)*q-(x-w)*r;
1935: q=2.0*(q-r);
1936: if (q > 0.0) p = -p;
1937: q=fabs(q);
1938: etemp=e;
1939: e=d;
1940: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1941: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1942: else {
1.224 brouard 1943: d=p/q;
1944: u=x+d;
1945: if (u-a < tol2 || b-u < tol2)
1946: d=SIGN(tol1,xm-x);
1.126 brouard 1947: }
1948: } else {
1949: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1950: }
1951: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1952: fu=(*f)(u);
1953: if (fu <= fx) {
1954: if (u >= x) a=x; else b=x;
1955: SHFT(v,w,x,u)
1.183 brouard 1956: SHFT(fv,fw,fx,fu)
1957: } else {
1958: if (u < x) a=u; else b=u;
1959: if (fu <= fw || w == x) {
1.224 brouard 1960: v=w;
1961: w=u;
1962: fv=fw;
1963: fw=fu;
1.183 brouard 1964: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1965: v=u;
1966: fv=fu;
1.183 brouard 1967: }
1968: }
1.126 brouard 1969: }
1970: nrerror("Too many iterations in brent");
1971: *xmin=x;
1972: return fx;
1973: }
1974:
1975: /****************** mnbrak ***********************/
1976:
1977: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1978: double (*func)(double))
1.183 brouard 1979: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1980: the downhill direction (defined by the function as evaluated at the initial points) and returns
1981: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1982: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1983: */
1.126 brouard 1984: double ulim,u,r,q, dum;
1985: double fu;
1.187 brouard 1986:
1987: double scale=10.;
1988: int iterscale=0;
1989:
1990: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1991: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1992:
1993:
1994: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1995: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1996: /* *bx = *ax - (*ax - *bx)/scale; */
1997: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1998: /* } */
1999:
1.126 brouard 2000: if (*fb > *fa) {
2001: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2002: SHFT(dum,*fb,*fa,dum)
2003: }
1.126 brouard 2004: *cx=(*bx)+GOLD*(*bx-*ax);
2005: *fc=(*func)(*cx);
1.183 brouard 2006: #ifdef DEBUG
1.224 brouard 2007: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2008: 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 2009: #endif
1.224 brouard 2010: 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 2011: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2012: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2013: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2014: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2015: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2016: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2017: fu=(*func)(u);
1.163 brouard 2018: #ifdef DEBUG
2019: /* f(x)=A(x-u)**2+f(u) */
2020: double A, fparabu;
2021: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2022: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2023: 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);
2024: 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 2025: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2026: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2027: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2028: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2029: #endif
1.184 brouard 2030: #ifdef MNBRAKORIGINAL
1.183 brouard 2031: #else
1.191 brouard 2032: /* if (fu > *fc) { */
2033: /* #ifdef DEBUG */
2034: /* printf("mnbrak4 fu > fc \n"); */
2035: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2036: /* #endif */
2037: /* /\* 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 *\\/ *\/ */
2038: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2039: /* dum=u; /\* Shifting c and u *\/ */
2040: /* u = *cx; */
2041: /* *cx = dum; */
2042: /* dum = fu; */
2043: /* fu = *fc; */
2044: /* *fc =dum; */
2045: /* } else { /\* end *\/ */
2046: /* #ifdef DEBUG */
2047: /* printf("mnbrak3 fu < fc \n"); */
2048: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2049: /* #endif */
2050: /* dum=u; /\* Shifting c and u *\/ */
2051: /* u = *cx; */
2052: /* *cx = dum; */
2053: /* dum = fu; */
2054: /* fu = *fc; */
2055: /* *fc =dum; */
2056: /* } */
1.224 brouard 2057: #ifdef DEBUGMNBRAK
2058: double A, fparabu;
2059: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2060: fparabu= *fa - A*(*ax-u)*(*ax-u);
2061: 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);
2062: 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 2063: #endif
1.191 brouard 2064: dum=u; /* Shifting c and u */
2065: u = *cx;
2066: *cx = dum;
2067: dum = fu;
2068: fu = *fc;
2069: *fc =dum;
1.183 brouard 2070: #endif
1.162 brouard 2071: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2072: #ifdef DEBUG
1.224 brouard 2073: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2074: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2075: #endif
1.126 brouard 2076: fu=(*func)(u);
2077: if (fu < *fc) {
1.183 brouard 2078: #ifdef DEBUG
1.224 brouard 2079: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2080: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2081: #endif
2082: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2083: SHFT(*fb,*fc,fu,(*func)(u))
2084: #ifdef DEBUG
2085: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2086: #endif
2087: }
1.162 brouard 2088: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2089: #ifdef DEBUG
1.224 brouard 2090: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2091: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2092: #endif
1.126 brouard 2093: u=ulim;
2094: fu=(*func)(u);
1.183 brouard 2095: } else { /* u could be left to b (if r > q parabola has a maximum) */
2096: #ifdef DEBUG
1.224 brouard 2097: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2098: 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 2099: #endif
1.126 brouard 2100: u=(*cx)+GOLD*(*cx-*bx);
2101: fu=(*func)(u);
1.224 brouard 2102: #ifdef DEBUG
2103: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2104: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2105: #endif
1.183 brouard 2106: } /* end tests */
1.126 brouard 2107: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2108: SHFT(*fa,*fb,*fc,fu)
2109: #ifdef DEBUG
1.224 brouard 2110: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2111: 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 2112: #endif
2113: } /* 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 2114: }
2115:
2116: /*************** linmin ************************/
1.162 brouard 2117: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2118: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2119: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2120: the value of func at the returned location p . This is actually all accomplished by calling the
2121: routines mnbrak and brent .*/
1.126 brouard 2122: int ncom;
2123: double *pcom,*xicom;
2124: double (*nrfunc)(double []);
2125:
1.224 brouard 2126: #ifdef LINMINORIGINAL
1.126 brouard 2127: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2128: #else
2129: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2130: #endif
1.126 brouard 2131: {
2132: double brent(double ax, double bx, double cx,
2133: double (*f)(double), double tol, double *xmin);
2134: double f1dim(double x);
2135: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2136: double *fc, double (*func)(double));
2137: int j;
2138: double xx,xmin,bx,ax;
2139: double fx,fb,fa;
1.187 brouard 2140:
1.203 brouard 2141: #ifdef LINMINORIGINAL
2142: #else
2143: double scale=10., axs, xxs; /* Scale added for infinity */
2144: #endif
2145:
1.126 brouard 2146: ncom=n;
2147: pcom=vector(1,n);
2148: xicom=vector(1,n);
2149: nrfunc=func;
2150: for (j=1;j<=n;j++) {
2151: pcom[j]=p[j];
1.202 brouard 2152: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2153: }
1.187 brouard 2154:
1.203 brouard 2155: #ifdef LINMINORIGINAL
2156: xx=1.;
2157: #else
2158: axs=0.0;
2159: xxs=1.;
2160: do{
2161: xx= xxs;
2162: #endif
1.187 brouard 2163: ax=0.;
2164: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2165: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2166: /* 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)) */
2167: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2168: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2169: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2170: /* 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 2171: #ifdef LINMINORIGINAL
2172: #else
2173: if (fx != fx){
1.224 brouard 2174: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2175: printf("|");
2176: fprintf(ficlog,"|");
1.203 brouard 2177: #ifdef DEBUGLINMIN
1.224 brouard 2178: 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 2179: #endif
2180: }
1.224 brouard 2181: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2182: #endif
2183:
1.191 brouard 2184: #ifdef DEBUGLINMIN
2185: 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 2186: 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 2187: #endif
1.224 brouard 2188: #ifdef LINMINORIGINAL
2189: #else
2190: if(fb == fx){ /* Flat function in the direction */
2191: xmin=xx;
2192: *flat=1;
2193: }else{
2194: *flat=0;
2195: #endif
2196: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2197: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2198: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2199: /* fmin = f(p[j] + xmin * xi[j]) */
2200: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2201: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2202: #ifdef DEBUG
1.224 brouard 2203: 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);
2204: 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);
2205: #endif
2206: #ifdef LINMINORIGINAL
2207: #else
2208: }
1.126 brouard 2209: #endif
1.191 brouard 2210: #ifdef DEBUGLINMIN
2211: printf("linmin end ");
1.202 brouard 2212: fprintf(ficlog,"linmin end ");
1.191 brouard 2213: #endif
1.126 brouard 2214: for (j=1;j<=n;j++) {
1.203 brouard 2215: #ifdef LINMINORIGINAL
2216: xi[j] *= xmin;
2217: #else
2218: #ifdef DEBUGLINMIN
2219: if(xxs <1.0)
2220: printf(" before xi[%d]=%12.8f", j,xi[j]);
2221: #endif
2222: 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) */
2223: #ifdef DEBUGLINMIN
2224: if(xxs <1.0)
2225: 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 );
2226: #endif
2227: #endif
1.187 brouard 2228: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2229: }
1.191 brouard 2230: #ifdef DEBUGLINMIN
1.203 brouard 2231: printf("\n");
1.191 brouard 2232: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2233: 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 2234: for (j=1;j<=n;j++) {
1.202 brouard 2235: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2236: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2237: if(j % ncovmodel == 0){
1.191 brouard 2238: printf("\n");
1.202 brouard 2239: fprintf(ficlog,"\n");
2240: }
1.191 brouard 2241: }
1.203 brouard 2242: #else
1.191 brouard 2243: #endif
1.126 brouard 2244: free_vector(xicom,1,n);
2245: free_vector(pcom,1,n);
2246: }
2247:
2248:
2249: /*************** powell ************************/
1.162 brouard 2250: /*
2251: Minimization of a function func of n variables. Input consists of an initial starting point
2252: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2253: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2254: such that failure to decrease by more than this amount on one iteration signals doneness. On
2255: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2256: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2257: */
1.224 brouard 2258: #ifdef LINMINORIGINAL
2259: #else
2260: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2261: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2262: #endif
1.126 brouard 2263: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2264: double (*func)(double []))
2265: {
1.224 brouard 2266: #ifdef LINMINORIGINAL
2267: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2268: double (*func)(double []));
1.224 brouard 2269: #else
1.241 brouard 2270: void linmin(double p[], double xi[], int n, double *fret,
2271: double (*func)(double []),int *flat);
1.224 brouard 2272: #endif
1.239 brouard 2273: int i,ibig,j,jk,k;
1.126 brouard 2274: double del,t,*pt,*ptt,*xit;
1.181 brouard 2275: double directest;
1.126 brouard 2276: double fp,fptt;
2277: double *xits;
2278: int niterf, itmp;
1.224 brouard 2279: #ifdef LINMINORIGINAL
2280: #else
2281:
2282: flatdir=ivector(1,n);
2283: for (j=1;j<=n;j++) flatdir[j]=0;
2284: #endif
1.126 brouard 2285:
2286: pt=vector(1,n);
2287: ptt=vector(1,n);
2288: xit=vector(1,n);
2289: xits=vector(1,n);
2290: *fret=(*func)(p);
2291: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2292: rcurr_time = time(NULL);
1.126 brouard 2293: for (*iter=1;;++(*iter)) {
1.187 brouard 2294: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2295: ibig=0;
2296: del=0.0;
1.157 brouard 2297: rlast_time=rcurr_time;
2298: /* (void) gettimeofday(&curr_time,&tzp); */
2299: rcurr_time = time(NULL);
2300: curr_time = *localtime(&rcurr_time);
2301: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2302: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2303: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2304: for (i=1;i<=n;i++) {
1.126 brouard 2305: fprintf(ficrespow," %.12lf", p[i]);
2306: }
1.239 brouard 2307: fprintf(ficrespow,"\n");fflush(ficrespow);
2308: printf("\n#model= 1 + age ");
2309: fprintf(ficlog,"\n#model= 1 + age ");
2310: if(nagesqr==1){
1.241 brouard 2311: printf(" + age*age ");
2312: fprintf(ficlog," + age*age ");
1.239 brouard 2313: }
2314: for(j=1;j <=ncovmodel-2;j++){
2315: if(Typevar[j]==0) {
2316: printf(" + V%d ",Tvar[j]);
2317: fprintf(ficlog," + V%d ",Tvar[j]);
2318: }else if(Typevar[j]==1) {
2319: printf(" + V%d*age ",Tvar[j]);
2320: fprintf(ficlog," + V%d*age ",Tvar[j]);
2321: }else if(Typevar[j]==2) {
2322: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2323: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2324: }
2325: }
1.126 brouard 2326: printf("\n");
1.239 brouard 2327: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2328: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2329: fprintf(ficlog,"\n");
1.239 brouard 2330: for(i=1,jk=1; i <=nlstate; i++){
2331: for(k=1; k <=(nlstate+ndeath); k++){
2332: if (k != i) {
2333: printf("%d%d ",i,k);
2334: fprintf(ficlog,"%d%d ",i,k);
2335: for(j=1; j <=ncovmodel; j++){
2336: printf("%12.7f ",p[jk]);
2337: fprintf(ficlog,"%12.7f ",p[jk]);
2338: jk++;
2339: }
2340: printf("\n");
2341: fprintf(ficlog,"\n");
2342: }
2343: }
2344: }
1.241 brouard 2345: if(*iter <=3 && *iter >1){
1.157 brouard 2346: tml = *localtime(&rcurr_time);
2347: strcpy(strcurr,asctime(&tml));
2348: rforecast_time=rcurr_time;
1.126 brouard 2349: itmp = strlen(strcurr);
2350: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2351: strcurr[itmp-1]='\0';
1.162 brouard 2352: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2353: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2354: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2355: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2356: forecast_time = *localtime(&rforecast_time);
2357: strcpy(strfor,asctime(&forecast_time));
2358: itmp = strlen(strfor);
2359: if(strfor[itmp-1]=='\n')
2360: strfor[itmp-1]='\0';
2361: 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);
2362: 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 2363: }
2364: }
1.187 brouard 2365: for (i=1;i<=n;i++) { /* For each direction i */
2366: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2367: fptt=(*fret);
2368: #ifdef DEBUG
1.203 brouard 2369: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2370: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2371: #endif
1.203 brouard 2372: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2373: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2374: #ifdef LINMINORIGINAL
1.188 brouard 2375: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2376: #else
2377: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2378: flatdir[i]=flat; /* Function is vanishing in that direction i */
2379: #endif
2380: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2381: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2382: /* because that direction will be replaced unless the gain del is small */
2383: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2384: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2385: /* with the new direction. */
2386: del=fabs(fptt-(*fret));
2387: ibig=i;
1.126 brouard 2388: }
2389: #ifdef DEBUG
2390: printf("%d %.12e",i,(*fret));
2391: fprintf(ficlog,"%d %.12e",i,(*fret));
2392: for (j=1;j<=n;j++) {
1.224 brouard 2393: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2394: printf(" x(%d)=%.12e",j,xit[j]);
2395: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2396: }
2397: for(j=1;j<=n;j++) {
1.225 brouard 2398: printf(" p(%d)=%.12e",j,p[j]);
2399: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2400: }
2401: printf("\n");
2402: fprintf(ficlog,"\n");
2403: #endif
1.187 brouard 2404: } /* end loop on each direction i */
2405: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2406: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2407: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2408: for(j=1;j<=n;j++) {
1.225 brouard 2409: if(flatdir[j] >0){
2410: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2411: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2412: }
2413: /* printf("\n"); */
2414: /* fprintf(ficlog,"\n"); */
2415: }
1.243 brouard 2416: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2417: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2418: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2419: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2420: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2421: /* decreased of more than 3.84 */
2422: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2423: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2424: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2425:
1.188 brouard 2426: /* Starting the program with initial values given by a former maximization will simply change */
2427: /* the scales of the directions and the directions, because the are reset to canonical directions */
2428: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2429: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2430: #ifdef DEBUG
2431: int k[2],l;
2432: k[0]=1;
2433: k[1]=-1;
2434: printf("Max: %.12e",(*func)(p));
2435: fprintf(ficlog,"Max: %.12e",(*func)(p));
2436: for (j=1;j<=n;j++) {
2437: printf(" %.12e",p[j]);
2438: fprintf(ficlog," %.12e",p[j]);
2439: }
2440: printf("\n");
2441: fprintf(ficlog,"\n");
2442: for(l=0;l<=1;l++) {
2443: for (j=1;j<=n;j++) {
2444: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2445: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2446: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2447: }
2448: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2449: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2450: }
2451: #endif
2452:
1.224 brouard 2453: #ifdef LINMINORIGINAL
2454: #else
2455: free_ivector(flatdir,1,n);
2456: #endif
1.126 brouard 2457: free_vector(xit,1,n);
2458: free_vector(xits,1,n);
2459: free_vector(ptt,1,n);
2460: free_vector(pt,1,n);
2461: return;
1.192 brouard 2462: } /* enough precision */
1.240 brouard 2463: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2464: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2465: ptt[j]=2.0*p[j]-pt[j];
2466: xit[j]=p[j]-pt[j];
2467: pt[j]=p[j];
2468: }
1.181 brouard 2469: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2470: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2471: if (*iter <=4) {
1.225 brouard 2472: #else
2473: #endif
1.224 brouard 2474: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2475: #else
1.161 brouard 2476: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2477: #endif
1.162 brouard 2478: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2479: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2480: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2481: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2482: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2483: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2484: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2485: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2486: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2487: /* Even if f3 <f1, directest can be negative and t >0 */
2488: /* mu² and del² are equal when f3=f1 */
2489: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2490: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2491: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2492: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2493: #ifdef NRCORIGINAL
2494: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2495: #else
2496: 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 2497: t= t- del*SQR(fp-fptt);
1.183 brouard 2498: #endif
1.202 brouard 2499: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2500: #ifdef DEBUG
1.181 brouard 2501: 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);
2502: 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 2503: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2504: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2505: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2506: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2507: 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);
2508: 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);
2509: #endif
1.183 brouard 2510: #ifdef POWELLORIGINAL
2511: if (t < 0.0) { /* Then we use it for new direction */
2512: #else
1.182 brouard 2513: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2514: 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 2515: 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 2516: 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 2517: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2518: }
1.181 brouard 2519: if (directest < 0.0) { /* Then we use it for new direction */
2520: #endif
1.191 brouard 2521: #ifdef DEBUGLINMIN
1.234 brouard 2522: printf("Before linmin in direction P%d-P0\n",n);
2523: for (j=1;j<=n;j++) {
2524: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2525: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2526: if(j % ncovmodel == 0){
2527: printf("\n");
2528: fprintf(ficlog,"\n");
2529: }
2530: }
1.224 brouard 2531: #endif
2532: #ifdef LINMINORIGINAL
1.234 brouard 2533: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2534: #else
1.234 brouard 2535: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2536: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2537: #endif
1.234 brouard 2538:
1.191 brouard 2539: #ifdef DEBUGLINMIN
1.234 brouard 2540: for (j=1;j<=n;j++) {
2541: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2542: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2543: if(j % ncovmodel == 0){
2544: printf("\n");
2545: fprintf(ficlog,"\n");
2546: }
2547: }
1.224 brouard 2548: #endif
1.234 brouard 2549: for (j=1;j<=n;j++) {
2550: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2551: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2552: }
1.224 brouard 2553: #ifdef LINMINORIGINAL
2554: #else
1.234 brouard 2555: for (j=1, flatd=0;j<=n;j++) {
2556: if(flatdir[j]>0)
2557: flatd++;
2558: }
2559: if(flatd >0){
1.255 brouard 2560: printf("%d flat directions: ",flatd);
2561: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2562: for (j=1;j<=n;j++) {
2563: if(flatdir[j]>0){
2564: printf("%d ",j);
2565: fprintf(ficlog,"%d ",j);
2566: }
2567: }
2568: printf("\n");
2569: fprintf(ficlog,"\n");
2570: }
1.191 brouard 2571: #endif
1.234 brouard 2572: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2573: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2574:
1.126 brouard 2575: #ifdef DEBUG
1.234 brouard 2576: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2577: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2578: for(j=1;j<=n;j++){
2579: printf(" %lf",xit[j]);
2580: fprintf(ficlog," %lf",xit[j]);
2581: }
2582: printf("\n");
2583: fprintf(ficlog,"\n");
1.126 brouard 2584: #endif
1.192 brouard 2585: } /* end of t or directest negative */
1.224 brouard 2586: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2587: #else
1.234 brouard 2588: } /* end if (fptt < fp) */
1.192 brouard 2589: #endif
1.225 brouard 2590: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2591: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2592: #else
1.224 brouard 2593: #endif
1.234 brouard 2594: } /* loop iteration */
1.126 brouard 2595: }
1.234 brouard 2596:
1.126 brouard 2597: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2598:
1.235 brouard 2599: 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 2600: {
1.279 brouard 2601: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2602: * (and selected quantitative values in nres)
2603: * by left multiplying the unit
2604: * matrix by transitions matrix until convergence is reached with precision ftolpl
2605: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2606: * Wx is row vector: population in state 1, population in state 2, population dead
2607: * or prevalence in state 1, prevalence in state 2, 0
2608: * newm is the matrix after multiplications, its rows are identical at a factor.
2609: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2610: * Output is prlim.
2611: * Initial matrix pimij
2612: */
1.206 brouard 2613: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2614: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2615: /* 0, 0 , 1} */
2616: /*
2617: * and after some iteration: */
2618: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2619: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2620: /* 0, 0 , 1} */
2621: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2622: /* {0.51571254859325999, 0.4842874514067399, */
2623: /* 0.51326036147820708, 0.48673963852179264} */
2624: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2625:
1.126 brouard 2626: int i, ii,j,k;
1.209 brouard 2627: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2628: /* double **matprod2(); */ /* test */
1.218 brouard 2629: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2630: double **newm;
1.209 brouard 2631: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2632: int ncvloop=0;
1.288 brouard 2633: int first=0;
1.169 brouard 2634:
1.209 brouard 2635: min=vector(1,nlstate);
2636: max=vector(1,nlstate);
2637: meandiff=vector(1,nlstate);
2638:
1.218 brouard 2639: /* Starting with matrix unity */
1.126 brouard 2640: for (ii=1;ii<=nlstate+ndeath;ii++)
2641: for (j=1;j<=nlstate+ndeath;j++){
2642: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2643: }
1.169 brouard 2644:
2645: cov[1]=1.;
2646:
2647: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2648: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2649: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2650: ncvloop++;
1.126 brouard 2651: newm=savm;
2652: /* Covariates have to be included here again */
1.138 brouard 2653: cov[2]=agefin;
1.187 brouard 2654: if(nagesqr==1)
2655: cov[3]= agefin*agefin;;
1.234 brouard 2656: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2657: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2658: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2659: /* 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 2660: }
2661: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2662: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2663: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2664: /* 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 2665: }
1.237 brouard 2666: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2667: if(Dummy[Tvar[Tage[k]]]){
2668: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2669: } else{
1.235 brouard 2670: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2671: }
1.235 brouard 2672: /* 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 2673: }
1.237 brouard 2674: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2675: /* 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 2676: if(Dummy[Tvard[k][1]==0]){
2677: if(Dummy[Tvard[k][2]==0]){
2678: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2679: }else{
2680: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2681: }
2682: }else{
2683: if(Dummy[Tvard[k][2]==0]){
2684: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2685: }else{
2686: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2687: }
2688: }
1.234 brouard 2689: }
1.138 brouard 2690: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2691: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2692: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2693: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2694: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2695: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2696: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2697:
1.126 brouard 2698: savm=oldm;
2699: oldm=newm;
1.209 brouard 2700:
2701: for(j=1; j<=nlstate; j++){
2702: max[j]=0.;
2703: min[j]=1.;
2704: }
2705: for(i=1;i<=nlstate;i++){
2706: sumnew=0;
2707: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2708: for(j=1; j<=nlstate; j++){
2709: prlim[i][j]= newm[i][j]/(1-sumnew);
2710: max[j]=FMAX(max[j],prlim[i][j]);
2711: min[j]=FMIN(min[j],prlim[i][j]);
2712: }
2713: }
2714:
1.126 brouard 2715: maxmax=0.;
1.209 brouard 2716: for(j=1; j<=nlstate; j++){
2717: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2718: maxmax=FMAX(maxmax,meandiff[j]);
2719: /* 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 2720: } /* j loop */
1.203 brouard 2721: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2722: /* 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 2723: if(maxmax < ftolpl){
1.209 brouard 2724: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2725: free_vector(min,1,nlstate);
2726: free_vector(max,1,nlstate);
2727: free_vector(meandiff,1,nlstate);
1.126 brouard 2728: return prlim;
2729: }
1.288 brouard 2730: } /* agefin loop */
1.208 brouard 2731: /* After some age loop it doesn't converge */
1.288 brouard 2732: if(!first){
2733: first=1;
2734: 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);
2735: }
2736: 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);
2737:
1.209 brouard 2738: /* 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); */
2739: free_vector(min,1,nlstate);
2740: free_vector(max,1,nlstate);
2741: free_vector(meandiff,1,nlstate);
1.208 brouard 2742:
1.169 brouard 2743: return prlim; /* should not reach here */
1.126 brouard 2744: }
2745:
1.217 brouard 2746:
2747: /**** Back Prevalence limit (stable or period prevalence) ****************/
2748:
1.218 brouard 2749: /* 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) */
2750: /* 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 2751: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2752: {
1.264 brouard 2753: /* 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 2754: matrix by transitions matrix until convergence is reached with precision ftolpl */
2755: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2756: /* Wx is row vector: population in state 1, population in state 2, population dead */
2757: /* or prevalence in state 1, prevalence in state 2, 0 */
2758: /* newm is the matrix after multiplications, its rows are identical at a factor */
2759: /* Initial matrix pimij */
2760: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2761: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2762: /* 0, 0 , 1} */
2763: /*
2764: * and after some iteration: */
2765: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2766: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2767: /* 0, 0 , 1} */
2768: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2769: /* {0.51571254859325999, 0.4842874514067399, */
2770: /* 0.51326036147820708, 0.48673963852179264} */
2771: /* If we start from prlim again, prlim tends to a constant matrix */
2772:
2773: int i, ii,j,k;
1.247 brouard 2774: int first=0;
1.217 brouard 2775: double *min, *max, *meandiff, maxmax,sumnew=0.;
2776: /* double **matprod2(); */ /* test */
2777: double **out, cov[NCOVMAX+1], **bmij();
2778: double **newm;
1.218 brouard 2779: double **dnewm, **doldm, **dsavm; /* for use */
2780: double **oldm, **savm; /* for use */
2781:
1.217 brouard 2782: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2783: int ncvloop=0;
2784:
2785: min=vector(1,nlstate);
2786: max=vector(1,nlstate);
2787: meandiff=vector(1,nlstate);
2788:
1.266 brouard 2789: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2790: oldm=oldms; savm=savms;
2791:
2792: /* Starting with matrix unity */
2793: for (ii=1;ii<=nlstate+ndeath;ii++)
2794: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2795: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2796: }
2797:
2798: cov[1]=1.;
2799:
2800: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2801: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2802: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2803: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2804: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2805: ncvloop++;
1.218 brouard 2806: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2807: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2808: /* Covariates have to be included here again */
2809: cov[2]=agefin;
2810: if(nagesqr==1)
2811: cov[3]= agefin*agefin;;
1.242 brouard 2812: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2813: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2814: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2815: /* 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 2816: }
2817: /* for (k=1; k<=cptcovn;k++) { */
2818: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2819: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2820: /* /\* 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])]); *\/ */
2821: /* } */
2822: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2823: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2824: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2825: /* 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]); */
2826: }
2827: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2828: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2829: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2830: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2831: for (k=1; k<=cptcovage;k++){ /* For product with age */
2832: if(Dummy[Tvar[Tage[k]]]){
2833: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2834: } else{
2835: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2836: }
2837: /* 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]); */
2838: }
2839: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2840: /* 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]); */
2841: if(Dummy[Tvard[k][1]==0]){
2842: if(Dummy[Tvard[k][2]==0]){
2843: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2844: }else{
2845: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2846: }
2847: }else{
2848: if(Dummy[Tvard[k][2]==0]){
2849: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2850: }else{
2851: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2852: }
2853: }
1.217 brouard 2854: }
2855:
2856: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2857: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2858: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2859: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2860: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2861: /* ij should be linked to the correct index of cov */
2862: /* age and covariate values ij are in 'cov', but we need to pass
2863: * ij for the observed prevalence at age and status and covariate
2864: * number: prevacurrent[(int)agefin][ii][ij]
2865: */
2866: /* 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 *\/ */
2867: /* 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 *\/ */
2868: 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 2869: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2870: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2871: /* for(i=1; i<=nlstate+ndeath; i++) { */
2872: /* printf("%d newm= ",i); */
2873: /* for(j=1;j<=nlstate+ndeath;j++) { */
2874: /* printf("%f ",newm[i][j]); */
2875: /* } */
2876: /* printf("oldm * "); */
2877: /* for(j=1;j<=nlstate+ndeath;j++) { */
2878: /* printf("%f ",oldm[i][j]); */
2879: /* } */
1.268 brouard 2880: /* printf(" bmmij "); */
1.266 brouard 2881: /* for(j=1;j<=nlstate+ndeath;j++) { */
2882: /* printf("%f ",pmmij[i][j]); */
2883: /* } */
2884: /* printf("\n"); */
2885: /* } */
2886: /* } */
1.217 brouard 2887: savm=oldm;
2888: oldm=newm;
1.266 brouard 2889:
1.217 brouard 2890: for(j=1; j<=nlstate; j++){
2891: max[j]=0.;
2892: min[j]=1.;
2893: }
2894: for(j=1; j<=nlstate; j++){
2895: for(i=1;i<=nlstate;i++){
1.234 brouard 2896: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2897: bprlim[i][j]= newm[i][j];
2898: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2899: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2900: }
2901: }
1.218 brouard 2902:
1.217 brouard 2903: maxmax=0.;
2904: for(i=1; i<=nlstate; i++){
2905: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2906: maxmax=FMAX(maxmax,meandiff[i]);
2907: /* 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 2908: } /* i loop */
1.217 brouard 2909: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2910: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2911: if(maxmax < ftolpl){
1.220 brouard 2912: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2913: free_vector(min,1,nlstate);
2914: free_vector(max,1,nlstate);
2915: free_vector(meandiff,1,nlstate);
2916: return bprlim;
2917: }
1.288 brouard 2918: } /* agefin loop */
1.217 brouard 2919: /* After some age loop it doesn't converge */
1.288 brouard 2920: if(!first){
1.247 brouard 2921: first=1;
2922: 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\
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: }
2925: 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 2926: 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);
2927: /* 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); */
2928: free_vector(min,1,nlstate);
2929: free_vector(max,1,nlstate);
2930: free_vector(meandiff,1,nlstate);
2931:
2932: return bprlim; /* should not reach here */
2933: }
2934:
1.126 brouard 2935: /*************** transition probabilities ***************/
2936:
2937: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2938: {
1.138 brouard 2939: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2940: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2941: model to the ncovmodel covariates (including constant and age).
2942: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2943: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2944: ncth covariate in the global vector x is given by the formula:
2945: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2946: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2947: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2948: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2949: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2950: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2951: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2952: */
2953: double s1, lnpijopii;
1.126 brouard 2954: /*double t34;*/
1.164 brouard 2955: int i,j, nc, ii, jj;
1.126 brouard 2956:
1.223 brouard 2957: for(i=1; i<= nlstate; i++){
2958: for(j=1; j<i;j++){
2959: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2960: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2961: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2962: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2963: }
2964: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2965: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2966: }
2967: for(j=i+1; j<=nlstate+ndeath;j++){
2968: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2969: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2970: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2971: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2972: }
2973: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2974: }
2975: }
1.218 brouard 2976:
1.223 brouard 2977: for(i=1; i<= nlstate; i++){
2978: s1=0;
2979: for(j=1; j<i; j++){
2980: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2981: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2982: }
2983: for(j=i+1; j<=nlstate+ndeath; j++){
2984: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2985: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2986: }
2987: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2988: ps[i][i]=1./(s1+1.);
2989: /* Computing other pijs */
2990: for(j=1; j<i; j++)
2991: ps[i][j]= exp(ps[i][j])*ps[i][i];
2992: for(j=i+1; j<=nlstate+ndeath; j++)
2993: ps[i][j]= exp(ps[i][j])*ps[i][i];
2994: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2995: } /* end i */
1.218 brouard 2996:
1.223 brouard 2997: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2998: for(jj=1; jj<= nlstate+ndeath; jj++){
2999: ps[ii][jj]=0;
3000: ps[ii][ii]=1;
3001: }
3002: }
1.294 brouard 3003:
3004:
1.223 brouard 3005: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3006: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3007: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3008: /* } */
3009: /* printf("\n "); */
3010: /* } */
3011: /* printf("\n ");printf("%lf ",cov[2]);*/
3012: /*
3013: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3014: goto end;*/
1.266 brouard 3015: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3016: }
3017:
1.218 brouard 3018: /*************** backward transition probabilities ***************/
3019:
3020: /* 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 ) */
3021: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3022: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3023: {
1.266 brouard 3024: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
3025: * 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 3026: */
1.218 brouard 3027: int i, ii, j,k;
1.222 brouard 3028:
3029: double **out, **pmij();
3030: double sumnew=0.;
1.218 brouard 3031: double agefin;
1.292 brouard 3032: 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 3033: double **dnewm, **dsavm, **doldm;
3034: double **bbmij;
3035:
1.218 brouard 3036: doldm=ddoldms; /* global pointers */
1.222 brouard 3037: dnewm=ddnewms;
3038: dsavm=ddsavms;
3039:
3040: agefin=cov[2];
1.268 brouard 3041: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3042: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3043: the observed prevalence (with this covariate ij) at beginning of transition */
3044: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3045:
3046: /* P_x */
1.266 brouard 3047: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3048: /* outputs pmmij which is a stochastic matrix in row */
3049:
3050: /* Diag(w_x) */
1.292 brouard 3051: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3052: sumnew=0.;
1.269 brouard 3053: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3054: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3055: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3056: sumnew+=prevacurrent[(int)agefin][ii][ij];
3057: }
3058: if(sumnew >0.01){ /* At least some value in the prevalence */
3059: for (ii=1;ii<=nlstate+ndeath;ii++){
3060: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3061: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3062: }
3063: }else{
3064: for (ii=1;ii<=nlstate+ndeath;ii++){
3065: for (j=1;j<=nlstate+ndeath;j++)
3066: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3067: }
3068: /* if(sumnew <0.9){ */
3069: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3070: /* } */
3071: }
3072: k3=0.0; /* We put the last diagonal to 0 */
3073: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3074: doldm[ii][ii]= k3;
3075: }
3076: /* End doldm, At the end doldm is diag[(w_i)] */
3077:
1.292 brouard 3078: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3079: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3080:
1.292 brouard 3081: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3082: /* 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 3083: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3084: sumnew=0.;
1.222 brouard 3085: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3086: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3087: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3088: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3089: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3090: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3091: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3092: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3093: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3094: /* }else */
1.268 brouard 3095: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3096: } /*End ii */
3097: } /* 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 */
3098:
1.292 brouard 3099: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3100: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3101: /* end bmij */
1.266 brouard 3102: return ps; /*pointer is unchanged */
1.218 brouard 3103: }
1.217 brouard 3104: /*************** transition probabilities ***************/
3105:
1.218 brouard 3106: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3107: {
3108: /* According to parameters values stored in x and the covariate's values stored in cov,
3109: computes the probability to be observed in state j being in state i by appying the
3110: model to the ncovmodel covariates (including constant and age).
3111: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3112: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3113: ncth covariate in the global vector x is given by the formula:
3114: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3115: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3116: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3117: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3118: Outputs ps[i][j] the probability to be observed in j being in j according to
3119: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3120: */
3121: double s1, lnpijopii;
3122: /*double t34;*/
3123: int i,j, nc, ii, jj;
3124:
1.234 brouard 3125: for(i=1; i<= nlstate; i++){
3126: for(j=1; j<i;j++){
3127: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3128: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3129: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3130: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3131: }
3132: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3133: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3134: }
3135: for(j=i+1; j<=nlstate+ndeath;j++){
3136: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3137: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3138: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3139: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3140: }
3141: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3142: }
3143: }
3144:
3145: for(i=1; i<= nlstate; i++){
3146: s1=0;
3147: for(j=1; j<i; j++){
3148: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3149: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3150: }
3151: for(j=i+1; j<=nlstate+ndeath; j++){
3152: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3153: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3154: }
3155: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3156: ps[i][i]=1./(s1+1.);
3157: /* Computing other pijs */
3158: for(j=1; j<i; j++)
3159: ps[i][j]= exp(ps[i][j])*ps[i][i];
3160: for(j=i+1; j<=nlstate+ndeath; j++)
3161: ps[i][j]= exp(ps[i][j])*ps[i][i];
3162: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3163: } /* end i */
3164:
3165: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3166: for(jj=1; jj<= nlstate+ndeath; jj++){
3167: ps[ii][jj]=0;
3168: ps[ii][ii]=1;
3169: }
3170: }
1.296 brouard 3171: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3172: for(jj=1; jj<= nlstate+ndeath; jj++){
3173: s1=0.;
3174: for(ii=1; ii<= nlstate+ndeath; ii++){
3175: s1+=ps[ii][jj];
3176: }
3177: for(ii=1; ii<= nlstate; ii++){
3178: ps[ii][jj]=ps[ii][jj]/s1;
3179: }
3180: }
3181: /* Transposition */
3182: for(jj=1; jj<= nlstate+ndeath; jj++){
3183: for(ii=jj; ii<= nlstate+ndeath; ii++){
3184: s1=ps[ii][jj];
3185: ps[ii][jj]=ps[jj][ii];
3186: ps[jj][ii]=s1;
3187: }
3188: }
3189: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3190: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3191: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3192: /* } */
3193: /* printf("\n "); */
3194: /* } */
3195: /* printf("\n ");printf("%lf ",cov[2]);*/
3196: /*
3197: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3198: goto end;*/
3199: return ps;
1.217 brouard 3200: }
3201:
3202:
1.126 brouard 3203: /**************** Product of 2 matrices ******************/
3204:
1.145 brouard 3205: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3206: {
3207: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3208: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3209: /* in, b, out are matrice of pointers which should have been initialized
3210: before: only the contents of out is modified. The function returns
3211: a pointer to pointers identical to out */
1.145 brouard 3212: int i, j, k;
1.126 brouard 3213: for(i=nrl; i<= nrh; i++)
1.145 brouard 3214: for(k=ncolol; k<=ncoloh; k++){
3215: out[i][k]=0.;
3216: for(j=ncl; j<=nch; j++)
3217: out[i][k] +=in[i][j]*b[j][k];
3218: }
1.126 brouard 3219: return out;
3220: }
3221:
3222:
3223: /************* Higher Matrix Product ***************/
3224:
1.235 brouard 3225: 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 3226: {
1.218 brouard 3227: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3228: 'nhstepm*hstepm*stepm' months (i.e. until
3229: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3230: nhstepm*hstepm matrices.
3231: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3232: (typically every 2 years instead of every month which is too big
3233: for the memory).
3234: Model is determined by parameters x and covariates have to be
3235: included manually here.
3236:
3237: */
3238:
3239: int i, j, d, h, k;
1.131 brouard 3240: double **out, cov[NCOVMAX+1];
1.126 brouard 3241: double **newm;
1.187 brouard 3242: double agexact;
1.214 brouard 3243: double agebegin, ageend;
1.126 brouard 3244:
3245: /* Hstepm could be zero and should return the unit matrix */
3246: for (i=1;i<=nlstate+ndeath;i++)
3247: for (j=1;j<=nlstate+ndeath;j++){
3248: oldm[i][j]=(i==j ? 1.0 : 0.0);
3249: po[i][j][0]=(i==j ? 1.0 : 0.0);
3250: }
3251: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3252: for(h=1; h <=nhstepm; h++){
3253: for(d=1; d <=hstepm; d++){
3254: newm=savm;
3255: /* Covariates have to be included here again */
3256: cov[1]=1.;
1.214 brouard 3257: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3258: cov[2]=agexact;
3259: if(nagesqr==1)
1.227 brouard 3260: cov[3]= agexact*agexact;
1.235 brouard 3261: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3262: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3263: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3264: /* 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)); */
3265: }
3266: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3267: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3268: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3269: /* 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]); */
3270: }
3271: for (k=1; k<=cptcovage;k++){
3272: if(Dummy[Tvar[Tage[k]]]){
3273: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3274: } else{
3275: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3276: }
3277: /* 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]); */
3278: }
3279: for (k=1; k<=cptcovprod;k++){ /* */
3280: /* 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]); */
3281: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3282: }
3283: /* for (k=1; k<=cptcovn;k++) */
3284: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3285: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3286: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3287: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3288: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3289:
3290:
1.126 brouard 3291: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3292: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3293: /* right multiplication of oldm by the current matrix */
1.126 brouard 3294: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3295: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3296: /* if((int)age == 70){ */
3297: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3298: /* for(i=1; i<=nlstate+ndeath; i++) { */
3299: /* printf("%d pmmij ",i); */
3300: /* for(j=1;j<=nlstate+ndeath;j++) { */
3301: /* printf("%f ",pmmij[i][j]); */
3302: /* } */
3303: /* printf(" oldm "); */
3304: /* for(j=1;j<=nlstate+ndeath;j++) { */
3305: /* printf("%f ",oldm[i][j]); */
3306: /* } */
3307: /* printf("\n"); */
3308: /* } */
3309: /* } */
1.126 brouard 3310: savm=oldm;
3311: oldm=newm;
3312: }
3313: for(i=1; i<=nlstate+ndeath; i++)
3314: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3315: po[i][j][h]=newm[i][j];
3316: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3317: }
1.128 brouard 3318: /*printf("h=%d ",h);*/
1.126 brouard 3319: } /* end h */
1.267 brouard 3320: /* printf("\n H=%d \n",h); */
1.126 brouard 3321: return po;
3322: }
3323:
1.217 brouard 3324: /************* Higher Back Matrix Product ***************/
1.218 brouard 3325: /* 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 3326: 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 3327: {
1.266 brouard 3328: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3329: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3330: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3331: nhstepm*hstepm matrices.
3332: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3333: (typically every 2 years instead of every month which is too big
1.217 brouard 3334: for the memory).
1.218 brouard 3335: Model is determined by parameters x and covariates have to be
1.266 brouard 3336: included manually here. Then we use a call to bmij(x and cov)
3337: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3338: */
1.217 brouard 3339:
3340: int i, j, d, h, k;
1.266 brouard 3341: double **out, cov[NCOVMAX+1], **bmij();
3342: double **newm, ***newmm;
1.217 brouard 3343: double agexact;
3344: double agebegin, ageend;
1.222 brouard 3345: double **oldm, **savm;
1.217 brouard 3346:
1.266 brouard 3347: newmm=po; /* To be saved */
3348: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3349: /* Hstepm could be zero and should return the unit matrix */
3350: for (i=1;i<=nlstate+ndeath;i++)
3351: for (j=1;j<=nlstate+ndeath;j++){
3352: oldm[i][j]=(i==j ? 1.0 : 0.0);
3353: po[i][j][0]=(i==j ? 1.0 : 0.0);
3354: }
3355: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3356: for(h=1; h <=nhstepm; h++){
3357: for(d=1; d <=hstepm; d++){
3358: newm=savm;
3359: /* Covariates have to be included here again */
3360: cov[1]=1.;
1.271 brouard 3361: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3362: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3363: cov[2]=agexact;
3364: if(nagesqr==1)
1.222 brouard 3365: cov[3]= agexact*agexact;
1.266 brouard 3366: for (k=1; k<=cptcovn;k++){
3367: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3368: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3369: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3370: /* 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)); */
3371: }
1.267 brouard 3372: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3373: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3374: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3375: /* 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]); */
3376: }
3377: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3378: if(Dummy[Tvar[Tage[k]]]){
3379: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3380: } else{
3381: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3382: }
3383: /* 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]); */
3384: }
3385: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3386: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3387: }
1.217 brouard 3388: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3389: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3390:
1.218 brouard 3391: /* Careful transposed matrix */
1.266 brouard 3392: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3393: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3394: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3395: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3396: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3397: /* if((int)age == 70){ */
3398: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3399: /* for(i=1; i<=nlstate+ndeath; i++) { */
3400: /* printf("%d pmmij ",i); */
3401: /* for(j=1;j<=nlstate+ndeath;j++) { */
3402: /* printf("%f ",pmmij[i][j]); */
3403: /* } */
3404: /* printf(" oldm "); */
3405: /* for(j=1;j<=nlstate+ndeath;j++) { */
3406: /* printf("%f ",oldm[i][j]); */
3407: /* } */
3408: /* printf("\n"); */
3409: /* } */
3410: /* } */
3411: savm=oldm;
3412: oldm=newm;
3413: }
3414: for(i=1; i<=nlstate+ndeath; i++)
3415: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3416: po[i][j][h]=newm[i][j];
1.268 brouard 3417: /* if(h==nhstepm) */
3418: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3419: }
1.268 brouard 3420: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3421: } /* end h */
1.268 brouard 3422: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3423: return po;
3424: }
3425:
3426:
1.162 brouard 3427: #ifdef NLOPT
3428: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3429: double fret;
3430: double *xt;
3431: int j;
3432: myfunc_data *d2 = (myfunc_data *) pd;
3433: /* xt = (p1-1); */
3434: xt=vector(1,n);
3435: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3436:
3437: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3438: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3439: printf("Function = %.12lf ",fret);
3440: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3441: printf("\n");
3442: free_vector(xt,1,n);
3443: return fret;
3444: }
3445: #endif
1.126 brouard 3446:
3447: /*************** log-likelihood *************/
3448: double func( double *x)
3449: {
1.226 brouard 3450: int i, ii, j, k, mi, d, kk;
3451: int ioffset=0;
3452: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3453: double **out;
3454: double lli; /* Individual log likelihood */
3455: int s1, s2;
1.228 brouard 3456: 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 3457: double bbh, survp;
3458: long ipmx;
3459: double agexact;
3460: /*extern weight */
3461: /* We are differentiating ll according to initial status */
3462: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3463: /*for(i=1;i<imx;i++)
3464: printf(" %d\n",s[4][i]);
3465: */
1.162 brouard 3466:
1.226 brouard 3467: ++countcallfunc;
1.162 brouard 3468:
1.226 brouard 3469: cov[1]=1.;
1.126 brouard 3470:
1.226 brouard 3471: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3472: ioffset=0;
1.226 brouard 3473: if(mle==1){
3474: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3475: /* Computes the values of the ncovmodel covariates of the model
3476: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3477: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3478: to be observed in j being in i according to the model.
3479: */
1.243 brouard 3480: ioffset=2+nagesqr ;
1.233 brouard 3481: /* Fixed */
1.234 brouard 3482: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3483: 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)*/
3484: }
1.226 brouard 3485: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3486: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3487: has been calculated etc */
3488: /* For an individual i, wav[i] gives the number of effective waves */
3489: /* We compute the contribution to Likelihood of each effective transition
3490: mw[mi][i] is real wave of the mi th effectve wave */
3491: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3492: s2=s[mw[mi+1][i]][i];
3493: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3494: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3495: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3496: */
3497: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3498: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3499: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3500: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3501: }
3502: for (ii=1;ii<=nlstate+ndeath;ii++)
3503: for (j=1;j<=nlstate+ndeath;j++){
3504: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3505: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3506: }
3507: for(d=0; d<dh[mi][i]; d++){
3508: newm=savm;
3509: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3510: cov[2]=agexact;
3511: if(nagesqr==1)
3512: cov[3]= agexact*agexact; /* Should be changed here */
3513: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3514: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3515: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3516: else
3517: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3518: }
3519: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3520: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3521: savm=oldm;
3522: oldm=newm;
3523: } /* end mult */
3524:
3525: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3526: /* But now since version 0.9 we anticipate for bias at large stepm.
3527: * If stepm is larger than one month (smallest stepm) and if the exact delay
3528: * (in months) between two waves is not a multiple of stepm, we rounded to
3529: * the nearest (and in case of equal distance, to the lowest) interval but now
3530: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3531: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3532: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3533: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3534: * -stepm/2 to stepm/2 .
3535: * For stepm=1 the results are the same as for previous versions of Imach.
3536: * For stepm > 1 the results are less biased than in previous versions.
3537: */
1.234 brouard 3538: s1=s[mw[mi][i]][i];
3539: s2=s[mw[mi+1][i]][i];
3540: bbh=(double)bh[mi][i]/(double)stepm;
3541: /* bias bh is positive if real duration
3542: * is higher than the multiple of stepm and negative otherwise.
3543: */
3544: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3545: if( s2 > nlstate){
3546: /* i.e. if s2 is a death state and if the date of death is known
3547: then the contribution to the likelihood is the probability to
3548: die between last step unit time and current step unit time,
3549: which is also equal to probability to die before dh
3550: minus probability to die before dh-stepm .
3551: In version up to 0.92 likelihood was computed
3552: as if date of death was unknown. Death was treated as any other
3553: health state: the date of the interview describes the actual state
3554: and not the date of a change in health state. The former idea was
3555: to consider that at each interview the state was recorded
3556: (healthy, disable or death) and IMaCh was corrected; but when we
3557: introduced the exact date of death then we should have modified
3558: the contribution of an exact death to the likelihood. This new
3559: contribution is smaller and very dependent of the step unit
3560: stepm. It is no more the probability to die between last interview
3561: and month of death but the probability to survive from last
3562: interview up to one month before death multiplied by the
3563: probability to die within a month. Thanks to Chris
3564: Jackson for correcting this bug. Former versions increased
3565: mortality artificially. The bad side is that we add another loop
3566: which slows down the processing. The difference can be up to 10%
3567: lower mortality.
3568: */
3569: /* If, at the beginning of the maximization mostly, the
3570: cumulative probability or probability to be dead is
3571: constant (ie = 1) over time d, the difference is equal to
3572: 0. out[s1][3] = savm[s1][3]: probability, being at state
3573: s1 at precedent wave, to be dead a month before current
3574: wave is equal to probability, being at state s1 at
3575: precedent wave, to be dead at mont of the current
3576: wave. Then the observed probability (that this person died)
3577: is null according to current estimated parameter. In fact,
3578: it should be very low but not zero otherwise the log go to
3579: infinity.
3580: */
1.183 brouard 3581: /* #ifdef INFINITYORIGINAL */
3582: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3583: /* #else */
3584: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3585: /* lli=log(mytinydouble); */
3586: /* else */
3587: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3588: /* #endif */
1.226 brouard 3589: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3590:
1.226 brouard 3591: } else if ( s2==-1 ) { /* alive */
3592: for (j=1,survp=0. ; j<=nlstate; j++)
3593: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3594: /*survp += out[s1][j]; */
3595: lli= log(survp);
3596: }
3597: else if (s2==-4) {
3598: for (j=3,survp=0. ; j<=nlstate; j++)
3599: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3600: lli= log(survp);
3601: }
3602: else if (s2==-5) {
3603: for (j=1,survp=0. ; j<=2; j++)
3604: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3605: lli= log(survp);
3606: }
3607: else{
3608: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3609: /* 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 */
3610: }
3611: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3612: /*if(lli ==000.0)*/
3613: /*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); */
3614: ipmx +=1;
3615: sw += weight[i];
3616: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3617: /* if (lli < log(mytinydouble)){ */
3618: /* 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); */
3619: /* 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]); */
3620: /* } */
3621: } /* end of wave */
3622: } /* end of individual */
3623: } else if(mle==2){
3624: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3625: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3626: for(mi=1; mi<= wav[i]-1; mi++){
3627: for (ii=1;ii<=nlstate+ndeath;ii++)
3628: for (j=1;j<=nlstate+ndeath;j++){
3629: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3630: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3631: }
3632: for(d=0; d<=dh[mi][i]; d++){
3633: newm=savm;
3634: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3635: cov[2]=agexact;
3636: if(nagesqr==1)
3637: cov[3]= agexact*agexact;
3638: for (kk=1; kk<=cptcovage;kk++) {
3639: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3640: }
3641: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3642: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3643: savm=oldm;
3644: oldm=newm;
3645: } /* end mult */
3646:
3647: s1=s[mw[mi][i]][i];
3648: s2=s[mw[mi+1][i]][i];
3649: bbh=(double)bh[mi][i]/(double)stepm;
3650: 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 */
3651: ipmx +=1;
3652: sw += weight[i];
3653: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3654: } /* end of wave */
3655: } /* end of individual */
3656: } else if(mle==3){ /* exponential inter-extrapolation */
3657: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3658: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3659: for(mi=1; mi<= wav[i]-1; mi++){
3660: for (ii=1;ii<=nlstate+ndeath;ii++)
3661: for (j=1;j<=nlstate+ndeath;j++){
3662: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3663: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3664: }
3665: for(d=0; d<dh[mi][i]; d++){
3666: newm=savm;
3667: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3668: cov[2]=agexact;
3669: if(nagesqr==1)
3670: cov[3]= agexact*agexact;
3671: for (kk=1; kk<=cptcovage;kk++) {
3672: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3673: }
3674: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3675: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3676: savm=oldm;
3677: oldm=newm;
3678: } /* end mult */
3679:
3680: s1=s[mw[mi][i]][i];
3681: s2=s[mw[mi+1][i]][i];
3682: bbh=(double)bh[mi][i]/(double)stepm;
3683: 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 */
3684: ipmx +=1;
3685: sw += weight[i];
3686: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3687: } /* end of wave */
3688: } /* end of individual */
3689: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3690: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3691: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3692: for(mi=1; mi<= wav[i]-1; mi++){
3693: for (ii=1;ii<=nlstate+ndeath;ii++)
3694: for (j=1;j<=nlstate+ndeath;j++){
3695: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3696: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3697: }
3698: for(d=0; d<dh[mi][i]; d++){
3699: newm=savm;
3700: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3701: cov[2]=agexact;
3702: if(nagesqr==1)
3703: cov[3]= agexact*agexact;
3704: for (kk=1; kk<=cptcovage;kk++) {
3705: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3706: }
1.126 brouard 3707:
1.226 brouard 3708: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3709: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3710: savm=oldm;
3711: oldm=newm;
3712: } /* end mult */
3713:
3714: s1=s[mw[mi][i]][i];
3715: s2=s[mw[mi+1][i]][i];
3716: if( s2 > nlstate){
3717: lli=log(out[s1][s2] - savm[s1][s2]);
3718: } else if ( s2==-1 ) { /* alive */
3719: for (j=1,survp=0. ; j<=nlstate; j++)
3720: survp += out[s1][j];
3721: lli= log(survp);
3722: }else{
3723: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3724: }
3725: ipmx +=1;
3726: sw += weight[i];
3727: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3728: /* 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 3729: } /* end of wave */
3730: } /* end of individual */
3731: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3732: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3733: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3734: for(mi=1; mi<= wav[i]-1; mi++){
3735: for (ii=1;ii<=nlstate+ndeath;ii++)
3736: for (j=1;j<=nlstate+ndeath;j++){
3737: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3738: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3739: }
3740: for(d=0; d<dh[mi][i]; d++){
3741: newm=savm;
3742: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3743: cov[2]=agexact;
3744: if(nagesqr==1)
3745: cov[3]= agexact*agexact;
3746: for (kk=1; kk<=cptcovage;kk++) {
3747: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3748: }
1.126 brouard 3749:
1.226 brouard 3750: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3751: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3752: savm=oldm;
3753: oldm=newm;
3754: } /* end mult */
3755:
3756: s1=s[mw[mi][i]][i];
3757: s2=s[mw[mi+1][i]][i];
3758: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3759: ipmx +=1;
3760: sw += weight[i];
3761: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3762: /*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]);*/
3763: } /* end of wave */
3764: } /* end of individual */
3765: } /* End of if */
3766: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3767: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3768: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3769: return -l;
1.126 brouard 3770: }
3771:
3772: /*************** log-likelihood *************/
3773: double funcone( double *x)
3774: {
1.228 brouard 3775: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3776: int i, ii, j, k, mi, d, kk;
1.228 brouard 3777: int ioffset=0;
1.131 brouard 3778: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3779: double **out;
3780: double lli; /* Individual log likelihood */
3781: double llt;
3782: int s1, s2;
1.228 brouard 3783: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3784:
1.126 brouard 3785: double bbh, survp;
1.187 brouard 3786: double agexact;
1.214 brouard 3787: double agebegin, ageend;
1.126 brouard 3788: /*extern weight */
3789: /* We are differentiating ll according to initial status */
3790: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3791: /*for(i=1;i<imx;i++)
3792: printf(" %d\n",s[4][i]);
3793: */
3794: cov[1]=1.;
3795:
3796: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3797: ioffset=0;
3798: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3799: /* ioffset=2+nagesqr+cptcovage; */
3800: ioffset=2+nagesqr;
1.232 brouard 3801: /* Fixed */
1.224 brouard 3802: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3803: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3804: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3805: 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)*/
3806: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3807: /* cov[2+6]=covar[Tvar[6]][i]; */
3808: /* cov[2+6]=covar[2][i]; V2 */
3809: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3810: /* cov[2+7]=covar[Tvar[7]][i]; */
3811: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3812: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3813: /* cov[2+9]=covar[Tvar[9]][i]; */
3814: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3815: }
1.232 brouard 3816: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3817: /* 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?)*\/ */
3818: /* } */
1.231 brouard 3819: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3820: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3821: /* } */
1.225 brouard 3822:
1.233 brouard 3823:
3824: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3825: /* Wave varying (but not age varying) */
3826: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3827: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3828: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3829: }
1.232 brouard 3830: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3831: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3832: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3833: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3834: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3835: /* 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 3836: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3837: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3838: /* /\* 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]); *\/ */
3839: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3840: /* } */
1.126 brouard 3841: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3842: for (j=1;j<=nlstate+ndeath;j++){
3843: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3844: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3845: }
1.214 brouard 3846:
3847: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3848: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3849: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3850: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3851: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3852: and mw[mi+1][i]. dh depends on stepm.*/
3853: newm=savm;
1.247 brouard 3854: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3855: cov[2]=agexact;
3856: if(nagesqr==1)
3857: cov[3]= agexact*agexact;
3858: for (kk=1; kk<=cptcovage;kk++) {
3859: if(!FixedV[Tvar[Tage[kk]]])
3860: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3861: else
3862: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3863: }
3864: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3865: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3866: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3867: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3868: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3869: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3870: savm=oldm;
3871: oldm=newm;
1.126 brouard 3872: } /* end mult */
3873:
3874: s1=s[mw[mi][i]][i];
3875: s2=s[mw[mi+1][i]][i];
1.217 brouard 3876: /* if(s2==-1){ */
1.268 brouard 3877: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3878: /* /\* exit(1); *\/ */
3879: /* } */
1.126 brouard 3880: bbh=(double)bh[mi][i]/(double)stepm;
3881: /* bias is positive if real duration
3882: * is higher than the multiple of stepm and negative otherwise.
3883: */
3884: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3885: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3886: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3887: for (j=1,survp=0. ; j<=nlstate; j++)
3888: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3889: lli= log(survp);
1.126 brouard 3890: }else if (mle==1){
1.242 brouard 3891: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3892: } else if(mle==2){
1.242 brouard 3893: 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 3894: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3895: 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 3896: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3897: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3898: } else{ /* mle=0 back to 1 */
1.242 brouard 3899: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3900: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3901: } /* End of if */
3902: ipmx +=1;
3903: sw += weight[i];
3904: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3905: /*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 3906: if(globpr){
1.246 brouard 3907: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3908: %11.6f %11.6f %11.6f ", \
1.242 brouard 3909: 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 3910: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3911: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3912: llt +=ll[k]*gipmx/gsw;
3913: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3914: }
3915: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3916: }
1.232 brouard 3917: } /* end of wave */
3918: } /* end of individual */
3919: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3920: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3921: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3922: if(globpr==0){ /* First time we count the contributions and weights */
3923: gipmx=ipmx;
3924: gsw=sw;
3925: }
3926: return -l;
1.126 brouard 3927: }
3928:
3929:
3930: /*************** function likelione ***********/
1.292 brouard 3931: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 3932: {
3933: /* This routine should help understanding what is done with
3934: the selection of individuals/waves and
3935: to check the exact contribution to the likelihood.
3936: Plotting could be done.
3937: */
3938: int k;
3939:
3940: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3941: strcpy(fileresilk,"ILK_");
1.202 brouard 3942: strcat(fileresilk,fileresu);
1.126 brouard 3943: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3944: printf("Problem with resultfile: %s\n", fileresilk);
3945: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3946: }
1.214 brouard 3947: 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");
3948: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3949: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3950: for(k=1; k<=nlstate; k++)
3951: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3952: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3953: }
3954:
1.292 brouard 3955: *fretone=(*func)(p);
1.126 brouard 3956: if(*globpri !=0){
3957: fclose(ficresilk);
1.205 brouard 3958: if (mle ==0)
3959: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3960: else if(mle >=1)
3961: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3962: 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 3963: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3964:
3965: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3966: 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 3967: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3968: }
1.207 brouard 3969: 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 3970: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3971: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3972: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3973: fflush(fichtm);
1.205 brouard 3974: }
1.126 brouard 3975: return;
3976: }
3977:
3978:
3979: /*********** Maximum Likelihood Estimation ***************/
3980:
3981: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3982: {
1.165 brouard 3983: int i,j, iter=0;
1.126 brouard 3984: double **xi;
3985: double fret;
3986: double fretone; /* Only one call to likelihood */
3987: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3988:
3989: #ifdef NLOPT
3990: int creturn;
3991: nlopt_opt opt;
3992: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3993: double *lb;
3994: double minf; /* the minimum objective value, upon return */
3995: double * p1; /* Shifted parameters from 0 instead of 1 */
3996: myfunc_data dinst, *d = &dinst;
3997: #endif
3998:
3999:
1.126 brouard 4000: xi=matrix(1,npar,1,npar);
4001: for (i=1;i<=npar;i++)
4002: for (j=1;j<=npar;j++)
4003: xi[i][j]=(i==j ? 1.0 : 0.0);
4004: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4005: strcpy(filerespow,"POW_");
1.126 brouard 4006: strcat(filerespow,fileres);
4007: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4008: printf("Problem with resultfile: %s\n", filerespow);
4009: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4010: }
4011: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4012: for (i=1;i<=nlstate;i++)
4013: for(j=1;j<=nlstate+ndeath;j++)
4014: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4015: fprintf(ficrespow,"\n");
1.162 brouard 4016: #ifdef POWELL
1.126 brouard 4017: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4018: #endif
1.126 brouard 4019:
1.162 brouard 4020: #ifdef NLOPT
4021: #ifdef NEWUOA
4022: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4023: #else
4024: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4025: #endif
4026: lb=vector(0,npar-1);
4027: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4028: nlopt_set_lower_bounds(opt, lb);
4029: nlopt_set_initial_step1(opt, 0.1);
4030:
4031: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4032: d->function = func;
4033: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4034: nlopt_set_min_objective(opt, myfunc, d);
4035: nlopt_set_xtol_rel(opt, ftol);
4036: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4037: printf("nlopt failed! %d\n",creturn);
4038: }
4039: else {
4040: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4041: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4042: iter=1; /* not equal */
4043: }
4044: nlopt_destroy(opt);
4045: #endif
1.126 brouard 4046: free_matrix(xi,1,npar,1,npar);
4047: fclose(ficrespow);
1.203 brouard 4048: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4049: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4050: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4051:
4052: }
4053:
4054: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4055: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4056: {
4057: double **a,**y,*x,pd;
1.203 brouard 4058: /* double **hess; */
1.164 brouard 4059: int i, j;
1.126 brouard 4060: int *indx;
4061:
4062: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4063: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4064: void lubksb(double **a, int npar, int *indx, double b[]) ;
4065: void ludcmp(double **a, int npar, int *indx, double *d) ;
4066: double gompertz(double p[]);
1.203 brouard 4067: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4068:
4069: printf("\nCalculation of the hessian matrix. Wait...\n");
4070: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4071: for (i=1;i<=npar;i++){
1.203 brouard 4072: printf("%d-",i);fflush(stdout);
4073: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4074:
4075: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4076:
4077: /* printf(" %f ",p[i]);
4078: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4079: }
4080:
4081: for (i=1;i<=npar;i++) {
4082: for (j=1;j<=npar;j++) {
4083: if (j>i) {
1.203 brouard 4084: printf(".%d-%d",i,j);fflush(stdout);
4085: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4086: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4087:
4088: hess[j][i]=hess[i][j];
4089: /*printf(" %lf ",hess[i][j]);*/
4090: }
4091: }
4092: }
4093: printf("\n");
4094: fprintf(ficlog,"\n");
4095:
4096: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4097: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4098:
4099: a=matrix(1,npar,1,npar);
4100: y=matrix(1,npar,1,npar);
4101: x=vector(1,npar);
4102: indx=ivector(1,npar);
4103: for (i=1;i<=npar;i++)
4104: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4105: ludcmp(a,npar,indx,&pd);
4106:
4107: for (j=1;j<=npar;j++) {
4108: for (i=1;i<=npar;i++) x[i]=0;
4109: x[j]=1;
4110: lubksb(a,npar,indx,x);
4111: for (i=1;i<=npar;i++){
4112: matcov[i][j]=x[i];
4113: }
4114: }
4115:
4116: printf("\n#Hessian matrix#\n");
4117: fprintf(ficlog,"\n#Hessian matrix#\n");
4118: for (i=1;i<=npar;i++) {
4119: for (j=1;j<=npar;j++) {
1.203 brouard 4120: printf("%.6e ",hess[i][j]);
4121: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4122: }
4123: printf("\n");
4124: fprintf(ficlog,"\n");
4125: }
4126:
1.203 brouard 4127: /* printf("\n#Covariance matrix#\n"); */
4128: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4129: /* for (i=1;i<=npar;i++) { */
4130: /* for (j=1;j<=npar;j++) { */
4131: /* printf("%.6e ",matcov[i][j]); */
4132: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4133: /* } */
4134: /* printf("\n"); */
4135: /* fprintf(ficlog,"\n"); */
4136: /* } */
4137:
1.126 brouard 4138: /* Recompute Inverse */
1.203 brouard 4139: /* for (i=1;i<=npar;i++) */
4140: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4141: /* ludcmp(a,npar,indx,&pd); */
4142:
4143: /* printf("\n#Hessian matrix recomputed#\n"); */
4144:
4145: /* for (j=1;j<=npar;j++) { */
4146: /* for (i=1;i<=npar;i++) x[i]=0; */
4147: /* x[j]=1; */
4148: /* lubksb(a,npar,indx,x); */
4149: /* for (i=1;i<=npar;i++){ */
4150: /* y[i][j]=x[i]; */
4151: /* printf("%.3e ",y[i][j]); */
4152: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4153: /* } */
4154: /* printf("\n"); */
4155: /* fprintf(ficlog,"\n"); */
4156: /* } */
4157:
4158: /* Verifying the inverse matrix */
4159: #ifdef DEBUGHESS
4160: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4161:
1.203 brouard 4162: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4163: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4164:
4165: for (j=1;j<=npar;j++) {
4166: for (i=1;i<=npar;i++){
1.203 brouard 4167: printf("%.2f ",y[i][j]);
4168: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4169: }
4170: printf("\n");
4171: fprintf(ficlog,"\n");
4172: }
1.203 brouard 4173: #endif
1.126 brouard 4174:
4175: free_matrix(a,1,npar,1,npar);
4176: free_matrix(y,1,npar,1,npar);
4177: free_vector(x,1,npar);
4178: free_ivector(indx,1,npar);
1.203 brouard 4179: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4180:
4181:
4182: }
4183:
4184: /*************** hessian matrix ****************/
4185: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4186: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4187: int i;
4188: int l=1, lmax=20;
1.203 brouard 4189: double k1,k2, res, fx;
1.132 brouard 4190: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4191: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4192: int k=0,kmax=10;
4193: double l1;
4194:
4195: fx=func(x);
4196: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4197: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4198: l1=pow(10,l);
4199: delts=delt;
4200: for(k=1 ; k <kmax; k=k+1){
4201: delt = delta*(l1*k);
4202: p2[theta]=x[theta] +delt;
1.145 brouard 4203: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4204: p2[theta]=x[theta]-delt;
4205: k2=func(p2)-fx;
4206: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4207: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4208:
1.203 brouard 4209: #ifdef DEBUGHESSII
1.126 brouard 4210: 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);
4211: 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);
4212: #endif
4213: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4214: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4215: k=kmax;
4216: }
4217: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4218: k=kmax; l=lmax*10;
1.126 brouard 4219: }
4220: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4221: delts=delt;
4222: }
1.203 brouard 4223: } /* End loop k */
1.126 brouard 4224: }
4225: delti[theta]=delts;
4226: return res;
4227:
4228: }
4229:
1.203 brouard 4230: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4231: {
4232: int i;
1.164 brouard 4233: int l=1, lmax=20;
1.126 brouard 4234: double k1,k2,k3,k4,res,fx;
1.132 brouard 4235: double p2[MAXPARM+1];
1.203 brouard 4236: int k, kmax=1;
4237: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4238:
4239: int firstime=0;
1.203 brouard 4240:
1.126 brouard 4241: fx=func(x);
1.203 brouard 4242: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4243: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4244: p2[thetai]=x[thetai]+delti[thetai]*k;
4245: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4246: k1=func(p2)-fx;
4247:
1.203 brouard 4248: p2[thetai]=x[thetai]+delti[thetai]*k;
4249: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4250: k2=func(p2)-fx;
4251:
1.203 brouard 4252: p2[thetai]=x[thetai]-delti[thetai]*k;
4253: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4254: k3=func(p2)-fx;
4255:
1.203 brouard 4256: p2[thetai]=x[thetai]-delti[thetai]*k;
4257: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4258: k4=func(p2)-fx;
1.203 brouard 4259: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4260: if(k1*k2*k3*k4 <0.){
1.208 brouard 4261: firstime=1;
1.203 brouard 4262: kmax=kmax+10;
1.208 brouard 4263: }
4264: if(kmax >=10 || firstime ==1){
1.246 brouard 4265: 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);
4266: 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 4267: 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);
4268: 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);
4269: }
4270: #ifdef DEBUGHESSIJ
4271: v1=hess[thetai][thetai];
4272: v2=hess[thetaj][thetaj];
4273: cv12=res;
4274: /* Computing eigen value of Hessian matrix */
4275: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4276: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4277: if ((lc2 <0) || (lc1 <0) ){
4278: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4279: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4280: 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);
4281: 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);
4282: }
1.126 brouard 4283: #endif
4284: }
4285: return res;
4286: }
4287:
1.203 brouard 4288: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4289: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4290: /* { */
4291: /* int i; */
4292: /* int l=1, lmax=20; */
4293: /* double k1,k2,k3,k4,res,fx; */
4294: /* double p2[MAXPARM+1]; */
4295: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4296: /* int k=0,kmax=10; */
4297: /* double l1; */
4298:
4299: /* fx=func(x); */
4300: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4301: /* l1=pow(10,l); */
4302: /* delts=delt; */
4303: /* for(k=1 ; k <kmax; k=k+1){ */
4304: /* delt = delti*(l1*k); */
4305: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4306: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4307: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4308: /* k1=func(p2)-fx; */
4309:
4310: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4311: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4312: /* k2=func(p2)-fx; */
4313:
4314: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4315: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4316: /* k3=func(p2)-fx; */
4317:
4318: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4319: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4320: /* k4=func(p2)-fx; */
4321: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4322: /* #ifdef DEBUGHESSIJ */
4323: /* 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); */
4324: /* 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); */
4325: /* #endif */
4326: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4327: /* k=kmax; */
4328: /* } */
4329: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4330: /* k=kmax; l=lmax*10; */
4331: /* } */
4332: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4333: /* delts=delt; */
4334: /* } */
4335: /* } /\* End loop k *\/ */
4336: /* } */
4337: /* delti[theta]=delts; */
4338: /* return res; */
4339: /* } */
4340:
4341:
1.126 brouard 4342: /************** Inverse of matrix **************/
4343: void ludcmp(double **a, int n, int *indx, double *d)
4344: {
4345: int i,imax,j,k;
4346: double big,dum,sum,temp;
4347: double *vv;
4348:
4349: vv=vector(1,n);
4350: *d=1.0;
4351: for (i=1;i<=n;i++) {
4352: big=0.0;
4353: for (j=1;j<=n;j++)
4354: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4355: if (big == 0.0){
4356: printf(" Singular Hessian matrix at row %d:\n",i);
4357: for (j=1;j<=n;j++) {
4358: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4359: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4360: }
4361: fflush(ficlog);
4362: fclose(ficlog);
4363: nrerror("Singular matrix in routine ludcmp");
4364: }
1.126 brouard 4365: vv[i]=1.0/big;
4366: }
4367: for (j=1;j<=n;j++) {
4368: for (i=1;i<j;i++) {
4369: sum=a[i][j];
4370: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4371: a[i][j]=sum;
4372: }
4373: big=0.0;
4374: for (i=j;i<=n;i++) {
4375: sum=a[i][j];
4376: for (k=1;k<j;k++)
4377: sum -= a[i][k]*a[k][j];
4378: a[i][j]=sum;
4379: if ( (dum=vv[i]*fabs(sum)) >= big) {
4380: big=dum;
4381: imax=i;
4382: }
4383: }
4384: if (j != imax) {
4385: for (k=1;k<=n;k++) {
4386: dum=a[imax][k];
4387: a[imax][k]=a[j][k];
4388: a[j][k]=dum;
4389: }
4390: *d = -(*d);
4391: vv[imax]=vv[j];
4392: }
4393: indx[j]=imax;
4394: if (a[j][j] == 0.0) a[j][j]=TINY;
4395: if (j != n) {
4396: dum=1.0/(a[j][j]);
4397: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4398: }
4399: }
4400: free_vector(vv,1,n); /* Doesn't work */
4401: ;
4402: }
4403:
4404: void lubksb(double **a, int n, int *indx, double b[])
4405: {
4406: int i,ii=0,ip,j;
4407: double sum;
4408:
4409: for (i=1;i<=n;i++) {
4410: ip=indx[i];
4411: sum=b[ip];
4412: b[ip]=b[i];
4413: if (ii)
4414: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4415: else if (sum) ii=i;
4416: b[i]=sum;
4417: }
4418: for (i=n;i>=1;i--) {
4419: sum=b[i];
4420: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4421: b[i]=sum/a[i][i];
4422: }
4423: }
4424:
4425: void pstamp(FILE *fichier)
4426: {
1.196 brouard 4427: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4428: }
4429:
1.297 brouard 4430: void date2dmy(double date,double *day, double *month, double *year){
4431: double yp=0., yp1=0., yp2=0.;
4432:
4433: yp1=modf(date,&yp);/* extracts integral of date in yp and
4434: fractional in yp1 */
4435: *year=yp;
4436: yp2=modf((yp1*12),&yp);
4437: *month=yp;
4438: yp1=modf((yp2*30.5),&yp);
4439: *day=yp;
4440: if(*day==0) *day=1;
4441: if(*month==0) *month=1;
4442: }
4443:
1.253 brouard 4444:
4445:
1.126 brouard 4446: /************ Frequencies ********************/
1.251 brouard 4447: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4448: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4449: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4450: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4451:
1.265 brouard 4452: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4453: int iind=0, iage=0;
4454: int mi; /* Effective wave */
4455: int first;
4456: double ***freq; /* Frequencies */
1.268 brouard 4457: 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 */
4458: 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 4459: double *meanq, *stdq, *idq;
1.226 brouard 4460: double **meanqt;
4461: double *pp, **prop, *posprop, *pospropt;
4462: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4463: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4464: double agebegin, ageend;
4465:
4466: pp=vector(1,nlstate);
1.251 brouard 4467: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4468: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4469: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4470: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4471: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4472: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4473: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4474: meanqt=matrix(1,lastpass,1,nqtveff);
4475: strcpy(fileresp,"P_");
4476: strcat(fileresp,fileresu);
4477: /*strcat(fileresphtm,fileresu);*/
4478: if((ficresp=fopen(fileresp,"w"))==NULL) {
4479: printf("Problem with prevalence resultfile: %s\n", fileresp);
4480: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4481: exit(0);
4482: }
1.240 brouard 4483:
1.226 brouard 4484: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4485: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4486: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4487: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4488: fflush(ficlog);
4489: exit(70);
4490: }
4491: else{
4492: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4493: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4494: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4495: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4496: }
1.237 brouard 4497: 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 4498:
1.226 brouard 4499: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4500: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4501: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4502: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4503: fflush(ficlog);
4504: exit(70);
1.240 brouard 4505: } else{
1.226 brouard 4506: 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 4507: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4508: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4509: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4510: }
1.240 brouard 4511: 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);
4512:
1.253 brouard 4513: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4514: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4515: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4516: j1=0;
1.126 brouard 4517:
1.227 brouard 4518: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4519: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4520: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4521:
4522:
1.226 brouard 4523: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4524: reference=low_education V1=0,V2=0
4525: med_educ V1=1 V2=0,
4526: high_educ V1=0 V2=1
4527: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4528: */
1.249 brouard 4529: dateintsum=0;
4530: k2cpt=0;
4531:
1.253 brouard 4532: if(cptcoveff == 0 )
1.265 brouard 4533: nl=1; /* Constant and age model only */
1.253 brouard 4534: else
4535: nl=2;
1.265 brouard 4536:
4537: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4538: /* Loop on nj=1 or 2 if dummy covariates j!=0
4539: * Loop on j1(1 to 2**cptcoveff) covariate combination
4540: * freq[s1][s2][iage] =0.
4541: * Loop on iind
4542: * ++freq[s1][s2][iage] weighted
4543: * end iind
4544: * if covariate and j!0
4545: * headers Variable on one line
4546: * endif cov j!=0
4547: * header of frequency table by age
4548: * Loop on age
4549: * pp[s1]+=freq[s1][s2][iage] weighted
4550: * pos+=freq[s1][s2][iage] weighted
4551: * Loop on s1 initial state
4552: * fprintf(ficresp
4553: * end s1
4554: * end age
4555: * if j!=0 computes starting values
4556: * end compute starting values
4557: * end j1
4558: * end nl
4559: */
1.253 brouard 4560: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4561: if(nj==1)
4562: j=0; /* First pass for the constant */
1.265 brouard 4563: else{
1.253 brouard 4564: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4565: }
1.251 brouard 4566: first=1;
1.265 brouard 4567: 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 4568: posproptt=0.;
4569: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4570: scanf("%d", i);*/
4571: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4572: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4573: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4574: freq[i][s2][m]=0;
1.251 brouard 4575:
4576: for (i=1; i<=nlstate; i++) {
1.240 brouard 4577: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4578: prop[i][m]=0;
4579: posprop[i]=0;
4580: pospropt[i]=0;
4581: }
1.283 brouard 4582: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4583: idq[z1]=0.;
4584: meanq[z1]=0.;
4585: stdq[z1]=0.;
1.283 brouard 4586: }
4587: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4588: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4589: /* meanqt[m][z1]=0.; */
4590: /* } */
4591: /* } */
1.251 brouard 4592: /* dateintsum=0; */
4593: /* k2cpt=0; */
4594:
1.265 brouard 4595: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4596: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4597: bool=1;
4598: if(j !=0){
4599: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4600: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4601: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4602: /* if(Tvaraff[z1] ==-20){ */
4603: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4604: /* }else if(Tvaraff[z1] ==-10){ */
4605: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4606: /* }else */
4607: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4608: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4609: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4610: /* 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",
4611: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4612: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4613: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4614: } /* Onlyf fixed */
4615: } /* end z1 */
4616: } /* cptcovn > 0 */
4617: } /* end any */
4618: }/* end j==0 */
1.265 brouard 4619: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4620: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4621: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4622: m=mw[mi][iind];
4623: if(j!=0){
4624: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4625: for (z1=1; z1<=cptcoveff; z1++) {
4626: if( Fixed[Tmodelind[z1]]==1){
4627: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4628: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4629: value is -1, we don't select. It differs from the
4630: constant and age model which counts them. */
4631: bool=0; /* not selected */
4632: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4633: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4634: bool=0;
4635: }
4636: }
4637: }
4638: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4639: } /* end j==0 */
4640: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4641: if(bool==1){ /*Selected */
1.251 brouard 4642: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4643: and mw[mi+1][iind]. dh depends on stepm. */
4644: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4645: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4646: if(m >=firstpass && m <=lastpass){
4647: k2=anint[m][iind]+(mint[m][iind]/12.);
4648: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4649: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4650: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4651: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4652: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4653: if (m<lastpass) {
4654: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4655: /* 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]); */
4656: if(s[m][iind]==-1)
4657: 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.));
4658: 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 4659: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean */
4660: idq[z1]=idq[z1]+weight[iind];
4661: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4662: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4663: }
1.251 brouard 4664: /* if((int)agev[m][iind] == 55) */
4665: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4666: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4667: 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 4668: }
1.251 brouard 4669: } /* end if between passes */
4670: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4671: dateintsum=dateintsum+k2; /* on all covariates ?*/
4672: k2cpt++;
4673: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4674: }
1.251 brouard 4675: }else{
4676: bool=1;
4677: }/* end bool 2 */
4678: } /* end m */
1.284 brouard 4679: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4680: /* idq[z1]=idq[z1]+weight[iind]; */
4681: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4682: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4683: /* } */
1.251 brouard 4684: } /* end bool */
4685: } /* end iind = 1 to imx */
4686: /* prop[s][age] is feeded for any initial and valid live state as well as
4687: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4688:
4689:
4690: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4691: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4692: pstamp(ficresp);
1.251 brouard 4693: if (cptcoveff>0 && j!=0){
1.265 brouard 4694: pstamp(ficresp);
1.251 brouard 4695: printf( "\n#********** Variable ");
4696: fprintf(ficresp, "\n#********** Variable ");
4697: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4698: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4699: fprintf(ficlog, "\n#********** Variable ");
4700: for (z1=1; z1<=cptcoveff; z1++){
4701: if(!FixedV[Tvaraff[z1]]){
4702: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4703: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4704: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4705: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4706: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4707: }else{
1.251 brouard 4708: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4709: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4710: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4711: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4712: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4713: }
4714: }
4715: printf( "**********\n#");
4716: fprintf(ficresp, "**********\n#");
4717: fprintf(ficresphtm, "**********</h3>\n");
4718: fprintf(ficresphtmfr, "**********</h3>\n");
4719: fprintf(ficlog, "**********\n");
4720: }
1.284 brouard 4721: /*
4722: Printing means of quantitative variables if any
4723: */
4724: for (z1=1; z1<= nqfveff; z1++) {
1.285 brouard 4725: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.0f individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.284 brouard 4726: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
4727: if(weightopt==1){
4728: printf(" Weighted mean and standard deviation of");
4729: fprintf(ficlog," Weighted mean and standard deviation of");
4730: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4731: }
1.285 brouard 4732: 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]));
4733: 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]));
4734: 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 4735: }
4736: /* for (z1=1; z1<= nqtveff; z1++) { */
4737: /* for(m=1;m<=lastpass;m++){ */
4738: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4739: /* } */
4740: /* } */
1.283 brouard 4741:
1.251 brouard 4742: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4743: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4744: fprintf(ficresp, " Age");
4745: 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 4746: for(i=1; i<=nlstate;i++) {
1.265 brouard 4747: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4748: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4749: }
1.265 brouard 4750: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4751: fprintf(ficresphtm, "\n");
4752:
4753: /* Header of frequency table by age */
4754: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4755: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4756: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4757: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4758: if(s2!=0 && m!=0)
4759: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4760: }
1.226 brouard 4761: }
1.251 brouard 4762: fprintf(ficresphtmfr, "\n");
4763:
4764: /* For each age */
4765: for(iage=iagemin; iage <= iagemax+3; iage++){
4766: fprintf(ficresphtm,"<tr>");
4767: if(iage==iagemax+1){
4768: fprintf(ficlog,"1");
4769: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4770: }else if(iage==iagemax+2){
4771: fprintf(ficlog,"0");
4772: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4773: }else if(iage==iagemax+3){
4774: fprintf(ficlog,"Total");
4775: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4776: }else{
1.240 brouard 4777: if(first==1){
1.251 brouard 4778: first=0;
4779: printf("See log file for details...\n");
4780: }
4781: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4782: fprintf(ficlog,"Age %d", iage);
4783: }
1.265 brouard 4784: for(s1=1; s1 <=nlstate ; s1++){
4785: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4786: pp[s1] += freq[s1][m][iage];
1.251 brouard 4787: }
1.265 brouard 4788: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4789: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4790: pos += freq[s1][m][iage];
4791: if(pp[s1]>=1.e-10){
1.251 brouard 4792: if(first==1){
1.265 brouard 4793: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4794: }
1.265 brouard 4795: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4796: }else{
4797: if(first==1)
1.265 brouard 4798: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4799: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4800: }
4801: }
4802:
1.265 brouard 4803: for(s1=1; s1 <=nlstate ; s1++){
4804: /* posprop[s1]=0; */
4805: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4806: pp[s1] += freq[s1][m][iage];
4807: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4808:
4809: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4810: pos += pp[s1]; /* pos is the total number of transitions until this age */
4811: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4812: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4813: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4814: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4815: }
4816:
4817: /* Writing ficresp */
4818: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4819: if( iage <= iagemax){
4820: fprintf(ficresp," %d",iage);
4821: }
4822: }else if( nj==2){
4823: if( iage <= iagemax){
4824: fprintf(ficresp," %d",iage);
4825: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4826: }
1.240 brouard 4827: }
1.265 brouard 4828: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4829: if(pos>=1.e-5){
1.251 brouard 4830: if(first==1)
1.265 brouard 4831: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4832: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4833: }else{
4834: if(first==1)
1.265 brouard 4835: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4836: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4837: }
4838: if( iage <= iagemax){
4839: if(pos>=1.e-5){
1.265 brouard 4840: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4841: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4842: }else if( nj==2){
4843: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4844: }
4845: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4846: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4847: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4848: } else{
4849: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4850: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4851: }
1.240 brouard 4852: }
1.265 brouard 4853: pospropt[s1] +=posprop[s1];
4854: } /* end loop s1 */
1.251 brouard 4855: /* pospropt=0.; */
1.265 brouard 4856: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4857: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4858: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4859: if(first==1){
1.265 brouard 4860: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4861: }
1.265 brouard 4862: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4863: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4864: }
1.265 brouard 4865: if(s1!=0 && m!=0)
4866: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4867: }
1.265 brouard 4868: } /* end loop s1 */
1.251 brouard 4869: posproptt=0.;
1.265 brouard 4870: for(s1=1; s1 <=nlstate; s1++){
4871: posproptt += pospropt[s1];
1.251 brouard 4872: }
4873: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4874: fprintf(ficresphtm,"</tr>\n");
4875: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4876: if(iage <= iagemax)
4877: fprintf(ficresp,"\n");
1.240 brouard 4878: }
1.251 brouard 4879: if(first==1)
4880: printf("Others in log...\n");
4881: fprintf(ficlog,"\n");
4882: } /* end loop age iage */
1.265 brouard 4883:
1.251 brouard 4884: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4885: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4886: if(posproptt < 1.e-5){
1.265 brouard 4887: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4888: }else{
1.265 brouard 4889: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4890: }
1.226 brouard 4891: }
1.251 brouard 4892: fprintf(ficresphtm,"</tr>\n");
4893: fprintf(ficresphtm,"</table>\n");
4894: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4895: if(posproptt < 1.e-5){
1.251 brouard 4896: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4897: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4898: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4899: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4900: invalidvarcomb[j1]=1;
1.226 brouard 4901: }else{
1.251 brouard 4902: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4903: invalidvarcomb[j1]=0;
1.226 brouard 4904: }
1.251 brouard 4905: fprintf(ficresphtmfr,"</table>\n");
4906: fprintf(ficlog,"\n");
4907: if(j!=0){
4908: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4909: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4910: for(k=1; k <=(nlstate+ndeath); k++){
4911: if (k != i) {
1.265 brouard 4912: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4913: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4914: if(j1==1){ /* All dummy covariates to zero */
4915: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4916: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4917: printf("%d%d ",i,k);
4918: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4919: 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]));
4920: 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]));
4921: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4922: }
1.253 brouard 4923: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4924: for(iage=iagemin; iage <= iagemax+3; iage++){
4925: x[iage]= (double)iage;
4926: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4927: /* 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 4928: }
1.268 brouard 4929: /* Some are not finite, but linreg will ignore these ages */
4930: no=0;
1.253 brouard 4931: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4932: pstart[s1]=b;
4933: pstart[s1-1]=a;
1.252 brouard 4934: }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 */
4935: 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]);
4936: 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 4937: 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 4938: printf("%d%d ",i,k);
4939: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4940: 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 4941: }else{ /* Other cases, like quantitative fixed or varying covariates */
4942: ;
4943: }
4944: /* printf("%12.7f )", param[i][jj][k]); */
4945: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4946: s1++;
1.251 brouard 4947: } /* end jj */
4948: } /* end k!= i */
4949: } /* end k */
1.265 brouard 4950: } /* end i, s1 */
1.251 brouard 4951: } /* end j !=0 */
4952: } /* end selected combination of covariate j1 */
4953: if(j==0){ /* We can estimate starting values from the occurences in each case */
4954: printf("#Freqsummary: Starting values for the constants:\n");
4955: fprintf(ficlog,"\n");
1.265 brouard 4956: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4957: for(k=1; k <=(nlstate+ndeath); k++){
4958: if (k != i) {
4959: printf("%d%d ",i,k);
4960: fprintf(ficlog,"%d%d ",i,k);
4961: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4962: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4963: if(jj==1){ /* Age has to be done */
1.265 brouard 4964: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4965: 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]));
4966: 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 4967: }
4968: /* printf("%12.7f )", param[i][jj][k]); */
4969: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4970: s1++;
1.250 brouard 4971: }
1.251 brouard 4972: printf("\n");
4973: fprintf(ficlog,"\n");
1.250 brouard 4974: }
4975: }
1.284 brouard 4976: } /* end of state i */
1.251 brouard 4977: printf("#Freqsummary\n");
4978: fprintf(ficlog,"\n");
1.265 brouard 4979: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4980: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4981: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
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]);
4984: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4985: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4986: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4987: /* } */
4988: }
1.265 brouard 4989: } /* end loop s1 */
1.251 brouard 4990:
4991: printf("\n");
4992: fprintf(ficlog,"\n");
4993: } /* end j=0 */
1.249 brouard 4994: } /* end j */
1.252 brouard 4995:
1.253 brouard 4996: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4997: for(i=1, jk=1; i <=nlstate; i++){
4998: for(j=1; j <=nlstate+ndeath; j++){
4999: if(j!=i){
5000: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5001: printf("%1d%1d",i,j);
5002: fprintf(ficparo,"%1d%1d",i,j);
5003: for(k=1; k<=ncovmodel;k++){
5004: /* printf(" %lf",param[i][j][k]); */
5005: /* fprintf(ficparo," %lf",param[i][j][k]); */
5006: p[jk]=pstart[jk];
5007: printf(" %f ",pstart[jk]);
5008: fprintf(ficparo," %f ",pstart[jk]);
5009: jk++;
5010: }
5011: printf("\n");
5012: fprintf(ficparo,"\n");
5013: }
5014: }
5015: }
5016: } /* end mle=-2 */
1.226 brouard 5017: dateintmean=dateintsum/k2cpt;
1.296 brouard 5018: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5019:
1.226 brouard 5020: fclose(ficresp);
5021: fclose(ficresphtm);
5022: fclose(ficresphtmfr);
1.283 brouard 5023: free_vector(idq,1,nqfveff);
1.226 brouard 5024: free_vector(meanq,1,nqfveff);
1.284 brouard 5025: free_vector(stdq,1,nqfveff);
1.226 brouard 5026: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5027: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5028: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5029: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5030: free_vector(pospropt,1,nlstate);
5031: free_vector(posprop,1,nlstate);
1.251 brouard 5032: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5033: free_vector(pp,1,nlstate);
5034: /* End of freqsummary */
5035: }
1.126 brouard 5036:
1.268 brouard 5037: /* Simple linear regression */
5038: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5039:
5040: /* y=a+bx regression */
5041: double sumx = 0.0; /* sum of x */
5042: double sumx2 = 0.0; /* sum of x**2 */
5043: double sumxy = 0.0; /* sum of x * y */
5044: double sumy = 0.0; /* sum of y */
5045: double sumy2 = 0.0; /* sum of y**2 */
5046: double sume2 = 0.0; /* sum of square or residuals */
5047: double yhat;
5048:
5049: double denom=0;
5050: int i;
5051: int ne=*no;
5052:
5053: for ( i=ifi, ne=0;i<=ila;i++) {
5054: if(!isfinite(x[i]) || !isfinite(y[i])){
5055: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5056: continue;
5057: }
5058: ne=ne+1;
5059: sumx += x[i];
5060: sumx2 += x[i]*x[i];
5061: sumxy += x[i] * y[i];
5062: sumy += y[i];
5063: sumy2 += y[i]*y[i];
5064: denom = (ne * sumx2 - sumx*sumx);
5065: /* 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); */
5066: }
5067:
5068: denom = (ne * sumx2 - sumx*sumx);
5069: if (denom == 0) {
5070: // vertical, slope m is infinity
5071: *b = INFINITY;
5072: *a = 0;
5073: if (r) *r = 0;
5074: return 1;
5075: }
5076:
5077: *b = (ne * sumxy - sumx * sumy) / denom;
5078: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5079: if (r!=NULL) {
5080: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5081: sqrt((sumx2 - sumx*sumx/ne) *
5082: (sumy2 - sumy*sumy/ne));
5083: }
5084: *no=ne;
5085: for ( i=ifi, ne=0;i<=ila;i++) {
5086: if(!isfinite(x[i]) || !isfinite(y[i])){
5087: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5088: continue;
5089: }
5090: ne=ne+1;
5091: yhat = y[i] - *a -*b* x[i];
5092: sume2 += yhat * yhat ;
5093:
5094: denom = (ne * sumx2 - sumx*sumx);
5095: /* 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); */
5096: }
5097: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5098: *sa= *sb * sqrt(sumx2/ne);
5099:
5100: return 0;
5101: }
5102:
1.126 brouard 5103: /************ Prevalence ********************/
1.227 brouard 5104: 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)
5105: {
5106: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5107: in each health status at the date of interview (if between dateprev1 and dateprev2).
5108: We still use firstpass and lastpass as another selection.
5109: */
1.126 brouard 5110:
1.227 brouard 5111: int i, m, jk, j1, bool, z1,j, iv;
5112: int mi; /* Effective wave */
5113: int iage;
5114: double agebegin, ageend;
5115:
5116: double **prop;
5117: double posprop;
5118: double y2; /* in fractional years */
5119: int iagemin, iagemax;
5120: int first; /** to stop verbosity which is redirected to log file */
5121:
5122: iagemin= (int) agemin;
5123: iagemax= (int) agemax;
5124: /*pp=vector(1,nlstate);*/
1.251 brouard 5125: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5126: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5127: j1=0;
1.222 brouard 5128:
1.227 brouard 5129: /*j=cptcoveff;*/
5130: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5131:
1.288 brouard 5132: first=0;
1.227 brouard 5133: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5134: for (i=1; i<=nlstate; i++)
1.251 brouard 5135: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5136: prop[i][iage]=0.0;
5137: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5138: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5139: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5140:
5141: for (i=1; i<=imx; i++) { /* Each individual */
5142: bool=1;
5143: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5144: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5145: m=mw[mi][i];
5146: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5147: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5148: for (z1=1; z1<=cptcoveff; z1++){
5149: if( Fixed[Tmodelind[z1]]==1){
5150: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5151: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5152: bool=0;
5153: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5154: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5155: bool=0;
5156: }
5157: }
5158: if(bool==1){ /* Otherwise we skip that wave/person */
5159: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5160: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5161: if(m >=firstpass && m <=lastpass){
5162: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5163: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5164: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5165: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5166: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5167: 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);
5168: exit(1);
5169: }
5170: if (s[m][i]>0 && s[m][i]<=nlstate) {
5171: /*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]]);*/
5172: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5173: prop[s[m][i]][iagemax+3] += weight[i];
5174: } /* end valid statuses */
5175: } /* end selection of dates */
5176: } /* end selection of waves */
5177: } /* end bool */
5178: } /* end wave */
5179: } /* end individual */
5180: for(i=iagemin; i <= iagemax+3; i++){
5181: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5182: posprop += prop[jk][i];
5183: }
5184:
5185: for(jk=1; jk <=nlstate ; jk++){
5186: if( i <= iagemax){
5187: if(posprop>=1.e-5){
5188: probs[i][jk][j1]= prop[jk][i]/posprop;
5189: } else{
1.288 brouard 5190: if(!first){
5191: first=1;
1.266 brouard 5192: 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]);
5193: }else{
1.288 brouard 5194: 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 5195: }
5196: }
5197: }
5198: }/* end jk */
5199: }/* end i */
1.222 brouard 5200: /*} *//* end i1 */
1.227 brouard 5201: } /* end j1 */
1.222 brouard 5202:
1.227 brouard 5203: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5204: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5205: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5206: } /* End of prevalence */
1.126 brouard 5207:
5208: /************* Waves Concatenation ***************/
5209:
5210: 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)
5211: {
1.298 brouard 5212: /* 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 5213: Death is a valid wave (if date is known).
5214: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5215: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5216: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5217: */
1.126 brouard 5218:
1.224 brouard 5219: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5220: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5221: double sum=0., jmean=0.;*/
1.224 brouard 5222: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5223: int j, k=0,jk, ju, jl;
5224: double sum=0.;
5225: first=0;
1.214 brouard 5226: firstwo=0;
1.217 brouard 5227: firsthree=0;
1.218 brouard 5228: firstfour=0;
1.164 brouard 5229: jmin=100000;
1.126 brouard 5230: jmax=-1;
5231: jmean=0.;
1.224 brouard 5232:
5233: /* Treating live states */
1.214 brouard 5234: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5235: mi=0; /* First valid wave */
1.227 brouard 5236: mli=0; /* Last valid wave */
1.126 brouard 5237: m=firstpass;
1.214 brouard 5238: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5239: 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 */
5240: mli=m-1;/* mw[++mi][i]=m-1; */
5241: }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 */
5242: mw[++mi][i]=m;
5243: mli=m;
1.224 brouard 5244: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5245: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5246: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5247: }
1.227 brouard 5248: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5249: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5250: break;
1.224 brouard 5251: #else
1.227 brouard 5252: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5253: if(firsthree == 0){
1.262 brouard 5254: 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 5255: firsthree=1;
5256: }
1.262 brouard 5257: 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 5258: mw[++mi][i]=m;
5259: mli=m;
5260: }
5261: if(s[m][i]==-2){ /* Vital status is really unknown */
5262: nbwarn++;
5263: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5264: 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);
5265: 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);
5266: }
5267: break;
5268: }
5269: break;
1.224 brouard 5270: #endif
1.227 brouard 5271: }/* End m >= lastpass */
1.126 brouard 5272: }/* end while */
1.224 brouard 5273:
1.227 brouard 5274: /* 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 5275: /* After last pass */
1.224 brouard 5276: /* Treating death states */
1.214 brouard 5277: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5278: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5279: /* } */
1.126 brouard 5280: mi++; /* Death is another wave */
5281: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5282: /* Only death is a correct wave */
1.126 brouard 5283: mw[mi][i]=m;
1.257 brouard 5284: } /* else not in a death state */
1.224 brouard 5285: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5286: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5287: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5288: 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 */
5289: nbwarn++;
5290: if(firstfiv==0){
5291: 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 );
5292: firstfiv=1;
5293: }else{
5294: 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 );
5295: }
5296: }else{ /* Death occured afer last wave potential bias */
5297: nberr++;
5298: if(firstwo==0){
1.257 brouard 5299: 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 5300: firstwo=1;
5301: }
1.257 brouard 5302: 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 5303: }
1.257 brouard 5304: }else{ /* if date of interview is unknown */
1.227 brouard 5305: /* death is known but not confirmed by death status at any wave */
5306: if(firstfour==0){
5307: 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 );
5308: firstfour=1;
5309: }
5310: 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 5311: }
1.224 brouard 5312: } /* end if date of death is known */
5313: #endif
5314: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5315: /* wav[i]=mw[mi][i]; */
1.126 brouard 5316: if(mi==0){
5317: nbwarn++;
5318: if(first==0){
1.227 brouard 5319: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5320: first=1;
1.126 brouard 5321: }
5322: if(first==1){
1.227 brouard 5323: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5324: }
5325: } /* end mi==0 */
5326: } /* End individuals */
1.214 brouard 5327: /* wav and mw are no more changed */
1.223 brouard 5328:
1.214 brouard 5329:
1.126 brouard 5330: for(i=1; i<=imx; i++){
5331: for(mi=1; mi<wav[i];mi++){
5332: if (stepm <=0)
1.227 brouard 5333: dh[mi][i]=1;
1.126 brouard 5334: else{
1.260 brouard 5335: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5336: if (agedc[i] < 2*AGESUP) {
5337: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5338: if(j==0) j=1; /* Survives at least one month after exam */
5339: else if(j<0){
5340: nberr++;
5341: 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]);
5342: j=1; /* Temporary Dangerous patch */
5343: 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);
5344: 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]);
5345: 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);
5346: }
5347: k=k+1;
5348: if (j >= jmax){
5349: jmax=j;
5350: ijmax=i;
5351: }
5352: if (j <= jmin){
5353: jmin=j;
5354: ijmin=i;
5355: }
5356: sum=sum+j;
5357: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5358: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5359: }
5360: }
5361: else{
5362: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5363: /* 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 5364:
1.227 brouard 5365: k=k+1;
5366: if (j >= jmax) {
5367: jmax=j;
5368: ijmax=i;
5369: }
5370: else if (j <= jmin){
5371: jmin=j;
5372: ijmin=i;
5373: }
5374: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5375: /*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]);*/
5376: if(j<0){
5377: nberr++;
5378: 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]);
5379: 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]);
5380: }
5381: sum=sum+j;
5382: }
5383: jk= j/stepm;
5384: jl= j -jk*stepm;
5385: ju= j -(jk+1)*stepm;
5386: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5387: if(jl==0){
5388: dh[mi][i]=jk;
5389: bh[mi][i]=0;
5390: }else{ /* We want a negative bias in order to only have interpolation ie
5391: * to avoid the price of an extra matrix product in likelihood */
5392: dh[mi][i]=jk+1;
5393: bh[mi][i]=ju;
5394: }
5395: }else{
5396: if(jl <= -ju){
5397: dh[mi][i]=jk;
5398: bh[mi][i]=jl; /* bias is positive if real duration
5399: * is higher than the multiple of stepm and negative otherwise.
5400: */
5401: }
5402: else{
5403: dh[mi][i]=jk+1;
5404: bh[mi][i]=ju;
5405: }
5406: if(dh[mi][i]==0){
5407: dh[mi][i]=1; /* At least one step */
5408: bh[mi][i]=ju; /* At least one step */
5409: /* 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);*/
5410: }
5411: } /* end if mle */
1.126 brouard 5412: }
5413: } /* end wave */
5414: }
5415: jmean=sum/k;
5416: 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 5417: 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 5418: }
1.126 brouard 5419:
5420: /*********** Tricode ****************************/
1.220 brouard 5421: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5422: {
5423: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5424: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5425: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5426: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5427: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5428: */
1.130 brouard 5429:
1.242 brouard 5430: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5431: int modmaxcovj=0; /* Modality max of covariates j */
5432: int cptcode=0; /* Modality max of covariates j */
5433: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5434:
5435:
1.242 brouard 5436: /* cptcoveff=0; */
5437: /* *cptcov=0; */
1.126 brouard 5438:
1.242 brouard 5439: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5440: for (k=1; k <= maxncov; k++)
5441: for(j=1; j<=2; j++)
5442: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5443:
1.242 brouard 5444: /* Loop on covariates without age and products and no quantitative variable */
5445: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5446: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5447: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5448: switch(Fixed[k]) {
5449: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5450: 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*/
5451: ij=(int)(covar[Tvar[k]][i]);
5452: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5453: * If product of Vn*Vm, still boolean *:
5454: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5455: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5456: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5457: modality of the nth covariate of individual i. */
5458: if (ij > modmaxcovj)
5459: modmaxcovj=ij;
5460: else if (ij < modmincovj)
5461: modmincovj=ij;
1.287 brouard 5462: if (ij <0 || ij >1 ){
5463: printf("Information, IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5464: fprintf(ficlog,"Information, currently IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5465: }
5466: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5467: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5468: exit(1);
5469: }else
5470: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5471: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5472: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5473: /* getting the maximum value of the modality of the covariate
5474: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5475: female ies 1, then modmaxcovj=1.
5476: */
5477: } /* end for loop on individuals i */
5478: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5479: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5480: cptcode=modmaxcovj;
5481: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5482: /*for (i=0; i<=cptcode; i++) {*/
5483: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5484: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5485: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5486: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5487: if( j != -1){
5488: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5489: covariate for which somebody answered excluding
5490: undefined. Usually 2: 0 and 1. */
5491: }
5492: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5493: covariate for which somebody answered including
5494: undefined. Usually 3: -1, 0 and 1. */
5495: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5496: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5497: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5498:
1.242 brouard 5499: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5500: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5501: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5502: /* modmincovj=3; modmaxcovj = 7; */
5503: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5504: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5505: /* defining two dummy variables: variables V1_1 and V1_2.*/
5506: /* nbcode[Tvar[j]][ij]=k; */
5507: /* nbcode[Tvar[j]][1]=0; */
5508: /* nbcode[Tvar[j]][2]=1; */
5509: /* nbcode[Tvar[j]][3]=2; */
5510: /* To be continued (not working yet). */
5511: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5512:
5513: /* 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*/
5514: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5515: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5516: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5517: /*, could be restored in the future */
5518: 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 5519: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5520: break;
5521: }
5522: ij++;
1.287 brouard 5523: 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 5524: cptcode = ij; /* New max modality for covar j */
5525: } /* end of loop on modality i=-1 to 1 or more */
5526: break;
5527: case 1: /* Testing on varying covariate, could be simple and
5528: * should look at waves or product of fixed *
5529: * varying. No time to test -1, assuming 0 and 1 only */
5530: ij=0;
5531: for(i=0; i<=1;i++){
5532: nbcode[Tvar[k]][++ij]=i;
5533: }
5534: break;
5535: default:
5536: break;
5537: } /* end switch */
5538: } /* end dummy test */
1.287 brouard 5539: } /* 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 5540:
5541: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5542: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5543: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5544: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5545: 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 */
5546: 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 */
5547: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5548: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5549:
5550: ij=0;
5551: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5552: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5553: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5554: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5555: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5556: /* If product not in single variable we don't print results */
5557: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5558: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5559: 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*/
5560: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5561: 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 */
5562: if(Fixed[k]!=0)
5563: anyvaryingduminmodel=1;
5564: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5565: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5566: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5567: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5568: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5569: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5570: }
5571: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5572: /* ij--; */
5573: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5574: *cptcov=ij; /*Number of total real effective covariates: effective
5575: * because they can be excluded from the model and real
5576: * if in the model but excluded because missing values, but how to get k from ij?*/
5577: for(j=ij+1; j<= cptcovt; j++){
5578: Tvaraff[j]=0;
5579: Tmodelind[j]=0;
5580: }
5581: for(j=ntveff+1; j<= cptcovt; j++){
5582: TmodelInvind[j]=0;
5583: }
5584: /* To be sorted */
5585: ;
5586: }
1.126 brouard 5587:
1.145 brouard 5588:
1.126 brouard 5589: /*********** Health Expectancies ****************/
5590:
1.235 brouard 5591: 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 5592:
5593: {
5594: /* Health expectancies, no variances */
1.164 brouard 5595: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5596: int nhstepma, nstepma; /* Decreasing with age */
5597: double age, agelim, hf;
5598: double ***p3mat;
5599: double eip;
5600:
1.238 brouard 5601: /* pstamp(ficreseij); */
1.126 brouard 5602: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5603: fprintf(ficreseij,"# Age");
5604: for(i=1; i<=nlstate;i++){
5605: for(j=1; j<=nlstate;j++){
5606: fprintf(ficreseij," e%1d%1d ",i,j);
5607: }
5608: fprintf(ficreseij," e%1d. ",i);
5609: }
5610: fprintf(ficreseij,"\n");
5611:
5612:
5613: if(estepm < stepm){
5614: printf ("Problem %d lower than %d\n",estepm, stepm);
5615: }
5616: else hstepm=estepm;
5617: /* We compute the life expectancy from trapezoids spaced every estepm months
5618: * This is mainly to measure the difference between two models: for example
5619: * if stepm=24 months pijx are given only every 2 years and by summing them
5620: * we are calculating an estimate of the Life Expectancy assuming a linear
5621: * progression in between and thus overestimating or underestimating according
5622: * to the curvature of the survival function. If, for the same date, we
5623: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5624: * to compare the new estimate of Life expectancy with the same linear
5625: * hypothesis. A more precise result, taking into account a more precise
5626: * curvature will be obtained if estepm is as small as stepm. */
5627:
5628: /* For example we decided to compute the life expectancy with the smallest unit */
5629: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5630: nhstepm is the number of hstepm from age to agelim
5631: nstepm is the number of stepm from age to agelin.
1.270 brouard 5632: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5633: and note for a fixed period like estepm months */
5634: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5635: survival function given by stepm (the optimization length). Unfortunately it
5636: means that if the survival funtion is printed only each two years of age and if
5637: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5638: results. So we changed our mind and took the option of the best precision.
5639: */
5640: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5641:
5642: agelim=AGESUP;
5643: /* If stepm=6 months */
5644: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5645: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5646:
5647: /* nhstepm age range expressed in number of stepm */
5648: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5649: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5650: /* if (stepm >= YEARM) hstepm=1;*/
5651: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5652: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5653:
5654: for (age=bage; age<=fage; age ++){
5655: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5656: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5657: /* if (stepm >= YEARM) hstepm=1;*/
5658: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5659:
5660: /* If stepm=6 months */
5661: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5662: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5663:
1.235 brouard 5664: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5665:
5666: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5667:
5668: printf("%d|",(int)age);fflush(stdout);
5669: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5670:
5671: /* Computing expectancies */
5672: for(i=1; i<=nlstate;i++)
5673: for(j=1; j<=nlstate;j++)
5674: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5675: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5676:
5677: /* 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]);*/
5678:
5679: }
5680:
5681: fprintf(ficreseij,"%3.0f",age );
5682: for(i=1; i<=nlstate;i++){
5683: eip=0;
5684: for(j=1; j<=nlstate;j++){
5685: eip +=eij[i][j][(int)age];
5686: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5687: }
5688: fprintf(ficreseij,"%9.4f", eip );
5689: }
5690: fprintf(ficreseij,"\n");
5691:
5692: }
5693: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5694: printf("\n");
5695: fprintf(ficlog,"\n");
5696:
5697: }
5698:
1.235 brouard 5699: 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 5700:
5701: {
5702: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5703: to initial status i, ei. .
1.126 brouard 5704: */
5705: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5706: int nhstepma, nstepma; /* Decreasing with age */
5707: double age, agelim, hf;
5708: double ***p3matp, ***p3matm, ***varhe;
5709: double **dnewm,**doldm;
5710: double *xp, *xm;
5711: double **gp, **gm;
5712: double ***gradg, ***trgradg;
5713: int theta;
5714:
5715: double eip, vip;
5716:
5717: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5718: xp=vector(1,npar);
5719: xm=vector(1,npar);
5720: dnewm=matrix(1,nlstate*nlstate,1,npar);
5721: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5722:
5723: pstamp(ficresstdeij);
5724: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5725: fprintf(ficresstdeij,"# Age");
5726: for(i=1; i<=nlstate;i++){
5727: for(j=1; j<=nlstate;j++)
5728: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5729: fprintf(ficresstdeij," e%1d. ",i);
5730: }
5731: fprintf(ficresstdeij,"\n");
5732:
5733: pstamp(ficrescveij);
5734: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5735: fprintf(ficrescveij,"# Age");
5736: for(i=1; i<=nlstate;i++)
5737: for(j=1; j<=nlstate;j++){
5738: cptj= (j-1)*nlstate+i;
5739: for(i2=1; i2<=nlstate;i2++)
5740: for(j2=1; j2<=nlstate;j2++){
5741: cptj2= (j2-1)*nlstate+i2;
5742: if(cptj2 <= cptj)
5743: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5744: }
5745: }
5746: fprintf(ficrescveij,"\n");
5747:
5748: if(estepm < stepm){
5749: printf ("Problem %d lower than %d\n",estepm, stepm);
5750: }
5751: else hstepm=estepm;
5752: /* We compute the life expectancy from trapezoids spaced every estepm months
5753: * This is mainly to measure the difference between two models: for example
5754: * if stepm=24 months pijx are given only every 2 years and by summing them
5755: * we are calculating an estimate of the Life Expectancy assuming a linear
5756: * progression in between and thus overestimating or underestimating according
5757: * to the curvature of the survival function. If, for the same date, we
5758: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5759: * to compare the new estimate of Life expectancy with the same linear
5760: * hypothesis. A more precise result, taking into account a more precise
5761: * curvature will be obtained if estepm is as small as stepm. */
5762:
5763: /* For example we decided to compute the life expectancy with the smallest unit */
5764: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5765: nhstepm is the number of hstepm from age to agelim
5766: nstepm is the number of stepm from age to agelin.
5767: Look at hpijx to understand the reason of that which relies in memory size
5768: and note for a fixed period like estepm months */
5769: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5770: survival function given by stepm (the optimization length). Unfortunately it
5771: means that if the survival funtion is printed only each two years of age and if
5772: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5773: results. So we changed our mind and took the option of the best precision.
5774: */
5775: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5776:
5777: /* If stepm=6 months */
5778: /* nhstepm age range expressed in number of stepm */
5779: agelim=AGESUP;
5780: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5781: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5782: /* if (stepm >= YEARM) hstepm=1;*/
5783: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5784:
5785: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5786: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5787: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5788: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5789: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5790: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5791:
5792: for (age=bage; age<=fage; age ++){
5793: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5794: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5795: /* if (stepm >= YEARM) hstepm=1;*/
5796: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5797:
1.126 brouard 5798: /* If stepm=6 months */
5799: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5800: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5801:
5802: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5803:
1.126 brouard 5804: /* Computing Variances of health expectancies */
5805: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5806: decrease memory allocation */
5807: for(theta=1; theta <=npar; theta++){
5808: for(i=1; i<=npar; i++){
1.222 brouard 5809: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5810: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5811: }
1.235 brouard 5812: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5813: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5814:
1.126 brouard 5815: for(j=1; j<= nlstate; j++){
1.222 brouard 5816: for(i=1; i<=nlstate; i++){
5817: for(h=0; h<=nhstepm-1; h++){
5818: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5819: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5820: }
5821: }
1.126 brouard 5822: }
1.218 brouard 5823:
1.126 brouard 5824: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5825: for(h=0; h<=nhstepm-1; h++){
5826: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5827: }
1.126 brouard 5828: }/* End theta */
5829:
5830:
5831: for(h=0; h<=nhstepm-1; h++)
5832: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5833: for(theta=1; theta <=npar; theta++)
5834: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5835:
1.218 brouard 5836:
1.222 brouard 5837: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5838: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5839: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5840:
1.222 brouard 5841: printf("%d|",(int)age);fflush(stdout);
5842: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5843: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5844: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5845: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5846: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5847: for(ij=1;ij<=nlstate*nlstate;ij++)
5848: for(ji=1;ji<=nlstate*nlstate;ji++)
5849: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5850: }
5851: }
1.218 brouard 5852:
1.126 brouard 5853: /* Computing expectancies */
1.235 brouard 5854: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5855: for(i=1; i<=nlstate;i++)
5856: for(j=1; j<=nlstate;j++)
1.222 brouard 5857: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5858: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5859:
1.222 brouard 5860: /* 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 5861:
1.222 brouard 5862: }
1.269 brouard 5863:
5864: /* Standard deviation of expectancies ij */
1.126 brouard 5865: fprintf(ficresstdeij,"%3.0f",age );
5866: for(i=1; i<=nlstate;i++){
5867: eip=0.;
5868: vip=0.;
5869: for(j=1; j<=nlstate;j++){
1.222 brouard 5870: eip += eij[i][j][(int)age];
5871: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5872: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5873: 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 5874: }
5875: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5876: }
5877: fprintf(ficresstdeij,"\n");
1.218 brouard 5878:
1.269 brouard 5879: /* Variance of expectancies ij */
1.126 brouard 5880: fprintf(ficrescveij,"%3.0f",age );
5881: for(i=1; i<=nlstate;i++)
5882: for(j=1; j<=nlstate;j++){
1.222 brouard 5883: cptj= (j-1)*nlstate+i;
5884: for(i2=1; i2<=nlstate;i2++)
5885: for(j2=1; j2<=nlstate;j2++){
5886: cptj2= (j2-1)*nlstate+i2;
5887: if(cptj2 <= cptj)
5888: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5889: }
1.126 brouard 5890: }
5891: fprintf(ficrescveij,"\n");
1.218 brouard 5892:
1.126 brouard 5893: }
5894: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5895: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5896: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5897: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5898: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5899: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5900: printf("\n");
5901: fprintf(ficlog,"\n");
1.218 brouard 5902:
1.126 brouard 5903: free_vector(xm,1,npar);
5904: free_vector(xp,1,npar);
5905: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5906: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5907: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5908: }
1.218 brouard 5909:
1.126 brouard 5910: /************ Variance ******************/
1.235 brouard 5911: 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 5912: {
1.279 brouard 5913: /** Variance of health expectancies
5914: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5915: * double **newm;
5916: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5917: */
1.218 brouard 5918:
5919: /* int movingaverage(); */
5920: double **dnewm,**doldm;
5921: double **dnewmp,**doldmp;
5922: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 5923: int first=0;
1.218 brouard 5924: int k;
5925: double *xp;
1.279 brouard 5926: double **gp, **gm; /**< for var eij */
5927: double ***gradg, ***trgradg; /**< for var eij */
5928: double **gradgp, **trgradgp; /**< for var p point j */
5929: double *gpp, *gmp; /**< for var p point j */
5930: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5931: double ***p3mat;
5932: double age,agelim, hf;
5933: /* double ***mobaverage; */
5934: int theta;
5935: char digit[4];
5936: char digitp[25];
5937:
5938: char fileresprobmorprev[FILENAMELENGTH];
5939:
5940: if(popbased==1){
5941: if(mobilav!=0)
5942: strcpy(digitp,"-POPULBASED-MOBILAV_");
5943: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5944: }
5945: else
5946: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5947:
1.218 brouard 5948: /* if (mobilav!=0) { */
5949: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5950: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5951: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5952: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5953: /* } */
5954: /* } */
5955:
5956: strcpy(fileresprobmorprev,"PRMORPREV-");
5957: sprintf(digit,"%-d",ij);
5958: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5959: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5960: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5961: strcat(fileresprobmorprev,fileresu);
5962: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5963: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5964: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5965: }
5966: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5967: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5968: pstamp(ficresprobmorprev);
5969: 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 5970: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5971: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5972: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5973: }
5974: for(j=1;j<=cptcoveff;j++)
5975: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5976: fprintf(ficresprobmorprev,"\n");
5977:
1.218 brouard 5978: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5979: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5980: fprintf(ficresprobmorprev," p.%-d SE",j);
5981: for(i=1; i<=nlstate;i++)
5982: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5983: }
5984: fprintf(ficresprobmorprev,"\n");
5985:
5986: fprintf(ficgp,"\n# Routine varevsij");
5987: fprintf(ficgp,"\nunset title \n");
5988: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5989: 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");
5990: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 5991:
1.218 brouard 5992: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5993: pstamp(ficresvij);
5994: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5995: if(popbased==1)
5996: 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);
5997: else
5998: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5999: fprintf(ficresvij,"# Age");
6000: for(i=1; i<=nlstate;i++)
6001: for(j=1; j<=nlstate;j++)
6002: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6003: fprintf(ficresvij,"\n");
6004:
6005: xp=vector(1,npar);
6006: dnewm=matrix(1,nlstate,1,npar);
6007: doldm=matrix(1,nlstate,1,nlstate);
6008: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6009: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6010:
6011: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6012: gpp=vector(nlstate+1,nlstate+ndeath);
6013: gmp=vector(nlstate+1,nlstate+ndeath);
6014: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6015:
1.218 brouard 6016: if(estepm < stepm){
6017: printf ("Problem %d lower than %d\n",estepm, stepm);
6018: }
6019: else hstepm=estepm;
6020: /* For example we decided to compute the life expectancy with the smallest unit */
6021: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6022: nhstepm is the number of hstepm from age to agelim
6023: nstepm is the number of stepm from age to agelim.
6024: Look at function hpijx to understand why because of memory size limitations,
6025: we decided (b) to get a life expectancy respecting the most precise curvature of the
6026: survival function given by stepm (the optimization length). Unfortunately it
6027: means that if the survival funtion is printed every two years of age and if
6028: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6029: results. So we changed our mind and took the option of the best precision.
6030: */
6031: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6032: agelim = AGESUP;
6033: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6034: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6035: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6036: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6037: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6038: gp=matrix(0,nhstepm,1,nlstate);
6039: gm=matrix(0,nhstepm,1,nlstate);
6040:
6041:
6042: for(theta=1; theta <=npar; theta++){
6043: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6044: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6045: }
1.279 brouard 6046: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6047: * returns into prlim .
1.288 brouard 6048: */
1.242 brouard 6049: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6050:
6051: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6052: if (popbased==1) {
6053: if(mobilav ==0){
6054: for(i=1; i<=nlstate;i++)
6055: prlim[i][i]=probs[(int)age][i][ij];
6056: }else{ /* mobilav */
6057: for(i=1; i<=nlstate;i++)
6058: prlim[i][i]=mobaverage[(int)age][i][ij];
6059: }
6060: }
1.295 brouard 6061: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6062: */
6063: 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 6064: /**< 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 6065: * at horizon h in state j including mortality.
6066: */
1.218 brouard 6067: for(j=1; j<= nlstate; j++){
6068: for(h=0; h<=nhstepm; h++){
6069: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6070: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6071: }
6072: }
1.279 brouard 6073: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6074: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6075: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6076: */
6077: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6078: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6079: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6080: }
6081:
6082: /* Again with minus shift */
1.218 brouard 6083:
6084: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6085: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6086:
1.242 brouard 6087: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6088:
6089: if (popbased==1) {
6090: if(mobilav ==0){
6091: for(i=1; i<=nlstate;i++)
6092: prlim[i][i]=probs[(int)age][i][ij];
6093: }else{ /* mobilav */
6094: for(i=1; i<=nlstate;i++)
6095: prlim[i][i]=mobaverage[(int)age][i][ij];
6096: }
6097: }
6098:
1.235 brouard 6099: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6100:
6101: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6102: for(h=0; h<=nhstepm; h++){
6103: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6104: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6105: }
6106: }
6107: /* This for computing probability of death (h=1 means
6108: computed over hstepm matrices product = hstepm*stepm months)
6109: as a weighted average of prlim.
6110: */
6111: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6112: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6113: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6114: }
1.279 brouard 6115: /* end shifting computations */
6116:
6117: /**< Computing gradient matrix at horizon h
6118: */
1.218 brouard 6119: for(j=1; j<= nlstate; j++) /* vareij */
6120: for(h=0; h<=nhstepm; h++){
6121: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6122: }
1.279 brouard 6123: /**< Gradient of overall mortality p.3 (or p.j)
6124: */
6125: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6126: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6127: }
6128:
6129: } /* End theta */
1.279 brouard 6130:
6131: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6132: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6133:
6134: for(h=0; h<=nhstepm; h++) /* veij */
6135: for(j=1; j<=nlstate;j++)
6136: for(theta=1; theta <=npar; theta++)
6137: trgradg[h][j][theta]=gradg[h][theta][j];
6138:
6139: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6140: for(theta=1; theta <=npar; theta++)
6141: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6142: /**< as well as its transposed matrix
6143: */
1.218 brouard 6144:
6145: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6146: for(i=1;i<=nlstate;i++)
6147: for(j=1;j<=nlstate;j++)
6148: vareij[i][j][(int)age] =0.;
1.279 brouard 6149:
6150: /* Computing trgradg by matcov by gradg at age and summing over h
6151: * and k (nhstepm) formula 15 of article
6152: * Lievre-Brouard-Heathcote
6153: */
6154:
1.218 brouard 6155: for(h=0;h<=nhstepm;h++){
6156: for(k=0;k<=nhstepm;k++){
6157: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6158: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6159: for(i=1;i<=nlstate;i++)
6160: for(j=1;j<=nlstate;j++)
6161: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6162: }
6163: }
6164:
1.279 brouard 6165: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6166: * p.j overall mortality formula 49 but computed directly because
6167: * we compute the grad (wix pijx) instead of grad (pijx),even if
6168: * wix is independent of theta.
6169: */
1.218 brouard 6170: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6171: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6172: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6173: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6174: varppt[j][i]=doldmp[j][i];
6175: /* end ppptj */
6176: /* x centered again */
6177:
1.242 brouard 6178: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6179:
6180: if (popbased==1) {
6181: if(mobilav ==0){
6182: for(i=1; i<=nlstate;i++)
6183: prlim[i][i]=probs[(int)age][i][ij];
6184: }else{ /* mobilav */
6185: for(i=1; i<=nlstate;i++)
6186: prlim[i][i]=mobaverage[(int)age][i][ij];
6187: }
6188: }
6189:
6190: /* This for computing probability of death (h=1 means
6191: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6192: as a weighted average of prlim.
6193: */
1.235 brouard 6194: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6195: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6196: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6197: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6198: }
6199: /* end probability of death */
6200:
6201: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6202: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6203: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6204: for(i=1; i<=nlstate;i++){
6205: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6206: }
6207: }
6208: fprintf(ficresprobmorprev,"\n");
6209:
6210: fprintf(ficresvij,"%.0f ",age );
6211: for(i=1; i<=nlstate;i++)
6212: for(j=1; j<=nlstate;j++){
6213: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6214: }
6215: fprintf(ficresvij,"\n");
6216: free_matrix(gp,0,nhstepm,1,nlstate);
6217: free_matrix(gm,0,nhstepm,1,nlstate);
6218: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6219: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6220: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6221: } /* End age */
6222: free_vector(gpp,nlstate+1,nlstate+ndeath);
6223: free_vector(gmp,nlstate+1,nlstate+ndeath);
6224: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6225: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6226: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6227: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6228: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6229: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6230: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6231: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6232: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6233: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6234: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6235: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6236: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6237: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6238: 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);
6239: /* 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 6240: */
1.218 brouard 6241: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6242: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6243:
1.218 brouard 6244: free_vector(xp,1,npar);
6245: free_matrix(doldm,1,nlstate,1,nlstate);
6246: free_matrix(dnewm,1,nlstate,1,npar);
6247: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6248: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6249: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6250: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6251: fclose(ficresprobmorprev);
6252: fflush(ficgp);
6253: fflush(fichtm);
6254: } /* end varevsij */
1.126 brouard 6255:
6256: /************ Variance of prevlim ******************/
1.269 brouard 6257: 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 6258: {
1.205 brouard 6259: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6260: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6261:
1.268 brouard 6262: double **dnewmpar,**doldm;
1.126 brouard 6263: int i, j, nhstepm, hstepm;
6264: double *xp;
6265: double *gp, *gm;
6266: double **gradg, **trgradg;
1.208 brouard 6267: double **mgm, **mgp;
1.126 brouard 6268: double age,agelim;
6269: int theta;
6270:
6271: pstamp(ficresvpl);
1.288 brouard 6272: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6273: fprintf(ficresvpl,"# Age ");
6274: if(nresult >=1)
6275: fprintf(ficresvpl," Result# ");
1.126 brouard 6276: for(i=1; i<=nlstate;i++)
6277: fprintf(ficresvpl," %1d-%1d",i,i);
6278: fprintf(ficresvpl,"\n");
6279:
6280: xp=vector(1,npar);
1.268 brouard 6281: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6282: doldm=matrix(1,nlstate,1,nlstate);
6283:
6284: hstepm=1*YEARM; /* Every year of age */
6285: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6286: agelim = AGESUP;
6287: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6288: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6289: if (stepm >= YEARM) hstepm=1;
6290: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6291: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6292: mgp=matrix(1,npar,1,nlstate);
6293: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6294: gp=vector(1,nlstate);
6295: gm=vector(1,nlstate);
6296:
6297: for(theta=1; theta <=npar; theta++){
6298: for(i=1; i<=npar; i++){ /* Computes gradient */
6299: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6300: }
1.288 brouard 6301: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6302: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6303: /* else */
6304: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6305: for(i=1;i<=nlstate;i++){
1.126 brouard 6306: gp[i] = prlim[i][i];
1.208 brouard 6307: mgp[theta][i] = prlim[i][i];
6308: }
1.126 brouard 6309: for(i=1; i<=npar; i++) /* Computes gradient */
6310: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6311: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6312: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6313: /* else */
6314: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6315: for(i=1;i<=nlstate;i++){
1.126 brouard 6316: gm[i] = prlim[i][i];
1.208 brouard 6317: mgm[theta][i] = prlim[i][i];
6318: }
1.126 brouard 6319: for(i=1;i<=nlstate;i++)
6320: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6321: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6322: } /* End theta */
6323:
6324: trgradg =matrix(1,nlstate,1,npar);
6325:
6326: for(j=1; j<=nlstate;j++)
6327: for(theta=1; theta <=npar; theta++)
6328: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6329: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6330: /* printf("\nmgm mgp %d ",(int)age); */
6331: /* for(j=1; j<=nlstate;j++){ */
6332: /* printf(" %d ",j); */
6333: /* for(theta=1; theta <=npar; theta++) */
6334: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6335: /* printf("\n "); */
6336: /* } */
6337: /* } */
6338: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6339: /* printf("\n gradg %d ",(int)age); */
6340: /* for(j=1; j<=nlstate;j++){ */
6341: /* printf("%d ",j); */
6342: /* for(theta=1; theta <=npar; theta++) */
6343: /* printf("%d %lf ",theta,gradg[theta][j]); */
6344: /* printf("\n "); */
6345: /* } */
6346: /* } */
1.126 brouard 6347:
6348: for(i=1;i<=nlstate;i++)
6349: varpl[i][(int)age] =0.;
1.209 brouard 6350: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
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: }else{
1.268 brouard 6354: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6355: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6356: }
1.126 brouard 6357: for(i=1;i<=nlstate;i++)
6358: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6359:
6360: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6361: if(nresult >=1)
6362: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6363: for(i=1; i<=nlstate;i++){
1.126 brouard 6364: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6365: /* for(j=1;j<=nlstate;j++) */
6366: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6367: }
1.126 brouard 6368: fprintf(ficresvpl,"\n");
6369: free_vector(gp,1,nlstate);
6370: free_vector(gm,1,nlstate);
1.208 brouard 6371: free_matrix(mgm,1,npar,1,nlstate);
6372: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6373: free_matrix(gradg,1,npar,1,nlstate);
6374: free_matrix(trgradg,1,nlstate,1,npar);
6375: } /* End age */
6376:
6377: free_vector(xp,1,npar);
6378: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6379: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6380:
6381: }
6382:
6383:
6384: /************ Variance of backprevalence limit ******************/
1.269 brouard 6385: 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 6386: {
6387: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6388: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6389:
6390: double **dnewmpar,**doldm;
6391: int i, j, nhstepm, hstepm;
6392: double *xp;
6393: double *gp, *gm;
6394: double **gradg, **trgradg;
6395: double **mgm, **mgp;
6396: double age,agelim;
6397: int theta;
6398:
6399: pstamp(ficresvbl);
6400: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6401: fprintf(ficresvbl,"# Age ");
6402: if(nresult >=1)
6403: fprintf(ficresvbl," Result# ");
6404: for(i=1; i<=nlstate;i++)
6405: fprintf(ficresvbl," %1d-%1d",i,i);
6406: fprintf(ficresvbl,"\n");
6407:
6408: xp=vector(1,npar);
6409: dnewmpar=matrix(1,nlstate,1,npar);
6410: doldm=matrix(1,nlstate,1,nlstate);
6411:
6412: hstepm=1*YEARM; /* Every year of age */
6413: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6414: agelim = AGEINF;
6415: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6416: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6417: if (stepm >= YEARM) hstepm=1;
6418: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6419: gradg=matrix(1,npar,1,nlstate);
6420: mgp=matrix(1,npar,1,nlstate);
6421: mgm=matrix(1,npar,1,nlstate);
6422: gp=vector(1,nlstate);
6423: gm=vector(1,nlstate);
6424:
6425: for(theta=1; theta <=npar; theta++){
6426: for(i=1; i<=npar; i++){ /* Computes gradient */
6427: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6428: }
6429: if(mobilavproj > 0 )
6430: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6431: else
6432: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6433: for(i=1;i<=nlstate;i++){
6434: gp[i] = bprlim[i][i];
6435: mgp[theta][i] = bprlim[i][i];
6436: }
6437: for(i=1; i<=npar; i++) /* Computes gradient */
6438: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6439: if(mobilavproj > 0 )
6440: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6441: else
6442: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6443: for(i=1;i<=nlstate;i++){
6444: gm[i] = bprlim[i][i];
6445: mgm[theta][i] = bprlim[i][i];
6446: }
6447: for(i=1;i<=nlstate;i++)
6448: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6449: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6450: } /* End theta */
6451:
6452: trgradg =matrix(1,nlstate,1,npar);
6453:
6454: for(j=1; j<=nlstate;j++)
6455: for(theta=1; theta <=npar; theta++)
6456: trgradg[j][theta]=gradg[theta][j];
6457: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6458: /* printf("\nmgm mgp %d ",(int)age); */
6459: /* for(j=1; j<=nlstate;j++){ */
6460: /* printf(" %d ",j); */
6461: /* for(theta=1; theta <=npar; theta++) */
6462: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6463: /* printf("\n "); */
6464: /* } */
6465: /* } */
6466: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6467: /* printf("\n gradg %d ",(int)age); */
6468: /* for(j=1; j<=nlstate;j++){ */
6469: /* printf("%d ",j); */
6470: /* for(theta=1; theta <=npar; theta++) */
6471: /* printf("%d %lf ",theta,gradg[theta][j]); */
6472: /* printf("\n "); */
6473: /* } */
6474: /* } */
6475:
6476: for(i=1;i<=nlstate;i++)
6477: varbpl[i][(int)age] =0.;
6478: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6479: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6480: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6481: }else{
6482: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6483: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6484: }
6485: for(i=1;i<=nlstate;i++)
6486: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6487:
6488: fprintf(ficresvbl,"%.0f ",age );
6489: if(nresult >=1)
6490: fprintf(ficresvbl,"%d ",nres );
6491: for(i=1; i<=nlstate;i++)
6492: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6493: fprintf(ficresvbl,"\n");
6494: free_vector(gp,1,nlstate);
6495: free_vector(gm,1,nlstate);
6496: free_matrix(mgm,1,npar,1,nlstate);
6497: free_matrix(mgp,1,npar,1,nlstate);
6498: free_matrix(gradg,1,npar,1,nlstate);
6499: free_matrix(trgradg,1,nlstate,1,npar);
6500: } /* End age */
6501:
6502: free_vector(xp,1,npar);
6503: free_matrix(doldm,1,nlstate,1,npar);
6504: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6505:
6506: }
6507:
6508: /************ Variance of one-step probabilities ******************/
6509: 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 6510: {
6511: int i, j=0, k1, l1, tj;
6512: int k2, l2, j1, z1;
6513: int k=0, l;
6514: int first=1, first1, first2;
6515: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6516: double **dnewm,**doldm;
6517: double *xp;
6518: double *gp, *gm;
6519: double **gradg, **trgradg;
6520: double **mu;
6521: double age, cov[NCOVMAX+1];
6522: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6523: int theta;
6524: char fileresprob[FILENAMELENGTH];
6525: char fileresprobcov[FILENAMELENGTH];
6526: char fileresprobcor[FILENAMELENGTH];
6527: double ***varpij;
6528:
6529: strcpy(fileresprob,"PROB_");
6530: strcat(fileresprob,fileres);
6531: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6532: printf("Problem with resultfile: %s\n", fileresprob);
6533: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6534: }
6535: strcpy(fileresprobcov,"PROBCOV_");
6536: strcat(fileresprobcov,fileresu);
6537: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6538: printf("Problem with resultfile: %s\n", fileresprobcov);
6539: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6540: }
6541: strcpy(fileresprobcor,"PROBCOR_");
6542: strcat(fileresprobcor,fileresu);
6543: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6544: printf("Problem with resultfile: %s\n", fileresprobcor);
6545: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6546: }
6547: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6548: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6549: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6550: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6551: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6552: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6553: pstamp(ficresprob);
6554: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6555: fprintf(ficresprob,"# Age");
6556: pstamp(ficresprobcov);
6557: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6558: fprintf(ficresprobcov,"# Age");
6559: pstamp(ficresprobcor);
6560: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6561: fprintf(ficresprobcor,"# Age");
1.126 brouard 6562:
6563:
1.222 brouard 6564: for(i=1; i<=nlstate;i++)
6565: for(j=1; j<=(nlstate+ndeath);j++){
6566: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6567: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6568: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6569: }
6570: /* fprintf(ficresprob,"\n");
6571: fprintf(ficresprobcov,"\n");
6572: fprintf(ficresprobcor,"\n");
6573: */
6574: xp=vector(1,npar);
6575: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6576: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6577: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6578: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6579: first=1;
6580: fprintf(ficgp,"\n# Routine varprob");
6581: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6582: fprintf(fichtm,"\n");
6583:
1.288 brouard 6584: 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 6585: 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);
6586: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6587: and drawn. It helps understanding how is the covariance between two incidences.\
6588: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6589: 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 6590: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6591: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6592: standard deviations wide on each axis. <br>\
6593: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6594: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6595: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6596:
1.222 brouard 6597: cov[1]=1;
6598: /* tj=cptcoveff; */
1.225 brouard 6599: tj = (int) pow(2,cptcoveff);
1.222 brouard 6600: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6601: j1=0;
1.224 brouard 6602: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6603: if (cptcovn>0) {
6604: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6605: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6606: fprintf(ficresprob, "**********\n#\n");
6607: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6608: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6609: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6610:
1.222 brouard 6611: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6612: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6613: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6614:
6615:
1.222 brouard 6616: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6617: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6618: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6619:
1.222 brouard 6620: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6621: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6622: fprintf(ficresprobcor, "**********\n#");
6623: if(invalidvarcomb[j1]){
6624: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6625: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6626: continue;
6627: }
6628: }
6629: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6630: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6631: gp=vector(1,(nlstate)*(nlstate+ndeath));
6632: gm=vector(1,(nlstate)*(nlstate+ndeath));
6633: for (age=bage; age<=fage; age ++){
6634: cov[2]=age;
6635: if(nagesqr==1)
6636: cov[3]= age*age;
6637: for (k=1; k<=cptcovn;k++) {
6638: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6639: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6640: * 1 1 1 1 1
6641: * 2 2 1 1 1
6642: * 3 1 2 1 1
6643: */
6644: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6645: }
6646: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6647: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6648: for (k=1; k<=cptcovprod;k++)
6649: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6650:
6651:
1.222 brouard 6652: for(theta=1; theta <=npar; theta++){
6653: for(i=1; i<=npar; i++)
6654: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6655:
1.222 brouard 6656: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6657:
1.222 brouard 6658: k=0;
6659: for(i=1; i<= (nlstate); i++){
6660: for(j=1; j<=(nlstate+ndeath);j++){
6661: k=k+1;
6662: gp[k]=pmmij[i][j];
6663: }
6664: }
1.220 brouard 6665:
1.222 brouard 6666: for(i=1; i<=npar; i++)
6667: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6668:
1.222 brouard 6669: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6670: k=0;
6671: for(i=1; i<=(nlstate); i++){
6672: for(j=1; j<=(nlstate+ndeath);j++){
6673: k=k+1;
6674: gm[k]=pmmij[i][j];
6675: }
6676: }
1.220 brouard 6677:
1.222 brouard 6678: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6679: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6680: }
1.126 brouard 6681:
1.222 brouard 6682: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6683: for(theta=1; theta <=npar; theta++)
6684: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6685:
1.222 brouard 6686: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6687: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6688:
1.222 brouard 6689: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6690:
1.222 brouard 6691: k=0;
6692: for(i=1; i<=(nlstate); i++){
6693: for(j=1; j<=(nlstate+ndeath);j++){
6694: k=k+1;
6695: mu[k][(int) age]=pmmij[i][j];
6696: }
6697: }
6698: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6699: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6700: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6701:
1.222 brouard 6702: /*printf("\n%d ",(int)age);
6703: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6704: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6705: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6706: }*/
1.220 brouard 6707:
1.222 brouard 6708: fprintf(ficresprob,"\n%d ",(int)age);
6709: fprintf(ficresprobcov,"\n%d ",(int)age);
6710: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6711:
1.222 brouard 6712: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6713: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6714: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6715: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6716: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6717: }
6718: i=0;
6719: for (k=1; k<=(nlstate);k++){
6720: for (l=1; l<=(nlstate+ndeath);l++){
6721: i++;
6722: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6723: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6724: for (j=1; j<=i;j++){
6725: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6726: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6727: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6728: }
6729: }
6730: }/* end of loop for state */
6731: } /* end of loop for age */
6732: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6733: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6734: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6735: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6736:
6737: /* Confidence intervalle of pij */
6738: /*
6739: fprintf(ficgp,"\nunset parametric;unset label");
6740: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6741: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6742: 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);
6743: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6744: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6745: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6746: */
6747:
6748: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6749: first1=1;first2=2;
6750: for (k2=1; k2<=(nlstate);k2++){
6751: for (l2=1; l2<=(nlstate+ndeath);l2++){
6752: if(l2==k2) continue;
6753: j=(k2-1)*(nlstate+ndeath)+l2;
6754: for (k1=1; k1<=(nlstate);k1++){
6755: for (l1=1; l1<=(nlstate+ndeath);l1++){
6756: if(l1==k1) continue;
6757: i=(k1-1)*(nlstate+ndeath)+l1;
6758: if(i<=j) continue;
6759: for (age=bage; age<=fage; age ++){
6760: if ((int)age %5==0){
6761: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6762: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6763: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6764: mu1=mu[i][(int) age]/stepm*YEARM ;
6765: mu2=mu[j][(int) age]/stepm*YEARM;
6766: c12=cv12/sqrt(v1*v2);
6767: /* Computing eigen value of matrix of covariance */
6768: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6769: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6770: if ((lc2 <0) || (lc1 <0) ){
6771: if(first2==1){
6772: first1=0;
6773: 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);
6774: }
6775: 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);
6776: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6777: /* lc2=fabs(lc2); */
6778: }
1.220 brouard 6779:
1.222 brouard 6780: /* Eigen vectors */
1.280 brouard 6781: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6782: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6783: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6784: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6785: }else
6786: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6787: /*v21=sqrt(1.-v11*v11); *//* error */
6788: v21=(lc1-v1)/cv12*v11;
6789: v12=-v21;
6790: v22=v11;
6791: tnalp=v21/v11;
6792: if(first1==1){
6793: first1=0;
6794: 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);
6795: }
6796: 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);
6797: /*printf(fignu*/
6798: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6799: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6800: if(first==1){
6801: first=0;
6802: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6803: fprintf(ficgp,"\nset parametric;unset label");
6804: 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);
6805: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6806: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6807: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6808: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6809: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6810: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6811: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6812: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6813: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6814: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6815: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6816: 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 6817: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6818: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6819: }else{
6820: first=0;
6821: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6822: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6823: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6824: 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 6825: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6826: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6827: }/* if first */
6828: } /* age mod 5 */
6829: } /* end loop age */
6830: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6831: first=1;
6832: } /*l12 */
6833: } /* k12 */
6834: } /*l1 */
6835: }/* k1 */
6836: } /* loop on combination of covariates j1 */
6837: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6838: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6839: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6840: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6841: free_vector(xp,1,npar);
6842: fclose(ficresprob);
6843: fclose(ficresprobcov);
6844: fclose(ficresprobcor);
6845: fflush(ficgp);
6846: fflush(fichtmcov);
6847: }
1.126 brouard 6848:
6849:
6850: /******************* Printing html file ***********/
1.201 brouard 6851: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6852: int lastpass, int stepm, int weightopt, char model[],\
6853: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 6854: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
6855: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
6856: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 6857: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6858:
6859: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6860: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6861: </ul>");
1.237 brouard 6862: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6863: </ul>", model);
1.214 brouard 6864: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6865: 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",
6866: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6867: 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 6868: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6869: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6870: fprintf(fichtm,"\
6871: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6872: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6873: fprintf(fichtm,"\
1.217 brouard 6874: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6875: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6876: fprintf(fichtm,"\
1.288 brouard 6877: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6878: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6879: fprintf(fichtm,"\
1.288 brouard 6880: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6881: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6882: fprintf(fichtm,"\
1.211 brouard 6883: - (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 6884: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6885: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6886: if(prevfcast==1){
6887: fprintf(fichtm,"\
6888: - Prevalence projections by age and states: \
1.201 brouard 6889: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6890: }
1.126 brouard 6891:
6892:
1.225 brouard 6893: m=pow(2,cptcoveff);
1.222 brouard 6894: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6895:
1.264 brouard 6896: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6897:
6898: jj1=0;
6899:
6900: fprintf(fichtm," \n<ul>");
6901: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6902: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6903: if(m != 1 && TKresult[nres]!= k1)
6904: continue;
6905: jj1++;
6906: if (cptcovn > 0) {
6907: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6908: for (cpt=1; cpt<=cptcoveff;cpt++){
6909: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6910: }
6911: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6912: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6913: }
6914: fprintf(fichtm,"\">");
6915:
6916: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6917: fprintf(fichtm,"************ Results for covariates");
6918: for (cpt=1; cpt<=cptcoveff;cpt++){
6919: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6920: }
6921: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6922: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6923: }
6924: if(invalidvarcomb[k1]){
6925: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6926: continue;
6927: }
6928: fprintf(fichtm,"</a></li>");
6929: } /* cptcovn >0 */
6930: }
6931: fprintf(fichtm," \n</ul>");
6932:
1.222 brouard 6933: jj1=0;
1.237 brouard 6934:
6935: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6936: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6937: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6938: continue;
1.220 brouard 6939:
1.222 brouard 6940: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6941: jj1++;
6942: if (cptcovn > 0) {
1.264 brouard 6943: fprintf(fichtm,"\n<p><a name=\"rescov");
6944: for (cpt=1; cpt<=cptcoveff;cpt++){
6945: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6946: }
6947: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6948: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6949: }
6950: fprintf(fichtm,"\"</a>");
6951:
1.222 brouard 6952: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6953: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6954: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6955: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6956: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6957: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6958: }
1.237 brouard 6959: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6960: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6961: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6962: }
6963:
1.230 brouard 6964: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6965: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6966: if(invalidvarcomb[k1]){
6967: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6968: printf("\nCombination (%d) ignored because no cases \n",k1);
6969: continue;
6970: }
6971: }
6972: /* aij, bij */
1.259 brouard 6973: 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 6974: <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 6975: /* Pij */
1.241 brouard 6976: 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> \
6977: <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 6978: /* Quasi-incidences */
6979: 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 6980: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6981: 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 6982: 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> \
6983: <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 6984: /* Survival functions (period) in state j */
6985: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6986: 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 6987: <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 6988: }
6989: /* State specific survival functions (period) */
6990: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6991: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
6992: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 6993: <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 6994: }
1.288 brouard 6995: /* Period (forward stable) prevalence in each health state */
1.222 brouard 6996: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6997: 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> \
6998: <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 6999: }
1.296 brouard 7000: if(prevbcast==1){
1.288 brouard 7001: /* Backward prevalence in each health state */
1.222 brouard 7002: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7003: 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 7004: <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 7005: }
1.217 brouard 7006: }
1.222 brouard 7007: if(prevfcast==1){
1.288 brouard 7008: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7009: for(cpt=1; cpt<=nlstate;cpt++){
1.288 brouard 7010: 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 7011: <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 7012: }
7013: }
1.296 brouard 7014: if(prevbcast==1){
1.268 brouard 7015: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7016: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7017: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7018: 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 \
7019: 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) \
7020: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
7021: <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 7022: }
7023: }
1.220 brouard 7024:
1.222 brouard 7025: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 7026: 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> \
7027: <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 7028: }
7029: /* } /\* end i1 *\/ */
7030: }/* End k1 */
7031: fprintf(fichtm,"</ul>");
1.126 brouard 7032:
1.222 brouard 7033: fprintf(fichtm,"\
1.126 brouard 7034: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7035: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7036: - 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 7037: But because parameters are usually highly correlated (a higher incidence of disability \
7038: and a higher incidence of recovery can give very close observed transition) it might \
7039: be very useful to look not only at linear confidence intervals estimated from the \
7040: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7041: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7042: covariance matrix of the one-step probabilities. \
7043: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7044:
1.222 brouard 7045: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7046: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7047: fprintf(fichtm,"\
1.126 brouard 7048: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7049: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7050:
1.222 brouard 7051: fprintf(fichtm,"\
1.126 brouard 7052: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7053: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7054: fprintf(fichtm,"\
1.126 brouard 7055: - 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): \
7056: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7057: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7058: fprintf(fichtm,"\
1.126 brouard 7059: - (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): \
7060: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7061: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7062: fprintf(fichtm,"\
1.288 brouard 7063: - 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 7064: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7065: fprintf(fichtm,"\
1.128 brouard 7066: - 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 7067: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7068: fprintf(fichtm,"\
1.288 brouard 7069: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7070: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7071:
7072: /* if(popforecast==1) fprintf(fichtm,"\n */
7073: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7074: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7075: /* <br>",fileres,fileres,fileres,fileres); */
7076: /* else */
7077: /* 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 7078: fflush(fichtm);
7079: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7080:
1.225 brouard 7081: m=pow(2,cptcoveff);
1.222 brouard 7082: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7083:
1.222 brouard 7084: jj1=0;
1.237 brouard 7085:
1.241 brouard 7086: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7087: for(k1=1; k1<=m;k1++){
1.253 brouard 7088: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7089: continue;
1.222 brouard 7090: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7091: jj1++;
1.126 brouard 7092: if (cptcovn > 0) {
7093: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7094: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7095: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7096: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7097: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7098: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7099: }
7100:
1.126 brouard 7101: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7102:
1.222 brouard 7103: if(invalidvarcomb[k1]){
7104: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7105: continue;
7106: }
1.126 brouard 7107: }
7108: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7109: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 7110: 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 7111: <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 7112: }
7113: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 7114: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
7115: true period expectancies (those weighted with period prevalences are also\
7116: drawn in addition to the population based expectancies computed using\
1.241 brouard 7117: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
7118: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7119: /* } /\* end i1 *\/ */
7120: }/* End k1 */
1.241 brouard 7121: }/* End nres */
1.222 brouard 7122: fprintf(fichtm,"</ul>");
7123: fflush(fichtm);
1.126 brouard 7124: }
7125:
7126: /******************* Gnuplot file **************/
1.296 brouard 7127: 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 7128:
7129: char dirfileres[132],optfileres[132];
1.264 brouard 7130: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7131: 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 7132: int lv=0, vlv=0, kl=0;
1.130 brouard 7133: int ng=0;
1.201 brouard 7134: int vpopbased;
1.223 brouard 7135: int ioffset; /* variable offset for columns */
1.270 brouard 7136: int iyearc=1; /* variable column for year of projection */
7137: int iagec=1; /* variable column for age of projection */
1.235 brouard 7138: int nres=0; /* Index of resultline */
1.266 brouard 7139: int istart=1; /* For starting graphs in projections */
1.219 brouard 7140:
1.126 brouard 7141: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7142: /* printf("Problem with file %s",optionfilegnuplot); */
7143: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7144: /* } */
7145:
7146: /*#ifdef windows */
7147: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7148: /*#endif */
1.225 brouard 7149: m=pow(2,cptcoveff);
1.126 brouard 7150:
1.274 brouard 7151: /* diagram of the model */
7152: fprintf(ficgp,"\n#Diagram of the model \n");
7153: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7154: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7155: 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);
7156:
7157: 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);
7158: fprintf(ficgp,"\n#show arrow\nunset label\n");
7159: 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);
7160: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7161: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7162: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7163: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7164:
1.202 brouard 7165: /* Contribution to likelihood */
7166: /* Plot the probability implied in the likelihood */
1.223 brouard 7167: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7168: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7169: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7170: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7171: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7172: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7173: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7174: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7175: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7176: 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));
7177: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7178: 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));
7179: for (i=1; i<= nlstate ; i ++) {
7180: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7181: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7182: 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);
7183: for (j=2; j<= nlstate+ndeath ; j ++) {
7184: 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);
7185: }
7186: fprintf(ficgp,";\nset out; unset ylabel;\n");
7187: }
7188: /* 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 */
7189: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7190: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7191: fprintf(ficgp,"\nset out;unset log\n");
7192: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7193:
1.126 brouard 7194: strcpy(dirfileres,optionfilefiname);
7195: strcpy(optfileres,"vpl");
1.223 brouard 7196: /* 1eme*/
1.238 brouard 7197: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7198: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7199: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7200: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7201: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7202: continue;
7203: /* We are interested in selected combination by the resultline */
1.246 brouard 7204: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7205: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7206: strcpy(gplotlabel,"(");
1.238 brouard 7207: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7208: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7209: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7210: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7211: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7212: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7213: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7214: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7215: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7216: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7217: }
7218: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7219: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7220: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7221: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7222: }
7223: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7224: /* printf("\n#\n"); */
1.238 brouard 7225: fprintf(ficgp,"\n#\n");
7226: if(invalidvarcomb[k1]){
1.260 brouard 7227: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7228: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7229: continue;
7230: }
1.235 brouard 7231:
1.241 brouard 7232: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7233: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7234: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7235: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7236: 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);
7237: /* 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); */
7238: /* k1-1 error should be nres-1*/
1.238 brouard 7239: for (i=1; i<= nlstate ; i ++) {
7240: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7241: else fprintf(ficgp," %%*lf (%%*lf)");
7242: }
1.288 brouard 7243: 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 7244: for (i=1; i<= nlstate ; i ++) {
7245: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7246: else fprintf(ficgp," %%*lf (%%*lf)");
7247: }
1.260 brouard 7248: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7249: for (i=1; i<= nlstate ; i ++) {
7250: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7251: else fprintf(ficgp," %%*lf (%%*lf)");
7252: }
1.265 brouard 7253: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7254:
7255: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7256: if(cptcoveff ==0){
1.271 brouard 7257: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7258: }else{
7259: kl=0;
7260: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7261: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7262: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7263: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7264: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7265: vlv= nbcode[Tvaraff[k]][lv];
7266: kl++;
7267: /* 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 *\/ */
7268: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7269: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7270: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7271: if(k==cptcoveff){
7272: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7273: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7274: }else{
7275: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7276: kl++;
7277: }
7278: } /* end covariate */
7279: } /* end if no covariate */
7280:
1.296 brouard 7281: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7282: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7283: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7284: if(cptcoveff ==0){
1.245 brouard 7285: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7286: }else{
7287: kl=0;
7288: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7289: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7290: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7291: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7292: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7293: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7294: kl++;
1.238 brouard 7295: /* 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 *\/ */
7296: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7297: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7298: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7299: if(k==cptcoveff){
1.245 brouard 7300: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7301: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7302: }else{
7303: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7304: kl++;
7305: }
7306: } /* end covariate */
7307: } /* end if no covariate */
1.296 brouard 7308: if(prevbcast == 1){
1.268 brouard 7309: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7310: /* k1-1 error should be nres-1*/
7311: for (i=1; i<= nlstate ; i ++) {
7312: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7313: else fprintf(ficgp," %%*lf (%%*lf)");
7314: }
1.271 brouard 7315: 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 7316: for (i=1; i<= nlstate ; i ++) {
7317: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7318: else fprintf(ficgp," %%*lf (%%*lf)");
7319: }
1.276 brouard 7320: 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 7321: for (i=1; i<= nlstate ; i ++) {
7322: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7323: else fprintf(ficgp," %%*lf (%%*lf)");
7324: }
1.274 brouard 7325: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7326: } /* end if backprojcast */
1.296 brouard 7327: } /* end if prevbcast */
1.276 brouard 7328: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7329: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7330: } /* nres */
1.201 brouard 7331: } /* k1 */
7332: } /* cpt */
1.235 brouard 7333:
7334:
1.126 brouard 7335: /*2 eme*/
1.238 brouard 7336: for (k1=1; k1<= m ; k1 ++){
7337: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7338: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7339: continue;
7340: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7341: strcpy(gplotlabel,"(");
1.238 brouard 7342: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7343: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7344: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7345: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7346: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7347: vlv= nbcode[Tvaraff[k]][lv];
7348: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7349: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7350: }
1.237 brouard 7351: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7352: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7353: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7354: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7355: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7356: }
1.264 brouard 7357: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7358: fprintf(ficgp,"\n#\n");
1.223 brouard 7359: if(invalidvarcomb[k1]){
7360: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7361: continue;
7362: }
1.219 brouard 7363:
1.241 brouard 7364: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7365: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7366: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7367: if(vpopbased==0){
1.238 brouard 7368: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7369: }else
1.238 brouard 7370: fprintf(ficgp,"\nreplot ");
7371: for (i=1; i<= nlstate+1 ; i ++) {
7372: k=2*i;
1.261 brouard 7373: 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 7374: for (j=1; j<= nlstate+1 ; j ++) {
7375: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7376: else fprintf(ficgp," %%*lf (%%*lf)");
7377: }
7378: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7379: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7380: 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 7381: for (j=1; j<= nlstate+1 ; j ++) {
7382: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7383: else fprintf(ficgp," %%*lf (%%*lf)");
7384: }
7385: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7386: 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 7387: for (j=1; j<= nlstate+1 ; j ++) {
7388: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7389: else fprintf(ficgp," %%*lf (%%*lf)");
7390: }
7391: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7392: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7393: } /* state */
7394: } /* vpopbased */
1.264 brouard 7395: 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 7396: } /* end nres */
7397: } /* k1 end 2 eme*/
7398:
7399:
7400: /*3eme*/
7401: for (k1=1; k1<= m ; k1 ++){
7402: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7403: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7404: continue;
7405:
7406: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7407: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7408: strcpy(gplotlabel,"(");
1.238 brouard 7409: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7410: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7411: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7412: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7413: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7414: vlv= nbcode[Tvaraff[k]][lv];
7415: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7416: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7417: }
7418: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7419: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7420: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7421: }
1.264 brouard 7422: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7423: fprintf(ficgp,"\n#\n");
7424: if(invalidvarcomb[k1]){
7425: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7426: continue;
7427: }
7428:
7429: /* k=2+nlstate*(2*cpt-2); */
7430: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7431: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7432: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7433: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7434: 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 7435: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7436: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7437: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7438: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7439: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7440: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7441:
1.238 brouard 7442: */
7443: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7444: 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 7445: /* 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 7446:
1.238 brouard 7447: }
1.261 brouard 7448: 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 7449: }
1.264 brouard 7450: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7451: } /* end nres */
7452: } /* end kl 3eme */
1.126 brouard 7453:
1.223 brouard 7454: /* 4eme */
1.201 brouard 7455: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7456: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7457: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7458: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7459: continue;
1.238 brouard 7460: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7461: strcpy(gplotlabel,"(");
1.238 brouard 7462: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7463: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7464: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7465: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7466: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7467: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7468: vlv= nbcode[Tvaraff[k]][lv];
7469: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7470: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7471: }
7472: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7473: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7474: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7475: }
1.264 brouard 7476: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7477: fprintf(ficgp,"\n#\n");
7478: if(invalidvarcomb[k1]){
7479: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7480: continue;
1.223 brouard 7481: }
1.238 brouard 7482:
1.241 brouard 7483: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7484: 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 7485: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7486: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7487: k=3;
7488: for (i=1; i<= nlstate ; i ++){
7489: if(i==1){
7490: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7491: }else{
7492: fprintf(ficgp,", '' ");
7493: }
7494: l=(nlstate+ndeath)*(i-1)+1;
7495: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7496: for (j=2; j<= nlstate+ndeath ; j ++)
7497: fprintf(ficgp,"+$%d",k+l+j-1);
7498: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7499: } /* nlstate */
1.264 brouard 7500: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7501: } /* end cpt state*/
7502: } /* end nres */
7503: } /* end covariate k1 */
7504:
1.220 brouard 7505: /* 5eme */
1.201 brouard 7506: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7507: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7508: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7509: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7510: continue;
1.238 brouard 7511: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7512: strcpy(gplotlabel,"(");
1.238 brouard 7513: 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);
7514: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7515: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7516: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7517: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7518: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7519: vlv= nbcode[Tvaraff[k]][lv];
7520: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7521: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7522: }
7523: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7524: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7525: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7526: }
1.264 brouard 7527: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7528: fprintf(ficgp,"\n#\n");
7529: if(invalidvarcomb[k1]){
7530: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7531: continue;
7532: }
1.227 brouard 7533:
1.241 brouard 7534: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7535: 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 7536: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7537: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7538: k=3;
7539: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7540: if(j==1)
7541: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7542: else
7543: fprintf(ficgp,", '' ");
7544: l=(nlstate+ndeath)*(cpt-1) +j;
7545: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7546: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7547: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7548: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7549: } /* nlstate */
7550: fprintf(ficgp,", '' ");
7551: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7552: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7553: l=(nlstate+ndeath)*(cpt-1) +j;
7554: if(j < nlstate)
7555: fprintf(ficgp,"$%d +",k+l);
7556: else
7557: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7558: }
1.264 brouard 7559: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7560: } /* end cpt state*/
7561: } /* end covariate */
7562: } /* end nres */
1.227 brouard 7563:
1.220 brouard 7564: /* 6eme */
1.202 brouard 7565: /* CV preval stable (period) for each covariate */
1.237 brouard 7566: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7567: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7568: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7569: continue;
1.255 brouard 7570: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7571: strcpy(gplotlabel,"(");
1.288 brouard 7572: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7573: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7574: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7575: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7576: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7577: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7578: vlv= nbcode[Tvaraff[k]][lv];
7579: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7580: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7581: }
1.237 brouard 7582: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7583: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7584: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7585: }
1.264 brouard 7586: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7587: fprintf(ficgp,"\n#\n");
1.223 brouard 7588: if(invalidvarcomb[k1]){
1.227 brouard 7589: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7590: continue;
1.223 brouard 7591: }
1.227 brouard 7592:
1.241 brouard 7593: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7594: 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 7595: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7596: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7597: k=3; /* Offset */
1.255 brouard 7598: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7599: if(i==1)
7600: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7601: else
7602: fprintf(ficgp,", '' ");
1.255 brouard 7603: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7604: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7605: for (j=2; j<= nlstate ; j ++)
7606: fprintf(ficgp,"+$%d",k+l+j-1);
7607: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7608: } /* nlstate */
1.264 brouard 7609: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7610: } /* end cpt state*/
7611: } /* end covariate */
1.227 brouard 7612:
7613:
1.220 brouard 7614: /* 7eme */
1.296 brouard 7615: if(prevbcast == 1){
1.288 brouard 7616: /* CV backward prevalence for each covariate */
1.237 brouard 7617: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7618: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7619: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7620: continue;
1.268 brouard 7621: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7622: strcpy(gplotlabel,"(");
1.288 brouard 7623: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7624: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7625: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7626: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7627: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7628: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7629: vlv= nbcode[Tvaraff[k]][lv];
7630: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7631: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7632: }
1.237 brouard 7633: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7634: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7635: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7636: }
1.264 brouard 7637: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7638: fprintf(ficgp,"\n#\n");
7639: if(invalidvarcomb[k1]){
7640: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7641: continue;
7642: }
7643:
1.241 brouard 7644: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7645: 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 7646: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7647: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7648: k=3; /* Offset */
1.268 brouard 7649: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7650: if(i==1)
7651: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7652: else
7653: fprintf(ficgp,", '' ");
7654: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7655: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7656: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7657: /* 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 7658: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7659: /* for (j=2; j<= nlstate ; j ++) */
7660: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7661: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7662: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7663: } /* nlstate */
1.264 brouard 7664: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7665: } /* end cpt state*/
7666: } /* end covariate */
1.296 brouard 7667: } /* End if prevbcast */
1.218 brouard 7668:
1.223 brouard 7669: /* 8eme */
1.218 brouard 7670: if(prevfcast==1){
1.288 brouard 7671: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7672:
1.237 brouard 7673: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7674: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7675: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7676: continue;
1.211 brouard 7677: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7678: strcpy(gplotlabel,"(");
1.288 brouard 7679: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7680: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7681: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7682: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7683: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7684: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7685: vlv= nbcode[Tvaraff[k]][lv];
7686: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7687: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7688: }
1.237 brouard 7689: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7690: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7691: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7692: }
1.264 brouard 7693: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7694: fprintf(ficgp,"\n#\n");
7695: if(invalidvarcomb[k1]){
7696: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7697: continue;
7698: }
7699:
7700: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7701: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7702: 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 7703: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7704: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7705:
7706: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7707: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7708: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7709: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7710: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7711: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7712: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7713: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7714: if(i==istart){
1.227 brouard 7715: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7716: }else{
7717: fprintf(ficgp,",\\\n '' ");
7718: }
7719: if(cptcoveff ==0){ /* No covariate */
7720: ioffset=2; /* Age is in 2 */
7721: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7722: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7723: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7724: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7725: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7726: if(i==nlstate+1){
1.270 brouard 7727: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7728: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7729: fprintf(ficgp,",\\\n '' ");
7730: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7731: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7732: offyear, \
1.268 brouard 7733: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7734: }else
1.227 brouard 7735: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7736: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7737: }else{ /* more than 2 covariates */
1.270 brouard 7738: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7739: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7740: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7741: iyearc=ioffset-1;
7742: iagec=ioffset;
1.227 brouard 7743: fprintf(ficgp," u %d:(",ioffset);
7744: kl=0;
7745: strcpy(gplotcondition,"(");
7746: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7747: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7748: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7749: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7750: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7751: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7752: kl++;
7753: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7754: kl++;
7755: if(k <cptcoveff && cptcoveff>1)
7756: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7757: }
7758: strcpy(gplotcondition+strlen(gplotcondition),")");
7759: /* 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 *\/ */
7760: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7761: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7762: /* '' 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*/
7763: if(i==nlstate+1){
1.270 brouard 7764: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7765: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7766: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7767: fprintf(ficgp," u %d:(",iagec);
7768: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7769: iyearc, iagec, offyear, \
7770: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7771: /* '' 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 7772: }else{
7773: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7774: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7775: }
7776: } /* end if covariate */
7777: } /* nlstate */
1.264 brouard 7778: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7779: } /* end cpt state*/
7780: } /* end covariate */
7781: } /* End if prevfcast */
1.227 brouard 7782:
1.296 brouard 7783: if(prevbcast==1){
1.268 brouard 7784: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7785:
7786: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7787: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7788: if(m != 1 && TKresult[nres]!= k1)
7789: continue;
7790: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7791: strcpy(gplotlabel,"(");
7792: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7793: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7794: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7795: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7796: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7797: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7798: vlv= nbcode[Tvaraff[k]][lv];
7799: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7800: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7801: }
7802: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7803: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7804: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7805: }
7806: strcpy(gplotlabel+strlen(gplotlabel),")");
7807: fprintf(ficgp,"\n#\n");
7808: if(invalidvarcomb[k1]){
7809: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7810: continue;
7811: }
7812:
7813: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7814: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7815: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7816: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7817: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7818:
7819: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7820: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7821: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7822: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7823: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7824: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7825: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7826: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7827: if(i==istart){
7828: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7829: }else{
7830: fprintf(ficgp,",\\\n '' ");
7831: }
7832: if(cptcoveff ==0){ /* No covariate */
7833: ioffset=2; /* Age is in 2 */
7834: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7835: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7836: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7837: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7838: fprintf(ficgp," u %d:(", ioffset);
7839: if(i==nlstate+1){
1.270 brouard 7840: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7841: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7842: fprintf(ficgp,",\\\n '' ");
7843: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7844: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7845: offbyear, \
7846: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7847: }else
7848: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7849: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7850: }else{ /* more than 2 covariates */
1.270 brouard 7851: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7852: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7853: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7854: iyearc=ioffset-1;
7855: iagec=ioffset;
1.268 brouard 7856: fprintf(ficgp," u %d:(",ioffset);
7857: kl=0;
7858: strcpy(gplotcondition,"(");
7859: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7860: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7861: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7862: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7863: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7864: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7865: kl++;
7866: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7867: kl++;
7868: if(k <cptcoveff && cptcoveff>1)
7869: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7870: }
7871: strcpy(gplotcondition+strlen(gplotcondition),")");
7872: /* 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 *\/ */
7873: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7874: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7875: /* '' 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*/
7876: if(i==nlstate+1){
1.270 brouard 7877: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7878: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7879: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7880: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7881: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7882: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7883: iyearc,iagec,offbyear, \
7884: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7885: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7886: }else{
7887: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7888: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7889: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7890: }
7891: } /* end if covariate */
7892: } /* nlstate */
7893: fprintf(ficgp,"\nset out; unset label;\n");
7894: } /* end cpt state*/
7895: } /* end covariate */
1.296 brouard 7896: } /* End if prevbcast */
1.268 brouard 7897:
1.227 brouard 7898:
1.238 brouard 7899: /* 9eme writing MLE parameters */
7900: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7901: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7902: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7903: for(k=1; k <=(nlstate+ndeath); k++){
7904: if (k != i) {
1.227 brouard 7905: fprintf(ficgp,"# current state %d\n",k);
7906: for(j=1; j <=ncovmodel; j++){
7907: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7908: jk++;
7909: }
7910: fprintf(ficgp,"\n");
1.126 brouard 7911: }
7912: }
1.223 brouard 7913: }
1.187 brouard 7914: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7915:
1.145 brouard 7916: /*goto avoid;*/
1.238 brouard 7917: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7918: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7919: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7920: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7921: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7922: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7923: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7924: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7925: fprintf(ficgp,"# p11=1/(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,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7928: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7929: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7930: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7931: fprintf(ficgp,"#\n");
1.223 brouard 7932: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7933: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7934: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7935: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7936: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7937: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7938: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7939: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7940: continue;
1.264 brouard 7941: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7942: strcpy(gplotlabel,"(");
1.276 brouard 7943: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7944: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7945: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7946: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7947: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7948: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7949: vlv= nbcode[Tvaraff[k]][lv];
7950: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7951: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7952: }
1.237 brouard 7953: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7954: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7955: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7956: }
1.264 brouard 7957: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7958: fprintf(ficgp,"\n#\n");
1.264 brouard 7959: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7960: fprintf(ficgp,"\nset key outside ");
7961: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7962: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7963: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7964: if (ng==1){
7965: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7966: fprintf(ficgp,"\nunset log y");
7967: }else if (ng==2){
7968: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7969: fprintf(ficgp,"\nset log y");
7970: }else if (ng==3){
7971: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7972: fprintf(ficgp,"\nset log y");
7973: }else
7974: fprintf(ficgp,"\nunset title ");
7975: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7976: i=1;
7977: for(k2=1; k2<=nlstate; k2++) {
7978: k3=i;
7979: for(k=1; k<=(nlstate+ndeath); k++) {
7980: if (k != k2){
7981: switch( ng) {
7982: case 1:
7983: if(nagesqr==0)
7984: fprintf(ficgp," p%d+p%d*x",i,i+1);
7985: else /* nagesqr =1 */
7986: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7987: break;
7988: case 2: /* ng=2 */
7989: if(nagesqr==0)
7990: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7991: else /* nagesqr =1 */
7992: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7993: break;
7994: case 3:
7995: if(nagesqr==0)
7996: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7997: else /* nagesqr =1 */
7998: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7999: break;
8000: }
8001: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8002: ijp=1; /* product no age */
8003: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8004: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8005: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8006: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8007: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8008: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8009: if(DummyV[j]==0){
8010: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8011: }else{ /* quantitative */
8012: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8013: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8014: }
8015: ij++;
1.237 brouard 8016: }
1.268 brouard 8017: }
8018: }else if(cptcovprod >0){
8019: if(j==Tprod[ijp]) { /* */
8020: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8021: if(ijp <=cptcovprod) { /* Product */
8022: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8023: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8024: /* 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)]); */
8025: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8026: }else{ /* Vn is dummy and Vm is quanti */
8027: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8028: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8029: }
8030: }else{ /* Vn*Vm Vn is quanti */
8031: if(DummyV[Tvard[ijp][2]]==0){
8032: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8033: }else{ /* Both quanti */
8034: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8035: }
1.237 brouard 8036: }
1.268 brouard 8037: ijp++;
1.237 brouard 8038: }
1.268 brouard 8039: } /* end Tprod */
1.237 brouard 8040: } else{ /* simple covariate */
1.264 brouard 8041: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8042: if(Dummy[j]==0){
8043: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8044: }else{ /* quantitative */
8045: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8046: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8047: }
1.237 brouard 8048: } /* end simple */
8049: } /* end j */
1.223 brouard 8050: }else{
8051: i=i-ncovmodel;
8052: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8053: fprintf(ficgp," (1.");
8054: }
1.227 brouard 8055:
1.223 brouard 8056: if(ng != 1){
8057: fprintf(ficgp,")/(1");
1.227 brouard 8058:
1.264 brouard 8059: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8060: if(nagesqr==0)
1.264 brouard 8061: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8062: else /* nagesqr =1 */
1.264 brouard 8063: 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 8064:
1.223 brouard 8065: ij=1;
8066: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8067: if(cptcovage >0){
8068: if((j-2)==Tage[ij]) { /* Bug valgrind */
8069: if(ij <=cptcovage) { /* Bug valgrind */
8070: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8071: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8072: ij++;
8073: }
8074: }
8075: }else
8076: 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 8077: }
8078: fprintf(ficgp,")");
8079: }
8080: fprintf(ficgp,")");
8081: if(ng ==2)
1.276 brouard 8082: 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 8083: else /* ng= 3 */
1.276 brouard 8084: 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 8085: }else{ /* end ng <> 1 */
8086: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8087: 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 8088: }
8089: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8090: fprintf(ficgp,",");
8091: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8092: fprintf(ficgp,",");
8093: i=i+ncovmodel;
8094: } /* end k */
8095: } /* end k2 */
1.276 brouard 8096: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8097: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8098: } /* end k1 */
1.223 brouard 8099: } /* end ng */
8100: /* avoid: */
8101: fflush(ficgp);
1.126 brouard 8102: } /* end gnuplot */
8103:
8104:
8105: /*************** Moving average **************/
1.219 brouard 8106: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8107: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8108:
1.222 brouard 8109: int i, cpt, cptcod;
8110: int modcovmax =1;
8111: int mobilavrange, mob;
8112: int iage=0;
1.288 brouard 8113: int firstA1=0, firstA2=0;
1.222 brouard 8114:
1.266 brouard 8115: double sum=0., sumr=0.;
1.222 brouard 8116: double age;
1.266 brouard 8117: double *sumnewp, *sumnewm, *sumnewmr;
8118: double *agemingood, *agemaxgood;
8119: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8120:
8121:
1.278 brouard 8122: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8123: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8124:
8125: sumnewp = vector(1,ncovcombmax);
8126: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8127: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8128: agemingood = vector(1,ncovcombmax);
1.266 brouard 8129: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8130: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8131: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8132:
8133: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8134: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8135: sumnewp[cptcod]=0.;
1.266 brouard 8136: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8137: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8138: }
8139: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8140:
1.266 brouard 8141: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8142: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8143: else mobilavrange=mobilav;
8144: for (age=bage; age<=fage; age++)
8145: for (i=1; i<=nlstate;i++)
8146: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8147: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8148: /* We keep the original values on the extreme ages bage, fage and for
8149: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8150: we use a 5 terms etc. until the borders are no more concerned.
8151: */
8152: for (mob=3;mob <=mobilavrange;mob=mob+2){
8153: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8154: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8155: sumnewm[cptcod]=0.;
8156: for (i=1; i<=nlstate;i++){
1.222 brouard 8157: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8158: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8159: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8160: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8161: }
8162: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8163: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8164: } /* end i */
8165: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8166: } /* end cptcod */
1.222 brouard 8167: }/* end age */
8168: }/* end mob */
1.266 brouard 8169: }else{
8170: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8171: return -1;
1.266 brouard 8172: }
8173:
8174: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8175: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8176: if(invalidvarcomb[cptcod]){
8177: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8178: continue;
8179: }
1.219 brouard 8180:
1.266 brouard 8181: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8182: sumnewm[cptcod]=0.;
8183: sumnewmr[cptcod]=0.;
8184: for (i=1; i<=nlstate;i++){
8185: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8186: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8187: }
8188: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8189: agemingoodr[cptcod]=age;
8190: }
8191: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8192: agemingood[cptcod]=age;
8193: }
8194: } /* age */
8195: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8196: sumnewm[cptcod]=0.;
1.266 brouard 8197: sumnewmr[cptcod]=0.;
1.222 brouard 8198: for (i=1; i<=nlstate;i++){
8199: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8200: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8201: }
8202: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8203: agemaxgoodr[cptcod]=age;
1.222 brouard 8204: }
8205: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8206: agemaxgood[cptcod]=age;
8207: }
8208: } /* age */
8209: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8210: /* but they will change */
1.288 brouard 8211: firstA1=0;firstA2=0;
1.266 brouard 8212: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8213: sumnewm[cptcod]=0.;
8214: sumnewmr[cptcod]=0.;
8215: for (i=1; i<=nlstate;i++){
8216: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8217: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8218: }
8219: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8220: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8221: agemaxgoodr[cptcod]=age; /* age min */
8222: for (i=1; i<=nlstate;i++)
8223: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8224: }else{ /* bad we change the value with the values of good ages */
8225: for (i=1; i<=nlstate;i++){
8226: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8227: } /* i */
8228: } /* end bad */
8229: }else{
8230: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8231: agemaxgood[cptcod]=age;
8232: }else{ /* bad we change the value with the values of good ages */
8233: for (i=1; i<=nlstate;i++){
8234: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8235: } /* i */
8236: } /* end bad */
8237: }/* end else */
8238: sum=0.;sumr=0.;
8239: for (i=1; i<=nlstate;i++){
8240: sum+=mobaverage[(int)age][i][cptcod];
8241: sumr+=probs[(int)age][i][cptcod];
8242: }
8243: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8244: if(!firstA1){
8245: firstA1=1;
8246: 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);
8247: }
8248: 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 8249: } /* end bad */
8250: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8251: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8252: if(!firstA2){
8253: firstA2=1;
8254: 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);
8255: }
8256: 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 8257: } /* end bad */
8258: }/* age */
1.266 brouard 8259:
8260: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8261: sumnewm[cptcod]=0.;
1.266 brouard 8262: sumnewmr[cptcod]=0.;
1.222 brouard 8263: for (i=1; i<=nlstate;i++){
8264: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8265: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8266: }
8267: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8268: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8269: agemingoodr[cptcod]=age;
8270: for (i=1; i<=nlstate;i++)
8271: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8272: }else{ /* bad we change the value with the values of good ages */
8273: for (i=1; i<=nlstate;i++){
8274: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8275: } /* i */
8276: } /* end bad */
8277: }else{
8278: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8279: agemingood[cptcod]=age;
8280: }else{ /* bad */
8281: for (i=1; i<=nlstate;i++){
8282: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8283: } /* i */
8284: } /* end bad */
8285: }/* end else */
8286: sum=0.;sumr=0.;
8287: for (i=1; i<=nlstate;i++){
8288: sum+=mobaverage[(int)age][i][cptcod];
8289: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8290: }
1.266 brouard 8291: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8292: 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 8293: } /* end bad */
8294: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8295: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8296: 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 8297: } /* end bad */
8298: }/* age */
1.266 brouard 8299:
1.222 brouard 8300:
8301: for (age=bage; age<=fage; age++){
1.235 brouard 8302: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8303: sumnewp[cptcod]=0.;
8304: sumnewm[cptcod]=0.;
8305: for (i=1; i<=nlstate;i++){
8306: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8307: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8308: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8309: }
8310: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8311: }
8312: /* printf("\n"); */
8313: /* } */
1.266 brouard 8314:
1.222 brouard 8315: /* brutal averaging */
1.266 brouard 8316: /* for (i=1; i<=nlstate;i++){ */
8317: /* for (age=1; age<=bage; age++){ */
8318: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8319: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8320: /* } */
8321: /* for (age=fage; age<=AGESUP; age++){ */
8322: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8323: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8324: /* } */
8325: /* } /\* end i status *\/ */
8326: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8327: /* for (age=1; age<=AGESUP; age++){ */
8328: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8329: /* mobaverage[(int)age][i][cptcod]=0.; */
8330: /* } */
8331: /* } */
1.222 brouard 8332: }/* end cptcod */
1.266 brouard 8333: free_vector(agemaxgoodr,1, ncovcombmax);
8334: free_vector(agemaxgood,1, ncovcombmax);
8335: free_vector(agemingood,1, ncovcombmax);
8336: free_vector(agemingoodr,1, ncovcombmax);
8337: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8338: free_vector(sumnewm,1, ncovcombmax);
8339: free_vector(sumnewp,1, ncovcombmax);
8340: return 0;
8341: }/* End movingaverage */
1.218 brouard 8342:
1.126 brouard 8343:
1.296 brouard 8344:
1.126 brouard 8345: /************** Forecasting ******************/
1.296 brouard 8346: /* 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)*/
8347: 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){
8348: /* dateintemean, mean date of interviews
8349: dateprojd, year, month, day of starting projection
8350: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8351: agemin, agemax range of age
8352: dateprev1 dateprev2 range of dates during which prevalence is computed
8353: */
1.296 brouard 8354: /* double anprojd, mprojd, jprojd; */
8355: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8356: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8357: double agec; /* generic age */
1.296 brouard 8358: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8359: double *popeffectif,*popcount;
8360: double ***p3mat;
1.218 brouard 8361: /* double ***mobaverage; */
1.126 brouard 8362: char fileresf[FILENAMELENGTH];
8363:
8364: agelim=AGESUP;
1.211 brouard 8365: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8366: in each health status at the date of interview (if between dateprev1 and dateprev2).
8367: We still use firstpass and lastpass as another selection.
8368: */
1.214 brouard 8369: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8370: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8371:
1.201 brouard 8372: strcpy(fileresf,"F_");
8373: strcat(fileresf,fileresu);
1.126 brouard 8374: if((ficresf=fopen(fileresf,"w"))==NULL) {
8375: printf("Problem with forecast resultfile: %s\n", fileresf);
8376: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8377: }
1.235 brouard 8378: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8379: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8380:
1.225 brouard 8381: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8382:
8383:
8384: stepsize=(int) (stepm+YEARM-1)/YEARM;
8385: if (stepm<=12) stepsize=1;
8386: if(estepm < stepm){
8387: printf ("Problem %d lower than %d\n",estepm, stepm);
8388: }
1.270 brouard 8389: else{
8390: hstepm=estepm;
8391: }
8392: if(estepm > stepm){ /* Yes every two year */
8393: stepsize=2;
8394: }
1.296 brouard 8395: hstepm=hstepm/stepm;
1.126 brouard 8396:
1.296 brouard 8397:
8398: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8399: /* fractional in yp1 *\/ */
8400: /* aintmean=yp; */
8401: /* yp2=modf((yp1*12),&yp); */
8402: /* mintmean=yp; */
8403: /* yp1=modf((yp2*30.5),&yp); */
8404: /* jintmean=yp; */
8405: /* if(jintmean==0) jintmean=1; */
8406: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8407:
1.296 brouard 8408:
8409: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8410: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8411: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8412: i1=pow(2,cptcoveff);
1.126 brouard 8413: if (cptcovn < 1){i1=1;}
8414:
1.296 brouard 8415: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8416:
8417: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8418:
1.126 brouard 8419: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8420: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8421: for(k=1; k<=i1;k++){
1.253 brouard 8422: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8423: continue;
1.227 brouard 8424: if(invalidvarcomb[k]){
8425: printf("\nCombination (%d) projection ignored because no cases \n",k);
8426: continue;
8427: }
8428: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8429: for(j=1;j<=cptcoveff;j++) {
8430: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8431: }
1.235 brouard 8432: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8433: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8434: }
1.227 brouard 8435: fprintf(ficresf," yearproj age");
8436: for(j=1; j<=nlstate+ndeath;j++){
8437: for(i=1; i<=nlstate;i++)
8438: fprintf(ficresf," p%d%d",i,j);
8439: fprintf(ficresf," wp.%d",j);
8440: }
1.296 brouard 8441: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8442: fprintf(ficresf,"\n");
1.296 brouard 8443: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8444: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8445: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8446: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8447: nhstepm = nhstepm/hstepm;
8448: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8449: oldm=oldms;savm=savms;
1.268 brouard 8450: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8451: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8452: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8453: for (h=0; h<=nhstepm; h++){
8454: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8455: break;
8456: }
8457: }
8458: fprintf(ficresf,"\n");
8459: for(j=1;j<=cptcoveff;j++)
8460: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8461: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8462:
8463: for(j=1; j<=nlstate+ndeath;j++) {
8464: ppij=0.;
8465: for(i=1; i<=nlstate;i++) {
1.278 brouard 8466: if (mobilav>=1)
8467: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8468: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8469: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8470: }
1.268 brouard 8471: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8472: } /* end i */
8473: fprintf(ficresf," %.3f", ppij);
8474: }/* end j */
1.227 brouard 8475: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8476: } /* end agec */
1.266 brouard 8477: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8478: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8479: } /* end yearp */
8480: } /* end k */
1.219 brouard 8481:
1.126 brouard 8482: fclose(ficresf);
1.215 brouard 8483: printf("End of Computing forecasting \n");
8484: fprintf(ficlog,"End of Computing forecasting\n");
8485:
1.126 brouard 8486: }
8487:
1.269 brouard 8488: /************** Back Forecasting ******************/
1.296 brouard 8489: /* 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){ */
8490: 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){
8491: /* back1, year, month, day of starting backprojection
1.267 brouard 8492: agemin, agemax range of age
8493: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8494: anback2 year of end of backprojection (same day and month as back1).
8495: prevacurrent and prev are prevalences.
1.267 brouard 8496: */
8497: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8498: double agec; /* generic age */
1.296 brouard 8499: double agelim, ppij, ppi, yp,yp1,yp2,jintmean,mintmean,aintmean;
1.267 brouard 8500: double *popeffectif,*popcount;
8501: double ***p3mat;
8502: /* double ***mobaverage; */
8503: char fileresfb[FILENAMELENGTH];
8504:
1.268 brouard 8505: agelim=AGEINF;
1.267 brouard 8506: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8507: in each health status at the date of interview (if between dateprev1 and dateprev2).
8508: We still use firstpass and lastpass as another selection.
8509: */
8510: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8511: /* firstpass, lastpass, stepm, weightopt, model); */
8512:
8513: /*Do we need to compute prevalence again?*/
8514:
8515: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8516:
8517: strcpy(fileresfb,"FB_");
8518: strcat(fileresfb,fileresu);
8519: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8520: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8521: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8522: }
8523: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8524: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8525:
8526: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8527:
8528:
8529: stepsize=(int) (stepm+YEARM-1)/YEARM;
8530: if (stepm<=12) stepsize=1;
8531: if(estepm < stepm){
8532: printf ("Problem %d lower than %d\n",estepm, stepm);
8533: }
1.270 brouard 8534: else{
8535: hstepm=estepm;
8536: }
8537: if(estepm >= stepm){ /* Yes every two year */
8538: stepsize=2;
8539: }
1.267 brouard 8540:
8541: hstepm=hstepm/stepm;
1.296 brouard 8542: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8543: /* fractional in yp1 *\/ */
8544: /* aintmean=yp; */
8545: /* yp2=modf((yp1*12),&yp); */
8546: /* mintmean=yp; */
8547: /* yp1=modf((yp2*30.5),&yp); */
8548: /* jintmean=yp; */
8549: /* if(jintmean==0) jintmean=1; */
8550: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8551:
8552: i1=pow(2,cptcoveff);
8553: if (cptcovn < 1){i1=1;}
8554:
1.296 brouard 8555: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8556: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8557:
8558: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8559:
8560: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8561: for(k=1; k<=i1;k++){
8562: if(i1 != 1 && TKresult[nres]!= k)
8563: continue;
8564: if(invalidvarcomb[k]){
8565: printf("\nCombination (%d) projection ignored because no cases \n",k);
8566: continue;
8567: }
1.268 brouard 8568: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8569: for(j=1;j<=cptcoveff;j++) {
8570: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8571: }
8572: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8573: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8574: }
8575: fprintf(ficresfb," yearbproj age");
8576: for(j=1; j<=nlstate+ndeath;j++){
8577: for(i=1; i<=nlstate;i++)
1.268 brouard 8578: fprintf(ficresfb," b%d%d",i,j);
8579: fprintf(ficresfb," b.%d",j);
1.267 brouard 8580: }
1.296 brouard 8581: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8582: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8583: fprintf(ficresfb,"\n");
1.296 brouard 8584: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8585: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8586: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8587: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8588: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8589: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8590: nhstepm = nhstepm/hstepm;
8591: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8592: oldm=oldms;savm=savms;
1.268 brouard 8593: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8594: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8595: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8596: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8597: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8598: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8599: for (h=0; h<=nhstepm; h++){
1.268 brouard 8600: if (h*hstepm/YEARM*stepm ==-yearp) {
8601: break;
8602: }
8603: }
8604: fprintf(ficresfb,"\n");
8605: for(j=1;j<=cptcoveff;j++)
8606: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8607: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8608: for(i=1; i<=nlstate+ndeath;i++) {
8609: ppij=0.;ppi=0.;
8610: for(j=1; j<=nlstate;j++) {
8611: /* if (mobilav==1) */
1.269 brouard 8612: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8613: ppi=ppi+prevacurrent[(int)agec][j][k];
8614: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8615: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8616: /* else { */
8617: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8618: /* } */
1.268 brouard 8619: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8620: } /* end j */
8621: if(ppi <0.99){
8622: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8623: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8624: }
8625: fprintf(ficresfb," %.3f", ppij);
8626: }/* end j */
1.267 brouard 8627: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8628: } /* end agec */
8629: } /* end yearp */
8630: } /* end k */
1.217 brouard 8631:
1.267 brouard 8632: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8633:
1.267 brouard 8634: fclose(ficresfb);
8635: printf("End of Computing Back forecasting \n");
8636: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8637:
1.267 brouard 8638: }
1.217 brouard 8639:
1.269 brouard 8640: /* Variance of prevalence limit: varprlim */
8641: 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 8642: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8643:
8644: char fileresvpl[FILENAMELENGTH];
8645: FILE *ficresvpl;
8646: double **oldm, **savm;
8647: double **varpl; /* Variances of prevalence limits by age */
8648: int i1, k, nres, j ;
8649:
8650: strcpy(fileresvpl,"VPL_");
8651: strcat(fileresvpl,fileresu);
8652: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8653: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8654: exit(0);
8655: }
1.288 brouard 8656: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8657: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8658:
8659: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8660: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8661:
8662: i1=pow(2,cptcoveff);
8663: if (cptcovn < 1){i1=1;}
8664:
8665: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8666: for(k=1; k<=i1;k++){
8667: if(i1 != 1 && TKresult[nres]!= k)
8668: continue;
8669: fprintf(ficresvpl,"\n#****** ");
8670: printf("\n#****** ");
8671: fprintf(ficlog,"\n#****** ");
8672: for(j=1;j<=cptcoveff;j++) {
8673: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8674: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8675: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8676: }
8677: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8678: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8679: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8680: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8681: }
8682: fprintf(ficresvpl,"******\n");
8683: printf("******\n");
8684: fprintf(ficlog,"******\n");
8685:
8686: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8687: oldm=oldms;savm=savms;
8688: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8689: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8690: /*}*/
8691: }
8692:
8693: fclose(ficresvpl);
1.288 brouard 8694: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8695: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8696:
8697: }
8698: /* Variance of back prevalence: varbprlim */
8699: 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){
8700: /*------- Variance of back (stable) prevalence------*/
8701:
8702: char fileresvbl[FILENAMELENGTH];
8703: FILE *ficresvbl;
8704:
8705: double **oldm, **savm;
8706: double **varbpl; /* Variances of back prevalence limits by age */
8707: int i1, k, nres, j ;
8708:
8709: strcpy(fileresvbl,"VBL_");
8710: strcat(fileresvbl,fileresu);
8711: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8712: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8713: exit(0);
8714: }
8715: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8716: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8717:
8718:
8719: i1=pow(2,cptcoveff);
8720: if (cptcovn < 1){i1=1;}
8721:
8722: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8723: for(k=1; k<=i1;k++){
8724: if(i1 != 1 && TKresult[nres]!= k)
8725: continue;
8726: fprintf(ficresvbl,"\n#****** ");
8727: printf("\n#****** ");
8728: fprintf(ficlog,"\n#****** ");
8729: for(j=1;j<=cptcoveff;j++) {
8730: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8731: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8732: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8733: }
8734: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8735: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8736: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8737: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8738: }
8739: fprintf(ficresvbl,"******\n");
8740: printf("******\n");
8741: fprintf(ficlog,"******\n");
8742:
8743: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8744: oldm=oldms;savm=savms;
8745:
8746: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8747: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8748: /*}*/
8749: }
8750:
8751: fclose(ficresvbl);
8752: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8753: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8754:
8755: } /* End of varbprlim */
8756:
1.126 brouard 8757: /************** Forecasting *****not tested NB*************/
1.227 brouard 8758: /* 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 8759:
1.227 brouard 8760: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8761: /* int *popage; */
8762: /* double calagedatem, agelim, kk1, kk2; */
8763: /* double *popeffectif,*popcount; */
8764: /* double ***p3mat,***tabpop,***tabpopprev; */
8765: /* /\* double ***mobaverage; *\/ */
8766: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8767:
1.227 brouard 8768: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8769: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8770: /* agelim=AGESUP; */
8771: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8772:
1.227 brouard 8773: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8774:
8775:
1.227 brouard 8776: /* strcpy(filerespop,"POP_"); */
8777: /* strcat(filerespop,fileresu); */
8778: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8779: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8780: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8781: /* } */
8782: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8783: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8784:
1.227 brouard 8785: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8786:
1.227 brouard 8787: /* /\* if (mobilav!=0) { *\/ */
8788: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8789: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8790: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8791: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8792: /* /\* } *\/ */
8793: /* /\* } *\/ */
1.126 brouard 8794:
1.227 brouard 8795: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8796: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8797:
1.227 brouard 8798: /* agelim=AGESUP; */
1.126 brouard 8799:
1.227 brouard 8800: /* hstepm=1; */
8801: /* hstepm=hstepm/stepm; */
1.218 brouard 8802:
1.227 brouard 8803: /* if (popforecast==1) { */
8804: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8805: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8806: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8807: /* } */
8808: /* popage=ivector(0,AGESUP); */
8809: /* popeffectif=vector(0,AGESUP); */
8810: /* popcount=vector(0,AGESUP); */
1.126 brouard 8811:
1.227 brouard 8812: /* i=1; */
8813: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8814:
1.227 brouard 8815: /* imx=i; */
8816: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8817: /* } */
1.218 brouard 8818:
1.227 brouard 8819: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8820: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8821: /* k=k+1; */
8822: /* fprintf(ficrespop,"\n#******"); */
8823: /* for(j=1;j<=cptcoveff;j++) { */
8824: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8825: /* } */
8826: /* fprintf(ficrespop,"******\n"); */
8827: /* fprintf(ficrespop,"# Age"); */
8828: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8829: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8830:
1.227 brouard 8831: /* for (cpt=0; cpt<=0;cpt++) { */
8832: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8833:
1.227 brouard 8834: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8835: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8836: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8837:
1.227 brouard 8838: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8839: /* oldm=oldms;savm=savms; */
8840: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8841:
1.227 brouard 8842: /* for (h=0; h<=nhstepm; h++){ */
8843: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8844: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8845: /* } */
8846: /* for(j=1; j<=nlstate+ndeath;j++) { */
8847: /* kk1=0.;kk2=0; */
8848: /* for(i=1; i<=nlstate;i++) { */
8849: /* if (mobilav==1) */
8850: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8851: /* else { */
8852: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8853: /* } */
8854: /* } */
8855: /* if (h==(int)(calagedatem+12*cpt)){ */
8856: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8857: /* /\*fprintf(ficrespop," %.3f", kk1); */
8858: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8859: /* } */
8860: /* } */
8861: /* for(i=1; i<=nlstate;i++){ */
8862: /* kk1=0.; */
8863: /* for(j=1; j<=nlstate;j++){ */
8864: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8865: /* } */
8866: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8867: /* } */
1.218 brouard 8868:
1.227 brouard 8869: /* if (h==(int)(calagedatem+12*cpt)) */
8870: /* for(j=1; j<=nlstate;j++) */
8871: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8872: /* } */
8873: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8874: /* } */
8875: /* } */
1.218 brouard 8876:
1.227 brouard 8877: /* /\******\/ */
1.218 brouard 8878:
1.227 brouard 8879: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8880: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8881: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8882: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8883: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8884:
1.227 brouard 8885: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8886: /* oldm=oldms;savm=savms; */
8887: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8888: /* for (h=0; h<=nhstepm; h++){ */
8889: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8890: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8891: /* } */
8892: /* for(j=1; j<=nlstate+ndeath;j++) { */
8893: /* kk1=0.;kk2=0; */
8894: /* for(i=1; i<=nlstate;i++) { */
8895: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8896: /* } */
8897: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8898: /* } */
8899: /* } */
8900: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8901: /* } */
8902: /* } */
8903: /* } */
8904: /* } */
1.218 brouard 8905:
1.227 brouard 8906: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8907:
1.227 brouard 8908: /* if (popforecast==1) { */
8909: /* free_ivector(popage,0,AGESUP); */
8910: /* free_vector(popeffectif,0,AGESUP); */
8911: /* free_vector(popcount,0,AGESUP); */
8912: /* } */
8913: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8914: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8915: /* fclose(ficrespop); */
8916: /* } /\* End of popforecast *\/ */
1.218 brouard 8917:
1.126 brouard 8918: int fileappend(FILE *fichier, char *optionfich)
8919: {
8920: if((fichier=fopen(optionfich,"a"))==NULL) {
8921: printf("Problem with file: %s\n", optionfich);
8922: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8923: return (0);
8924: }
8925: fflush(fichier);
8926: return (1);
8927: }
8928:
8929:
8930: /**************** function prwizard **********************/
8931: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8932: {
8933:
8934: /* Wizard to print covariance matrix template */
8935:
1.164 brouard 8936: char ca[32], cb[32];
8937: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8938: int numlinepar;
8939:
8940: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8941: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8942: for(i=1; i <=nlstate; i++){
8943: jj=0;
8944: for(j=1; j <=nlstate+ndeath; j++){
8945: if(j==i) continue;
8946: jj++;
8947: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8948: printf("%1d%1d",i,j);
8949: fprintf(ficparo,"%1d%1d",i,j);
8950: for(k=1; k<=ncovmodel;k++){
8951: /* printf(" %lf",param[i][j][k]); */
8952: /* fprintf(ficparo," %lf",param[i][j][k]); */
8953: printf(" 0.");
8954: fprintf(ficparo," 0.");
8955: }
8956: printf("\n");
8957: fprintf(ficparo,"\n");
8958: }
8959: }
8960: printf("# Scales (for hessian or gradient estimation)\n");
8961: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8962: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8963: for(i=1; i <=nlstate; i++){
8964: jj=0;
8965: for(j=1; j <=nlstate+ndeath; j++){
8966: if(j==i) continue;
8967: jj++;
8968: fprintf(ficparo,"%1d%1d",i,j);
8969: printf("%1d%1d",i,j);
8970: fflush(stdout);
8971: for(k=1; k<=ncovmodel;k++){
8972: /* printf(" %le",delti3[i][j][k]); */
8973: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8974: printf(" 0.");
8975: fprintf(ficparo," 0.");
8976: }
8977: numlinepar++;
8978: printf("\n");
8979: fprintf(ficparo,"\n");
8980: }
8981: }
8982: printf("# Covariance matrix\n");
8983: /* # 121 Var(a12)\n\ */
8984: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8985: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8986: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8987: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8988: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8989: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8990: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8991: fflush(stdout);
8992: fprintf(ficparo,"# Covariance matrix\n");
8993: /* # 121 Var(a12)\n\ */
8994: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8995: /* # ...\n\ */
8996: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8997:
8998: for(itimes=1;itimes<=2;itimes++){
8999: jj=0;
9000: for(i=1; i <=nlstate; i++){
9001: for(j=1; j <=nlstate+ndeath; j++){
9002: if(j==i) continue;
9003: for(k=1; k<=ncovmodel;k++){
9004: jj++;
9005: ca[0]= k+'a'-1;ca[1]='\0';
9006: if(itimes==1){
9007: printf("#%1d%1d%d",i,j,k);
9008: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9009: }else{
9010: printf("%1d%1d%d",i,j,k);
9011: fprintf(ficparo,"%1d%1d%d",i,j,k);
9012: /* printf(" %.5le",matcov[i][j]); */
9013: }
9014: ll=0;
9015: for(li=1;li <=nlstate; li++){
9016: for(lj=1;lj <=nlstate+ndeath; lj++){
9017: if(lj==li) continue;
9018: for(lk=1;lk<=ncovmodel;lk++){
9019: ll++;
9020: if(ll<=jj){
9021: cb[0]= lk +'a'-1;cb[1]='\0';
9022: if(ll<jj){
9023: if(itimes==1){
9024: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9025: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9026: }else{
9027: printf(" 0.");
9028: fprintf(ficparo," 0.");
9029: }
9030: }else{
9031: if(itimes==1){
9032: printf(" Var(%s%1d%1d)",ca,i,j);
9033: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9034: }else{
9035: printf(" 0.");
9036: fprintf(ficparo," 0.");
9037: }
9038: }
9039: }
9040: } /* end lk */
9041: } /* end lj */
9042: } /* end li */
9043: printf("\n");
9044: fprintf(ficparo,"\n");
9045: numlinepar++;
9046: } /* end k*/
9047: } /*end j */
9048: } /* end i */
9049: } /* end itimes */
9050:
9051: } /* end of prwizard */
9052: /******************* Gompertz Likelihood ******************************/
9053: double gompertz(double x[])
9054: {
9055: double A,B,L=0.0,sump=0.,num=0.;
9056: int i,n=0; /* n is the size of the sample */
9057:
1.220 brouard 9058: for (i=1;i<=imx ; i++) {
1.126 brouard 9059: sump=sump+weight[i];
9060: /* sump=sump+1;*/
9061: num=num+1;
9062: }
9063:
9064:
9065: /* for (i=0; i<=imx; i++)
9066: 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]);*/
9067:
9068: for (i=1;i<=imx ; i++)
9069: {
9070: if (cens[i] == 1 && wav[i]>1)
9071: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9072:
9073: if (cens[i] == 0 && wav[i]>1)
9074: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
9075: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
9076:
9077: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9078: if (wav[i] > 1 ) { /* ??? */
9079: L=L+A*weight[i];
9080: /* 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]);*/
9081: }
9082: }
9083:
9084: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9085:
9086: return -2*L*num/sump;
9087: }
9088:
1.136 brouard 9089: #ifdef GSL
9090: /******************* Gompertz_f Likelihood ******************************/
9091: double gompertz_f(const gsl_vector *v, void *params)
9092: {
9093: double A,B,LL=0.0,sump=0.,num=0.;
9094: double *x= (double *) v->data;
9095: int i,n=0; /* n is the size of the sample */
9096:
9097: for (i=0;i<=imx-1 ; i++) {
9098: sump=sump+weight[i];
9099: /* sump=sump+1;*/
9100: num=num+1;
9101: }
9102:
9103:
9104: /* for (i=0; i<=imx; i++)
9105: 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]);*/
9106: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9107: for (i=1;i<=imx ; i++)
9108: {
9109: if (cens[i] == 1 && wav[i]>1)
9110: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9111:
9112: if (cens[i] == 0 && wav[i]>1)
9113: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9114: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9115:
9116: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9117: if (wav[i] > 1 ) { /* ??? */
9118: LL=LL+A*weight[i];
9119: /* 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]);*/
9120: }
9121: }
9122:
9123: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9124: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9125:
9126: return -2*LL*num/sump;
9127: }
9128: #endif
9129:
1.126 brouard 9130: /******************* Printing html file ***********/
1.201 brouard 9131: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9132: int lastpass, int stepm, int weightopt, char model[],\
9133: int imx, double p[],double **matcov,double agemortsup){
9134: int i,k;
9135:
9136: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9137: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9138: for (i=1;i<=2;i++)
9139: 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 9140: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9141: fprintf(fichtm,"</ul>");
9142:
9143: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9144:
9145: 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>");
9146:
9147: for (k=agegomp;k<(agemortsup-2);k++)
9148: 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]);
9149:
9150:
9151: fflush(fichtm);
9152: }
9153:
9154: /******************* Gnuplot file **************/
1.201 brouard 9155: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9156:
9157: char dirfileres[132],optfileres[132];
1.164 brouard 9158:
1.126 brouard 9159: int ng;
9160:
9161:
9162: /*#ifdef windows */
9163: fprintf(ficgp,"cd \"%s\" \n",pathc);
9164: /*#endif */
9165:
9166:
9167: strcpy(dirfileres,optionfilefiname);
9168: strcpy(optfileres,"vpl");
1.199 brouard 9169: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9170: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9171: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9172: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9173: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9174:
9175: }
9176:
1.136 brouard 9177: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9178: {
1.126 brouard 9179:
1.136 brouard 9180: /*-------- data file ----------*/
9181: FILE *fic;
9182: char dummy[]=" ";
1.240 brouard 9183: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9184: int lstra;
1.136 brouard 9185: int linei, month, year,iout;
9186: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9187: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9188: char *stratrunc;
1.223 brouard 9189:
1.240 brouard 9190: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9191: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9192:
1.240 brouard 9193: for(v=1; v <=ncovcol;v++){
9194: DummyV[v]=0;
9195: FixedV[v]=0;
9196: }
9197: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9198: DummyV[v]=1;
9199: FixedV[v]=0;
9200: }
9201: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9202: DummyV[v]=0;
9203: FixedV[v]=1;
9204: }
9205: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9206: DummyV[v]=1;
9207: FixedV[v]=1;
9208: }
9209: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9210: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9211: 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]);
9212: }
1.126 brouard 9213:
1.136 brouard 9214: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9215: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9216: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9217: }
1.126 brouard 9218:
1.136 brouard 9219: i=1;
9220: linei=0;
9221: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9222: linei=linei+1;
9223: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9224: if(line[j] == '\t')
9225: line[j] = ' ';
9226: }
9227: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9228: ;
9229: };
9230: line[j+1]=0; /* Trims blanks at end of line */
9231: if(line[0]=='#'){
9232: fprintf(ficlog,"Comment line\n%s\n",line);
9233: printf("Comment line\n%s\n",line);
9234: continue;
9235: }
9236: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9237: strcpy(line, linetmp);
1.223 brouard 9238:
9239: /* Loops on waves */
9240: for (j=maxwav;j>=1;j--){
9241: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9242: cutv(stra, strb, line, ' ');
9243: if(strb[0]=='.') { /* Missing value */
9244: lval=-1;
9245: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9246: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9247: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9248: 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);
9249: 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);
9250: return 1;
9251: }
9252: }else{
9253: errno=0;
9254: /* what_kind_of_number(strb); */
9255: dval=strtod(strb,&endptr);
9256: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9257: /* if(strb != endptr && *endptr == '\0') */
9258: /* dval=dlval; */
9259: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9260: if( strb[0]=='\0' || (*endptr != '\0')){
9261: 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);
9262: 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);
9263: return 1;
9264: }
9265: cotqvar[j][iv][i]=dval;
9266: cotvar[j][ntv+iv][i]=dval;
9267: }
9268: strcpy(line,stra);
1.223 brouard 9269: }/* end loop ntqv */
1.225 brouard 9270:
1.223 brouard 9271: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9272: cutv(stra, strb, line, ' ');
9273: if(strb[0]=='.') { /* Missing value */
9274: lval=-1;
9275: }else{
9276: errno=0;
9277: lval=strtol(strb,&endptr,10);
9278: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9279: if( strb[0]=='\0' || (*endptr != '\0')){
9280: 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);
9281: 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);
9282: return 1;
9283: }
9284: }
9285: if(lval <-1 || lval >1){
9286: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9287: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9288: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9289: For example, for multinomial values like 1, 2 and 3,\n \
9290: build V1=0 V2=0 for the reference value (1),\n \
9291: V1=1 V2=0 for (2) \n \
1.223 brouard 9292: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9293: output of IMaCh is often meaningless.\n \
1.223 brouard 9294: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9295: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9296: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9297: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9298: For example, for multinomial values like 1, 2 and 3,\n \
9299: build V1=0 V2=0 for the reference value (1),\n \
9300: V1=1 V2=0 for (2) \n \
1.223 brouard 9301: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9302: output of IMaCh is often meaningless.\n \
1.223 brouard 9303: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9304: return 1;
9305: }
9306: cotvar[j][iv][i]=(double)(lval);
9307: strcpy(line,stra);
1.223 brouard 9308: }/* end loop ntv */
1.225 brouard 9309:
1.223 brouard 9310: /* Statuses at wave */
1.137 brouard 9311: cutv(stra, strb, line, ' ');
1.223 brouard 9312: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9313: lval=-1;
1.136 brouard 9314: }else{
1.238 brouard 9315: errno=0;
9316: lval=strtol(strb,&endptr,10);
9317: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9318: if( strb[0]=='\0' || (*endptr != '\0')){
9319: 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);
9320: 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);
9321: return 1;
9322: }
1.136 brouard 9323: }
1.225 brouard 9324:
1.136 brouard 9325: s[j][i]=lval;
1.225 brouard 9326:
1.223 brouard 9327: /* Date of Interview */
1.136 brouard 9328: strcpy(line,stra);
9329: cutv(stra, strb,line,' ');
1.169 brouard 9330: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9331: }
1.169 brouard 9332: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9333: month=99;
9334: year=9999;
1.136 brouard 9335: }else{
1.225 brouard 9336: 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);
9337: 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);
9338: return 1;
1.136 brouard 9339: }
9340: anint[j][i]= (double) year;
9341: mint[j][i]= (double)month;
9342: strcpy(line,stra);
1.223 brouard 9343: } /* End loop on waves */
1.225 brouard 9344:
1.223 brouard 9345: /* Date of death */
1.136 brouard 9346: cutv(stra, strb,line,' ');
1.169 brouard 9347: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9348: }
1.169 brouard 9349: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9350: month=99;
9351: year=9999;
9352: }else{
1.141 brouard 9353: 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 9354: 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);
9355: return 1;
1.136 brouard 9356: }
9357: andc[i]=(double) year;
9358: moisdc[i]=(double) month;
9359: strcpy(line,stra);
9360:
1.223 brouard 9361: /* Date of birth */
1.136 brouard 9362: cutv(stra, strb,line,' ');
1.169 brouard 9363: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9364: }
1.169 brouard 9365: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9366: month=99;
9367: year=9999;
9368: }else{
1.141 brouard 9369: 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);
9370: 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 9371: return 1;
1.136 brouard 9372: }
9373: if (year==9999) {
1.141 brouard 9374: 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);
9375: 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 9376: return 1;
9377:
1.136 brouard 9378: }
9379: annais[i]=(double)(year);
9380: moisnais[i]=(double)(month);
9381: strcpy(line,stra);
1.225 brouard 9382:
1.223 brouard 9383: /* Sample weight */
1.136 brouard 9384: cutv(stra, strb,line,' ');
9385: errno=0;
9386: dval=strtod(strb,&endptr);
9387: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9388: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9389: 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 9390: fflush(ficlog);
9391: return 1;
9392: }
9393: weight[i]=dval;
9394: strcpy(line,stra);
1.225 brouard 9395:
1.223 brouard 9396: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9397: cutv(stra, strb, line, ' ');
9398: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9399: lval=-1;
1.223 brouard 9400: }else{
1.225 brouard 9401: errno=0;
9402: /* what_kind_of_number(strb); */
9403: dval=strtod(strb,&endptr);
9404: /* if(strb != endptr && *endptr == '\0') */
9405: /* dval=dlval; */
9406: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9407: if( strb[0]=='\0' || (*endptr != '\0')){
9408: 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);
9409: 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);
9410: return 1;
9411: }
9412: coqvar[iv][i]=dval;
1.226 brouard 9413: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9414: }
9415: strcpy(line,stra);
9416: }/* end loop nqv */
1.136 brouard 9417:
1.223 brouard 9418: /* Covariate values */
1.136 brouard 9419: for (j=ncovcol;j>=1;j--){
9420: cutv(stra, strb,line,' ');
1.223 brouard 9421: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9422: lval=-1;
1.136 brouard 9423: }else{
1.225 brouard 9424: errno=0;
9425: lval=strtol(strb,&endptr,10);
9426: if( strb[0]=='\0' || (*endptr != '\0')){
9427: 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);
9428: 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);
9429: return 1;
9430: }
1.136 brouard 9431: }
9432: if(lval <-1 || lval >1){
1.225 brouard 9433: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9434: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9435: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9436: For example, for multinomial values like 1, 2 and 3,\n \
9437: build V1=0 V2=0 for the reference value (1),\n \
9438: V1=1 V2=0 for (2) \n \
1.136 brouard 9439: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9440: output of IMaCh is often meaningless.\n \
1.136 brouard 9441: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9442: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9443: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9444: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9445: For example, for multinomial values like 1, 2 and 3,\n \
9446: build V1=0 V2=0 for the reference value (1),\n \
9447: V1=1 V2=0 for (2) \n \
1.136 brouard 9448: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9449: output of IMaCh is often meaningless.\n \
1.136 brouard 9450: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9451: return 1;
1.136 brouard 9452: }
9453: covar[j][i]=(double)(lval);
9454: strcpy(line,stra);
9455: }
9456: lstra=strlen(stra);
1.225 brouard 9457:
1.136 brouard 9458: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9459: stratrunc = &(stra[lstra-9]);
9460: num[i]=atol(stratrunc);
9461: }
9462: else
9463: num[i]=atol(stra);
9464: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9465: 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;}*/
9466:
9467: i=i+1;
9468: } /* End loop reading data */
1.225 brouard 9469:
1.136 brouard 9470: *imax=i-1; /* Number of individuals */
9471: fclose(fic);
1.225 brouard 9472:
1.136 brouard 9473: return (0);
1.164 brouard 9474: /* endread: */
1.225 brouard 9475: printf("Exiting readdata: ");
9476: fclose(fic);
9477: return (1);
1.223 brouard 9478: }
1.126 brouard 9479:
1.234 brouard 9480: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9481: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9482: while (*p2 == ' ')
1.234 brouard 9483: p2++;
9484: /* while ((*p1++ = *p2++) !=0) */
9485: /* ; */
9486: /* do */
9487: /* while (*p2 == ' ') */
9488: /* p2++; */
9489: /* while (*p1++ == *p2++); */
9490: *stri=p2;
1.145 brouard 9491: }
9492:
1.235 brouard 9493: int decoderesult ( char resultline[], int nres)
1.230 brouard 9494: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9495: {
1.235 brouard 9496: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9497: char resultsav[MAXLINE];
1.234 brouard 9498: int resultmodel[MAXLINE];
9499: int modelresult[MAXLINE];
1.230 brouard 9500: char stra[80], strb[80], strc[80], strd[80],stre[80];
9501:
1.234 brouard 9502: removefirstspace(&resultline);
1.233 brouard 9503: printf("decoderesult:%s\n",resultline);
1.230 brouard 9504:
9505: if (strstr(resultline,"v") !=0){
9506: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9507: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9508: return 1;
9509: }
9510: trimbb(resultsav, resultline);
9511: if (strlen(resultsav) >1){
9512: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9513: }
1.253 brouard 9514: if(j == 0){ /* Resultline but no = */
9515: TKresult[nres]=0; /* Combination for the nresult and the model */
9516: return (0);
9517: }
9518:
1.234 brouard 9519: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9520: 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);
9521: 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);
9522: }
9523: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9524: if(nbocc(resultsav,'=') >1){
9525: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9526: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9527: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9528: }else
9529: cutl(strc,strd,resultsav,'=');
1.230 brouard 9530: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9531:
1.230 brouard 9532: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9533: Tvarsel[k]=atoi(strc);
9534: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9535: /* cptcovsel++; */
9536: if (nbocc(stra,'=') >0)
9537: strcpy(resultsav,stra); /* and analyzes it */
9538: }
1.235 brouard 9539: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9540: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9541: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9542: match=0;
1.236 brouard 9543: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9544: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9545: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9546: match=1;
9547: break;
9548: }
9549: }
9550: if(match == 0){
9551: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9552: }
9553: }
9554: }
1.235 brouard 9555: /* Checking for missing or useless values in comparison of current model needs */
9556: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9557: match=0;
1.235 brouard 9558: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9559: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9560: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9561: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9562: ++match;
9563: }
9564: }
9565: }
9566: if(match == 0){
9567: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9568: }else if(match > 1){
9569: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9570: }
9571: }
1.235 brouard 9572:
1.234 brouard 9573: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9574: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9575: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9576: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9577: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9578: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9579: /* 1 0 0 0 */
9580: /* 2 1 0 0 */
9581: /* 3 0 1 0 */
9582: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9583: /* 5 0 0 1 */
9584: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9585: /* 7 0 1 1 */
9586: /* 8 1 1 1 */
1.237 brouard 9587: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9588: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9589: /* V5*age V5 known which value for nres? */
9590: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9591: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9592: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9593: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9594: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9595: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9596: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9597: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9598: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9599: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9600: k4++;;
9601: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9602: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9603: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9604: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9605: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9606: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9607: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9608: k4q++;;
9609: }
9610: }
1.234 brouard 9611:
1.235 brouard 9612: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9613: return (0);
9614: }
1.235 brouard 9615:
1.230 brouard 9616: int decodemodel( char model[], int lastobs)
9617: /**< This routine decodes the model and returns:
1.224 brouard 9618: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9619: * - nagesqr = 1 if age*age in the model, otherwise 0.
9620: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9621: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9622: * - cptcovage number of covariates with age*products =2
9623: * - cptcovs number of simple covariates
9624: * - 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
9625: * which is a new column after the 9 (ncovcol) variables.
9626: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9627: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9628: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9629: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9630: */
1.136 brouard 9631: {
1.238 brouard 9632: int i, j, k, ks, v;
1.227 brouard 9633: int j1, k1, k2, k3, k4;
1.136 brouard 9634: char modelsav[80];
1.145 brouard 9635: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9636: char *strpt;
1.136 brouard 9637:
1.145 brouard 9638: /*removespace(model);*/
1.136 brouard 9639: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9640: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9641: if (strstr(model,"AGE") !=0){
1.192 brouard 9642: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9643: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9644: return 1;
9645: }
1.141 brouard 9646: if (strstr(model,"v") !=0){
9647: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9648: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9649: return 1;
9650: }
1.187 brouard 9651: strcpy(modelsav,model);
9652: if ((strpt=strstr(model,"age*age")) !=0){
9653: printf(" strpt=%s, model=%s\n",strpt, model);
9654: if(strpt != model){
1.234 brouard 9655: printf("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);
1.234 brouard 9658: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9659: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9660: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9661: return 1;
1.225 brouard 9662: }
1.187 brouard 9663: nagesqr=1;
9664: if (strstr(model,"+age*age") !=0)
1.234 brouard 9665: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9666: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9667: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9668: else
1.234 brouard 9669: substrchaine(modelsav, model, "age*age");
1.187 brouard 9670: }else
9671: nagesqr=0;
9672: if (strlen(modelsav) >1){
9673: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9674: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9675: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9676: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9677: * cst, age and age*age
9678: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9679: /* including age products which are counted in cptcovage.
9680: * but the covariates which are products must be treated
9681: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9682: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9683: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9684:
9685:
1.187 brouard 9686: /* Design
9687: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9688: * < ncovcol=8 >
9689: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9690: * k= 1 2 3 4 5 6 7 8
9691: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9692: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9693: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9694: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9695: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9696: * Tage[++cptcovage]=k
9697: * if products, new covar are created after ncovcol with k1
9698: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9699: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9700: * 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
9701: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9702: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9703: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9704: * < ncovcol=8 >
9705: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9706: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9707: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9708: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9709: * p Tprod[1]@2={ 6, 5}
9710: *p Tvard[1][1]@4= {7, 8, 5, 6}
9711: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9712: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9713: *How to reorganize?
9714: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9715: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9716: * {2, 1, 4, 8, 5, 6, 3, 7}
9717: * Struct []
9718: */
1.225 brouard 9719:
1.187 brouard 9720: /* This loop fills the array Tvar from the string 'model'.*/
9721: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9722: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9723: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9724: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9725: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9726: /* k=1 Tvar[1]=2 (from V2) */
9727: /* k=5 Tvar[5] */
9728: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9729: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9730: /* } */
1.198 brouard 9731: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9732: /*
9733: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9734: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9735: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9736: }
1.187 brouard 9737: cptcovage=0;
9738: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9739: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9740: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9741: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9742: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9743: /*scanf("%d",i);*/
9744: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9745: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9746: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9747: /* covar is not filled and then is empty */
9748: cptcovprod--;
9749: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9750: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9751: Typevar[k]=1; /* 1 for age product */
9752: cptcovage++; /* Sums the number of covariates which include age as a product */
9753: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9754: /*printf("stre=%s ", stre);*/
9755: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9756: cptcovprod--;
9757: cutl(stre,strb,strc,'V');
9758: Tvar[k]=atoi(stre);
9759: Typevar[k]=1; /* 1 for age product */
9760: cptcovage++;
9761: Tage[cptcovage]=k;
9762: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9763: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9764: cptcovn++;
9765: cptcovprodnoage++;k1++;
9766: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9767: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9768: because this model-covariate is a construction we invent a new column
9769: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9770: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9771: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9772: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9773: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9774: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9775: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9776: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9777: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9778: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9779: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9780: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9781: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9782: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9783: for (i=1; i<=lastobs;i++){
9784: /* Computes the new covariate which is a product of
9785: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9786: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9787: }
9788: } /* End age is not in the model */
9789: } /* End if model includes a product */
9790: else { /* no more sum */
9791: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9792: /* scanf("%d",i);*/
9793: cutl(strd,strc,strb,'V');
9794: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9795: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9796: Tvar[k]=atoi(strd);
9797: Typevar[k]=0; /* 0 for simple covariates */
9798: }
9799: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9800: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9801: scanf("%d",i);*/
1.187 brouard 9802: } /* end of loop + on total covariates */
9803: } /* end if strlen(modelsave == 0) age*age might exist */
9804: } /* end if strlen(model == 0) */
1.136 brouard 9805:
9806: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9807: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9808:
1.136 brouard 9809: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9810: printf("cptcovprod=%d ", cptcovprod);
9811: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9812: scanf("%d ",i);*/
9813:
9814:
1.230 brouard 9815: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9816: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9817: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9818: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9819: k = 1 2 3 4 5 6 7 8 9
9820: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9821: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9822: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9823: Dummy[k] 1 0 0 0 3 1 1 2 3
9824: Tmodelind[combination of covar]=k;
1.225 brouard 9825: */
9826: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9827: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9828: /* 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 9829: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9830: printf("Model=%s\n\
9831: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9832: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9833: 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);
9834: fprintf(ficlog,"Model=%s\n\
9835: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9836: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9837: 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 9838: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9839: 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 */
9840: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9841: Fixed[k]= 0;
9842: Dummy[k]= 0;
1.225 brouard 9843: ncoveff++;
1.232 brouard 9844: ncovf++;
1.234 brouard 9845: nsd++;
9846: modell[k].maintype= FTYPE;
9847: TvarsD[nsd]=Tvar[k];
9848: TvarsDind[nsd]=k;
9849: TvarF[ncovf]=Tvar[k];
9850: TvarFind[ncovf]=k;
9851: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9852: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9853: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9854: Fixed[k]= 0;
9855: Dummy[k]= 0;
9856: ncoveff++;
9857: ncovf++;
9858: modell[k].maintype= FTYPE;
9859: TvarF[ncovf]=Tvar[k];
9860: TvarFind[ncovf]=k;
1.230 brouard 9861: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9862: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9863: }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 9864: Fixed[k]= 0;
9865: Dummy[k]= 1;
1.230 brouard 9866: nqfveff++;
1.234 brouard 9867: modell[k].maintype= FTYPE;
9868: modell[k].subtype= FQ;
9869: nsq++;
9870: TvarsQ[nsq]=Tvar[k];
9871: TvarsQind[nsq]=k;
1.232 brouard 9872: ncovf++;
1.234 brouard 9873: TvarF[ncovf]=Tvar[k];
9874: TvarFind[ncovf]=k;
1.231 brouard 9875: 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 9876: 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 9877: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9878: Fixed[k]= 1;
9879: Dummy[k]= 0;
1.225 brouard 9880: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9881: modell[k].maintype= VTYPE;
9882: modell[k].subtype= VD;
9883: nsd++;
9884: TvarsD[nsd]=Tvar[k];
9885: TvarsDind[nsd]=k;
9886: ncovv++; /* Only simple time varying variables */
9887: TvarV[ncovv]=Tvar[k];
1.242 brouard 9888: 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 9889: 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 */
9890: 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 9891: 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);
9892: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9893: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9894: Fixed[k]= 1;
9895: Dummy[k]= 1;
9896: nqtveff++;
9897: modell[k].maintype= VTYPE;
9898: modell[k].subtype= VQ;
9899: ncovv++; /* Only simple time varying variables */
9900: nsq++;
9901: TvarsQ[nsq]=Tvar[k];
9902: TvarsQind[nsq]=k;
9903: TvarV[ncovv]=Tvar[k];
1.242 brouard 9904: 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 9905: 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 */
9906: 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 9907: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9908: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9909: 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 9910: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9911: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9912: ncova++;
9913: TvarA[ncova]=Tvar[k];
9914: TvarAind[ncova]=k;
1.231 brouard 9915: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9916: Fixed[k]= 2;
9917: Dummy[k]= 2;
9918: modell[k].maintype= ATYPE;
9919: modell[k].subtype= APFD;
9920: /* ncoveff++; */
1.227 brouard 9921: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9922: Fixed[k]= 2;
9923: Dummy[k]= 3;
9924: modell[k].maintype= ATYPE;
9925: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9926: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9927: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9928: Fixed[k]= 3;
9929: Dummy[k]= 2;
9930: modell[k].maintype= ATYPE;
9931: modell[k].subtype= APVD; /* Product age * varying dummy */
9932: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9933: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9934: Fixed[k]= 3;
9935: Dummy[k]= 3;
9936: modell[k].maintype= ATYPE;
9937: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9938: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9939: }
9940: }else if (Typevar[k] == 2) { /* product without age */
9941: k1=Tposprod[k];
9942: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9943: if(Tvard[k1][2] <=ncovcol){
9944: Fixed[k]= 1;
9945: Dummy[k]= 0;
9946: modell[k].maintype= FTYPE;
9947: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9948: ncovf++; /* Fixed variables without age */
9949: TvarF[ncovf]=Tvar[k];
9950: TvarFind[ncovf]=k;
9951: }else if(Tvard[k1][2] <=ncovcol+nqv){
9952: Fixed[k]= 0; /* or 2 ?*/
9953: Dummy[k]= 1;
9954: modell[k].maintype= FTYPE;
9955: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9956: ncovf++; /* Varying variables without age */
9957: TvarF[ncovf]=Tvar[k];
9958: TvarFind[ncovf]=k;
9959: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9960: Fixed[k]= 1;
9961: Dummy[k]= 0;
9962: modell[k].maintype= VTYPE;
9963: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9964: ncovv++; /* Varying variables without age */
9965: TvarV[ncovv]=Tvar[k];
9966: TvarVind[ncovv]=k;
9967: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9968: Fixed[k]= 1;
9969: Dummy[k]= 1;
9970: modell[k].maintype= VTYPE;
9971: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9972: ncovv++; /* Varying variables without age */
9973: TvarV[ncovv]=Tvar[k];
9974: TvarVind[ncovv]=k;
9975: }
1.227 brouard 9976: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9977: if(Tvard[k1][2] <=ncovcol){
9978: Fixed[k]= 0; /* or 2 ?*/
9979: Dummy[k]= 1;
9980: modell[k].maintype= FTYPE;
9981: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9982: ncovf++; /* Fixed variables without age */
9983: TvarF[ncovf]=Tvar[k];
9984: TvarFind[ncovf]=k;
9985: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9986: Fixed[k]= 1;
9987: Dummy[k]= 1;
9988: modell[k].maintype= VTYPE;
9989: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9990: ncovv++; /* Varying variables without age */
9991: TvarV[ncovv]=Tvar[k];
9992: TvarVind[ncovv]=k;
9993: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9994: Fixed[k]= 1;
9995: Dummy[k]= 1;
9996: modell[k].maintype= VTYPE;
9997: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9998: ncovv++; /* Varying variables without age */
9999: TvarV[ncovv]=Tvar[k];
10000: TvarVind[ncovv]=k;
10001: ncovv++; /* Varying variables without age */
10002: TvarV[ncovv]=Tvar[k];
10003: TvarVind[ncovv]=k;
10004: }
1.227 brouard 10005: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10006: if(Tvard[k1][2] <=ncovcol){
10007: Fixed[k]= 1;
10008: Dummy[k]= 1;
10009: modell[k].maintype= VTYPE;
10010: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10011: ncovv++; /* Varying variables without age */
10012: TvarV[ncovv]=Tvar[k];
10013: TvarVind[ncovv]=k;
10014: }else if(Tvard[k1][2] <=ncovcol+nqv){
10015: Fixed[k]= 1;
10016: Dummy[k]= 1;
10017: modell[k].maintype= VTYPE;
10018: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10019: ncovv++; /* Varying variables without age */
10020: TvarV[ncovv]=Tvar[k];
10021: TvarVind[ncovv]=k;
10022: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10023: Fixed[k]= 1;
10024: Dummy[k]= 0;
10025: modell[k].maintype= VTYPE;
10026: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10027: ncovv++; /* Varying variables without age */
10028: TvarV[ncovv]=Tvar[k];
10029: TvarVind[ncovv]=k;
10030: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10031: Fixed[k]= 1;
10032: Dummy[k]= 1;
10033: modell[k].maintype= VTYPE;
10034: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10035: ncovv++; /* Varying variables without age */
10036: TvarV[ncovv]=Tvar[k];
10037: TvarVind[ncovv]=k;
10038: }
1.227 brouard 10039: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10040: if(Tvard[k1][2] <=ncovcol){
10041: Fixed[k]= 1;
10042: Dummy[k]= 1;
10043: modell[k].maintype= VTYPE;
10044: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10045: ncovv++; /* Varying variables without age */
10046: TvarV[ncovv]=Tvar[k];
10047: TvarVind[ncovv]=k;
10048: }else if(Tvard[k1][2] <=ncovcol+nqv){
10049: Fixed[k]= 1;
10050: Dummy[k]= 1;
10051: modell[k].maintype= VTYPE;
10052: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10053: ncovv++; /* Varying variables without age */
10054: TvarV[ncovv]=Tvar[k];
10055: TvarVind[ncovv]=k;
10056: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10057: Fixed[k]= 1;
10058: Dummy[k]= 1;
10059: modell[k].maintype= VTYPE;
10060: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10061: ncovv++; /* Varying variables without age */
10062: TvarV[ncovv]=Tvar[k];
10063: TvarVind[ncovv]=k;
10064: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10065: Fixed[k]= 1;
10066: Dummy[k]= 1;
10067: modell[k].maintype= VTYPE;
10068: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10069: ncovv++; /* Varying variables without age */
10070: TvarV[ncovv]=Tvar[k];
10071: TvarVind[ncovv]=k;
10072: }
1.227 brouard 10073: }else{
1.240 brouard 10074: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10075: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10076: } /*end k1*/
1.225 brouard 10077: }else{
1.226 brouard 10078: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10079: 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 10080: }
1.227 brouard 10081: 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 10082: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10083: 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]);
10084: }
10085: /* Searching for doublons in the model */
10086: for(k1=1; k1<= cptcovt;k1++){
10087: for(k2=1; k2 <k1;k2++){
1.285 brouard 10088: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10089: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10090: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10091: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10092: 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]);
10093: 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 10094: return(1);
10095: }
10096: }else if (Typevar[k1] ==2){
10097: k3=Tposprod[k1];
10098: k4=Tposprod[k2];
10099: 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])) ){
10100: 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]]);
10101: 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);
10102: return(1);
10103: }
10104: }
1.227 brouard 10105: }
10106: }
1.225 brouard 10107: }
10108: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10109: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10110: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10111: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10112: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10113: /*endread:*/
1.225 brouard 10114: printf("Exiting decodemodel: ");
10115: return (1);
1.136 brouard 10116: }
10117:
1.169 brouard 10118: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10119: {/* Check ages at death */
1.136 brouard 10120: int i, m;
1.218 brouard 10121: int firstone=0;
10122:
1.136 brouard 10123: for (i=1; i<=imx; i++) {
10124: for(m=2; (m<= maxwav); m++) {
10125: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10126: anint[m][i]=9999;
1.216 brouard 10127: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10128: s[m][i]=-1;
1.136 brouard 10129: }
10130: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10131: *nberr = *nberr + 1;
1.218 brouard 10132: if(firstone == 0){
10133: firstone=1;
1.260 brouard 10134: 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 10135: }
1.262 brouard 10136: 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 10137: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10138: }
10139: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10140: (*nberr)++;
1.259 brouard 10141: 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 10142: 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 10143: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10144: }
10145: }
10146: }
10147:
10148: for (i=1; i<=imx; i++) {
10149: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10150: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10151: 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 10152: if (s[m][i] >= nlstate+1) {
1.169 brouard 10153: if(agedc[i]>0){
10154: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10155: agev[m][i]=agedc[i];
1.214 brouard 10156: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10157: }else {
1.136 brouard 10158: if ((int)andc[i]!=9999){
10159: nbwarn++;
10160: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10161: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10162: agev[m][i]=-1;
10163: }
10164: }
1.169 brouard 10165: } /* agedc > 0 */
1.214 brouard 10166: } /* end if */
1.136 brouard 10167: else if(s[m][i] !=9){ /* Standard case, age in fractional
10168: years but with the precision of a month */
10169: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10170: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10171: agev[m][i]=1;
10172: else if(agev[m][i] < *agemin){
10173: *agemin=agev[m][i];
10174: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10175: }
10176: else if(agev[m][i] >*agemax){
10177: *agemax=agev[m][i];
1.156 brouard 10178: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10179: }
10180: /*agev[m][i]=anint[m][i]-annais[i];*/
10181: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10182: } /* en if 9*/
1.136 brouard 10183: else { /* =9 */
1.214 brouard 10184: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10185: agev[m][i]=1;
10186: s[m][i]=-1;
10187: }
10188: }
1.214 brouard 10189: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10190: agev[m][i]=1;
1.214 brouard 10191: else{
10192: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10193: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10194: agev[m][i]=0;
10195: }
10196: } /* End for lastpass */
10197: }
1.136 brouard 10198:
10199: for (i=1; i<=imx; i++) {
10200: for(m=firstpass; (m<=lastpass); m++){
10201: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10202: (*nberr)++;
1.136 brouard 10203: 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);
10204: 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);
10205: return 1;
10206: }
10207: }
10208: }
10209:
10210: /*for (i=1; i<=imx; i++){
10211: for (m=firstpass; (m<lastpass); m++){
10212: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10213: }
10214:
10215: }*/
10216:
10217:
1.139 brouard 10218: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10219: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10220:
10221: return (0);
1.164 brouard 10222: /* endread:*/
1.136 brouard 10223: printf("Exiting calandcheckages: ");
10224: return (1);
10225: }
10226:
1.172 brouard 10227: #if defined(_MSC_VER)
10228: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10229: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10230: //#include "stdafx.h"
10231: //#include <stdio.h>
10232: //#include <tchar.h>
10233: //#include <windows.h>
10234: //#include <iostream>
10235: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10236:
10237: LPFN_ISWOW64PROCESS fnIsWow64Process;
10238:
10239: BOOL IsWow64()
10240: {
10241: BOOL bIsWow64 = FALSE;
10242:
10243: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10244: // (HANDLE, PBOOL);
10245:
10246: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10247:
10248: HMODULE module = GetModuleHandle(_T("kernel32"));
10249: const char funcName[] = "IsWow64Process";
10250: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10251: GetProcAddress(module, funcName);
10252:
10253: if (NULL != fnIsWow64Process)
10254: {
10255: if (!fnIsWow64Process(GetCurrentProcess(),
10256: &bIsWow64))
10257: //throw std::exception("Unknown error");
10258: printf("Unknown error\n");
10259: }
10260: return bIsWow64 != FALSE;
10261: }
10262: #endif
1.177 brouard 10263:
1.191 brouard 10264: void syscompilerinfo(int logged)
1.292 brouard 10265: {
10266: #include <stdint.h>
10267:
10268: /* #include "syscompilerinfo.h"*/
1.185 brouard 10269: /* command line Intel compiler 32bit windows, XP compatible:*/
10270: /* /GS /W3 /Gy
10271: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10272: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10273: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10274: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10275: */
10276: /* 64 bits */
1.185 brouard 10277: /*
10278: /GS /W3 /Gy
10279: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10280: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10281: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10282: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10283: /* Optimization are useless and O3 is slower than O2 */
10284: /*
10285: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10286: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10287: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10288: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10289: */
1.186 brouard 10290: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10291: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10292: /PDB:"visual studio
10293: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10294: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10295: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10296: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10297: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10298: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10299: uiAccess='false'"
10300: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10301: /NOLOGO /TLBID:1
10302: */
1.292 brouard 10303:
10304:
1.177 brouard 10305: #if defined __INTEL_COMPILER
1.178 brouard 10306: #if defined(__GNUC__)
10307: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10308: #endif
1.177 brouard 10309: #elif defined(__GNUC__)
1.179 brouard 10310: #ifndef __APPLE__
1.174 brouard 10311: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10312: #endif
1.177 brouard 10313: struct utsname sysInfo;
1.178 brouard 10314: int cross = CROSS;
10315: if (cross){
10316: printf("Cross-");
1.191 brouard 10317: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10318: }
1.174 brouard 10319: #endif
10320:
1.191 brouard 10321: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10322: #if defined(__clang__)
1.191 brouard 10323: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10324: #endif
10325: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10326: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10327: #endif
10328: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10329: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10330: #endif
10331: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10332: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10333: #endif
10334: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10335: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10336: #endif
10337: #if defined(_MSC_VER)
1.191 brouard 10338: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10339: #endif
10340: #if defined(__PGI)
1.191 brouard 10341: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10342: #endif
10343: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10344: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10345: #endif
1.191 brouard 10346: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10347:
1.167 brouard 10348: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10349: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10350: // Windows (x64 and x86)
1.191 brouard 10351: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10352: #elif __unix__ // all unices, not all compilers
10353: // Unix
1.191 brouard 10354: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10355: #elif __linux__
10356: // linux
1.191 brouard 10357: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10358: #elif __APPLE__
1.174 brouard 10359: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10360: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10361: #endif
10362:
10363: /* __MINGW32__ */
10364: /* __CYGWIN__ */
10365: /* __MINGW64__ */
10366: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10367: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10368: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10369: /* _WIN64 // Defined for applications for Win64. */
10370: /* _M_X64 // Defined for compilations that target x64 processors. */
10371: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10372:
1.167 brouard 10373: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10374: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10375: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10376: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10377: #else
1.191 brouard 10378: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10379: #endif
10380:
1.169 brouard 10381: #if defined(__GNUC__)
10382: # if defined(__GNUC_PATCHLEVEL__)
10383: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10384: + __GNUC_MINOR__ * 100 \
10385: + __GNUC_PATCHLEVEL__)
10386: # else
10387: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10388: + __GNUC_MINOR__ * 100)
10389: # endif
1.174 brouard 10390: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10391: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10392:
10393: if (uname(&sysInfo) != -1) {
10394: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10395: 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 10396: }
10397: else
10398: perror("uname() error");
1.179 brouard 10399: //#ifndef __INTEL_COMPILER
10400: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10401: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10402: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10403: #endif
1.169 brouard 10404: #endif
1.172 brouard 10405:
1.286 brouard 10406: // void main ()
1.172 brouard 10407: // {
1.169 brouard 10408: #if defined(_MSC_VER)
1.174 brouard 10409: if (IsWow64()){
1.191 brouard 10410: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10411: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10412: }
10413: else{
1.191 brouard 10414: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10415: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10416: }
1.172 brouard 10417: // printf("\nPress Enter to continue...");
10418: // getchar();
10419: // }
10420:
1.169 brouard 10421: #endif
10422:
1.167 brouard 10423:
1.219 brouard 10424: }
1.136 brouard 10425:
1.219 brouard 10426: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10427: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10428: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10429: /* double ftolpl = 1.e-10; */
1.180 brouard 10430: double age, agebase, agelim;
1.203 brouard 10431: double tot;
1.180 brouard 10432:
1.202 brouard 10433: strcpy(filerespl,"PL_");
10434: strcat(filerespl,fileresu);
10435: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10436: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10437: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10438: }
1.288 brouard 10439: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10440: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10441: pstamp(ficrespl);
1.288 brouard 10442: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10443: fprintf(ficrespl,"#Age ");
10444: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10445: fprintf(ficrespl,"\n");
1.180 brouard 10446:
1.219 brouard 10447: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10448:
1.219 brouard 10449: agebase=ageminpar;
10450: agelim=agemaxpar;
1.180 brouard 10451:
1.227 brouard 10452: /* i1=pow(2,ncoveff); */
1.234 brouard 10453: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10454: if (cptcovn < 1){i1=1;}
1.180 brouard 10455:
1.238 brouard 10456: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10457: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10458: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10459: continue;
1.235 brouard 10460:
1.238 brouard 10461: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10462: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10463: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10464: /* k=k+1; */
10465: /* to clean */
10466: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10467: fprintf(ficrespl,"#******");
10468: printf("#******");
10469: fprintf(ficlog,"#******");
10470: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10471: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10472: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10473: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10474: }
10475: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10476: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10477: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10478: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10479: }
10480: fprintf(ficrespl,"******\n");
10481: printf("******\n");
10482: fprintf(ficlog,"******\n");
10483: if(invalidvarcomb[k]){
10484: printf("\nCombination (%d) ignored because no case \n",k);
10485: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10486: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10487: continue;
10488: }
1.219 brouard 10489:
1.238 brouard 10490: fprintf(ficrespl,"#Age ");
10491: for(j=1;j<=cptcoveff;j++) {
10492: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10493: }
10494: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10495: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10496:
1.238 brouard 10497: for (age=agebase; age<=agelim; age++){
10498: /* for (age=agebase; age<=agebase; age++){ */
10499: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10500: fprintf(ficrespl,"%.0f ",age );
10501: for(j=1;j<=cptcoveff;j++)
10502: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10503: tot=0.;
10504: for(i=1; i<=nlstate;i++){
10505: tot += prlim[i][i];
10506: fprintf(ficrespl," %.5f", prlim[i][i]);
10507: }
10508: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10509: } /* Age */
10510: /* was end of cptcod */
10511: } /* cptcov */
10512: } /* nres */
1.219 brouard 10513: return 0;
1.180 brouard 10514: }
10515:
1.218 brouard 10516: 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 10517: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10518:
10519: /* Computes the back prevalence limit for any combination of covariate values
10520: * at any age between ageminpar and agemaxpar
10521: */
1.235 brouard 10522: int i, j, k, i1, nres=0 ;
1.217 brouard 10523: /* double ftolpl = 1.e-10; */
10524: double age, agebase, agelim;
10525: double tot;
1.218 brouard 10526: /* double ***mobaverage; */
10527: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10528:
10529: strcpy(fileresplb,"PLB_");
10530: strcat(fileresplb,fileresu);
10531: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10532: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10533: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10534: }
1.288 brouard 10535: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10536: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10537: pstamp(ficresplb);
1.288 brouard 10538: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10539: fprintf(ficresplb,"#Age ");
10540: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10541: fprintf(ficresplb,"\n");
10542:
1.218 brouard 10543:
10544: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10545:
10546: agebase=ageminpar;
10547: agelim=agemaxpar;
10548:
10549:
1.227 brouard 10550: i1=pow(2,cptcoveff);
1.218 brouard 10551: if (cptcovn < 1){i1=1;}
1.227 brouard 10552:
1.238 brouard 10553: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10554: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10555: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10556: continue;
10557: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10558: fprintf(ficresplb,"#******");
10559: printf("#******");
10560: fprintf(ficlog,"#******");
10561: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10562: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10563: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10564: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10565: }
10566: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10567: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10568: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10569: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10570: }
10571: fprintf(ficresplb,"******\n");
10572: printf("******\n");
10573: fprintf(ficlog,"******\n");
10574: if(invalidvarcomb[k]){
10575: printf("\nCombination (%d) ignored because no cases \n",k);
10576: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10577: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10578: continue;
10579: }
1.218 brouard 10580:
1.238 brouard 10581: fprintf(ficresplb,"#Age ");
10582: for(j=1;j<=cptcoveff;j++) {
10583: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10584: }
10585: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10586: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10587:
10588:
1.238 brouard 10589: for (age=agebase; age<=agelim; age++){
10590: /* for (age=agebase; age<=agebase; age++){ */
10591: if(mobilavproj > 0){
10592: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10593: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10594: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10595: }else if (mobilavproj == 0){
10596: 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);
10597: 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);
10598: exit(1);
10599: }else{
10600: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10601: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10602: /* printf("TOTOT\n"); */
10603: /* exit(1); */
1.238 brouard 10604: }
10605: fprintf(ficresplb,"%.0f ",age );
10606: for(j=1;j<=cptcoveff;j++)
10607: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10608: tot=0.;
10609: for(i=1; i<=nlstate;i++){
10610: tot += bprlim[i][i];
10611: fprintf(ficresplb," %.5f", bprlim[i][i]);
10612: }
10613: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10614: } /* Age */
10615: /* was end of cptcod */
1.255 brouard 10616: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10617: } /* end of any combination */
10618: } /* end of nres */
1.218 brouard 10619: /* hBijx(p, bage, fage); */
10620: /* fclose(ficrespijb); */
10621:
10622: return 0;
1.217 brouard 10623: }
1.218 brouard 10624:
1.180 brouard 10625: int hPijx(double *p, int bage, int fage){
10626: /*------------- h Pij x at various ages ------------*/
10627:
10628: int stepsize;
10629: int agelim;
10630: int hstepm;
10631: int nhstepm;
1.235 brouard 10632: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10633:
10634: double agedeb;
10635: double ***p3mat;
10636:
1.201 brouard 10637: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10638: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10639: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10640: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10641: }
10642: printf("Computing pij: result on file '%s' \n", filerespij);
10643: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10644:
10645: stepsize=(int) (stepm+YEARM-1)/YEARM;
10646: /*if (stepm<=24) stepsize=2;*/
10647:
10648: agelim=AGESUP;
10649: hstepm=stepsize*YEARM; /* Every year of age */
10650: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10651:
1.180 brouard 10652: /* hstepm=1; aff par mois*/
10653: pstamp(ficrespij);
10654: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10655: i1= pow(2,cptcoveff);
1.218 brouard 10656: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10657: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10658: /* k=k+1; */
1.235 brouard 10659: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10660: for(k=1; k<=i1;k++){
1.253 brouard 10661: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10662: continue;
1.183 brouard 10663: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10664: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10665: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10666: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10667: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10668: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10669: }
1.183 brouard 10670: fprintf(ficrespij,"******\n");
10671:
10672: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10673: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10674: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10675:
10676: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10677:
1.183 brouard 10678: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10679: oldm=oldms;savm=savms;
1.235 brouard 10680: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10681: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10682: for(i=1; i<=nlstate;i++)
10683: for(j=1; j<=nlstate+ndeath;j++)
10684: fprintf(ficrespij," %1d-%1d",i,j);
10685: fprintf(ficrespij,"\n");
10686: for (h=0; h<=nhstepm; h++){
10687: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10688: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10689: for(i=1; i<=nlstate;i++)
10690: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10691: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10692: fprintf(ficrespij,"\n");
10693: }
1.183 brouard 10694: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10695: fprintf(ficrespij,"\n");
10696: }
1.180 brouard 10697: /*}*/
10698: }
1.218 brouard 10699: return 0;
1.180 brouard 10700: }
1.218 brouard 10701:
10702: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10703: /*------------- h Bij x at various ages ------------*/
10704:
10705: int stepsize;
1.218 brouard 10706: /* int agelim; */
10707: int ageminl;
1.217 brouard 10708: int hstepm;
10709: int nhstepm;
1.238 brouard 10710: int h, i, i1, j, k, nres;
1.218 brouard 10711:
1.217 brouard 10712: double agedeb;
10713: double ***p3mat;
1.218 brouard 10714:
10715: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10716: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10717: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10718: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10719: }
10720: printf("Computing pij back: result on file '%s' \n", filerespijb);
10721: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10722:
10723: stepsize=(int) (stepm+YEARM-1)/YEARM;
10724: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10725:
1.218 brouard 10726: /* agelim=AGESUP; */
1.289 brouard 10727: ageminl=AGEINF; /* was 30 */
1.218 brouard 10728: hstepm=stepsize*YEARM; /* Every year of age */
10729: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10730:
10731: /* hstepm=1; aff par mois*/
10732: pstamp(ficrespijb);
1.255 brouard 10733: 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 10734: i1= pow(2,cptcoveff);
1.218 brouard 10735: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10736: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10737: /* k=k+1; */
1.238 brouard 10738: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10739: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10740: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10741: continue;
10742: fprintf(ficrespijb,"\n#****** ");
10743: for(j=1;j<=cptcoveff;j++)
10744: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10745: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10746: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10747: }
10748: fprintf(ficrespijb,"******\n");
1.264 brouard 10749: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10750: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10751: continue;
10752: }
10753:
10754: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10755: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10756: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 10757: 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 */
10758: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 10759:
10760: /* nhstepm=nhstepm*YEARM; aff par mois*/
10761:
1.266 brouard 10762: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10763: /* and memory limitations if stepm is small */
10764:
1.238 brouard 10765: /* oldm=oldms;savm=savms; */
10766: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10767: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10768: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10769: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10770: for(i=1; i<=nlstate;i++)
10771: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10772: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10773: fprintf(ficrespijb,"\n");
1.238 brouard 10774: for (h=0; h<=nhstepm; h++){
10775: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10776: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10777: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10778: for(i=1; i<=nlstate;i++)
10779: for(j=1; j<=nlstate+ndeath;j++)
10780: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10781: fprintf(ficrespijb,"\n");
10782: }
10783: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10784: fprintf(ficrespijb,"\n");
10785: } /* end age deb */
10786: } /* end combination */
10787: } /* end nres */
1.218 brouard 10788: return 0;
10789: } /* hBijx */
1.217 brouard 10790:
1.180 brouard 10791:
1.136 brouard 10792: /***********************************************/
10793: /**************** Main Program *****************/
10794: /***********************************************/
10795:
10796: int main(int argc, char *argv[])
10797: {
10798: #ifdef GSL
10799: const gsl_multimin_fminimizer_type *T;
10800: size_t iteri = 0, it;
10801: int rval = GSL_CONTINUE;
10802: int status = GSL_SUCCESS;
10803: double ssval;
10804: #endif
10805: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 10806: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
10807: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 10808: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10809: int jj, ll, li, lj, lk;
1.136 brouard 10810: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10811: int num_filled;
1.136 brouard 10812: int itimes;
10813: int NDIM=2;
10814: int vpopbased=0;
1.235 brouard 10815: int nres=0;
1.258 brouard 10816: int endishere=0;
1.277 brouard 10817: int noffset=0;
1.274 brouard 10818: int ncurrv=0; /* Temporary variable */
10819:
1.164 brouard 10820: char ca[32], cb[32];
1.136 brouard 10821: /* FILE *fichtm; *//* Html File */
10822: /* FILE *ficgp;*/ /*Gnuplot File */
10823: struct stat info;
1.191 brouard 10824: double agedeb=0.;
1.194 brouard 10825:
10826: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10827: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10828:
1.165 brouard 10829: double fret;
1.191 brouard 10830: double dum=0.; /* Dummy variable */
1.136 brouard 10831: double ***p3mat;
1.218 brouard 10832: /* double ***mobaverage; */
1.164 brouard 10833:
10834: char line[MAXLINE];
1.197 brouard 10835: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10836:
1.234 brouard 10837: char modeltemp[MAXLINE];
1.230 brouard 10838: char resultline[MAXLINE];
10839:
1.136 brouard 10840: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10841: char *tok, *val; /* pathtot */
1.290 brouard 10842: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 10843: int c, h , cpt, c2;
1.191 brouard 10844: int jl=0;
10845: int i1, j1, jk, stepsize=0;
1.194 brouard 10846: int count=0;
10847:
1.164 brouard 10848: int *tab;
1.136 brouard 10849: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 10850: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
10851: /* double anprojf, mprojf, jprojf; */
10852: /* double jintmean,mintmean,aintmean; */
10853: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10854: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10855: double yrfproj= 10.0; /* Number of years of forward projections */
10856: double yrbproj= 10.0; /* Number of years of backward projections */
10857: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 10858: int mobilav=0,popforecast=0;
1.191 brouard 10859: int hstepm=0, nhstepm=0;
1.136 brouard 10860: int agemortsup;
10861: float sumlpop=0.;
10862: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10863: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10864:
1.191 brouard 10865: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10866: double ftolpl=FTOL;
10867: double **prlim;
1.217 brouard 10868: double **bprlim;
1.136 brouard 10869: double ***param; /* Matrix of parameters */
1.251 brouard 10870: double ***paramstart; /* Matrix of starting parameter values */
10871: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10872: double **matcov; /* Matrix of covariance */
1.203 brouard 10873: double **hess; /* Hessian matrix */
1.136 brouard 10874: double ***delti3; /* Scale */
10875: double *delti; /* Scale */
10876: double ***eij, ***vareij;
10877: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10878:
1.136 brouard 10879: double *epj, vepp;
1.164 brouard 10880:
1.273 brouard 10881: double dateprev1, dateprev2;
1.296 brouard 10882: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
10883: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
10884:
1.217 brouard 10885:
1.136 brouard 10886: double **ximort;
1.145 brouard 10887: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10888: int *dcwave;
10889:
1.164 brouard 10890: char z[1]="c";
1.136 brouard 10891:
10892: /*char *strt;*/
10893: char strtend[80];
1.126 brouard 10894:
1.164 brouard 10895:
1.126 brouard 10896: /* setlocale (LC_ALL, ""); */
10897: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10898: /* textdomain (PACKAGE); */
10899: /* setlocale (LC_CTYPE, ""); */
10900: /* setlocale (LC_MESSAGES, ""); */
10901:
10902: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10903: rstart_time = time(NULL);
10904: /* (void) gettimeofday(&start_time,&tzp);*/
10905: start_time = *localtime(&rstart_time);
1.126 brouard 10906: curr_time=start_time;
1.157 brouard 10907: /*tml = *localtime(&start_time.tm_sec);*/
10908: /* strcpy(strstart,asctime(&tml)); */
10909: strcpy(strstart,asctime(&start_time));
1.126 brouard 10910:
10911: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10912: /* tp.tm_sec = tp.tm_sec +86400; */
10913: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10914: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10915: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10916: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10917: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10918: /* strt=asctime(&tmg); */
10919: /* printf("Time(after) =%s",strstart); */
10920: /* (void) time (&time_value);
10921: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10922: * tm = *localtime(&time_value);
10923: * strstart=asctime(&tm);
10924: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10925: */
10926:
10927: nberr=0; /* Number of errors and warnings */
10928: nbwarn=0;
1.184 brouard 10929: #ifdef WIN32
10930: _getcwd(pathcd, size);
10931: #else
1.126 brouard 10932: getcwd(pathcd, size);
1.184 brouard 10933: #endif
1.191 brouard 10934: syscompilerinfo(0);
1.196 brouard 10935: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10936: if(argc <=1){
10937: printf("\nEnter the parameter file name: ");
1.205 brouard 10938: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10939: printf("ERROR Empty parameter file name\n");
10940: goto end;
10941: }
1.126 brouard 10942: i=strlen(pathr);
10943: if(pathr[i-1]=='\n')
10944: pathr[i-1]='\0';
1.156 brouard 10945: i=strlen(pathr);
1.205 brouard 10946: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10947: pathr[i-1]='\0';
1.205 brouard 10948: }
10949: i=strlen(pathr);
10950: if( i==0 ){
10951: printf("ERROR Empty parameter file name\n");
10952: goto end;
10953: }
10954: for (tok = pathr; tok != NULL; ){
1.126 brouard 10955: printf("Pathr |%s|\n",pathr);
10956: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10957: printf("val= |%s| pathr=%s\n",val,pathr);
10958: strcpy (pathtot, val);
10959: if(pathr[0] == '\0') break; /* Dirty */
10960: }
10961: }
1.281 brouard 10962: else if (argc<=2){
10963: strcpy(pathtot,argv[1]);
10964: }
1.126 brouard 10965: else{
10966: strcpy(pathtot,argv[1]);
1.281 brouard 10967: strcpy(z,argv[2]);
10968: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 10969: }
10970: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10971: /*cygwin_split_path(pathtot,path,optionfile);
10972: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10973: /* cutv(path,optionfile,pathtot,'\\');*/
10974:
10975: /* Split argv[0], imach program to get pathimach */
10976: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10977: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10978: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10979: /* strcpy(pathimach,argv[0]); */
10980: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10981: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10982: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10983: #ifdef WIN32
10984: _chdir(path); /* Can be a relative path */
10985: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10986: #else
1.126 brouard 10987: chdir(path); /* Can be a relative path */
1.184 brouard 10988: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10989: #endif
10990: printf("Current directory %s!\n",pathcd);
1.126 brouard 10991: strcpy(command,"mkdir ");
10992: strcat(command,optionfilefiname);
10993: if((outcmd=system(command)) != 0){
1.169 brouard 10994: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10995: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10996: /* fclose(ficlog); */
10997: /* exit(1); */
10998: }
10999: /* if((imk=mkdir(optionfilefiname))<0){ */
11000: /* perror("mkdir"); */
11001: /* } */
11002:
11003: /*-------- arguments in the command line --------*/
11004:
1.186 brouard 11005: /* Main Log file */
1.126 brouard 11006: strcat(filelog, optionfilefiname);
11007: strcat(filelog,".log"); /* */
11008: if((ficlog=fopen(filelog,"w"))==NULL) {
11009: printf("Problem with logfile %s\n",filelog);
11010: goto end;
11011: }
11012: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11013: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11014: fprintf(ficlog,"\nEnter the parameter file name: \n");
11015: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11016: path=%s \n\
11017: optionfile=%s\n\
11018: optionfilext=%s\n\
1.156 brouard 11019: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11020:
1.197 brouard 11021: syscompilerinfo(1);
1.167 brouard 11022:
1.126 brouard 11023: printf("Local time (at start):%s",strstart);
11024: fprintf(ficlog,"Local time (at start): %s",strstart);
11025: fflush(ficlog);
11026: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11027: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11028:
11029: /* */
11030: strcpy(fileres,"r");
11031: strcat(fileres, optionfilefiname);
1.201 brouard 11032: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11033: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11034: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11035:
1.186 brouard 11036: /* Main ---------arguments file --------*/
1.126 brouard 11037:
11038: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11039: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11040: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11041: fflush(ficlog);
1.149 brouard 11042: /* goto end; */
11043: exit(70);
1.126 brouard 11044: }
11045:
11046: strcpy(filereso,"o");
1.201 brouard 11047: strcat(filereso,fileresu);
1.126 brouard 11048: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11049: printf("Problem with Output resultfile: %s\n", filereso);
11050: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11051: fflush(ficlog);
11052: goto end;
11053: }
1.278 brouard 11054: /*-------- Rewriting parameter file ----------*/
11055: strcpy(rfileres,"r"); /* "Rparameterfile */
11056: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11057: strcat(rfileres,"."); /* */
11058: strcat(rfileres,optionfilext); /* Other files have txt extension */
11059: if((ficres =fopen(rfileres,"w"))==NULL) {
11060: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11061: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11062: fflush(ficlog);
11063: goto end;
11064: }
11065: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11066:
1.278 brouard 11067:
1.126 brouard 11068: /* Reads comments: lines beginning with '#' */
11069: numlinepar=0;
1.277 brouard 11070: /* Is it a BOM UTF-8 Windows file? */
11071: /* First parameter line */
1.197 brouard 11072: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11073: noffset=0;
11074: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11075: {
11076: noffset=noffset+3;
11077: printf("# File is an UTF8 Bom.\n"); // 0xBF
11078: }
11079: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
11080: {
11081: noffset=noffset+2;
11082: printf("# File is an UTF16BE BOM file\n");
11083: }
11084: else if( line[0] == 0 && line[1] == 0)
11085: {
11086: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11087: noffset=noffset+4;
11088: printf("# File is an UTF16BE BOM file\n");
11089: }
11090: } else{
11091: ;/*printf(" Not a BOM file\n");*/
11092: }
11093:
1.197 brouard 11094: /* If line starts with a # it is a comment */
1.277 brouard 11095: if (line[noffset] == '#') {
1.197 brouard 11096: numlinepar++;
11097: fputs(line,stdout);
11098: fputs(line,ficparo);
1.278 brouard 11099: fputs(line,ficres);
1.197 brouard 11100: fputs(line,ficlog);
11101: continue;
11102: }else
11103: break;
11104: }
11105: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11106: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11107: if (num_filled != 5) {
11108: printf("Should be 5 parameters\n");
1.283 brouard 11109: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11110: }
1.126 brouard 11111: numlinepar++;
1.197 brouard 11112: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11113: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11114: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11115: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11116: }
11117: /* Second parameter line */
11118: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11119: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11120: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11121: if (line[0] == '#') {
11122: numlinepar++;
1.283 brouard 11123: printf("%s",line);
11124: fprintf(ficres,"%s",line);
11125: fprintf(ficparo,"%s",line);
11126: fprintf(ficlog,"%s",line);
1.197 brouard 11127: continue;
11128: }else
11129: break;
11130: }
1.223 brouard 11131: 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", \
11132: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11133: if (num_filled != 11) {
11134: 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 11135: printf("but line=%s\n",line);
1.283 brouard 11136: 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");
11137: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11138: }
1.286 brouard 11139: if( lastpass > maxwav){
11140: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11141: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11142: fflush(ficlog);
11143: goto end;
11144: }
11145: 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 11146: 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 11147: 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 11148: 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 11149: }
1.203 brouard 11150: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11151: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11152: /* Third parameter line */
11153: while(fgets(line, MAXLINE, ficpar)) {
11154: /* If line starts with a # it is a comment */
11155: if (line[0] == '#') {
11156: numlinepar++;
1.283 brouard 11157: printf("%s",line);
11158: fprintf(ficres,"%s",line);
11159: fprintf(ficparo,"%s",line);
11160: fprintf(ficlog,"%s",line);
1.197 brouard 11161: continue;
11162: }else
11163: break;
11164: }
1.201 brouard 11165: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11166: if (num_filled != 1){
11167: printf("ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
11168: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
1.197 brouard 11169: model[0]='\0';
11170: goto end;
11171: }
11172: else{
11173: if (model[0]=='+'){
11174: for(i=1; i<=strlen(model);i++)
11175: modeltemp[i-1]=model[i];
1.201 brouard 11176: strcpy(model,modeltemp);
1.197 brouard 11177: }
11178: }
1.199 brouard 11179: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11180: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11181: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11182: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11183: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11184: }
11185: /* 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); */
11186: /* numlinepar=numlinepar+3; /\* In general *\/ */
11187: /* 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 11188: /* 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); */
11189: /* 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 11190: fflush(ficlog);
1.190 brouard 11191: /* if(model[0]=='#'|| model[0]== '\0'){ */
11192: if(model[0]=='#'){
1.279 brouard 11193: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11194: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11195: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11196: if(mle != -1){
1.279 brouard 11197: 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 11198: exit(1);
11199: }
11200: }
1.126 brouard 11201: while((c=getc(ficpar))=='#' && c!= EOF){
11202: ungetc(c,ficpar);
11203: fgets(line, MAXLINE, ficpar);
11204: numlinepar++;
1.195 brouard 11205: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11206: z[0]=line[1];
11207: }
11208: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11209: fputs(line, stdout);
11210: //puts(line);
1.126 brouard 11211: fputs(line,ficparo);
11212: fputs(line,ficlog);
11213: }
11214: ungetc(c,ficpar);
11215:
11216:
1.290 brouard 11217: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11218: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11219: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11220: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11221: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11222: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11223: v1+v2*age+v2*v3 makes cptcovn = 3
11224: */
11225: if (strlen(model)>1)
1.187 brouard 11226: 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 11227: else
1.187 brouard 11228: ncovmodel=2; /* Constant and age */
1.133 brouard 11229: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11230: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11231: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11232: 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);
11233: 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);
11234: fflush(stdout);
11235: fclose (ficlog);
11236: goto end;
11237: }
1.126 brouard 11238: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11239: delti=delti3[1][1];
11240: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11241: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11242: /* We could also provide initial parameters values giving by simple logistic regression
11243: * only one way, that is without matrix product. We will have nlstate maximizations */
11244: /* for(i=1;i<nlstate;i++){ */
11245: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11246: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11247: /* } */
1.126 brouard 11248: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11249: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11250: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11251: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11252: fclose (ficparo);
11253: fclose (ficlog);
11254: goto end;
11255: exit(0);
1.220 brouard 11256: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11257: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11258: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11259: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11260: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11261: matcov=matrix(1,npar,1,npar);
1.203 brouard 11262: hess=matrix(1,npar,1,npar);
1.220 brouard 11263: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11264: /* Read guessed parameters */
1.126 brouard 11265: /* Reads comments: lines beginning with '#' */
11266: while((c=getc(ficpar))=='#' && c!= EOF){
11267: ungetc(c,ficpar);
11268: fgets(line, MAXLINE, ficpar);
11269: numlinepar++;
1.141 brouard 11270: fputs(line,stdout);
1.126 brouard 11271: fputs(line,ficparo);
11272: fputs(line,ficlog);
11273: }
11274: ungetc(c,ficpar);
11275:
11276: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11277: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11278: for(i=1; i <=nlstate; i++){
1.234 brouard 11279: j=0;
1.126 brouard 11280: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11281: if(jj==i) continue;
11282: j++;
1.292 brouard 11283: while((c=getc(ficpar))=='#' && c!= EOF){
11284: ungetc(c,ficpar);
11285: fgets(line, MAXLINE, ficpar);
11286: numlinepar++;
11287: fputs(line,stdout);
11288: fputs(line,ficparo);
11289: fputs(line,ficlog);
11290: }
11291: ungetc(c,ficpar);
1.234 brouard 11292: fscanf(ficpar,"%1d%1d",&i1,&j1);
11293: if ((i1 != i) || (j1 != jj)){
11294: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11295: It might be a problem of design; if ncovcol and the model are correct\n \
11296: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11297: exit(1);
11298: }
11299: fprintf(ficparo,"%1d%1d",i1,j1);
11300: if(mle==1)
11301: printf("%1d%1d",i,jj);
11302: fprintf(ficlog,"%1d%1d",i,jj);
11303: for(k=1; k<=ncovmodel;k++){
11304: fscanf(ficpar," %lf",¶m[i][j][k]);
11305: if(mle==1){
11306: printf(" %lf",param[i][j][k]);
11307: fprintf(ficlog," %lf",param[i][j][k]);
11308: }
11309: else
11310: fprintf(ficlog," %lf",param[i][j][k]);
11311: fprintf(ficparo," %lf",param[i][j][k]);
11312: }
11313: fscanf(ficpar,"\n");
11314: numlinepar++;
11315: if(mle==1)
11316: printf("\n");
11317: fprintf(ficlog,"\n");
11318: fprintf(ficparo,"\n");
1.126 brouard 11319: }
11320: }
11321: fflush(ficlog);
1.234 brouard 11322:
1.251 brouard 11323: /* Reads parameters values */
1.126 brouard 11324: p=param[1][1];
1.251 brouard 11325: pstart=paramstart[1][1];
1.126 brouard 11326:
11327: /* Reads comments: lines beginning with '#' */
11328: while((c=getc(ficpar))=='#' && c!= EOF){
11329: ungetc(c,ficpar);
11330: fgets(line, MAXLINE, ficpar);
11331: numlinepar++;
1.141 brouard 11332: fputs(line,stdout);
1.126 brouard 11333: fputs(line,ficparo);
11334: fputs(line,ficlog);
11335: }
11336: ungetc(c,ficpar);
11337:
11338: for(i=1; i <=nlstate; i++){
11339: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11340: fscanf(ficpar,"%1d%1d",&i1,&j1);
11341: if ( (i1-i) * (j1-j) != 0){
11342: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11343: exit(1);
11344: }
11345: printf("%1d%1d",i,j);
11346: fprintf(ficparo,"%1d%1d",i1,j1);
11347: fprintf(ficlog,"%1d%1d",i1,j1);
11348: for(k=1; k<=ncovmodel;k++){
11349: fscanf(ficpar,"%le",&delti3[i][j][k]);
11350: printf(" %le",delti3[i][j][k]);
11351: fprintf(ficparo," %le",delti3[i][j][k]);
11352: fprintf(ficlog," %le",delti3[i][j][k]);
11353: }
11354: fscanf(ficpar,"\n");
11355: numlinepar++;
11356: printf("\n");
11357: fprintf(ficparo,"\n");
11358: fprintf(ficlog,"\n");
1.126 brouard 11359: }
11360: }
11361: fflush(ficlog);
1.234 brouard 11362:
1.145 brouard 11363: /* Reads covariance matrix */
1.126 brouard 11364: delti=delti3[1][1];
1.220 brouard 11365:
11366:
1.126 brouard 11367: /* 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 11368:
1.126 brouard 11369: /* Reads comments: lines beginning with '#' */
11370: while((c=getc(ficpar))=='#' && c!= EOF){
11371: ungetc(c,ficpar);
11372: fgets(line, MAXLINE, ficpar);
11373: numlinepar++;
1.141 brouard 11374: fputs(line,stdout);
1.126 brouard 11375: fputs(line,ficparo);
11376: fputs(line,ficlog);
11377: }
11378: ungetc(c,ficpar);
1.220 brouard 11379:
1.126 brouard 11380: matcov=matrix(1,npar,1,npar);
1.203 brouard 11381: hess=matrix(1,npar,1,npar);
1.131 brouard 11382: for(i=1; i <=npar; i++)
11383: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11384:
1.194 brouard 11385: /* Scans npar lines */
1.126 brouard 11386: for(i=1; i <=npar; i++){
1.226 brouard 11387: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11388: if(count != 3){
1.226 brouard 11389: printf("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: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11393: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11394: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11395: exit(1);
1.220 brouard 11396: }else{
1.226 brouard 11397: if(mle==1)
11398: printf("%1d%1d%d",i1,j1,jk);
11399: }
11400: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11401: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11402: for(j=1; j <=i; j++){
1.226 brouard 11403: fscanf(ficpar," %le",&matcov[i][j]);
11404: if(mle==1){
11405: printf(" %.5le",matcov[i][j]);
11406: }
11407: fprintf(ficlog," %.5le",matcov[i][j]);
11408: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11409: }
11410: fscanf(ficpar,"\n");
11411: numlinepar++;
11412: if(mle==1)
1.220 brouard 11413: printf("\n");
1.126 brouard 11414: fprintf(ficlog,"\n");
11415: fprintf(ficparo,"\n");
11416: }
1.194 brouard 11417: /* End of read covariance matrix npar lines */
1.126 brouard 11418: for(i=1; i <=npar; i++)
11419: for(j=i+1;j<=npar;j++)
1.226 brouard 11420: matcov[i][j]=matcov[j][i];
1.126 brouard 11421:
11422: if(mle==1)
11423: printf("\n");
11424: fprintf(ficlog,"\n");
11425:
11426: fflush(ficlog);
11427:
11428: } /* End of mle != -3 */
1.218 brouard 11429:
1.186 brouard 11430: /* Main data
11431: */
1.290 brouard 11432: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11433: /* num=lvector(1,n); */
11434: /* moisnais=vector(1,n); */
11435: /* annais=vector(1,n); */
11436: /* moisdc=vector(1,n); */
11437: /* andc=vector(1,n); */
11438: /* weight=vector(1,n); */
11439: /* agedc=vector(1,n); */
11440: /* cod=ivector(1,n); */
11441: /* for(i=1;i<=n;i++){ */
11442: num=lvector(firstobs,lastobs);
11443: moisnais=vector(firstobs,lastobs);
11444: annais=vector(firstobs,lastobs);
11445: moisdc=vector(firstobs,lastobs);
11446: andc=vector(firstobs,lastobs);
11447: weight=vector(firstobs,lastobs);
11448: agedc=vector(firstobs,lastobs);
11449: cod=ivector(firstobs,lastobs);
11450: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11451: num[i]=0;
11452: moisnais[i]=0;
11453: annais[i]=0;
11454: moisdc[i]=0;
11455: andc[i]=0;
11456: agedc[i]=0;
11457: cod[i]=0;
11458: weight[i]=1.0; /* Equal weights, 1 by default */
11459: }
1.290 brouard 11460: mint=matrix(1,maxwav,firstobs,lastobs);
11461: anint=matrix(1,maxwav,firstobs,lastobs);
11462: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11463: tab=ivector(1,NCOVMAX);
1.144 brouard 11464: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11465: 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 11466:
1.136 brouard 11467: /* Reads data from file datafile */
11468: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11469: goto end;
11470:
11471: /* Calculation of the number of parameters from char model */
1.234 brouard 11472: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11473: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11474: k=3 V4 Tvar[k=3]= 4 (from V4)
11475: k=2 V1 Tvar[k=2]= 1 (from V1)
11476: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11477: */
11478:
11479: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11480: TvarsDind=ivector(1,NCOVMAX); /* */
11481: TvarsD=ivector(1,NCOVMAX); /* */
11482: TvarsQind=ivector(1,NCOVMAX); /* */
11483: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11484: TvarF=ivector(1,NCOVMAX); /* */
11485: TvarFind=ivector(1,NCOVMAX); /* */
11486: TvarV=ivector(1,NCOVMAX); /* */
11487: TvarVind=ivector(1,NCOVMAX); /* */
11488: TvarA=ivector(1,NCOVMAX); /* */
11489: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11490: TvarFD=ivector(1,NCOVMAX); /* */
11491: TvarFDind=ivector(1,NCOVMAX); /* */
11492: TvarFQ=ivector(1,NCOVMAX); /* */
11493: TvarFQind=ivector(1,NCOVMAX); /* */
11494: TvarVD=ivector(1,NCOVMAX); /* */
11495: TvarVDind=ivector(1,NCOVMAX); /* */
11496: TvarVQ=ivector(1,NCOVMAX); /* */
11497: TvarVQind=ivector(1,NCOVMAX); /* */
11498:
1.230 brouard 11499: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11500: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11501: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11502: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11503: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11504: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11505: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11506: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11507: */
11508: /* For model-covariate k tells which data-covariate to use but
11509: because this model-covariate is a construction we invent a new column
11510: ncovcol + k1
11511: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11512: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11513: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11514: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11515: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11516: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11517: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11518: */
1.145 brouard 11519: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11520: 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 11521: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11522: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11523: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11524: 4 covariates (3 plus signs)
11525: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11526: */
1.230 brouard 11527: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11528: * individual dummy, fixed or varying:
11529: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11530: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11531: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11532: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11533: * Tmodelind[1]@9={9,0,3,2,}*/
11534: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11535: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11536: * individual quantitative, fixed or varying:
11537: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11538: * 3, 1, 0, 0, 0, 0, 0, 0},
11539: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11540: /* Main decodemodel */
11541:
1.187 brouard 11542:
1.223 brouard 11543: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11544: goto end;
11545:
1.137 brouard 11546: if((double)(lastobs-imx)/(double)imx > 1.10){
11547: nbwarn++;
11548: 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);
11549: 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);
11550: }
1.136 brouard 11551: /* if(mle==1){*/
1.137 brouard 11552: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11553: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11554: }
11555:
11556: /*-calculation of age at interview from date of interview and age at death -*/
11557: agev=matrix(1,maxwav,1,imx);
11558:
11559: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11560: goto end;
11561:
1.126 brouard 11562:
1.136 brouard 11563: agegomp=(int)agemin;
1.290 brouard 11564: free_vector(moisnais,firstobs,lastobs);
11565: free_vector(annais,firstobs,lastobs);
1.126 brouard 11566: /* free_matrix(mint,1,maxwav,1,n);
11567: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11568: /* free_vector(moisdc,1,n); */
11569: /* free_vector(andc,1,n); */
1.145 brouard 11570: /* */
11571:
1.126 brouard 11572: wav=ivector(1,imx);
1.214 brouard 11573: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11574: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11575: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11576: 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.*/
11577: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11578: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11579:
11580: /* Concatenates waves */
1.214 brouard 11581: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11582: Death is a valid wave (if date is known).
11583: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11584: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11585: and mw[mi+1][i]. dh depends on stepm.
11586: */
11587:
1.126 brouard 11588: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11589: /* Concatenates waves */
1.145 brouard 11590:
1.290 brouard 11591: free_vector(moisdc,firstobs,lastobs);
11592: free_vector(andc,firstobs,lastobs);
1.215 brouard 11593:
1.126 brouard 11594: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11595: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11596: ncodemax[1]=1;
1.145 brouard 11597: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11598: cptcoveff=0;
1.220 brouard 11599: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11600: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11601: }
11602:
11603: ncovcombmax=pow(2,cptcoveff);
11604: invalidvarcomb=ivector(1, ncovcombmax);
11605: for(i=1;i<ncovcombmax;i++)
11606: invalidvarcomb[i]=0;
11607:
1.211 brouard 11608: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11609: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11610: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11611:
1.200 brouard 11612: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11613: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11614: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11615: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11616: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11617: * (currently 0 or 1) in the data.
11618: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11619: * corresponding modality (h,j).
11620: */
11621:
1.145 brouard 11622: h=0;
11623: /*if (cptcovn > 0) */
1.126 brouard 11624: m=pow(2,cptcoveff);
11625:
1.144 brouard 11626: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11627: * For k=4 covariates, h goes from 1 to m=2**k
11628: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11629: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11630: * h\k 1 2 3 4
1.143 brouard 11631: *______________________________
11632: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11633: * 2 2 1 1 1
11634: * 3 i=2 1 2 1 1
11635: * 4 2 2 1 1
11636: * 5 i=3 1 i=2 1 2 1
11637: * 6 2 1 2 1
11638: * 7 i=4 1 2 2 1
11639: * 8 2 2 2 1
1.197 brouard 11640: * 9 i=5 1 i=3 1 i=2 1 2
11641: * 10 2 1 1 2
11642: * 11 i=6 1 2 1 2
11643: * 12 2 2 1 2
11644: * 13 i=7 1 i=4 1 2 2
11645: * 14 2 1 2 2
11646: * 15 i=8 1 2 2 2
11647: * 16 2 2 2 2
1.143 brouard 11648: */
1.212 brouard 11649: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11650: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11651: * and the value of each covariate?
11652: * V1=1, V2=1, V3=2, V4=1 ?
11653: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11654: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11655: * In order to get the real value in the data, we use nbcode
11656: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11657: * We are keeping this crazy system in order to be able (in the future?)
11658: * to have more than 2 values (0 or 1) for a covariate.
11659: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11660: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11661: * bbbbbbbb
11662: * 76543210
11663: * h-1 00000101 (6-1=5)
1.219 brouard 11664: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11665: * &
11666: * 1 00000001 (1)
1.219 brouard 11667: * 00000000 = 1 & ((h-1) >> (k-1))
11668: * +1= 00000001 =1
1.211 brouard 11669: *
11670: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11671: * h' 1101 =2^3+2^2+0x2^1+2^0
11672: * >>k' 11
11673: * & 00000001
11674: * = 00000001
11675: * +1 = 00000010=2 = codtabm(14,3)
11676: * Reverse h=6 and m=16?
11677: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11678: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11679: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11680: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11681: * V3=decodtabm(14,3,2**4)=2
11682: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11683: *(h-1) >> (j-1) 0011 =13 >> 2
11684: * &1 000000001
11685: * = 000000001
11686: * +1= 000000010 =2
11687: * 2211
11688: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11689: * V3=2
1.220 brouard 11690: * codtabm and decodtabm are identical
1.211 brouard 11691: */
11692:
1.145 brouard 11693:
11694: free_ivector(Ndum,-1,NCOVMAX);
11695:
11696:
1.126 brouard 11697:
1.186 brouard 11698: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11699: strcpy(optionfilegnuplot,optionfilefiname);
11700: if(mle==-3)
1.201 brouard 11701: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11702: strcat(optionfilegnuplot,".gp");
11703:
11704: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11705: printf("Problem with file %s",optionfilegnuplot);
11706: }
11707: else{
1.204 brouard 11708: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11709: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11710: //fprintf(ficgp,"set missing 'NaNq'\n");
11711: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11712: }
11713: /* fclose(ficgp);*/
1.186 brouard 11714:
11715:
11716: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11717:
11718: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11719: if(mle==-3)
1.201 brouard 11720: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11721: strcat(optionfilehtm,".htm");
11722: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11723: printf("Problem with %s \n",optionfilehtm);
11724: exit(0);
1.126 brouard 11725: }
11726:
11727: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11728: strcat(optionfilehtmcov,"-cov.htm");
11729: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11730: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11731: }
11732: else{
11733: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11734: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11735: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11736: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11737: }
11738:
1.213 brouard 11739: 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 11740: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11741: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11742: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11743: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11744: \n\
11745: <hr size=\"2\" color=\"#EC5E5E\">\
11746: <ul><li><h4>Parameter files</h4>\n\
11747: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11748: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11749: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11750: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11751: - Date and time at start: %s</ul>\n",\
11752: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11753: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11754: fileres,fileres,\
11755: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11756: fflush(fichtm);
11757:
11758: strcpy(pathr,path);
11759: strcat(pathr,optionfilefiname);
1.184 brouard 11760: #ifdef WIN32
11761: _chdir(optionfilefiname); /* Move to directory named optionfile */
11762: #else
1.126 brouard 11763: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11764: #endif
11765:
1.126 brouard 11766:
1.220 brouard 11767: /* Calculates basic frequencies. Computes observed prevalence at single age
11768: and for any valid combination of covariates
1.126 brouard 11769: and prints on file fileres'p'. */
1.251 brouard 11770: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11771: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11772:
11773: fprintf(fichtm,"\n");
1.286 brouard 11774: 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 11775: ftol, stepm);
11776: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11777: ncurrv=1;
11778: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11779: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11780: ncurrv=i;
11781: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11782: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 11783: ncurrv=i;
11784: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11785: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 11786: ncurrv=i;
11787: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11788: 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", \
11789: nlstate, ndeath, maxwav, mle, weightopt);
11790:
11791: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11792: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11793:
11794:
11795: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11796: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11797: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11798: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11799: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11800: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11801: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11802: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11803: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11804:
1.126 brouard 11805: /* For Powell, parameters are in a vector p[] starting at p[1]
11806: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11807: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11808:
11809: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11810: /* For mortality only */
1.126 brouard 11811: if (mle==-3){
1.136 brouard 11812: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11813: for(i=1;i<=NDIM;i++)
11814: for(j=1;j<=NDIM;j++)
11815: ximort[i][j]=0.;
1.186 brouard 11816: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 11817: cens=ivector(firstobs,lastobs);
11818: ageexmed=vector(firstobs,lastobs);
11819: agecens=vector(firstobs,lastobs);
11820: dcwave=ivector(firstobs,lastobs);
1.223 brouard 11821:
1.126 brouard 11822: for (i=1; i<=imx; i++){
11823: dcwave[i]=-1;
11824: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11825: if (s[m][i]>nlstate) {
11826: dcwave[i]=m;
11827: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11828: break;
11829: }
1.126 brouard 11830: }
1.226 brouard 11831:
1.126 brouard 11832: for (i=1; i<=imx; i++) {
11833: if (wav[i]>0){
1.226 brouard 11834: ageexmed[i]=agev[mw[1][i]][i];
11835: j=wav[i];
11836: agecens[i]=1.;
11837:
11838: if (ageexmed[i]> 1 && wav[i] > 0){
11839: agecens[i]=agev[mw[j][i]][i];
11840: cens[i]= 1;
11841: }else if (ageexmed[i]< 1)
11842: cens[i]= -1;
11843: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11844: cens[i]=0 ;
1.126 brouard 11845: }
11846: else cens[i]=-1;
11847: }
11848:
11849: for (i=1;i<=NDIM;i++) {
11850: for (j=1;j<=NDIM;j++)
1.226 brouard 11851: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11852: }
11853:
1.145 brouard 11854: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11855: /*printf("%lf %lf", p[1], p[2]);*/
11856:
11857:
1.136 brouard 11858: #ifdef GSL
11859: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11860: #else
1.126 brouard 11861: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11862: #endif
1.201 brouard 11863: strcpy(filerespow,"POW-MORT_");
11864: strcat(filerespow,fileresu);
1.126 brouard 11865: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11866: printf("Problem with resultfile: %s\n", filerespow);
11867: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11868: }
1.136 brouard 11869: #ifdef GSL
11870: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11871: #else
1.126 brouard 11872: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11873: #endif
1.126 brouard 11874: /* for (i=1;i<=nlstate;i++)
11875: for(j=1;j<=nlstate+ndeath;j++)
11876: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11877: */
11878: fprintf(ficrespow,"\n");
1.136 brouard 11879: #ifdef GSL
11880: /* gsl starts here */
11881: T = gsl_multimin_fminimizer_nmsimplex;
11882: gsl_multimin_fminimizer *sfm = NULL;
11883: gsl_vector *ss, *x;
11884: gsl_multimin_function minex_func;
11885:
11886: /* Initial vertex size vector */
11887: ss = gsl_vector_alloc (NDIM);
11888:
11889: if (ss == NULL){
11890: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11891: }
11892: /* Set all step sizes to 1 */
11893: gsl_vector_set_all (ss, 0.001);
11894:
11895: /* Starting point */
1.126 brouard 11896:
1.136 brouard 11897: x = gsl_vector_alloc (NDIM);
11898:
11899: if (x == NULL){
11900: gsl_vector_free(ss);
11901: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11902: }
11903:
11904: /* Initialize method and iterate */
11905: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11906: /* gsl_vector_set(x, 0, 0.0268); */
11907: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11908: gsl_vector_set(x, 0, p[1]);
11909: gsl_vector_set(x, 1, p[2]);
11910:
11911: minex_func.f = &gompertz_f;
11912: minex_func.n = NDIM;
11913: minex_func.params = (void *)&p; /* ??? */
11914:
11915: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11916: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11917:
11918: printf("Iterations beginning .....\n\n");
11919: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11920:
11921: iteri=0;
11922: while (rval == GSL_CONTINUE){
11923: iteri++;
11924: status = gsl_multimin_fminimizer_iterate(sfm);
11925:
11926: if (status) printf("error: %s\n", gsl_strerror (status));
11927: fflush(0);
11928:
11929: if (status)
11930: break;
11931:
11932: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11933: ssval = gsl_multimin_fminimizer_size (sfm);
11934:
11935: if (rval == GSL_SUCCESS)
11936: printf ("converged to a local maximum at\n");
11937:
11938: printf("%5d ", iteri);
11939: for (it = 0; it < NDIM; it++){
11940: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11941: }
11942: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11943: }
11944:
11945: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11946:
11947: gsl_vector_free(x); /* initial values */
11948: gsl_vector_free(ss); /* inital step size */
11949: for (it=0; it<NDIM; it++){
11950: p[it+1]=gsl_vector_get(sfm->x,it);
11951: fprintf(ficrespow," %.12lf", p[it]);
11952: }
11953: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11954: #endif
11955: #ifdef POWELL
11956: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11957: #endif
1.126 brouard 11958: fclose(ficrespow);
11959:
1.203 brouard 11960: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11961:
11962: for(i=1; i <=NDIM; i++)
11963: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11964: matcov[i][j]=matcov[j][i];
1.126 brouard 11965:
11966: printf("\nCovariance matrix\n ");
1.203 brouard 11967: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11968: for(i=1; i <=NDIM; i++) {
11969: for(j=1;j<=NDIM;j++){
1.220 brouard 11970: printf("%f ",matcov[i][j]);
11971: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11972: }
1.203 brouard 11973: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11974: }
11975:
11976: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11977: for (i=1;i<=NDIM;i++) {
1.126 brouard 11978: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11979: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11980: }
1.126 brouard 11981: lsurv=vector(1,AGESUP);
11982: lpop=vector(1,AGESUP);
11983: tpop=vector(1,AGESUP);
11984: lsurv[agegomp]=100000;
11985:
11986: for (k=agegomp;k<=AGESUP;k++) {
11987: agemortsup=k;
11988: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11989: }
11990:
11991: for (k=agegomp;k<agemortsup;k++)
11992: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11993:
11994: for (k=agegomp;k<agemortsup;k++){
11995: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11996: sumlpop=sumlpop+lpop[k];
11997: }
11998:
11999: tpop[agegomp]=sumlpop;
12000: for (k=agegomp;k<(agemortsup-3);k++){
12001: /* tpop[k+1]=2;*/
12002: tpop[k+1]=tpop[k]-lpop[k];
12003: }
12004:
12005:
12006: printf("\nAge lx qx dx Lx Tx e(x)\n");
12007: for (k=agegomp;k<(agemortsup-2);k++)
12008: 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]);
12009:
12010:
12011: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12012: ageminpar=50;
12013: agemaxpar=100;
1.194 brouard 12014: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12015: printf("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);
12018: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12019: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12020: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12021: }else{
12022: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12023: 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 12024: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12025: }
1.201 brouard 12026: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12027: stepm, weightopt,\
12028: model,imx,p,matcov,agemortsup);
12029:
12030: free_vector(lsurv,1,AGESUP);
12031: free_vector(lpop,1,AGESUP);
12032: free_vector(tpop,1,AGESUP);
1.220 brouard 12033: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12034: free_ivector(dcwave,firstobs,lastobs);
12035: free_vector(agecens,firstobs,lastobs);
12036: free_vector(ageexmed,firstobs,lastobs);
12037: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12038: #ifdef GSL
1.136 brouard 12039: #endif
1.186 brouard 12040: } /* Endof if mle==-3 mortality only */
1.205 brouard 12041: /* Standard */
12042: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12043: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12044: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12045: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12046: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12047: for (k=1; k<=npar;k++)
12048: printf(" %d %8.5f",k,p[k]);
12049: printf("\n");
1.205 brouard 12050: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12051: /* mlikeli uses func not funcone */
1.247 brouard 12052: /* for(i=1;i<nlstate;i++){ */
12053: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12054: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12055: /* } */
1.205 brouard 12056: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12057: }
12058: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12059: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12060: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12061: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12062: }
12063: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12064: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12065: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12066: for (k=1; k<=npar;k++)
12067: printf(" %d %8.5f",k,p[k]);
12068: printf("\n");
12069:
12070: /*--------- results files --------------*/
1.283 brouard 12071: /* 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 12072:
12073:
12074: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12075: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12076: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12077: for(i=1,jk=1; i <=nlstate; i++){
12078: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12079: if (k != i) {
12080: printf("%d%d ",i,k);
12081: fprintf(ficlog,"%d%d ",i,k);
12082: fprintf(ficres,"%1d%1d ",i,k);
12083: for(j=1; j <=ncovmodel; j++){
12084: printf("%12.7f ",p[jk]);
12085: fprintf(ficlog,"%12.7f ",p[jk]);
12086: fprintf(ficres,"%12.7f ",p[jk]);
12087: jk++;
12088: }
12089: printf("\n");
12090: fprintf(ficlog,"\n");
12091: fprintf(ficres,"\n");
12092: }
1.126 brouard 12093: }
12094: }
1.203 brouard 12095: if(mle != 0){
12096: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12097: ftolhess=ftol; /* Usually correct */
1.203 brouard 12098: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12099: 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");
12100: 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");
12101: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12102: for(k=1; k <=(nlstate+ndeath); k++){
12103: if (k != i) {
12104: printf("%d%d ",i,k);
12105: fprintf(ficlog,"%d%d ",i,k);
12106: for(j=1; j <=ncovmodel; j++){
12107: 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]));
12108: 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]));
12109: jk++;
12110: }
12111: printf("\n");
12112: fprintf(ficlog,"\n");
12113: }
12114: }
1.193 brouard 12115: }
1.203 brouard 12116: } /* end of hesscov and Wald tests */
1.225 brouard 12117:
1.203 brouard 12118: /* */
1.126 brouard 12119: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12120: printf("# Scales (for hessian or gradient estimation)\n");
12121: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12122: for(i=1,jk=1; i <=nlstate; i++){
12123: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12124: if (j!=i) {
12125: fprintf(ficres,"%1d%1d",i,j);
12126: printf("%1d%1d",i,j);
12127: fprintf(ficlog,"%1d%1d",i,j);
12128: for(k=1; k<=ncovmodel;k++){
12129: printf(" %.5e",delti[jk]);
12130: fprintf(ficlog," %.5e",delti[jk]);
12131: fprintf(ficres," %.5e",delti[jk]);
12132: jk++;
12133: }
12134: printf("\n");
12135: fprintf(ficlog,"\n");
12136: fprintf(ficres,"\n");
12137: }
1.126 brouard 12138: }
12139: }
12140:
12141: 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 12142: if(mle >= 1) /* To big for the screen */
1.126 brouard 12143: 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");
12144: 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");
12145: /* # 121 Var(a12)\n\ */
12146: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12147: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12148: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12149: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12150: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12151: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12152: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12153:
12154:
12155: /* Just to have a covariance matrix which will be more understandable
12156: even is we still don't want to manage dictionary of variables
12157: */
12158: for(itimes=1;itimes<=2;itimes++){
12159: jj=0;
12160: for(i=1; i <=nlstate; i++){
1.225 brouard 12161: for(j=1; j <=nlstate+ndeath; j++){
12162: if(j==i) continue;
12163: for(k=1; k<=ncovmodel;k++){
12164: jj++;
12165: ca[0]= k+'a'-1;ca[1]='\0';
12166: if(itimes==1){
12167: if(mle>=1)
12168: printf("#%1d%1d%d",i,j,k);
12169: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12170: fprintf(ficres,"#%1d%1d%d",i,j,k);
12171: }else{
12172: if(mle>=1)
12173: printf("%1d%1d%d",i,j,k);
12174: fprintf(ficlog,"%1d%1d%d",i,j,k);
12175: fprintf(ficres,"%1d%1d%d",i,j,k);
12176: }
12177: ll=0;
12178: for(li=1;li <=nlstate; li++){
12179: for(lj=1;lj <=nlstate+ndeath; lj++){
12180: if(lj==li) continue;
12181: for(lk=1;lk<=ncovmodel;lk++){
12182: ll++;
12183: if(ll<=jj){
12184: cb[0]= lk +'a'-1;cb[1]='\0';
12185: if(ll<jj){
12186: if(itimes==1){
12187: if(mle>=1)
12188: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12189: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12190: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12191: }else{
12192: if(mle>=1)
12193: printf(" %.5e",matcov[jj][ll]);
12194: fprintf(ficlog," %.5e",matcov[jj][ll]);
12195: fprintf(ficres," %.5e",matcov[jj][ll]);
12196: }
12197: }else{
12198: if(itimes==1){
12199: if(mle>=1)
12200: printf(" Var(%s%1d%1d)",ca,i,j);
12201: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12202: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12203: }else{
12204: if(mle>=1)
12205: printf(" %.7e",matcov[jj][ll]);
12206: fprintf(ficlog," %.7e",matcov[jj][ll]);
12207: fprintf(ficres," %.7e",matcov[jj][ll]);
12208: }
12209: }
12210: }
12211: } /* end lk */
12212: } /* end lj */
12213: } /* end li */
12214: if(mle>=1)
12215: printf("\n");
12216: fprintf(ficlog,"\n");
12217: fprintf(ficres,"\n");
12218: numlinepar++;
12219: } /* end k*/
12220: } /*end j */
1.126 brouard 12221: } /* end i */
12222: } /* end itimes */
12223:
12224: fflush(ficlog);
12225: fflush(ficres);
1.225 brouard 12226: while(fgets(line, MAXLINE, ficpar)) {
12227: /* If line starts with a # it is a comment */
12228: if (line[0] == '#') {
12229: numlinepar++;
12230: fputs(line,stdout);
12231: fputs(line,ficparo);
12232: fputs(line,ficlog);
1.299 ! brouard 12233: fputs(line,ficres);
1.225 brouard 12234: continue;
12235: }else
12236: break;
12237: }
12238:
1.209 brouard 12239: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12240: /* ungetc(c,ficpar); */
12241: /* fgets(line, MAXLINE, ficpar); */
12242: /* fputs(line,stdout); */
12243: /* fputs(line,ficparo); */
12244: /* } */
12245: /* ungetc(c,ficpar); */
1.126 brouard 12246:
12247: estepm=0;
1.209 brouard 12248: 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 12249:
12250: if (num_filled != 6) {
12251: 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);
12252: 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);
12253: goto end;
12254: }
12255: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12256: }
12257: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12258: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12259:
1.209 brouard 12260: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12261: if (estepm==0 || estepm < stepm) estepm=stepm;
12262: if (fage <= 2) {
12263: bage = ageminpar;
12264: fage = agemaxpar;
12265: }
12266:
12267: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12268: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12269: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12270:
1.186 brouard 12271: /* Other stuffs, more or less useful */
1.254 brouard 12272: while(fgets(line, MAXLINE, ficpar)) {
12273: /* If line starts with a # it is a comment */
12274: if (line[0] == '#') {
12275: numlinepar++;
12276: fputs(line,stdout);
12277: fputs(line,ficparo);
12278: fputs(line,ficlog);
1.299 ! brouard 12279: fputs(line,ficres);
1.254 brouard 12280: continue;
12281: }else
12282: break;
12283: }
12284:
12285: 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){
12286:
12287: if (num_filled != 7) {
12288: 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);
12289: 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);
12290: goto end;
12291: }
12292: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12293: 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);
12294: 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);
12295: 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 12296: }
1.254 brouard 12297:
12298: while(fgets(line, MAXLINE, ficpar)) {
12299: /* If line starts with a # it is a comment */
12300: if (line[0] == '#') {
12301: numlinepar++;
12302: fputs(line,stdout);
12303: fputs(line,ficparo);
12304: fputs(line,ficlog);
1.299 ! brouard 12305: fputs(line,ficres);
1.254 brouard 12306: continue;
12307: }else
12308: break;
1.126 brouard 12309: }
12310:
12311:
12312: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12313: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12314:
1.254 brouard 12315: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12316: if (num_filled != 1) {
12317: 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);
12318: 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);
12319: goto end;
12320: }
12321: printf("pop_based=%d\n",popbased);
12322: fprintf(ficlog,"pop_based=%d\n",popbased);
12323: fprintf(ficparo,"pop_based=%d\n",popbased);
12324: fprintf(ficres,"pop_based=%d\n",popbased);
12325: }
12326:
1.258 brouard 12327: /* Results */
12328: nresult=0;
12329: do{
12330: if(!fgets(line, MAXLINE, ficpar)){
12331: endishere=1;
12332: parameterline=14;
12333: }else if (line[0] == '#') {
12334: /* If line starts with a # it is a comment */
1.254 brouard 12335: numlinepar++;
12336: fputs(line,stdout);
12337: fputs(line,ficparo);
12338: fputs(line,ficlog);
1.299 ! brouard 12339: fputs(line,ficres);
1.254 brouard 12340: continue;
1.258 brouard 12341: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12342: parameterline=11;
1.296 brouard 12343: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12344: parameterline=12;
12345: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12346: parameterline=13;
12347: else{
12348: parameterline=14;
1.254 brouard 12349: }
1.258 brouard 12350: switch (parameterline){
12351: case 11:
1.296 brouard 12352: 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)){
12353: 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 12354: 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);
12355: 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);
12356: 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);
12357: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12358: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12359: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12360: prvforecast = 1;
12361: }
12362: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.299 ! brouard 12363: printf("prevforecast=%d yearsfproj=%lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
! 12364: fprintf(ficlog,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
! 12365: fprintf(ficres,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12366: prvforecast = 2;
12367: }
12368: else {
12369: 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);
12370: 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);
12371: goto end;
1.258 brouard 12372: }
1.254 brouard 12373: break;
1.258 brouard 12374: case 12:
1.296 brouard 12375: 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)){
12376: 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);
12377: 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);
12378: 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);
12379: 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);
12380: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12381: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12382: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12383: prvbackcast = 1;
12384: }
12385: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.299 ! brouard 12386: printf("prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
! 12387: fprintf(ficlog,"prevbackcast=%d yearsfproj=%lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
! 12388: fprintf(ficres,"prevbackcast=%d yearsfproj=%lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12389: prvbackcast = 2;
12390: }
12391: else {
12392: 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);
12393: 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);
12394: goto end;
1.258 brouard 12395: }
1.230 brouard 12396: break;
1.258 brouard 12397: case 13:
12398: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12399: if (num_filled == 0){
12400: resultline[0]='\0';
12401: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12402: 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);
12403: break;
12404: } else if (num_filled != 1){
12405: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12406: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12407: }
12408: nresult++; /* Sum of resultlines */
12409: printf("Result %d: result=%s\n",nresult, resultline);
12410: if(nresult > MAXRESULTLINES){
12411: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12412: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12413: goto end;
12414: }
12415: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12416: fprintf(ficparo,"result: %s\n",resultline);
12417: fprintf(ficres,"result: %s\n",resultline);
12418: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12419: break;
1.258 brouard 12420: case 14:
1.259 brouard 12421: if(ncovmodel >2 && nresult==0 ){
12422: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12423: goto end;
12424: }
1.259 brouard 12425: break;
1.258 brouard 12426: default:
12427: nresult=1;
12428: decoderesult(".",nresult ); /* No covariate */
12429: }
12430: } /* End switch parameterline */
12431: }while(endishere==0); /* End do */
1.126 brouard 12432:
1.230 brouard 12433: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12434: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12435:
12436: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12437: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12438: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12439: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12440: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12441: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12442: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12443: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12444: }else{
1.270 brouard 12445: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12446: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12447: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12448: if(prvforecast==1){
12449: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12450: jprojd=jproj1;
12451: mprojd=mproj1;
12452: anprojd=anproj1;
12453: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12454: jprojf=jproj2;
12455: mprojf=mproj2;
12456: anprojf=anproj2;
12457: } else if(prvforecast == 2){
12458: dateprojd=dateintmean;
12459: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12460: dateprojf=dateintmean+yrfproj;
12461: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12462: }
12463: if(prvbackcast==1){
12464: datebackd=(jback1+12*mback1+365*anback1)/365;
12465: jbackd=jback1;
12466: mbackd=mback1;
12467: anbackd=anback1;
12468: datebackf=(jback2+12*mback2+365*anback2)/365;
12469: jbackf=jback2;
12470: mbackf=mback2;
12471: anbackf=anback2;
12472: } else if(prvbackcast == 2){
12473: datebackd=dateintmean;
12474: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12475: datebackf=dateintmean-yrbproj;
12476: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12477: }
12478:
12479: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12480: }
12481: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12482: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12483: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12484:
1.225 brouard 12485: /*------------ free_vector -------------*/
12486: /* chdir(path); */
1.220 brouard 12487:
1.215 brouard 12488: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12489: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12490: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12491: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12492: free_lvector(num,firstobs,lastobs);
12493: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12494: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12495: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12496: fclose(ficparo);
12497: fclose(ficres);
1.220 brouard 12498:
12499:
1.186 brouard 12500: /* Other results (useful)*/
1.220 brouard 12501:
12502:
1.126 brouard 12503: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12504: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12505: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12506: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12507: fclose(ficrespl);
12508:
12509: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12510: /*#include "hpijx.h"*/
12511: hPijx(p, bage, fage);
1.145 brouard 12512: fclose(ficrespij);
1.227 brouard 12513:
1.220 brouard 12514: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12515: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12516: k=1;
1.126 brouard 12517: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12518:
1.269 brouard 12519: /* Prevalence for each covariate combination in probs[age][status][cov] */
12520: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12521: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12522: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12523: for(k=1;k<=ncovcombmax;k++)
12524: probs[i][j][k]=0.;
1.269 brouard 12525: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12526: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12527: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12528: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12529: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12530: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12531: for(k=1;k<=ncovcombmax;k++)
12532: mobaverages[i][j][k]=0.;
1.219 brouard 12533: mobaverage=mobaverages;
12534: if (mobilav!=0) {
1.235 brouard 12535: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12536: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12537: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12538: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12539: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12540: }
1.269 brouard 12541: } else if (mobilavproj !=0) {
1.235 brouard 12542: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12543: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12544: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12545: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12546: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12547: }
1.269 brouard 12548: }else{
12549: printf("Internal error moving average\n");
12550: fflush(stdout);
12551: exit(1);
1.219 brouard 12552: }
12553: }/* end if moving average */
1.227 brouard 12554:
1.126 brouard 12555: /*---------- Forecasting ------------------*/
1.296 brouard 12556: if(prevfcast==1){
12557: /* /\* if(stepm ==1){*\/ */
12558: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12559: /*This done previously after freqsummary.*/
12560: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
12561: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
12562:
12563: /* } else if (prvforecast==2){ */
12564: /* /\* if(stepm ==1){*\/ */
12565: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12566: /* } */
12567: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
12568: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 12569: }
1.269 brouard 12570:
1.296 brouard 12571: /* Prevbcasting */
12572: if(prevbcast==1){
1.219 brouard 12573: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12574: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12575: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12576:
12577: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12578:
12579: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12580:
1.219 brouard 12581: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12582: fclose(ficresplb);
12583:
1.222 brouard 12584: hBijx(p, bage, fage, mobaverage);
12585: fclose(ficrespijb);
1.219 brouard 12586:
1.296 brouard 12587: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
12588: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
12589: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
12590: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
12591: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
12592: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
12593:
12594:
1.269 brouard 12595: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12596:
12597:
1.269 brouard 12598: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12599: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12600: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12601: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 12602: } /* end Prevbcasting */
1.268 brouard 12603:
1.186 brouard 12604:
12605: /* ------ Other prevalence ratios------------ */
1.126 brouard 12606:
1.215 brouard 12607: free_ivector(wav,1,imx);
12608: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12609: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12610: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12611:
12612:
1.127 brouard 12613: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12614:
1.201 brouard 12615: strcpy(filerese,"E_");
12616: strcat(filerese,fileresu);
1.126 brouard 12617: if((ficreseij=fopen(filerese,"w"))==NULL) {
12618: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12619: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12620: }
1.208 brouard 12621: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12622: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12623:
12624: pstamp(ficreseij);
1.219 brouard 12625:
1.235 brouard 12626: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12627: if (cptcovn < 1){i1=1;}
12628:
12629: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12630: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12631: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12632: continue;
1.219 brouard 12633: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12634: printf("\n#****** ");
1.225 brouard 12635: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12636: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12637: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12638: }
12639: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12640: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12641: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12642: }
12643: fprintf(ficreseij,"******\n");
1.235 brouard 12644: printf("******\n");
1.219 brouard 12645:
12646: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12647: oldm=oldms;savm=savms;
1.235 brouard 12648: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12649:
1.219 brouard 12650: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12651: }
12652: fclose(ficreseij);
1.208 brouard 12653: printf("done evsij\n");fflush(stdout);
12654: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12655:
1.218 brouard 12656:
1.227 brouard 12657: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12658:
1.201 brouard 12659: strcpy(filerest,"T_");
12660: strcat(filerest,fileresu);
1.127 brouard 12661: if((ficrest=fopen(filerest,"w"))==NULL) {
12662: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12663: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12664: }
1.208 brouard 12665: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12666: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12667: strcpy(fileresstde,"STDE_");
12668: strcat(fileresstde,fileresu);
1.126 brouard 12669: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12670: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12671: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12672: }
1.227 brouard 12673: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12674: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12675:
1.201 brouard 12676: strcpy(filerescve,"CVE_");
12677: strcat(filerescve,fileresu);
1.126 brouard 12678: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12679: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12680: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12681: }
1.227 brouard 12682: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12683: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12684:
1.201 brouard 12685: strcpy(fileresv,"V_");
12686: strcat(fileresv,fileresu);
1.126 brouard 12687: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12688: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12689: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12690: }
1.227 brouard 12691: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12692: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12693:
1.235 brouard 12694: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12695: if (cptcovn < 1){i1=1;}
12696:
12697: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12698: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12699: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12700: continue;
1.242 brouard 12701: printf("\n#****** Result for:");
12702: fprintf(ficrest,"\n#****** Result for:");
12703: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12704: for(j=1;j<=cptcoveff;j++){
12705: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12706: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12707: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12708: }
1.235 brouard 12709: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12710: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12711: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12712: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12713: }
1.208 brouard 12714: fprintf(ficrest,"******\n");
1.227 brouard 12715: fprintf(ficlog,"******\n");
12716: printf("******\n");
1.208 brouard 12717:
12718: fprintf(ficresstdeij,"\n#****** ");
12719: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12720: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12721: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12722: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12723: }
1.235 brouard 12724: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12725: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12726: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12727: }
1.208 brouard 12728: fprintf(ficresstdeij,"******\n");
12729: fprintf(ficrescveij,"******\n");
12730:
12731: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12732: /* pstamp(ficresvij); */
1.225 brouard 12733: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12734: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12735: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12736: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12737: }
1.208 brouard 12738: fprintf(ficresvij,"******\n");
12739:
12740: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12741: oldm=oldms;savm=savms;
1.235 brouard 12742: printf(" cvevsij ");
12743: fprintf(ficlog, " cvevsij ");
12744: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12745: printf(" end cvevsij \n ");
12746: fprintf(ficlog, " end cvevsij \n ");
12747:
12748: /*
12749: */
12750: /* goto endfree; */
12751:
12752: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12753: pstamp(ficrest);
12754:
1.269 brouard 12755: epj=vector(1,nlstate+1);
1.208 brouard 12756: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12757: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12758: cptcod= 0; /* To be deleted */
12759: printf("varevsij vpopbased=%d \n",vpopbased);
12760: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12761: 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 12762: 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 ");
12763: if(vpopbased==1)
12764: 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);
12765: else
1.288 brouard 12766: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12767: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12768: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12769: fprintf(ficrest,"\n");
12770: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 12771: printf("Computing age specific forward period (stable) prevalences in each health state \n");
12772: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12773: for(age=bage; age <=fage ;age++){
1.235 brouard 12774: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12775: if (vpopbased==1) {
12776: if(mobilav ==0){
12777: for(i=1; i<=nlstate;i++)
12778: prlim[i][i]=probs[(int)age][i][k];
12779: }else{ /* mobilav */
12780: for(i=1; i<=nlstate;i++)
12781: prlim[i][i]=mobaverage[(int)age][i][k];
12782: }
12783: }
1.219 brouard 12784:
1.227 brouard 12785: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12786: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12787: /* printf(" age %4.0f ",age); */
12788: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12789: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12790: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12791: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12792: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12793: }
12794: epj[nlstate+1] +=epj[j];
12795: }
12796: /* printf(" age %4.0f \n",age); */
1.219 brouard 12797:
1.227 brouard 12798: for(i=1, vepp=0.;i <=nlstate;i++)
12799: for(j=1;j <=nlstate;j++)
12800: vepp += vareij[i][j][(int)age];
12801: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12802: for(j=1;j <=nlstate;j++){
12803: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12804: }
12805: fprintf(ficrest,"\n");
12806: }
1.208 brouard 12807: } /* End vpopbased */
1.269 brouard 12808: free_vector(epj,1,nlstate+1);
1.208 brouard 12809: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12810: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12811: printf("done selection\n");fflush(stdout);
12812: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12813:
1.235 brouard 12814: } /* End k selection */
1.227 brouard 12815:
12816: printf("done State-specific expectancies\n");fflush(stdout);
12817: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12818:
1.288 brouard 12819: /* variance-covariance of forward period prevalence*/
1.269 brouard 12820: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12821:
1.227 brouard 12822:
1.290 brouard 12823: free_vector(weight,firstobs,lastobs);
1.227 brouard 12824: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 12825: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
12826: free_matrix(anint,1,maxwav,firstobs,lastobs);
12827: free_matrix(mint,1,maxwav,firstobs,lastobs);
12828: free_ivector(cod,firstobs,lastobs);
1.227 brouard 12829: free_ivector(tab,1,NCOVMAX);
12830: fclose(ficresstdeij);
12831: fclose(ficrescveij);
12832: fclose(ficresvij);
12833: fclose(ficrest);
12834: fclose(ficpar);
12835:
12836:
1.126 brouard 12837: /*---------- End : free ----------------*/
1.219 brouard 12838: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12839: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12840: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12841: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12842: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12843: } /* mle==-3 arrives here for freeing */
1.227 brouard 12844: /* endfree:*/
12845: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12846: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12847: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 12848: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
12849: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
12850: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
12851: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 12852: free_matrix(matcov,1,npar,1,npar);
12853: free_matrix(hess,1,npar,1,npar);
12854: /*free_vector(delti,1,npar);*/
12855: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12856: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12857: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12858: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12859:
12860: free_ivector(ncodemax,1,NCOVMAX);
12861: free_ivector(ncodemaxwundef,1,NCOVMAX);
12862: free_ivector(Dummy,-1,NCOVMAX);
12863: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12864: free_ivector(DummyV,1,NCOVMAX);
12865: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12866: free_ivector(Typevar,-1,NCOVMAX);
12867: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12868: free_ivector(TvarsQ,1,NCOVMAX);
12869: free_ivector(TvarsQind,1,NCOVMAX);
12870: free_ivector(TvarsD,1,NCOVMAX);
12871: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12872: free_ivector(TvarFD,1,NCOVMAX);
12873: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12874: free_ivector(TvarF,1,NCOVMAX);
12875: free_ivector(TvarFind,1,NCOVMAX);
12876: free_ivector(TvarV,1,NCOVMAX);
12877: free_ivector(TvarVind,1,NCOVMAX);
12878: free_ivector(TvarA,1,NCOVMAX);
12879: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12880: free_ivector(TvarFQ,1,NCOVMAX);
12881: free_ivector(TvarFQind,1,NCOVMAX);
12882: free_ivector(TvarVD,1,NCOVMAX);
12883: free_ivector(TvarVDind,1,NCOVMAX);
12884: free_ivector(TvarVQ,1,NCOVMAX);
12885: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12886: free_ivector(Tvarsel,1,NCOVMAX);
12887: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12888: free_ivector(Tposprod,1,NCOVMAX);
12889: free_ivector(Tprod,1,NCOVMAX);
12890: free_ivector(Tvaraff,1,NCOVMAX);
12891: free_ivector(invalidvarcomb,1,ncovcombmax);
12892: free_ivector(Tage,1,NCOVMAX);
12893: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12894: free_ivector(TmodelInvind,1,NCOVMAX);
12895: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12896:
12897: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12898: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12899: fflush(fichtm);
12900: fflush(ficgp);
12901:
1.227 brouard 12902:
1.126 brouard 12903: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12904: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12905: 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 12906: }else{
12907: printf("End of Imach\n");
12908: fprintf(ficlog,"End of Imach\n");
12909: }
12910: printf("See log file on %s\n",filelog);
12911: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12912: /*(void) gettimeofday(&end_time,&tzp);*/
12913: rend_time = time(NULL);
12914: end_time = *localtime(&rend_time);
12915: /* tml = *localtime(&end_time.tm_sec); */
12916: strcpy(strtend,asctime(&end_time));
1.126 brouard 12917: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12918: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12919: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12920:
1.157 brouard 12921: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12922: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12923: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12924: /* printf("Total time was %d uSec.\n", total_usecs);*/
12925: /* if(fileappend(fichtm,optionfilehtm)){ */
12926: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12927: fclose(fichtm);
12928: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12929: fclose(fichtmcov);
12930: fclose(ficgp);
12931: fclose(ficlog);
12932: /*------ End -----------*/
1.227 brouard 12933:
1.281 brouard 12934:
12935: /* Executes gnuplot */
1.227 brouard 12936:
12937: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12938: #ifdef WIN32
1.227 brouard 12939: if (_chdir(pathcd) != 0)
12940: printf("Can't move to directory %s!\n",path);
12941: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12942: #else
1.227 brouard 12943: if(chdir(pathcd) != 0)
12944: printf("Can't move to directory %s!\n", path);
12945: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12946: #endif
1.126 brouard 12947: printf("Current directory %s!\n",pathcd);
12948: /*strcat(plotcmd,CHARSEPARATOR);*/
12949: sprintf(plotcmd,"gnuplot");
1.157 brouard 12950: #ifdef _WIN32
1.126 brouard 12951: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12952: #endif
12953: if(!stat(plotcmd,&info)){
1.158 brouard 12954: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12955: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12956: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12957: }else
12958: strcpy(pplotcmd,plotcmd);
1.157 brouard 12959: #ifdef __unix
1.126 brouard 12960: strcpy(plotcmd,GNUPLOTPROGRAM);
12961: if(!stat(plotcmd,&info)){
1.158 brouard 12962: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12963: }else
12964: strcpy(pplotcmd,plotcmd);
12965: #endif
12966: }else
12967: strcpy(pplotcmd,plotcmd);
12968:
12969: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12970: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 12971: strcpy(pplotcmd,plotcmd);
1.227 brouard 12972:
1.126 brouard 12973: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 12974: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12975: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12976: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 12977: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 12978: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 12979: strcpy(plotcmd,pplotcmd);
12980: }
1.126 brouard 12981: }
1.158 brouard 12982: printf(" Successful, please wait...");
1.126 brouard 12983: while (z[0] != 'q') {
12984: /* chdir(path); */
1.154 brouard 12985: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12986: scanf("%s",z);
12987: /* if (z[0] == 'c') system("./imach"); */
12988: if (z[0] == 'e') {
1.158 brouard 12989: #ifdef __APPLE__
1.152 brouard 12990: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12991: #elif __linux
12992: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12993: #else
1.152 brouard 12994: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12995: #endif
12996: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12997: system(pplotcmd);
1.126 brouard 12998: }
12999: else if (z[0] == 'g') system(plotcmd);
13000: else if (z[0] == 'q') exit(0);
13001: }
1.227 brouard 13002: end:
1.126 brouard 13003: while (z[0] != 'q') {
1.195 brouard 13004: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13005: scanf("%s",z);
13006: }
1.283 brouard 13007: printf("End\n");
1.282 brouard 13008: exit(0);
1.126 brouard 13009: }
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