Annotation of imach/src/imach.c, revision 1.279
1.279 ! brouard 1: /* $Id: imach.c,v 1.278 2017/07/19 14:09:02 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.279 ! brouard 4: Revision 1.278 2017/07/19 14:09:02 brouard
! 5: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
! 6:
1.278 brouard 7: Revision 1.277 2017/07/17 08:53:49 brouard
8: Summary: BOM files can be read now
9:
1.277 brouard 10: Revision 1.276 2017/06/30 15:48:31 brouard
11: Summary: Graphs improvements
12:
1.276 brouard 13: Revision 1.275 2017/06/30 13:39:33 brouard
14: Summary: Saito's color
15:
1.275 brouard 16: Revision 1.274 2017/06/29 09:47:08 brouard
17: Summary: Version 0.99r14
18:
1.274 brouard 19: Revision 1.273 2017/06/27 11:06:02 brouard
20: Summary: More documentation on projections
21:
1.273 brouard 22: Revision 1.272 2017/06/27 10:22:40 brouard
23: Summary: Color of backprojection changed from 6 to 5(yellow)
24:
1.272 brouard 25: Revision 1.271 2017/06/27 10:17:50 brouard
26: Summary: Some bug with rint
27:
1.271 brouard 28: Revision 1.270 2017/05/24 05:45:29 brouard
29: *** empty log message ***
30:
1.270 brouard 31: Revision 1.269 2017/05/23 08:39:25 brouard
32: Summary: Code into subroutine, cleanings
33:
1.269 brouard 34: Revision 1.268 2017/05/18 20:09:32 brouard
35: Summary: backprojection and confidence intervals of backprevalence
36:
1.268 brouard 37: Revision 1.267 2017/05/13 10:25:05 brouard
38: Summary: temporary save for backprojection
39:
1.267 brouard 40: Revision 1.266 2017/05/13 07:26:12 brouard
41: Summary: Version 0.99r13 (improvements and bugs fixed)
42:
1.266 brouard 43: Revision 1.265 2017/04/26 16:22:11 brouard
44: Summary: imach 0.99r13 Some bugs fixed
45:
1.265 brouard 46: Revision 1.264 2017/04/26 06:01:29 brouard
47: Summary: Labels in graphs
48:
1.264 brouard 49: Revision 1.263 2017/04/24 15:23:15 brouard
50: Summary: to save
51:
1.263 brouard 52: Revision 1.262 2017/04/18 16:48:12 brouard
53: *** empty log message ***
54:
1.262 brouard 55: Revision 1.261 2017/04/05 10:14:09 brouard
56: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
57:
1.261 brouard 58: Revision 1.260 2017/04/04 17:46:59 brouard
59: Summary: Gnuplot indexations fixed (humm)
60:
1.260 brouard 61: Revision 1.259 2017/04/04 13:01:16 brouard
62: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
63:
1.259 brouard 64: Revision 1.258 2017/04/03 10:17:47 brouard
65: Summary: Version 0.99r12
66:
67: Some cleanings, conformed with updated documentation.
68:
1.258 brouard 69: Revision 1.257 2017/03/29 16:53:30 brouard
70: Summary: Temp
71:
1.257 brouard 72: Revision 1.256 2017/03/27 05:50:23 brouard
73: Summary: Temporary
74:
1.256 brouard 75: Revision 1.255 2017/03/08 16:02:28 brouard
76: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
77:
1.255 brouard 78: Revision 1.254 2017/03/08 07:13:00 brouard
79: Summary: Fixing data parameter line
80:
1.254 brouard 81: Revision 1.253 2016/12/15 11:59:41 brouard
82: Summary: 0.99 in progress
83:
1.253 brouard 84: Revision 1.252 2016/09/15 21:15:37 brouard
85: *** empty log message ***
86:
1.252 brouard 87: Revision 1.251 2016/09/15 15:01:13 brouard
88: Summary: not working
89:
1.251 brouard 90: Revision 1.250 2016/09/08 16:07:27 brouard
91: Summary: continue
92:
1.250 brouard 93: Revision 1.249 2016/09/07 17:14:18 brouard
94: Summary: Starting values from frequencies
95:
1.249 brouard 96: Revision 1.248 2016/09/07 14:10:18 brouard
97: *** empty log message ***
98:
1.248 brouard 99: Revision 1.247 2016/09/02 11:11:21 brouard
100: *** empty log message ***
101:
1.247 brouard 102: Revision 1.246 2016/09/02 08:49:22 brouard
103: *** empty log message ***
104:
1.246 brouard 105: Revision 1.245 2016/09/02 07:25:01 brouard
106: *** empty log message ***
107:
1.245 brouard 108: Revision 1.244 2016/09/02 07:17:34 brouard
109: *** empty log message ***
110:
1.244 brouard 111: Revision 1.243 2016/09/02 06:45:35 brouard
112: *** empty log message ***
113:
1.243 brouard 114: Revision 1.242 2016/08/30 15:01:20 brouard
115: Summary: Fixing a lots
116:
1.242 brouard 117: Revision 1.241 2016/08/29 17:17:25 brouard
118: Summary: gnuplot problem in Back projection to fix
119:
1.241 brouard 120: Revision 1.240 2016/08/29 07:53:18 brouard
121: Summary: Better
122:
1.240 brouard 123: Revision 1.239 2016/08/26 15:51:03 brouard
124: Summary: Improvement in Powell output in order to copy and paste
125:
126: Author:
127:
1.239 brouard 128: Revision 1.238 2016/08/26 14:23:35 brouard
129: Summary: Starting tests of 0.99
130:
1.238 brouard 131: Revision 1.237 2016/08/26 09:20:19 brouard
132: Summary: to valgrind
133:
1.237 brouard 134: Revision 1.236 2016/08/25 10:50:18 brouard
135: *** empty log message ***
136:
1.236 brouard 137: Revision 1.235 2016/08/25 06:59:23 brouard
138: *** empty log message ***
139:
1.235 brouard 140: Revision 1.234 2016/08/23 16:51:20 brouard
141: *** empty log message ***
142:
1.234 brouard 143: Revision 1.233 2016/08/23 07:40:50 brouard
144: Summary: not working
145:
1.233 brouard 146: Revision 1.232 2016/08/22 14:20:21 brouard
147: Summary: not working
148:
1.232 brouard 149: Revision 1.231 2016/08/22 07:17:15 brouard
150: Summary: not working
151:
1.231 brouard 152: Revision 1.230 2016/08/22 06:55:53 brouard
153: Summary: Not working
154:
1.230 brouard 155: Revision 1.229 2016/07/23 09:45:53 brouard
156: Summary: Completing for func too
157:
1.229 brouard 158: Revision 1.228 2016/07/22 17:45:30 brouard
159: Summary: Fixing some arrays, still debugging
160:
1.227 brouard 161: Revision 1.226 2016/07/12 18:42:34 brouard
162: Summary: temp
163:
1.226 brouard 164: Revision 1.225 2016/07/12 08:40:03 brouard
165: Summary: saving but not running
166:
1.225 brouard 167: Revision 1.224 2016/07/01 13:16:01 brouard
168: Summary: Fixes
169:
1.224 brouard 170: Revision 1.223 2016/02/19 09:23:35 brouard
171: Summary: temporary
172:
1.223 brouard 173: Revision 1.222 2016/02/17 08:14:50 brouard
174: Summary: Probably last 0.98 stable version 0.98r6
175:
1.222 brouard 176: Revision 1.221 2016/02/15 23:35:36 brouard
177: Summary: minor bug
178:
1.220 brouard 179: Revision 1.219 2016/02/15 00:48:12 brouard
180: *** empty log message ***
181:
1.219 brouard 182: Revision 1.218 2016/02/12 11:29:23 brouard
183: Summary: 0.99 Back projections
184:
1.218 brouard 185: Revision 1.217 2015/12/23 17:18:31 brouard
186: Summary: Experimental backcast
187:
1.217 brouard 188: Revision 1.216 2015/12/18 17:32:11 brouard
189: Summary: 0.98r4 Warning and status=-2
190:
191: Version 0.98r4 is now:
192: - displaying an error when status is -1, date of interview unknown and date of death known;
193: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
194: Older changes concerning s=-2, dating from 2005 have been supersed.
195:
1.216 brouard 196: Revision 1.215 2015/12/16 08:52:24 brouard
197: Summary: 0.98r4 working
198:
1.215 brouard 199: Revision 1.214 2015/12/16 06:57:54 brouard
200: Summary: temporary not working
201:
1.214 brouard 202: Revision 1.213 2015/12/11 18:22:17 brouard
203: Summary: 0.98r4
204:
1.213 brouard 205: Revision 1.212 2015/11/21 12:47:24 brouard
206: Summary: minor typo
207:
1.212 brouard 208: Revision 1.211 2015/11/21 12:41:11 brouard
209: Summary: 0.98r3 with some graph of projected cross-sectional
210:
211: Author: Nicolas Brouard
212:
1.211 brouard 213: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 214: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 215: Summary: Adding ftolpl parameter
216: Author: N Brouard
217:
218: We had difficulties to get smoothed confidence intervals. It was due
219: to the period prevalence which wasn't computed accurately. The inner
220: parameter ftolpl is now an outer parameter of the .imach parameter
221: file after estepm. If ftolpl is small 1.e-4 and estepm too,
222: computation are long.
223:
1.209 brouard 224: Revision 1.208 2015/11/17 14:31:57 brouard
225: Summary: temporary
226:
1.208 brouard 227: Revision 1.207 2015/10/27 17:36:57 brouard
228: *** empty log message ***
229:
1.207 brouard 230: Revision 1.206 2015/10/24 07:14:11 brouard
231: *** empty log message ***
232:
1.206 brouard 233: Revision 1.205 2015/10/23 15:50:53 brouard
234: Summary: 0.98r3 some clarification for graphs on likelihood contributions
235:
1.205 brouard 236: Revision 1.204 2015/10/01 16:20:26 brouard
237: Summary: Some new graphs of contribution to likelihood
238:
1.204 brouard 239: Revision 1.203 2015/09/30 17:45:14 brouard
240: Summary: looking at better estimation of the hessian
241:
242: Also a better criteria for convergence to the period prevalence And
243: therefore adding the number of years needed to converge. (The
244: prevalence in any alive state shold sum to one
245:
1.203 brouard 246: Revision 1.202 2015/09/22 19:45:16 brouard
247: Summary: Adding some overall graph on contribution to likelihood. Might change
248:
1.202 brouard 249: Revision 1.201 2015/09/15 17:34:58 brouard
250: Summary: 0.98r0
251:
252: - Some new graphs like suvival functions
253: - Some bugs fixed like model=1+age+V2.
254:
1.201 brouard 255: Revision 1.200 2015/09/09 16:53:55 brouard
256: Summary: Big bug thanks to Flavia
257:
258: Even model=1+age+V2. did not work anymore
259:
1.200 brouard 260: Revision 1.199 2015/09/07 14:09:23 brouard
261: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
262:
1.199 brouard 263: Revision 1.198 2015/09/03 07:14:39 brouard
264: Summary: 0.98q5 Flavia
265:
1.198 brouard 266: Revision 1.197 2015/09/01 18:24:39 brouard
267: *** empty log message ***
268:
1.197 brouard 269: Revision 1.196 2015/08/18 23:17:52 brouard
270: Summary: 0.98q5
271:
1.196 brouard 272: Revision 1.195 2015/08/18 16:28:39 brouard
273: Summary: Adding a hack for testing purpose
274:
275: After reading the title, ftol and model lines, if the comment line has
276: a q, starting with #q, the answer at the end of the run is quit. It
277: permits to run test files in batch with ctest. The former workaround was
278: $ echo q | imach foo.imach
279:
1.195 brouard 280: Revision 1.194 2015/08/18 13:32:00 brouard
281: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
282:
1.194 brouard 283: Revision 1.193 2015/08/04 07:17:42 brouard
284: Summary: 0.98q4
285:
1.193 brouard 286: Revision 1.192 2015/07/16 16:49:02 brouard
287: Summary: Fixing some outputs
288:
1.192 brouard 289: Revision 1.191 2015/07/14 10:00:33 brouard
290: Summary: Some fixes
291:
1.191 brouard 292: Revision 1.190 2015/05/05 08:51:13 brouard
293: Summary: Adding digits in output parameters (7 digits instead of 6)
294:
295: Fix 1+age+.
296:
1.190 brouard 297: Revision 1.189 2015/04/30 14:45:16 brouard
298: Summary: 0.98q2
299:
1.189 brouard 300: Revision 1.188 2015/04/30 08:27:53 brouard
301: *** empty log message ***
302:
1.188 brouard 303: Revision 1.187 2015/04/29 09:11:15 brouard
304: *** empty log message ***
305:
1.187 brouard 306: Revision 1.186 2015/04/23 12:01:52 brouard
307: Summary: V1*age is working now, version 0.98q1
308:
309: Some codes had been disabled in order to simplify and Vn*age was
310: working in the optimization phase, ie, giving correct MLE parameters,
311: but, as usual, outputs were not correct and program core dumped.
312:
1.186 brouard 313: Revision 1.185 2015/03/11 13:26:42 brouard
314: Summary: Inclusion of compile and links command line for Intel Compiler
315:
1.185 brouard 316: Revision 1.184 2015/03/11 11:52:39 brouard
317: Summary: Back from Windows 8. Intel Compiler
318:
1.184 brouard 319: Revision 1.183 2015/03/10 20:34:32 brouard
320: Summary: 0.98q0, trying with directest, mnbrak fixed
321:
322: We use directest instead of original Powell test; probably no
323: incidence on the results, but better justifications;
324: We fixed Numerical Recipes mnbrak routine which was wrong and gave
325: wrong results.
326:
1.183 brouard 327: Revision 1.182 2015/02/12 08:19:57 brouard
328: Summary: Trying to keep directest which seems simpler and more general
329: Author: Nicolas Brouard
330:
1.182 brouard 331: Revision 1.181 2015/02/11 23:22:24 brouard
332: Summary: Comments on Powell added
333:
334: Author:
335:
1.181 brouard 336: Revision 1.180 2015/02/11 17:33:45 brouard
337: Summary: Finishing move from main to function (hpijx and prevalence_limit)
338:
1.180 brouard 339: Revision 1.179 2015/01/04 09:57:06 brouard
340: Summary: back to OS/X
341:
1.179 brouard 342: Revision 1.178 2015/01/04 09:35:48 brouard
343: *** empty log message ***
344:
1.178 brouard 345: Revision 1.177 2015/01/03 18:40:56 brouard
346: Summary: Still testing ilc32 on OSX
347:
1.177 brouard 348: Revision 1.176 2015/01/03 16:45:04 brouard
349: *** empty log message ***
350:
1.176 brouard 351: Revision 1.175 2015/01/03 16:33:42 brouard
352: *** empty log message ***
353:
1.175 brouard 354: Revision 1.174 2015/01/03 16:15:49 brouard
355: Summary: Still in cross-compilation
356:
1.174 brouard 357: Revision 1.173 2015/01/03 12:06:26 brouard
358: Summary: trying to detect cross-compilation
359:
1.173 brouard 360: Revision 1.172 2014/12/27 12:07:47 brouard
361: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
362:
1.172 brouard 363: Revision 1.171 2014/12/23 13:26:59 brouard
364: Summary: Back from Visual C
365:
366: Still problem with utsname.h on Windows
367:
1.171 brouard 368: Revision 1.170 2014/12/23 11:17:12 brouard
369: Summary: Cleaning some \%% back to %%
370:
371: The escape was mandatory for a specific compiler (which one?), but too many warnings.
372:
1.170 brouard 373: Revision 1.169 2014/12/22 23:08:31 brouard
374: Summary: 0.98p
375:
376: Outputs some informations on compiler used, OS etc. Testing on different platforms.
377:
1.169 brouard 378: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 379: Summary: update
1.169 brouard 380:
1.168 brouard 381: Revision 1.167 2014/12/22 13:50:56 brouard
382: Summary: Testing uname and compiler version and if compiled 32 or 64
383:
384: Testing on Linux 64
385:
1.167 brouard 386: Revision 1.166 2014/12/22 11:40:47 brouard
387: *** empty log message ***
388:
1.166 brouard 389: Revision 1.165 2014/12/16 11:20:36 brouard
390: Summary: After compiling on Visual C
391:
392: * imach.c (Module): Merging 1.61 to 1.162
393:
1.165 brouard 394: Revision 1.164 2014/12/16 10:52:11 brouard
395: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
396:
397: * imach.c (Module): Merging 1.61 to 1.162
398:
1.164 brouard 399: Revision 1.163 2014/12/16 10:30:11 brouard
400: * imach.c (Module): Merging 1.61 to 1.162
401:
1.163 brouard 402: Revision 1.162 2014/09/25 11:43:39 brouard
403: Summary: temporary backup 0.99!
404:
1.162 brouard 405: Revision 1.1 2014/09/16 11:06:58 brouard
406: Summary: With some code (wrong) for nlopt
407:
408: Author:
409:
410: Revision 1.161 2014/09/15 20:41:41 brouard
411: Summary: Problem with macro SQR on Intel compiler
412:
1.161 brouard 413: Revision 1.160 2014/09/02 09:24:05 brouard
414: *** empty log message ***
415:
1.160 brouard 416: Revision 1.159 2014/09/01 10:34:10 brouard
417: Summary: WIN32
418: Author: Brouard
419:
1.159 brouard 420: Revision 1.158 2014/08/27 17:11:51 brouard
421: *** empty log message ***
422:
1.158 brouard 423: Revision 1.157 2014/08/27 16:26:55 brouard
424: Summary: Preparing windows Visual studio version
425: Author: Brouard
426:
427: In order to compile on Visual studio, time.h is now correct and time_t
428: and tm struct should be used. difftime should be used but sometimes I
429: just make the differences in raw time format (time(&now).
430: Trying to suppress #ifdef LINUX
431: Add xdg-open for __linux in order to open default browser.
432:
1.157 brouard 433: Revision 1.156 2014/08/25 20:10:10 brouard
434: *** empty log message ***
435:
1.156 brouard 436: Revision 1.155 2014/08/25 18:32:34 brouard
437: Summary: New compile, minor changes
438: Author: Brouard
439:
1.155 brouard 440: Revision 1.154 2014/06/20 17:32:08 brouard
441: Summary: Outputs now all graphs of convergence to period prevalence
442:
1.154 brouard 443: Revision 1.153 2014/06/20 16:45:46 brouard
444: Summary: If 3 live state, convergence to period prevalence on same graph
445: Author: Brouard
446:
1.153 brouard 447: Revision 1.152 2014/06/18 17:54:09 brouard
448: Summary: open browser, use gnuplot on same dir than imach if not found in the path
449:
1.152 brouard 450: Revision 1.151 2014/06/18 16:43:30 brouard
451: *** empty log message ***
452:
1.151 brouard 453: Revision 1.150 2014/06/18 16:42:35 brouard
454: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
455: Author: brouard
456:
1.150 brouard 457: Revision 1.149 2014/06/18 15:51:14 brouard
458: Summary: Some fixes in parameter files errors
459: Author: Nicolas Brouard
460:
1.149 brouard 461: Revision 1.148 2014/06/17 17:38:48 brouard
462: Summary: Nothing new
463: Author: Brouard
464:
465: Just a new packaging for OS/X version 0.98nS
466:
1.148 brouard 467: Revision 1.147 2014/06/16 10:33:11 brouard
468: *** empty log message ***
469:
1.147 brouard 470: Revision 1.146 2014/06/16 10:20:28 brouard
471: Summary: Merge
472: Author: Brouard
473:
474: Merge, before building revised version.
475:
1.146 brouard 476: Revision 1.145 2014/06/10 21:23:15 brouard
477: Summary: Debugging with valgrind
478: Author: Nicolas Brouard
479:
480: Lot of changes in order to output the results with some covariates
481: After the Edimburgh REVES conference 2014, it seems mandatory to
482: improve the code.
483: No more memory valgrind error but a lot has to be done in order to
484: continue the work of splitting the code into subroutines.
485: Also, decodemodel has been improved. Tricode is still not
486: optimal. nbcode should be improved. Documentation has been added in
487: the source code.
488:
1.144 brouard 489: Revision 1.143 2014/01/26 09:45:38 brouard
490: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
491:
492: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
493: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
494:
1.143 brouard 495: Revision 1.142 2014/01/26 03:57:36 brouard
496: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
497:
498: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
499:
1.142 brouard 500: Revision 1.141 2014/01/26 02:42:01 brouard
501: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
502:
1.141 brouard 503: Revision 1.140 2011/09/02 10:37:54 brouard
504: Summary: times.h is ok with mingw32 now.
505:
1.140 brouard 506: Revision 1.139 2010/06/14 07:50:17 brouard
507: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
508: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
509:
1.139 brouard 510: Revision 1.138 2010/04/30 18:19:40 brouard
511: *** empty log message ***
512:
1.138 brouard 513: Revision 1.137 2010/04/29 18:11:38 brouard
514: (Module): Checking covariates for more complex models
515: than V1+V2. A lot of change to be done. Unstable.
516:
1.137 brouard 517: Revision 1.136 2010/04/26 20:30:53 brouard
518: (Module): merging some libgsl code. Fixing computation
519: of likelione (using inter/intrapolation if mle = 0) in order to
520: get same likelihood as if mle=1.
521: Some cleaning of code and comments added.
522:
1.136 brouard 523: Revision 1.135 2009/10/29 15:33:14 brouard
524: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
525:
1.135 brouard 526: Revision 1.134 2009/10/29 13:18:53 brouard
527: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
528:
1.134 brouard 529: Revision 1.133 2009/07/06 10:21:25 brouard
530: just nforces
531:
1.133 brouard 532: Revision 1.132 2009/07/06 08:22:05 brouard
533: Many tings
534:
1.132 brouard 535: Revision 1.131 2009/06/20 16:22:47 brouard
536: Some dimensions resccaled
537:
1.131 brouard 538: Revision 1.130 2009/05/26 06:44:34 brouard
539: (Module): Max Covariate is now set to 20 instead of 8. A
540: lot of cleaning with variables initialized to 0. Trying to make
541: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
542:
1.130 brouard 543: Revision 1.129 2007/08/31 13:49:27 lievre
544: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
545:
1.129 lievre 546: Revision 1.128 2006/06/30 13:02:05 brouard
547: (Module): Clarifications on computing e.j
548:
1.128 brouard 549: Revision 1.127 2006/04/28 18:11:50 brouard
550: (Module): Yes the sum of survivors was wrong since
551: imach-114 because nhstepm was no more computed in the age
552: loop. Now we define nhstepma in the age loop.
553: (Module): In order to speed up (in case of numerous covariates) we
554: compute health expectancies (without variances) in a first step
555: and then all the health expectancies with variances or standard
556: deviation (needs data from the Hessian matrices) which slows the
557: computation.
558: In the future we should be able to stop the program is only health
559: expectancies and graph are needed without standard deviations.
560:
1.127 brouard 561: Revision 1.126 2006/04/28 17:23:28 brouard
562: (Module): Yes the sum of survivors was wrong since
563: imach-114 because nhstepm was no more computed in the age
564: loop. Now we define nhstepma in the age loop.
565: Version 0.98h
566:
1.126 brouard 567: Revision 1.125 2006/04/04 15:20:31 lievre
568: Errors in calculation of health expectancies. Age was not initialized.
569: Forecasting file added.
570:
571: Revision 1.124 2006/03/22 17:13:53 lievre
572: Parameters are printed with %lf instead of %f (more numbers after the comma).
573: The log-likelihood is printed in the log file
574:
575: Revision 1.123 2006/03/20 10:52:43 brouard
576: * imach.c (Module): <title> changed, corresponds to .htm file
577: name. <head> headers where missing.
578:
579: * imach.c (Module): Weights can have a decimal point as for
580: English (a comma might work with a correct LC_NUMERIC environment,
581: otherwise the weight is truncated).
582: Modification of warning when the covariates values are not 0 or
583: 1.
584: Version 0.98g
585:
586: Revision 1.122 2006/03/20 09:45:41 brouard
587: (Module): Weights can have a decimal point as for
588: English (a comma might work with a correct LC_NUMERIC environment,
589: otherwise the weight is truncated).
590: Modification of warning when the covariates values are not 0 or
591: 1.
592: Version 0.98g
593:
594: Revision 1.121 2006/03/16 17:45:01 lievre
595: * imach.c (Module): Comments concerning covariates added
596:
597: * imach.c (Module): refinements in the computation of lli if
598: status=-2 in order to have more reliable computation if stepm is
599: not 1 month. Version 0.98f
600:
601: Revision 1.120 2006/03/16 15:10:38 lievre
602: (Module): refinements in the computation of lli if
603: status=-2 in order to have more reliable computation if stepm is
604: not 1 month. Version 0.98f
605:
606: Revision 1.119 2006/03/15 17:42:26 brouard
607: (Module): Bug if status = -2, the loglikelihood was
608: computed as likelihood omitting the logarithm. Version O.98e
609:
610: Revision 1.118 2006/03/14 18:20:07 brouard
611: (Module): varevsij Comments added explaining the second
612: table of variances if popbased=1 .
613: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
614: (Module): Function pstamp added
615: (Module): Version 0.98d
616:
617: Revision 1.117 2006/03/14 17:16:22 brouard
618: (Module): varevsij Comments added explaining the second
619: table of variances if popbased=1 .
620: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
621: (Module): Function pstamp added
622: (Module): Version 0.98d
623:
624: Revision 1.116 2006/03/06 10:29:27 brouard
625: (Module): Variance-covariance wrong links and
626: varian-covariance of ej. is needed (Saito).
627:
628: Revision 1.115 2006/02/27 12:17:45 brouard
629: (Module): One freematrix added in mlikeli! 0.98c
630:
631: Revision 1.114 2006/02/26 12:57:58 brouard
632: (Module): Some improvements in processing parameter
633: filename with strsep.
634:
635: Revision 1.113 2006/02/24 14:20:24 brouard
636: (Module): Memory leaks checks with valgrind and:
637: datafile was not closed, some imatrix were not freed and on matrix
638: allocation too.
639:
640: Revision 1.112 2006/01/30 09:55:26 brouard
641: (Module): Back to gnuplot.exe instead of wgnuplot.exe
642:
643: Revision 1.111 2006/01/25 20:38:18 brouard
644: (Module): Lots of cleaning and bugs added (Gompertz)
645: (Module): Comments can be added in data file. Missing date values
646: can be a simple dot '.'.
647:
648: Revision 1.110 2006/01/25 00:51:50 brouard
649: (Module): Lots of cleaning and bugs added (Gompertz)
650:
651: Revision 1.109 2006/01/24 19:37:15 brouard
652: (Module): Comments (lines starting with a #) are allowed in data.
653:
654: Revision 1.108 2006/01/19 18:05:42 lievre
655: Gnuplot problem appeared...
656: To be fixed
657:
658: Revision 1.107 2006/01/19 16:20:37 brouard
659: Test existence of gnuplot in imach path
660:
661: Revision 1.106 2006/01/19 13:24:36 brouard
662: Some cleaning and links added in html output
663:
664: Revision 1.105 2006/01/05 20:23:19 lievre
665: *** empty log message ***
666:
667: Revision 1.104 2005/09/30 16:11:43 lievre
668: (Module): sump fixed, loop imx fixed, and simplifications.
669: (Module): If the status is missing at the last wave but we know
670: that the person is alive, then we can code his/her status as -2
671: (instead of missing=-1 in earlier versions) and his/her
672: contributions to the likelihood is 1 - Prob of dying from last
673: health status (= 1-p13= p11+p12 in the easiest case of somebody in
674: the healthy state at last known wave). Version is 0.98
675:
676: Revision 1.103 2005/09/30 15:54:49 lievre
677: (Module): sump fixed, loop imx fixed, and simplifications.
678:
679: Revision 1.102 2004/09/15 17:31:30 brouard
680: Add the possibility to read data file including tab characters.
681:
682: Revision 1.101 2004/09/15 10:38:38 brouard
683: Fix on curr_time
684:
685: Revision 1.100 2004/07/12 18:29:06 brouard
686: Add version for Mac OS X. Just define UNIX in Makefile
687:
688: Revision 1.99 2004/06/05 08:57:40 brouard
689: *** empty log message ***
690:
691: Revision 1.98 2004/05/16 15:05:56 brouard
692: New version 0.97 . First attempt to estimate force of mortality
693: directly from the data i.e. without the need of knowing the health
694: state at each age, but using a Gompertz model: log u =a + b*age .
695: This is the basic analysis of mortality and should be done before any
696: other analysis, in order to test if the mortality estimated from the
697: cross-longitudinal survey is different from the mortality estimated
698: from other sources like vital statistic data.
699:
700: The same imach parameter file can be used but the option for mle should be -3.
701:
1.133 brouard 702: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 703: former routines in order to include the new code within the former code.
704:
705: The output is very simple: only an estimate of the intercept and of
706: the slope with 95% confident intervals.
707:
708: Current limitations:
709: A) Even if you enter covariates, i.e. with the
710: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
711: B) There is no computation of Life Expectancy nor Life Table.
712:
713: Revision 1.97 2004/02/20 13:25:42 lievre
714: Version 0.96d. Population forecasting command line is (temporarily)
715: suppressed.
716:
717: Revision 1.96 2003/07/15 15:38:55 brouard
718: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
719: rewritten within the same printf. Workaround: many printfs.
720:
721: Revision 1.95 2003/07/08 07:54:34 brouard
722: * imach.c (Repository):
723: (Repository): Using imachwizard code to output a more meaningful covariance
724: matrix (cov(a12,c31) instead of numbers.
725:
726: Revision 1.94 2003/06/27 13:00:02 brouard
727: Just cleaning
728:
729: Revision 1.93 2003/06/25 16:33:55 brouard
730: (Module): On windows (cygwin) function asctime_r doesn't
731: exist so I changed back to asctime which exists.
732: (Module): Version 0.96b
733:
734: Revision 1.92 2003/06/25 16:30:45 brouard
735: (Module): On windows (cygwin) function asctime_r doesn't
736: exist so I changed back to asctime which exists.
737:
738: Revision 1.91 2003/06/25 15:30:29 brouard
739: * imach.c (Repository): Duplicated warning errors corrected.
740: (Repository): Elapsed time after each iteration is now output. It
741: helps to forecast when convergence will be reached. Elapsed time
742: is stamped in powell. We created a new html file for the graphs
743: concerning matrix of covariance. It has extension -cov.htm.
744:
745: Revision 1.90 2003/06/24 12:34:15 brouard
746: (Module): Some bugs corrected for windows. Also, when
747: mle=-1 a template is output in file "or"mypar.txt with the design
748: of the covariance matrix to be input.
749:
750: Revision 1.89 2003/06/24 12:30:52 brouard
751: (Module): Some bugs corrected for windows. Also, when
752: mle=-1 a template is output in file "or"mypar.txt with the design
753: of the covariance matrix to be input.
754:
755: Revision 1.88 2003/06/23 17:54:56 brouard
756: * 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.
757:
758: Revision 1.87 2003/06/18 12:26:01 brouard
759: Version 0.96
760:
761: Revision 1.86 2003/06/17 20:04:08 brouard
762: (Module): Change position of html and gnuplot routines and added
763: routine fileappend.
764:
765: Revision 1.85 2003/06/17 13:12:43 brouard
766: * imach.c (Repository): Check when date of death was earlier that
767: current date of interview. It may happen when the death was just
768: prior to the death. In this case, dh was negative and likelihood
769: was wrong (infinity). We still send an "Error" but patch by
770: assuming that the date of death was just one stepm after the
771: interview.
772: (Repository): Because some people have very long ID (first column)
773: we changed int to long in num[] and we added a new lvector for
774: memory allocation. But we also truncated to 8 characters (left
775: truncation)
776: (Repository): No more line truncation errors.
777:
778: Revision 1.84 2003/06/13 21:44:43 brouard
779: * imach.c (Repository): Replace "freqsummary" at a correct
780: place. It differs from routine "prevalence" which may be called
781: many times. Probs is memory consuming and must be used with
782: parcimony.
783: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
784:
785: Revision 1.83 2003/06/10 13:39:11 lievre
786: *** empty log message ***
787:
788: Revision 1.82 2003/06/05 15:57:20 brouard
789: Add log in imach.c and fullversion number is now printed.
790:
791: */
792: /*
793: Interpolated Markov Chain
794:
795: Short summary of the programme:
796:
1.227 brouard 797: This program computes Healthy Life Expectancies or State-specific
798: (if states aren't health statuses) Expectancies from
799: cross-longitudinal data. Cross-longitudinal data consist in:
800:
801: -1- a first survey ("cross") where individuals from different ages
802: are interviewed on their health status or degree of disability (in
803: the case of a health survey which is our main interest)
804:
805: -2- at least a second wave of interviews ("longitudinal") which
806: measure each change (if any) in individual health status. Health
807: expectancies are computed from the time spent in each health state
808: according to a model. More health states you consider, more time is
809: necessary to reach the Maximum Likelihood of the parameters involved
810: in the model. The simplest model is the multinomial logistic model
811: where pij is the probability to be observed in state j at the second
812: wave conditional to be observed in state i at the first
813: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
814: etc , where 'age' is age and 'sex' is a covariate. If you want to
815: have a more complex model than "constant and age", you should modify
816: the program where the markup *Covariates have to be included here
817: again* invites you to do it. More covariates you add, slower the
1.126 brouard 818: convergence.
819:
820: The advantage of this computer programme, compared to a simple
821: multinomial logistic model, is clear when the delay between waves is not
822: identical for each individual. Also, if a individual missed an
823: intermediate interview, the information is lost, but taken into
824: account using an interpolation or extrapolation.
825:
826: hPijx is the probability to be observed in state i at age x+h
827: conditional to the observed state i at age x. The delay 'h' can be
828: split into an exact number (nh*stepm) of unobserved intermediate
829: states. This elementary transition (by month, quarter,
830: semester or year) is modelled as a multinomial logistic. The hPx
831: matrix is simply the matrix product of nh*stepm elementary matrices
832: and the contribution of each individual to the likelihood is simply
833: hPijx.
834:
835: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 836: of the life expectancies. It also computes the period (stable) prevalence.
837:
838: Back prevalence and projections:
1.227 brouard 839:
840: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
841: double agemaxpar, double ftolpl, int *ncvyearp, double
842: dateprev1,double dateprev2, int firstpass, int lastpass, int
843: mobilavproj)
844:
845: Computes the back prevalence limit for any combination of
846: covariate values k at any age between ageminpar and agemaxpar and
847: returns it in **bprlim. In the loops,
848:
849: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
850: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
851:
852: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 853: Computes for any combination of covariates k and any age between bage and fage
854: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
855: oldm=oldms;savm=savms;
1.227 brouard 856:
1.267 brouard 857: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 858: Computes the transition matrix starting at age 'age' over
859: 'nhstepm*hstepm*stepm' months (i.e. until
860: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 861: nhstepm*hstepm matrices.
862:
863: Returns p3mat[i][j][h] after calling
864: p3mat[i][j][h]=matprod2(newm,
865: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
866: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
867: oldm);
1.226 brouard 868:
869: Important routines
870:
871: - func (or funcone), computes logit (pij) distinguishing
872: o fixed variables (single or product dummies or quantitative);
873: o varying variables by:
874: (1) wave (single, product dummies, quantitative),
875: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
876: % fixed dummy (treated) or quantitative (not done because time-consuming);
877: % varying dummy (not done) or quantitative (not done);
878: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
879: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
880: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
881: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
882: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 883:
1.226 brouard 884:
885:
1.133 brouard 886: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
887: Institut national d'études démographiques, Paris.
1.126 brouard 888: This software have been partly granted by Euro-REVES, a concerted action
889: from the European Union.
890: It is copyrighted identically to a GNU software product, ie programme and
891: software can be distributed freely for non commercial use. Latest version
892: can be accessed at http://euroreves.ined.fr/imach .
893:
894: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
895: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
896:
897: **********************************************************************/
898: /*
899: main
900: read parameterfile
901: read datafile
902: concatwav
903: freqsummary
904: if (mle >= 1)
905: mlikeli
906: print results files
907: if mle==1
908: computes hessian
909: read end of parameter file: agemin, agemax, bage, fage, estepm
910: begin-prev-date,...
911: open gnuplot file
912: open html file
1.145 brouard 913: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
914: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
915: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
916: freexexit2 possible for memory heap.
917:
918: h Pij x | pij_nom ficrestpij
919: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
920: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
921: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
922:
923: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
924: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
925: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
926: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
927: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
928:
1.126 brouard 929: forecasting if prevfcast==1 prevforecast call prevalence()
930: health expectancies
931: Variance-covariance of DFLE
932: prevalence()
933: movingaverage()
934: varevsij()
935: if popbased==1 varevsij(,popbased)
936: total life expectancies
937: Variance of period (stable) prevalence
938: end
939: */
940:
1.187 brouard 941: /* #define DEBUG */
942: /* #define DEBUGBRENT */
1.203 brouard 943: /* #define DEBUGLINMIN */
944: /* #define DEBUGHESS */
945: #define DEBUGHESSIJ
1.224 brouard 946: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 947: #define POWELL /* Instead of NLOPT */
1.224 brouard 948: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 949: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
950: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 951:
952: #include <math.h>
953: #include <stdio.h>
954: #include <stdlib.h>
955: #include <string.h>
1.226 brouard 956: #include <ctype.h>
1.159 brouard 957:
958: #ifdef _WIN32
959: #include <io.h>
1.172 brouard 960: #include <windows.h>
961: #include <tchar.h>
1.159 brouard 962: #else
1.126 brouard 963: #include <unistd.h>
1.159 brouard 964: #endif
1.126 brouard 965:
966: #include <limits.h>
967: #include <sys/types.h>
1.171 brouard 968:
969: #if defined(__GNUC__)
970: #include <sys/utsname.h> /* Doesn't work on Windows */
971: #endif
972:
1.126 brouard 973: #include <sys/stat.h>
974: #include <errno.h>
1.159 brouard 975: /* extern int errno; */
1.126 brouard 976:
1.157 brouard 977: /* #ifdef LINUX */
978: /* #include <time.h> */
979: /* #include "timeval.h" */
980: /* #else */
981: /* #include <sys/time.h> */
982: /* #endif */
983:
1.126 brouard 984: #include <time.h>
985:
1.136 brouard 986: #ifdef GSL
987: #include <gsl/gsl_errno.h>
988: #include <gsl/gsl_multimin.h>
989: #endif
990:
1.167 brouard 991:
1.162 brouard 992: #ifdef NLOPT
993: #include <nlopt.h>
994: typedef struct {
995: double (* function)(double [] );
996: } myfunc_data ;
997: #endif
998:
1.126 brouard 999: /* #include <libintl.h> */
1000: /* #define _(String) gettext (String) */
1001:
1.251 brouard 1002: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1003:
1004: #define GNUPLOTPROGRAM "gnuplot"
1005: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1006: #define FILENAMELENGTH 132
1007:
1008: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1009: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1010:
1.144 brouard 1011: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1012: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1013:
1014: #define NINTERVMAX 8
1.144 brouard 1015: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1016: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1017: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1018: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1019: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1020: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1021: #define MAXN 20000
1.144 brouard 1022: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1023: /* #define AGESUP 130 */
1024: #define AGESUP 150
1.268 brouard 1025: #define AGEINF 0
1.218 brouard 1026: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1027: #define AGEBASE 40
1.194 brouard 1028: #define AGEOVERFLOW 1.e20
1.164 brouard 1029: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1030: #ifdef _WIN32
1031: #define DIRSEPARATOR '\\'
1032: #define CHARSEPARATOR "\\"
1033: #define ODIRSEPARATOR '/'
1034: #else
1.126 brouard 1035: #define DIRSEPARATOR '/'
1036: #define CHARSEPARATOR "/"
1037: #define ODIRSEPARATOR '\\'
1038: #endif
1039:
1.279 ! brouard 1040: /* $Id: imach.c,v 1.278 2017/07/19 14:09:02 brouard Exp $ */
1.126 brouard 1041: /* $State: Exp $ */
1.196 brouard 1042: #include "version.h"
1043: char version[]=__IMACH_VERSION__;
1.224 brouard 1044: char copyright[]="February 2016,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2018";
1.279 ! brouard 1045: char fullversion[]="$Revision: 1.278 $ $Date: 2017/07/19 14:09:02 $";
1.126 brouard 1046: char strstart[80];
1047: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1048: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1049: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1050: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1051: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1052: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1053: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1054: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1055: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1056: int cptcovprodnoage=0; /**< Number of covariate products without age */
1057: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1058: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1059: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1060: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1061: int nsd=0; /**< Total number of single dummy variables (output) */
1062: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1063: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1064: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1065: int ntveff=0; /**< ntveff number of effective time varying variables */
1066: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1067: int cptcov=0; /* Working variable */
1.218 brouard 1068: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1069: int npar=NPARMAX;
1070: int nlstate=2; /* Number of live states */
1071: int ndeath=1; /* Number of dead states */
1.130 brouard 1072: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1073: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1074: int popbased=0;
1075:
1076: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1077: int maxwav=0; /* Maxim number of waves */
1078: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1079: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1080: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1081: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1082: int mle=1, weightopt=0;
1.126 brouard 1083: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1084: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1085: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1086: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1087: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1088: int selected(int kvar); /* Is covariate kvar selected for printing results */
1089:
1.130 brouard 1090: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1091: double **matprod2(); /* test */
1.126 brouard 1092: double **oldm, **newm, **savm; /* Working pointers to matrices */
1093: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1094: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1095:
1.136 brouard 1096: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1097: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1098: FILE *ficlog, *ficrespow;
1.130 brouard 1099: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1100: double fretone; /* Only one call to likelihood */
1.130 brouard 1101: long ipmx=0; /* Number of contributions */
1.126 brouard 1102: double sw; /* Sum of weights */
1103: char filerespow[FILENAMELENGTH];
1104: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1105: FILE *ficresilk;
1106: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1107: FILE *ficresprobmorprev;
1108: FILE *fichtm, *fichtmcov; /* Html File */
1109: FILE *ficreseij;
1110: char filerese[FILENAMELENGTH];
1111: FILE *ficresstdeij;
1112: char fileresstde[FILENAMELENGTH];
1113: FILE *ficrescveij;
1114: char filerescve[FILENAMELENGTH];
1115: FILE *ficresvij;
1116: char fileresv[FILENAMELENGTH];
1.269 brouard 1117:
1.126 brouard 1118: char title[MAXLINE];
1.234 brouard 1119: char model[MAXLINE]; /**< The model line */
1.217 brouard 1120: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1121: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1122: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1123: char command[FILENAMELENGTH];
1124: int outcmd=0;
1125:
1.217 brouard 1126: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1127: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1128: char filelog[FILENAMELENGTH]; /* Log file */
1129: char filerest[FILENAMELENGTH];
1130: char fileregp[FILENAMELENGTH];
1131: char popfile[FILENAMELENGTH];
1132:
1133: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1134:
1.157 brouard 1135: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1136: /* struct timezone tzp; */
1137: /* extern int gettimeofday(); */
1138: struct tm tml, *gmtime(), *localtime();
1139:
1140: extern time_t time();
1141:
1142: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1143: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1144: struct tm tm;
1145:
1.126 brouard 1146: char strcurr[80], strfor[80];
1147:
1148: char *endptr;
1149: long lval;
1150: double dval;
1151:
1152: #define NR_END 1
1153: #define FREE_ARG char*
1154: #define FTOL 1.0e-10
1155:
1156: #define NRANSI
1.240 brouard 1157: #define ITMAX 200
1158: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1159:
1160: #define TOL 2.0e-4
1161:
1162: #define CGOLD 0.3819660
1163: #define ZEPS 1.0e-10
1164: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1165:
1166: #define GOLD 1.618034
1167: #define GLIMIT 100.0
1168: #define TINY 1.0e-20
1169:
1170: static double maxarg1,maxarg2;
1171: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1172: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1173:
1174: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1175: #define rint(a) floor(a+0.5)
1.166 brouard 1176: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1177: #define mytinydouble 1.0e-16
1.166 brouard 1178: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1179: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1180: /* static double dsqrarg; */
1181: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1182: static double sqrarg;
1183: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1184: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1185: int agegomp= AGEGOMP;
1186:
1187: int imx;
1188: int stepm=1;
1189: /* Stepm, step in month: minimum step interpolation*/
1190:
1191: int estepm;
1192: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1193:
1194: int m,nb;
1195: long *num;
1.197 brouard 1196: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1197: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1198: covariate for which somebody answered excluding
1199: undefined. Usually 2: 0 and 1. */
1200: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1201: covariate for which somebody answered including
1202: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1203: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1204: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1205: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1206: double *ageexmed,*agecens;
1207: double dateintmean=0;
1208:
1209: double *weight;
1210: int **s; /* Status */
1.141 brouard 1211: double *agedc;
1.145 brouard 1212: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1213: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1214: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1215: double **coqvar; /* Fixed quantitative covariate nqv */
1216: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1217: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1218: double idx;
1219: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1220: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1221: /*k 1 2 3 4 5 6 7 8 9 */
1222: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1223: /* Tndvar[k] 1 2 3 4 5 */
1224: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1225: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1226: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1227: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1228: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1229: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1230: /* Tprod[i]=k 4 7 */
1231: /* Tage[i]=k 5 8 */
1232: /* */
1233: /* Type */
1234: /* V 1 2 3 4 5 */
1235: /* F F V V V */
1236: /* D Q D D Q */
1237: /* */
1238: int *TvarsD;
1239: int *TvarsDind;
1240: int *TvarsQ;
1241: int *TvarsQind;
1242:
1.235 brouard 1243: #define MAXRESULTLINES 10
1244: int nresult=0;
1.258 brouard 1245: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1246: int TKresult[MAXRESULTLINES];
1.237 brouard 1247: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1248: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1249: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1250: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1251: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1252: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1253:
1.234 brouard 1254: /* 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 1255: 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 */
1256: 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 */
1257: 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 */
1258: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1259: 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 */
1260: 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 1261: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1262: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1263: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1264: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1265: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1266: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1267: 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 */
1268: 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 */
1269:
1.230 brouard 1270: int *Tvarsel; /**< Selected covariates for output */
1271: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1272: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1273: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1274: 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 1275: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1276: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1277: int *Tage;
1.227 brouard 1278: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1279: 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 1280: 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*/
1281: 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 1282: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1283: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1284: int **Tvard;
1285: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1286: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1287: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1288: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1289: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1290: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1291: double *lsurv, *lpop, *tpop;
1292:
1.231 brouard 1293: #define FD 1; /* Fixed dummy covariate */
1294: #define FQ 2; /* Fixed quantitative covariate */
1295: #define FP 3; /* Fixed product covariate */
1296: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1297: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1298: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1299: #define VD 10; /* Varying dummy covariate */
1300: #define VQ 11; /* Varying quantitative covariate */
1301: #define VP 12; /* Varying product covariate */
1302: #define VPDD 13; /* Varying product dummy*dummy covariate */
1303: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1304: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1305: #define APFD 16; /* Age product * fixed dummy covariate */
1306: #define APFQ 17; /* Age product * fixed quantitative covariate */
1307: #define APVD 18; /* Age product * varying dummy covariate */
1308: #define APVQ 19; /* Age product * varying quantitative covariate */
1309:
1310: #define FTYPE 1; /* Fixed covariate */
1311: #define VTYPE 2; /* Varying covariate (loop in wave) */
1312: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1313:
1314: struct kmodel{
1315: int maintype; /* main type */
1316: int subtype; /* subtype */
1317: };
1318: struct kmodel modell[NCOVMAX];
1319:
1.143 brouard 1320: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1321: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1322:
1323: /**************** split *************************/
1324: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1325: {
1326: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1327: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1328: */
1329: char *ss; /* pointer */
1.186 brouard 1330: int l1=0, l2=0; /* length counters */
1.126 brouard 1331:
1332: l1 = strlen(path ); /* length of path */
1333: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1334: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1335: if ( ss == NULL ) { /* no directory, so determine current directory */
1336: strcpy( name, path ); /* we got the fullname name because no directory */
1337: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1338: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1339: /* get current working directory */
1340: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1341: #ifdef WIN32
1342: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1343: #else
1344: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1345: #endif
1.126 brouard 1346: return( GLOCK_ERROR_GETCWD );
1347: }
1348: /* got dirc from getcwd*/
1349: printf(" DIRC = %s \n",dirc);
1.205 brouard 1350: } else { /* strip directory from path */
1.126 brouard 1351: ss++; /* after this, the filename */
1352: l2 = strlen( ss ); /* length of filename */
1353: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1354: strcpy( name, ss ); /* save file name */
1355: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1356: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1357: printf(" DIRC2 = %s \n",dirc);
1358: }
1359: /* We add a separator at the end of dirc if not exists */
1360: l1 = strlen( dirc ); /* length of directory */
1361: if( dirc[l1-1] != DIRSEPARATOR ){
1362: dirc[l1] = DIRSEPARATOR;
1363: dirc[l1+1] = 0;
1364: printf(" DIRC3 = %s \n",dirc);
1365: }
1366: ss = strrchr( name, '.' ); /* find last / */
1367: if (ss >0){
1368: ss++;
1369: strcpy(ext,ss); /* save extension */
1370: l1= strlen( name);
1371: l2= strlen(ss)+1;
1372: strncpy( finame, name, l1-l2);
1373: finame[l1-l2]= 0;
1374: }
1375:
1376: return( 0 ); /* we're done */
1377: }
1378:
1379:
1380: /******************************************/
1381:
1382: void replace_back_to_slash(char *s, char*t)
1383: {
1384: int i;
1385: int lg=0;
1386: i=0;
1387: lg=strlen(t);
1388: for(i=0; i<= lg; i++) {
1389: (s[i] = t[i]);
1390: if (t[i]== '\\') s[i]='/';
1391: }
1392: }
1393:
1.132 brouard 1394: char *trimbb(char *out, char *in)
1.137 brouard 1395: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1396: char *s;
1397: s=out;
1398: while (*in != '\0'){
1.137 brouard 1399: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1400: in++;
1401: }
1402: *out++ = *in++;
1403: }
1404: *out='\0';
1405: return s;
1406: }
1407:
1.187 brouard 1408: /* char *substrchaine(char *out, char *in, char *chain) */
1409: /* { */
1410: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1411: /* char *s, *t; */
1412: /* t=in;s=out; */
1413: /* while ((*in != *chain) && (*in != '\0')){ */
1414: /* *out++ = *in++; */
1415: /* } */
1416:
1417: /* /\* *in matches *chain *\/ */
1418: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1419: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1420: /* } */
1421: /* in--; chain--; */
1422: /* while ( (*in != '\0')){ */
1423: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1424: /* *out++ = *in++; */
1425: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1426: /* } */
1427: /* *out='\0'; */
1428: /* out=s; */
1429: /* return out; */
1430: /* } */
1431: char *substrchaine(char *out, char *in, char *chain)
1432: {
1433: /* Substract chain 'chain' from 'in', return and output 'out' */
1434: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1435:
1436: char *strloc;
1437:
1438: strcpy (out, in);
1439: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1440: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1441: if(strloc != NULL){
1442: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1443: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1444: /* strcpy (strloc, strloc +strlen(chain));*/
1445: }
1446: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1447: return out;
1448: }
1449:
1450:
1.145 brouard 1451: char *cutl(char *blocc, char *alocc, char *in, char occ)
1452: {
1.187 brouard 1453: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1454: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1455: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1456: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1457: */
1.160 brouard 1458: char *s, *t;
1.145 brouard 1459: t=in;s=in;
1460: while ((*in != occ) && (*in != '\0')){
1461: *alocc++ = *in++;
1462: }
1463: if( *in == occ){
1464: *(alocc)='\0';
1465: s=++in;
1466: }
1467:
1468: if (s == t) {/* occ not found */
1469: *(alocc-(in-s))='\0';
1470: in=s;
1471: }
1472: while ( *in != '\0'){
1473: *blocc++ = *in++;
1474: }
1475:
1476: *blocc='\0';
1477: return t;
1478: }
1.137 brouard 1479: char *cutv(char *blocc, char *alocc, char *in, char occ)
1480: {
1.187 brouard 1481: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1482: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1483: gives blocc="abcdef2ghi" and alocc="j".
1484: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1485: */
1486: char *s, *t;
1487: t=in;s=in;
1488: while (*in != '\0'){
1489: while( *in == occ){
1490: *blocc++ = *in++;
1491: s=in;
1492: }
1493: *blocc++ = *in++;
1494: }
1495: if (s == t) /* occ not found */
1496: *(blocc-(in-s))='\0';
1497: else
1498: *(blocc-(in-s)-1)='\0';
1499: in=s;
1500: while ( *in != '\0'){
1501: *alocc++ = *in++;
1502: }
1503:
1504: *alocc='\0';
1505: return s;
1506: }
1507:
1.126 brouard 1508: int nbocc(char *s, char occ)
1509: {
1510: int i,j=0;
1511: int lg=20;
1512: i=0;
1513: lg=strlen(s);
1514: for(i=0; i<= lg; i++) {
1.234 brouard 1515: if (s[i] == occ ) j++;
1.126 brouard 1516: }
1517: return j;
1518: }
1519:
1.137 brouard 1520: /* void cutv(char *u,char *v, char*t, char occ) */
1521: /* { */
1522: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1523: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1524: /* gives u="abcdef2ghi" and v="j" *\/ */
1525: /* int i,lg,j,p=0; */
1526: /* i=0; */
1527: /* lg=strlen(t); */
1528: /* for(j=0; j<=lg-1; j++) { */
1529: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1530: /* } */
1.126 brouard 1531:
1.137 brouard 1532: /* for(j=0; j<p; j++) { */
1533: /* (u[j] = t[j]); */
1534: /* } */
1535: /* u[p]='\0'; */
1.126 brouard 1536:
1.137 brouard 1537: /* for(j=0; j<= lg; j++) { */
1538: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1539: /* } */
1540: /* } */
1.126 brouard 1541:
1.160 brouard 1542: #ifdef _WIN32
1543: char * strsep(char **pp, const char *delim)
1544: {
1545: char *p, *q;
1546:
1547: if ((p = *pp) == NULL)
1548: return 0;
1549: if ((q = strpbrk (p, delim)) != NULL)
1550: {
1551: *pp = q + 1;
1552: *q = '\0';
1553: }
1554: else
1555: *pp = 0;
1556: return p;
1557: }
1558: #endif
1559:
1.126 brouard 1560: /********************** nrerror ********************/
1561:
1562: void nrerror(char error_text[])
1563: {
1564: fprintf(stderr,"ERREUR ...\n");
1565: fprintf(stderr,"%s\n",error_text);
1566: exit(EXIT_FAILURE);
1567: }
1568: /*********************** vector *******************/
1569: double *vector(int nl, int nh)
1570: {
1571: double *v;
1572: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1573: if (!v) nrerror("allocation failure in vector");
1574: return v-nl+NR_END;
1575: }
1576:
1577: /************************ free vector ******************/
1578: void free_vector(double*v, int nl, int nh)
1579: {
1580: free((FREE_ARG)(v+nl-NR_END));
1581: }
1582:
1583: /************************ivector *******************************/
1584: int *ivector(long nl,long nh)
1585: {
1586: int *v;
1587: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1588: if (!v) nrerror("allocation failure in ivector");
1589: return v-nl+NR_END;
1590: }
1591:
1592: /******************free ivector **************************/
1593: void free_ivector(int *v, long nl, long nh)
1594: {
1595: free((FREE_ARG)(v+nl-NR_END));
1596: }
1597:
1598: /************************lvector *******************************/
1599: long *lvector(long nl,long nh)
1600: {
1601: long *v;
1602: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1603: if (!v) nrerror("allocation failure in ivector");
1604: return v-nl+NR_END;
1605: }
1606:
1607: /******************free lvector **************************/
1608: void free_lvector(long *v, long nl, long nh)
1609: {
1610: free((FREE_ARG)(v+nl-NR_END));
1611: }
1612:
1613: /******************* imatrix *******************************/
1614: int **imatrix(long nrl, long nrh, long ncl, long nch)
1615: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1616: {
1617: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1618: int **m;
1619:
1620: /* allocate pointers to rows */
1621: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1622: if (!m) nrerror("allocation failure 1 in matrix()");
1623: m += NR_END;
1624: m -= nrl;
1625:
1626:
1627: /* allocate rows and set pointers to them */
1628: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1629: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1630: m[nrl] += NR_END;
1631: m[nrl] -= ncl;
1632:
1633: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1634:
1635: /* return pointer to array of pointers to rows */
1636: return m;
1637: }
1638:
1639: /****************** free_imatrix *************************/
1640: void free_imatrix(m,nrl,nrh,ncl,nch)
1641: int **m;
1642: long nch,ncl,nrh,nrl;
1643: /* free an int matrix allocated by imatrix() */
1644: {
1645: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1646: free((FREE_ARG) (m+nrl-NR_END));
1647: }
1648:
1649: /******************* matrix *******************************/
1650: double **matrix(long nrl, long nrh, long ncl, long nch)
1651: {
1652: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1653: double **m;
1654:
1655: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1656: if (!m) nrerror("allocation failure 1 in matrix()");
1657: m += NR_END;
1658: m -= nrl;
1659:
1660: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1661: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1662: m[nrl] += NR_END;
1663: m[nrl] -= ncl;
1664:
1665: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1666: return m;
1.145 brouard 1667: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1668: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1669: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1670: */
1671: }
1672:
1673: /*************************free matrix ************************/
1674: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1675: {
1676: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1677: free((FREE_ARG)(m+nrl-NR_END));
1678: }
1679:
1680: /******************* ma3x *******************************/
1681: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1682: {
1683: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1684: double ***m;
1685:
1686: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1687: if (!m) nrerror("allocation failure 1 in matrix()");
1688: m += NR_END;
1689: m -= nrl;
1690:
1691: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1692: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1693: m[nrl] += NR_END;
1694: m[nrl] -= ncl;
1695:
1696: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1697:
1698: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1699: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1700: m[nrl][ncl] += NR_END;
1701: m[nrl][ncl] -= nll;
1702: for (j=ncl+1; j<=nch; j++)
1703: m[nrl][j]=m[nrl][j-1]+nlay;
1704:
1705: for (i=nrl+1; i<=nrh; i++) {
1706: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1707: for (j=ncl+1; j<=nch; j++)
1708: m[i][j]=m[i][j-1]+nlay;
1709: }
1710: return m;
1711: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1712: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1713: */
1714: }
1715:
1716: /*************************free ma3x ************************/
1717: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1718: {
1719: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1720: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1721: free((FREE_ARG)(m+nrl-NR_END));
1722: }
1723:
1724: /*************** function subdirf ***********/
1725: char *subdirf(char fileres[])
1726: {
1727: /* Caution optionfilefiname is hidden */
1728: strcpy(tmpout,optionfilefiname);
1729: strcat(tmpout,"/"); /* Add to the right */
1730: strcat(tmpout,fileres);
1731: return tmpout;
1732: }
1733:
1734: /*************** function subdirf2 ***********/
1735: char *subdirf2(char fileres[], char *preop)
1736: {
1737:
1738: /* Caution optionfilefiname is hidden */
1739: strcpy(tmpout,optionfilefiname);
1740: strcat(tmpout,"/");
1741: strcat(tmpout,preop);
1742: strcat(tmpout,fileres);
1743: return tmpout;
1744: }
1745:
1746: /*************** function subdirf3 ***********/
1747: char *subdirf3(char fileres[], char *preop, char *preop2)
1748: {
1749:
1750: /* Caution optionfilefiname is hidden */
1751: strcpy(tmpout,optionfilefiname);
1752: strcat(tmpout,"/");
1753: strcat(tmpout,preop);
1754: strcat(tmpout,preop2);
1755: strcat(tmpout,fileres);
1756: return tmpout;
1757: }
1.213 brouard 1758:
1759: /*************** function subdirfext ***********/
1760: char *subdirfext(char fileres[], char *preop, char *postop)
1761: {
1762:
1763: strcpy(tmpout,preop);
1764: strcat(tmpout,fileres);
1765: strcat(tmpout,postop);
1766: return tmpout;
1767: }
1.126 brouard 1768:
1.213 brouard 1769: /*************** function subdirfext3 ***********/
1770: char *subdirfext3(char fileres[], char *preop, char *postop)
1771: {
1772:
1773: /* Caution optionfilefiname is hidden */
1774: strcpy(tmpout,optionfilefiname);
1775: strcat(tmpout,"/");
1776: strcat(tmpout,preop);
1777: strcat(tmpout,fileres);
1778: strcat(tmpout,postop);
1779: return tmpout;
1780: }
1781:
1.162 brouard 1782: char *asc_diff_time(long time_sec, char ascdiff[])
1783: {
1784: long sec_left, days, hours, minutes;
1785: days = (time_sec) / (60*60*24);
1786: sec_left = (time_sec) % (60*60*24);
1787: hours = (sec_left) / (60*60) ;
1788: sec_left = (sec_left) %(60*60);
1789: minutes = (sec_left) /60;
1790: sec_left = (sec_left) % (60);
1791: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1792: return ascdiff;
1793: }
1794:
1.126 brouard 1795: /***************** f1dim *************************/
1796: extern int ncom;
1797: extern double *pcom,*xicom;
1798: extern double (*nrfunc)(double []);
1799:
1800: double f1dim(double x)
1801: {
1802: int j;
1803: double f;
1804: double *xt;
1805:
1806: xt=vector(1,ncom);
1807: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1808: f=(*nrfunc)(xt);
1809: free_vector(xt,1,ncom);
1810: return f;
1811: }
1812:
1813: /*****************brent *************************/
1814: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1815: {
1816: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1817: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1818: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1819: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1820: * returned function value.
1821: */
1.126 brouard 1822: int iter;
1823: double a,b,d,etemp;
1.159 brouard 1824: double fu=0,fv,fw,fx;
1.164 brouard 1825: double ftemp=0.;
1.126 brouard 1826: double p,q,r,tol1,tol2,u,v,w,x,xm;
1827: double e=0.0;
1828:
1829: a=(ax < cx ? ax : cx);
1830: b=(ax > cx ? ax : cx);
1831: x=w=v=bx;
1832: fw=fv=fx=(*f)(x);
1833: for (iter=1;iter<=ITMAX;iter++) {
1834: xm=0.5*(a+b);
1835: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1836: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1837: printf(".");fflush(stdout);
1838: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1839: #ifdef DEBUGBRENT
1.126 brouard 1840: 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);
1841: 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);
1842: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1843: #endif
1844: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1845: *xmin=x;
1846: return fx;
1847: }
1848: ftemp=fu;
1849: if (fabs(e) > tol1) {
1850: r=(x-w)*(fx-fv);
1851: q=(x-v)*(fx-fw);
1852: p=(x-v)*q-(x-w)*r;
1853: q=2.0*(q-r);
1854: if (q > 0.0) p = -p;
1855: q=fabs(q);
1856: etemp=e;
1857: e=d;
1858: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1859: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1860: else {
1.224 brouard 1861: d=p/q;
1862: u=x+d;
1863: if (u-a < tol2 || b-u < tol2)
1864: d=SIGN(tol1,xm-x);
1.126 brouard 1865: }
1866: } else {
1867: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1868: }
1869: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1870: fu=(*f)(u);
1871: if (fu <= fx) {
1872: if (u >= x) a=x; else b=x;
1873: SHFT(v,w,x,u)
1.183 brouard 1874: SHFT(fv,fw,fx,fu)
1875: } else {
1876: if (u < x) a=u; else b=u;
1877: if (fu <= fw || w == x) {
1.224 brouard 1878: v=w;
1879: w=u;
1880: fv=fw;
1881: fw=fu;
1.183 brouard 1882: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1883: v=u;
1884: fv=fu;
1.183 brouard 1885: }
1886: }
1.126 brouard 1887: }
1888: nrerror("Too many iterations in brent");
1889: *xmin=x;
1890: return fx;
1891: }
1892:
1893: /****************** mnbrak ***********************/
1894:
1895: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1896: double (*func)(double))
1.183 brouard 1897: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1898: the downhill direction (defined by the function as evaluated at the initial points) and returns
1899: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1900: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1901: */
1.126 brouard 1902: double ulim,u,r,q, dum;
1903: double fu;
1.187 brouard 1904:
1905: double scale=10.;
1906: int iterscale=0;
1907:
1908: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1909: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1910:
1911:
1912: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1913: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1914: /* *bx = *ax - (*ax - *bx)/scale; */
1915: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1916: /* } */
1917:
1.126 brouard 1918: if (*fb > *fa) {
1919: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1920: SHFT(dum,*fb,*fa,dum)
1921: }
1.126 brouard 1922: *cx=(*bx)+GOLD*(*bx-*ax);
1923: *fc=(*func)(*cx);
1.183 brouard 1924: #ifdef DEBUG
1.224 brouard 1925: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1926: 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 1927: #endif
1.224 brouard 1928: 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 1929: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1930: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1931: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1932: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1933: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1934: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1935: fu=(*func)(u);
1.163 brouard 1936: #ifdef DEBUG
1937: /* f(x)=A(x-u)**2+f(u) */
1938: double A, fparabu;
1939: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1940: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1941: 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);
1942: 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 1943: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1944: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1945: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1946: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1947: #endif
1.184 brouard 1948: #ifdef MNBRAKORIGINAL
1.183 brouard 1949: #else
1.191 brouard 1950: /* if (fu > *fc) { */
1951: /* #ifdef DEBUG */
1952: /* printf("mnbrak4 fu > fc \n"); */
1953: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1954: /* #endif */
1955: /* /\* 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 *\\/ *\/ */
1956: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1957: /* dum=u; /\* Shifting c and u *\/ */
1958: /* u = *cx; */
1959: /* *cx = dum; */
1960: /* dum = fu; */
1961: /* fu = *fc; */
1962: /* *fc =dum; */
1963: /* } else { /\* end *\/ */
1964: /* #ifdef DEBUG */
1965: /* printf("mnbrak3 fu < fc \n"); */
1966: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1967: /* #endif */
1968: /* dum=u; /\* Shifting c and u *\/ */
1969: /* u = *cx; */
1970: /* *cx = dum; */
1971: /* dum = fu; */
1972: /* fu = *fc; */
1973: /* *fc =dum; */
1974: /* } */
1.224 brouard 1975: #ifdef DEBUGMNBRAK
1976: double A, fparabu;
1977: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1978: fparabu= *fa - A*(*ax-u)*(*ax-u);
1979: 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);
1980: 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 1981: #endif
1.191 brouard 1982: dum=u; /* Shifting c and u */
1983: u = *cx;
1984: *cx = dum;
1985: dum = fu;
1986: fu = *fc;
1987: *fc =dum;
1.183 brouard 1988: #endif
1.162 brouard 1989: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1990: #ifdef DEBUG
1.224 brouard 1991: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1992: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1993: #endif
1.126 brouard 1994: fu=(*func)(u);
1995: if (fu < *fc) {
1.183 brouard 1996: #ifdef DEBUG
1.224 brouard 1997: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1998: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1999: #endif
2000: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2001: SHFT(*fb,*fc,fu,(*func)(u))
2002: #ifdef DEBUG
2003: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2004: #endif
2005: }
1.162 brouard 2006: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2007: #ifdef DEBUG
1.224 brouard 2008: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2009: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2010: #endif
1.126 brouard 2011: u=ulim;
2012: fu=(*func)(u);
1.183 brouard 2013: } else { /* u could be left to b (if r > q parabola has a maximum) */
2014: #ifdef DEBUG
1.224 brouard 2015: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2016: 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 2017: #endif
1.126 brouard 2018: u=(*cx)+GOLD*(*cx-*bx);
2019: fu=(*func)(u);
1.224 brouard 2020: #ifdef DEBUG
2021: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2022: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2023: #endif
1.183 brouard 2024: } /* end tests */
1.126 brouard 2025: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2026: SHFT(*fa,*fb,*fc,fu)
2027: #ifdef DEBUG
1.224 brouard 2028: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2029: 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 2030: #endif
2031: } /* 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 2032: }
2033:
2034: /*************** linmin ************************/
1.162 brouard 2035: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2036: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2037: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2038: the value of func at the returned location p . This is actually all accomplished by calling the
2039: routines mnbrak and brent .*/
1.126 brouard 2040: int ncom;
2041: double *pcom,*xicom;
2042: double (*nrfunc)(double []);
2043:
1.224 brouard 2044: #ifdef LINMINORIGINAL
1.126 brouard 2045: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2046: #else
2047: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2048: #endif
1.126 brouard 2049: {
2050: double brent(double ax, double bx, double cx,
2051: double (*f)(double), double tol, double *xmin);
2052: double f1dim(double x);
2053: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2054: double *fc, double (*func)(double));
2055: int j;
2056: double xx,xmin,bx,ax;
2057: double fx,fb,fa;
1.187 brouard 2058:
1.203 brouard 2059: #ifdef LINMINORIGINAL
2060: #else
2061: double scale=10., axs, xxs; /* Scale added for infinity */
2062: #endif
2063:
1.126 brouard 2064: ncom=n;
2065: pcom=vector(1,n);
2066: xicom=vector(1,n);
2067: nrfunc=func;
2068: for (j=1;j<=n;j++) {
2069: pcom[j]=p[j];
1.202 brouard 2070: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2071: }
1.187 brouard 2072:
1.203 brouard 2073: #ifdef LINMINORIGINAL
2074: xx=1.;
2075: #else
2076: axs=0.0;
2077: xxs=1.;
2078: do{
2079: xx= xxs;
2080: #endif
1.187 brouard 2081: ax=0.;
2082: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2083: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2084: /* 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)) */
2085: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2086: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2087: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2088: /* 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 2089: #ifdef LINMINORIGINAL
2090: #else
2091: if (fx != fx){
1.224 brouard 2092: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2093: printf("|");
2094: fprintf(ficlog,"|");
1.203 brouard 2095: #ifdef DEBUGLINMIN
1.224 brouard 2096: 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 2097: #endif
2098: }
1.224 brouard 2099: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2100: #endif
2101:
1.191 brouard 2102: #ifdef DEBUGLINMIN
2103: 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 2104: 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 2105: #endif
1.224 brouard 2106: #ifdef LINMINORIGINAL
2107: #else
2108: if(fb == fx){ /* Flat function in the direction */
2109: xmin=xx;
2110: *flat=1;
2111: }else{
2112: *flat=0;
2113: #endif
2114: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2115: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2116: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2117: /* fmin = f(p[j] + xmin * xi[j]) */
2118: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2119: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2120: #ifdef DEBUG
1.224 brouard 2121: 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);
2122: 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);
2123: #endif
2124: #ifdef LINMINORIGINAL
2125: #else
2126: }
1.126 brouard 2127: #endif
1.191 brouard 2128: #ifdef DEBUGLINMIN
2129: printf("linmin end ");
1.202 brouard 2130: fprintf(ficlog,"linmin end ");
1.191 brouard 2131: #endif
1.126 brouard 2132: for (j=1;j<=n;j++) {
1.203 brouard 2133: #ifdef LINMINORIGINAL
2134: xi[j] *= xmin;
2135: #else
2136: #ifdef DEBUGLINMIN
2137: if(xxs <1.0)
2138: printf(" before xi[%d]=%12.8f", j,xi[j]);
2139: #endif
2140: 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) */
2141: #ifdef DEBUGLINMIN
2142: if(xxs <1.0)
2143: 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 );
2144: #endif
2145: #endif
1.187 brouard 2146: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2147: }
1.191 brouard 2148: #ifdef DEBUGLINMIN
1.203 brouard 2149: printf("\n");
1.191 brouard 2150: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2151: 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 2152: for (j=1;j<=n;j++) {
1.202 brouard 2153: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2154: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2155: if(j % ncovmodel == 0){
1.191 brouard 2156: printf("\n");
1.202 brouard 2157: fprintf(ficlog,"\n");
2158: }
1.191 brouard 2159: }
1.203 brouard 2160: #else
1.191 brouard 2161: #endif
1.126 brouard 2162: free_vector(xicom,1,n);
2163: free_vector(pcom,1,n);
2164: }
2165:
2166:
2167: /*************** powell ************************/
1.162 brouard 2168: /*
2169: Minimization of a function func of n variables. Input consists of an initial starting point
2170: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2171: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2172: such that failure to decrease by more than this amount on one iteration signals doneness. On
2173: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2174: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2175: */
1.224 brouard 2176: #ifdef LINMINORIGINAL
2177: #else
2178: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2179: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2180: #endif
1.126 brouard 2181: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2182: double (*func)(double []))
2183: {
1.224 brouard 2184: #ifdef LINMINORIGINAL
2185: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2186: double (*func)(double []));
1.224 brouard 2187: #else
1.241 brouard 2188: void linmin(double p[], double xi[], int n, double *fret,
2189: double (*func)(double []),int *flat);
1.224 brouard 2190: #endif
1.239 brouard 2191: int i,ibig,j,jk,k;
1.126 brouard 2192: double del,t,*pt,*ptt,*xit;
1.181 brouard 2193: double directest;
1.126 brouard 2194: double fp,fptt;
2195: double *xits;
2196: int niterf, itmp;
1.224 brouard 2197: #ifdef LINMINORIGINAL
2198: #else
2199:
2200: flatdir=ivector(1,n);
2201: for (j=1;j<=n;j++) flatdir[j]=0;
2202: #endif
1.126 brouard 2203:
2204: pt=vector(1,n);
2205: ptt=vector(1,n);
2206: xit=vector(1,n);
2207: xits=vector(1,n);
2208: *fret=(*func)(p);
2209: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2210: rcurr_time = time(NULL);
1.126 brouard 2211: for (*iter=1;;++(*iter)) {
1.187 brouard 2212: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2213: ibig=0;
2214: del=0.0;
1.157 brouard 2215: rlast_time=rcurr_time;
2216: /* (void) gettimeofday(&curr_time,&tzp); */
2217: rcurr_time = time(NULL);
2218: curr_time = *localtime(&rcurr_time);
2219: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2220: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2221: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2222: for (i=1;i<=n;i++) {
1.126 brouard 2223: fprintf(ficrespow," %.12lf", p[i]);
2224: }
1.239 brouard 2225: fprintf(ficrespow,"\n");fflush(ficrespow);
2226: printf("\n#model= 1 + age ");
2227: fprintf(ficlog,"\n#model= 1 + age ");
2228: if(nagesqr==1){
1.241 brouard 2229: printf(" + age*age ");
2230: fprintf(ficlog," + age*age ");
1.239 brouard 2231: }
2232: for(j=1;j <=ncovmodel-2;j++){
2233: if(Typevar[j]==0) {
2234: printf(" + V%d ",Tvar[j]);
2235: fprintf(ficlog," + V%d ",Tvar[j]);
2236: }else if(Typevar[j]==1) {
2237: printf(" + V%d*age ",Tvar[j]);
2238: fprintf(ficlog," + V%d*age ",Tvar[j]);
2239: }else if(Typevar[j]==2) {
2240: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2241: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2242: }
2243: }
1.126 brouard 2244: printf("\n");
1.239 brouard 2245: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2246: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2247: fprintf(ficlog,"\n");
1.239 brouard 2248: for(i=1,jk=1; i <=nlstate; i++){
2249: for(k=1; k <=(nlstate+ndeath); k++){
2250: if (k != i) {
2251: printf("%d%d ",i,k);
2252: fprintf(ficlog,"%d%d ",i,k);
2253: for(j=1; j <=ncovmodel; j++){
2254: printf("%12.7f ",p[jk]);
2255: fprintf(ficlog,"%12.7f ",p[jk]);
2256: jk++;
2257: }
2258: printf("\n");
2259: fprintf(ficlog,"\n");
2260: }
2261: }
2262: }
1.241 brouard 2263: if(*iter <=3 && *iter >1){
1.157 brouard 2264: tml = *localtime(&rcurr_time);
2265: strcpy(strcurr,asctime(&tml));
2266: rforecast_time=rcurr_time;
1.126 brouard 2267: itmp = strlen(strcurr);
2268: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2269: strcurr[itmp-1]='\0';
1.162 brouard 2270: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2271: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2272: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2273: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2274: forecast_time = *localtime(&rforecast_time);
2275: strcpy(strfor,asctime(&forecast_time));
2276: itmp = strlen(strfor);
2277: if(strfor[itmp-1]=='\n')
2278: strfor[itmp-1]='\0';
2279: 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);
2280: 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 2281: }
2282: }
1.187 brouard 2283: for (i=1;i<=n;i++) { /* For each direction i */
2284: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2285: fptt=(*fret);
2286: #ifdef DEBUG
1.203 brouard 2287: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2288: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2289: #endif
1.203 brouard 2290: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2291: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2292: #ifdef LINMINORIGINAL
1.188 brouard 2293: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2294: #else
2295: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2296: flatdir[i]=flat; /* Function is vanishing in that direction i */
2297: #endif
2298: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2299: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2300: /* because that direction will be replaced unless the gain del is small */
2301: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2302: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2303: /* with the new direction. */
2304: del=fabs(fptt-(*fret));
2305: ibig=i;
1.126 brouard 2306: }
2307: #ifdef DEBUG
2308: printf("%d %.12e",i,(*fret));
2309: fprintf(ficlog,"%d %.12e",i,(*fret));
2310: for (j=1;j<=n;j++) {
1.224 brouard 2311: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2312: printf(" x(%d)=%.12e",j,xit[j]);
2313: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2314: }
2315: for(j=1;j<=n;j++) {
1.225 brouard 2316: printf(" p(%d)=%.12e",j,p[j]);
2317: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2318: }
2319: printf("\n");
2320: fprintf(ficlog,"\n");
2321: #endif
1.187 brouard 2322: } /* end loop on each direction i */
2323: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2324: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2325: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2326: for(j=1;j<=n;j++) {
1.225 brouard 2327: if(flatdir[j] >0){
2328: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2329: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2330: }
2331: /* printf("\n"); */
2332: /* fprintf(ficlog,"\n"); */
2333: }
1.243 brouard 2334: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2335: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2336: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2337: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2338: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2339: /* decreased of more than 3.84 */
2340: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2341: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2342: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2343:
1.188 brouard 2344: /* Starting the program with initial values given by a former maximization will simply change */
2345: /* the scales of the directions and the directions, because the are reset to canonical directions */
2346: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2347: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2348: #ifdef DEBUG
2349: int k[2],l;
2350: k[0]=1;
2351: k[1]=-1;
2352: printf("Max: %.12e",(*func)(p));
2353: fprintf(ficlog,"Max: %.12e",(*func)(p));
2354: for (j=1;j<=n;j++) {
2355: printf(" %.12e",p[j]);
2356: fprintf(ficlog," %.12e",p[j]);
2357: }
2358: printf("\n");
2359: fprintf(ficlog,"\n");
2360: for(l=0;l<=1;l++) {
2361: for (j=1;j<=n;j++) {
2362: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2363: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2364: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2365: }
2366: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2367: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2368: }
2369: #endif
2370:
1.224 brouard 2371: #ifdef LINMINORIGINAL
2372: #else
2373: free_ivector(flatdir,1,n);
2374: #endif
1.126 brouard 2375: free_vector(xit,1,n);
2376: free_vector(xits,1,n);
2377: free_vector(ptt,1,n);
2378: free_vector(pt,1,n);
2379: return;
1.192 brouard 2380: } /* enough precision */
1.240 brouard 2381: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2382: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2383: ptt[j]=2.0*p[j]-pt[j];
2384: xit[j]=p[j]-pt[j];
2385: pt[j]=p[j];
2386: }
1.181 brouard 2387: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2388: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2389: if (*iter <=4) {
1.225 brouard 2390: #else
2391: #endif
1.224 brouard 2392: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2393: #else
1.161 brouard 2394: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2395: #endif
1.162 brouard 2396: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2397: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2398: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2399: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2400: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2401: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2402: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2403: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2404: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2405: /* Even if f3 <f1, directest can be negative and t >0 */
2406: /* mu² and del² are equal when f3=f1 */
2407: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2408: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2409: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2410: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2411: #ifdef NRCORIGINAL
2412: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2413: #else
2414: 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 2415: t= t- del*SQR(fp-fptt);
1.183 brouard 2416: #endif
1.202 brouard 2417: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2418: #ifdef DEBUG
1.181 brouard 2419: 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);
2420: 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 2421: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2422: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2423: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2424: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2425: 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);
2426: 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);
2427: #endif
1.183 brouard 2428: #ifdef POWELLORIGINAL
2429: if (t < 0.0) { /* Then we use it for new direction */
2430: #else
1.182 brouard 2431: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2432: 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 2433: 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 2434: 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 2435: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2436: }
1.181 brouard 2437: if (directest < 0.0) { /* Then we use it for new direction */
2438: #endif
1.191 brouard 2439: #ifdef DEBUGLINMIN
1.234 brouard 2440: printf("Before linmin in direction P%d-P0\n",n);
2441: for (j=1;j<=n;j++) {
2442: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2443: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2444: if(j % ncovmodel == 0){
2445: printf("\n");
2446: fprintf(ficlog,"\n");
2447: }
2448: }
1.224 brouard 2449: #endif
2450: #ifdef LINMINORIGINAL
1.234 brouard 2451: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2452: #else
1.234 brouard 2453: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2454: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2455: #endif
1.234 brouard 2456:
1.191 brouard 2457: #ifdef DEBUGLINMIN
1.234 brouard 2458: for (j=1;j<=n;j++) {
2459: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2460: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2461: if(j % ncovmodel == 0){
2462: printf("\n");
2463: fprintf(ficlog,"\n");
2464: }
2465: }
1.224 brouard 2466: #endif
1.234 brouard 2467: for (j=1;j<=n;j++) {
2468: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2469: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2470: }
1.224 brouard 2471: #ifdef LINMINORIGINAL
2472: #else
1.234 brouard 2473: for (j=1, flatd=0;j<=n;j++) {
2474: if(flatdir[j]>0)
2475: flatd++;
2476: }
2477: if(flatd >0){
1.255 brouard 2478: printf("%d flat directions: ",flatd);
2479: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2480: for (j=1;j<=n;j++) {
2481: if(flatdir[j]>0){
2482: printf("%d ",j);
2483: fprintf(ficlog,"%d ",j);
2484: }
2485: }
2486: printf("\n");
2487: fprintf(ficlog,"\n");
2488: }
1.191 brouard 2489: #endif
1.234 brouard 2490: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2491: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2492:
1.126 brouard 2493: #ifdef DEBUG
1.234 brouard 2494: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2495: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2496: for(j=1;j<=n;j++){
2497: printf(" %lf",xit[j]);
2498: fprintf(ficlog," %lf",xit[j]);
2499: }
2500: printf("\n");
2501: fprintf(ficlog,"\n");
1.126 brouard 2502: #endif
1.192 brouard 2503: } /* end of t or directest negative */
1.224 brouard 2504: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2505: #else
1.234 brouard 2506: } /* end if (fptt < fp) */
1.192 brouard 2507: #endif
1.225 brouard 2508: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2509: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2510: #else
1.224 brouard 2511: #endif
1.234 brouard 2512: } /* loop iteration */
1.126 brouard 2513: }
1.234 brouard 2514:
1.126 brouard 2515: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2516:
1.235 brouard 2517: 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 2518: {
1.279 ! brouard 2519: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
! 2520: * (and selected quantitative values in nres)
! 2521: * by left multiplying the unit
! 2522: * matrix by transitions matrix until convergence is reached with precision ftolpl
! 2523: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
! 2524: * Wx is row vector: population in state 1, population in state 2, population dead
! 2525: * or prevalence in state 1, prevalence in state 2, 0
! 2526: * newm is the matrix after multiplications, its rows are identical at a factor.
! 2527: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
! 2528: * Output is prlim.
! 2529: * Initial matrix pimij
! 2530: */
1.206 brouard 2531: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2532: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2533: /* 0, 0 , 1} */
2534: /*
2535: * and after some iteration: */
2536: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2537: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2538: /* 0, 0 , 1} */
2539: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2540: /* {0.51571254859325999, 0.4842874514067399, */
2541: /* 0.51326036147820708, 0.48673963852179264} */
2542: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2543:
1.126 brouard 2544: int i, ii,j,k;
1.209 brouard 2545: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2546: /* double **matprod2(); */ /* test */
1.218 brouard 2547: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2548: double **newm;
1.209 brouard 2549: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2550: int ncvloop=0;
1.169 brouard 2551:
1.209 brouard 2552: min=vector(1,nlstate);
2553: max=vector(1,nlstate);
2554: meandiff=vector(1,nlstate);
2555:
1.218 brouard 2556: /* Starting with matrix unity */
1.126 brouard 2557: for (ii=1;ii<=nlstate+ndeath;ii++)
2558: for (j=1;j<=nlstate+ndeath;j++){
2559: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2560: }
1.169 brouard 2561:
2562: cov[1]=1.;
2563:
2564: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2565: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2566: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2567: ncvloop++;
1.126 brouard 2568: newm=savm;
2569: /* Covariates have to be included here again */
1.138 brouard 2570: cov[2]=agefin;
1.187 brouard 2571: if(nagesqr==1)
2572: cov[3]= agefin*agefin;;
1.234 brouard 2573: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2574: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2575: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2576: /* 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 2577: }
2578: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2579: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2580: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2581: /* 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 2582: }
1.237 brouard 2583: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2584: if(Dummy[Tvar[Tage[k]]]){
2585: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2586: } else{
1.235 brouard 2587: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2588: }
1.235 brouard 2589: /* 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 2590: }
1.237 brouard 2591: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2592: /* 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 2593: if(Dummy[Tvard[k][1]==0]){
2594: if(Dummy[Tvard[k][2]==0]){
2595: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2596: }else{
2597: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2598: }
2599: }else{
2600: if(Dummy[Tvard[k][2]==0]){
2601: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2602: }else{
2603: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2604: }
2605: }
1.234 brouard 2606: }
1.138 brouard 2607: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2608: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2609: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2610: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2611: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2612: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2613: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2614:
1.126 brouard 2615: savm=oldm;
2616: oldm=newm;
1.209 brouard 2617:
2618: for(j=1; j<=nlstate; j++){
2619: max[j]=0.;
2620: min[j]=1.;
2621: }
2622: for(i=1;i<=nlstate;i++){
2623: sumnew=0;
2624: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2625: for(j=1; j<=nlstate; j++){
2626: prlim[i][j]= newm[i][j]/(1-sumnew);
2627: max[j]=FMAX(max[j],prlim[i][j]);
2628: min[j]=FMIN(min[j],prlim[i][j]);
2629: }
2630: }
2631:
1.126 brouard 2632: maxmax=0.;
1.209 brouard 2633: for(j=1; j<=nlstate; j++){
2634: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2635: maxmax=FMAX(maxmax,meandiff[j]);
2636: /* 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 2637: } /* j loop */
1.203 brouard 2638: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2639: /* 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 2640: if(maxmax < ftolpl){
1.209 brouard 2641: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2642: free_vector(min,1,nlstate);
2643: free_vector(max,1,nlstate);
2644: free_vector(meandiff,1,nlstate);
1.126 brouard 2645: return prlim;
2646: }
1.169 brouard 2647: } /* age loop */
1.208 brouard 2648: /* After some age loop it doesn't converge */
1.209 brouard 2649: printf("Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. \n\
1.208 brouard 2650: Earliest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
1.209 brouard 2651: /* 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); */
2652: free_vector(min,1,nlstate);
2653: free_vector(max,1,nlstate);
2654: free_vector(meandiff,1,nlstate);
1.208 brouard 2655:
1.169 brouard 2656: return prlim; /* should not reach here */
1.126 brouard 2657: }
2658:
1.217 brouard 2659:
2660: /**** Back Prevalence limit (stable or period prevalence) ****************/
2661:
1.218 brouard 2662: /* 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) */
2663: /* 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 2664: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2665: {
1.264 brouard 2666: /* 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 2667: matrix by transitions matrix until convergence is reached with precision ftolpl */
2668: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2669: /* Wx is row vector: population in state 1, population in state 2, population dead */
2670: /* or prevalence in state 1, prevalence in state 2, 0 */
2671: /* newm is the matrix after multiplications, its rows are identical at a factor */
2672: /* Initial matrix pimij */
2673: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2674: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2675: /* 0, 0 , 1} */
2676: /*
2677: * and after some iteration: */
2678: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2679: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2680: /* 0, 0 , 1} */
2681: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2682: /* {0.51571254859325999, 0.4842874514067399, */
2683: /* 0.51326036147820708, 0.48673963852179264} */
2684: /* If we start from prlim again, prlim tends to a constant matrix */
2685:
2686: int i, ii,j,k;
1.247 brouard 2687: int first=0;
1.217 brouard 2688: double *min, *max, *meandiff, maxmax,sumnew=0.;
2689: /* double **matprod2(); */ /* test */
2690: double **out, cov[NCOVMAX+1], **bmij();
2691: double **newm;
1.218 brouard 2692: double **dnewm, **doldm, **dsavm; /* for use */
2693: double **oldm, **savm; /* for use */
2694:
1.217 brouard 2695: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2696: int ncvloop=0;
2697:
2698: min=vector(1,nlstate);
2699: max=vector(1,nlstate);
2700: meandiff=vector(1,nlstate);
2701:
1.266 brouard 2702: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2703: oldm=oldms; savm=savms;
2704:
2705: /* Starting with matrix unity */
2706: for (ii=1;ii<=nlstate+ndeath;ii++)
2707: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2708: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2709: }
2710:
2711: cov[1]=1.;
2712:
2713: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2714: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2715: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2716: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2717: ncvloop++;
1.218 brouard 2718: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2719: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2720: /* Covariates have to be included here again */
2721: cov[2]=agefin;
2722: if(nagesqr==1)
2723: cov[3]= agefin*agefin;;
1.242 brouard 2724: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2725: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2726: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2727: /* 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 2728: }
2729: /* for (k=1; k<=cptcovn;k++) { */
2730: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2731: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2732: /* /\* 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])]); *\/ */
2733: /* } */
2734: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2735: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2736: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2737: /* 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]); */
2738: }
2739: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2740: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2741: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2742: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2743: for (k=1; k<=cptcovage;k++){ /* For product with age */
2744: if(Dummy[Tvar[Tage[k]]]){
2745: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2746: } else{
2747: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2748: }
2749: /* 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]); */
2750: }
2751: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2752: /* 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]); */
2753: if(Dummy[Tvard[k][1]==0]){
2754: if(Dummy[Tvard[k][2]==0]){
2755: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2756: }else{
2757: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2758: }
2759: }else{
2760: if(Dummy[Tvard[k][2]==0]){
2761: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2762: }else{
2763: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2764: }
2765: }
1.217 brouard 2766: }
2767:
2768: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2769: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2770: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2771: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2772: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2773: /* ij should be linked to the correct index of cov */
2774: /* age and covariate values ij are in 'cov', but we need to pass
2775: * ij for the observed prevalence at age and status and covariate
2776: * number: prevacurrent[(int)agefin][ii][ij]
2777: */
2778: /* 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 *\/ */
2779: /* 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 *\/ */
2780: 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 2781: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2782: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2783: /* for(i=1; i<=nlstate+ndeath; i++) { */
2784: /* printf("%d newm= ",i); */
2785: /* for(j=1;j<=nlstate+ndeath;j++) { */
2786: /* printf("%f ",newm[i][j]); */
2787: /* } */
2788: /* printf("oldm * "); */
2789: /* for(j=1;j<=nlstate+ndeath;j++) { */
2790: /* printf("%f ",oldm[i][j]); */
2791: /* } */
1.268 brouard 2792: /* printf(" bmmij "); */
1.266 brouard 2793: /* for(j=1;j<=nlstate+ndeath;j++) { */
2794: /* printf("%f ",pmmij[i][j]); */
2795: /* } */
2796: /* printf("\n"); */
2797: /* } */
2798: /* } */
1.217 brouard 2799: savm=oldm;
2800: oldm=newm;
1.266 brouard 2801:
1.217 brouard 2802: for(j=1; j<=nlstate; j++){
2803: max[j]=0.;
2804: min[j]=1.;
2805: }
2806: for(j=1; j<=nlstate; j++){
2807: for(i=1;i<=nlstate;i++){
1.234 brouard 2808: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2809: bprlim[i][j]= newm[i][j];
2810: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2811: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2812: }
2813: }
1.218 brouard 2814:
1.217 brouard 2815: maxmax=0.;
2816: for(i=1; i<=nlstate; i++){
2817: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2818: maxmax=FMAX(maxmax,meandiff[i]);
2819: /* 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 2820: } /* i loop */
1.217 brouard 2821: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2822: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2823: if(maxmax < ftolpl){
1.220 brouard 2824: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2825: free_vector(min,1,nlstate);
2826: free_vector(max,1,nlstate);
2827: free_vector(meandiff,1,nlstate);
2828: return bprlim;
2829: }
2830: } /* age loop */
2831: /* After some age loop it doesn't converge */
1.247 brouard 2832: if(first){
2833: first=1;
2834: 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\
2835: 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);
2836: }
2837: 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 2838: 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);
2839: /* 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); */
2840: free_vector(min,1,nlstate);
2841: free_vector(max,1,nlstate);
2842: free_vector(meandiff,1,nlstate);
2843:
2844: return bprlim; /* should not reach here */
2845: }
2846:
1.126 brouard 2847: /*************** transition probabilities ***************/
2848:
2849: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2850: {
1.138 brouard 2851: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2852: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2853: model to the ncovmodel covariates (including constant and age).
2854: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2855: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2856: ncth covariate in the global vector x is given by the formula:
2857: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2858: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2859: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2860: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2861: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2862: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2863: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2864: */
2865: double s1, lnpijopii;
1.126 brouard 2866: /*double t34;*/
1.164 brouard 2867: int i,j, nc, ii, jj;
1.126 brouard 2868:
1.223 brouard 2869: for(i=1; i<= nlstate; i++){
2870: for(j=1; j<i;j++){
2871: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2872: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2873: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2874: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2875: }
2876: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2877: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2878: }
2879: for(j=i+1; j<=nlstate+ndeath;j++){
2880: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2881: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2882: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2883: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2884: }
2885: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2886: }
2887: }
1.218 brouard 2888:
1.223 brouard 2889: for(i=1; i<= nlstate; i++){
2890: s1=0;
2891: for(j=1; j<i; j++){
2892: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2893: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2894: }
2895: for(j=i+1; j<=nlstate+ndeath; j++){
2896: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2897: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2898: }
2899: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2900: ps[i][i]=1./(s1+1.);
2901: /* Computing other pijs */
2902: for(j=1; j<i; j++)
2903: ps[i][j]= exp(ps[i][j])*ps[i][i];
2904: for(j=i+1; j<=nlstate+ndeath; j++)
2905: ps[i][j]= exp(ps[i][j])*ps[i][i];
2906: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2907: } /* end i */
1.218 brouard 2908:
1.223 brouard 2909: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2910: for(jj=1; jj<= nlstate+ndeath; jj++){
2911: ps[ii][jj]=0;
2912: ps[ii][ii]=1;
2913: }
2914: }
1.218 brouard 2915:
2916:
1.223 brouard 2917: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2918: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2919: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2920: /* } */
2921: /* printf("\n "); */
2922: /* } */
2923: /* printf("\n ");printf("%lf ",cov[2]);*/
2924: /*
2925: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2926: goto end;*/
1.266 brouard 2927: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2928: }
2929:
1.218 brouard 2930: /*************** backward transition probabilities ***************/
2931:
2932: /* 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 ) */
2933: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2934: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2935: {
1.266 brouard 2936: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2937: * 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 2938: */
1.218 brouard 2939: int i, ii, j,k;
1.222 brouard 2940:
2941: double **out, **pmij();
2942: double sumnew=0.;
1.218 brouard 2943: double agefin;
1.268 brouard 2944: double k3=0.; /* constant of the w_x diagonal matrixe (in order for B to sum to 1 even for death state) */
1.222 brouard 2945: double **dnewm, **dsavm, **doldm;
2946: double **bbmij;
2947:
1.218 brouard 2948: doldm=ddoldms; /* global pointers */
1.222 brouard 2949: dnewm=ddnewms;
2950: dsavm=ddsavms;
2951:
2952: agefin=cov[2];
1.268 brouard 2953: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2954: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2955: the observed prevalence (with this covariate ij) at beginning of transition */
2956: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2957:
2958: /* P_x */
1.266 brouard 2959: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2960: /* outputs pmmij which is a stochastic matrix in row */
2961:
2962: /* Diag(w_x) */
2963: /* Problem with prevacurrent which can be zero */
2964: sumnew=0.;
1.269 brouard 2965: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2966: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2967: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2968: sumnew+=prevacurrent[(int)agefin][ii][ij];
2969: }
2970: if(sumnew >0.01){ /* At least some value in the prevalence */
2971: for (ii=1;ii<=nlstate+ndeath;ii++){
2972: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2973: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2974: }
2975: }else{
2976: for (ii=1;ii<=nlstate+ndeath;ii++){
2977: for (j=1;j<=nlstate+ndeath;j++)
2978: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
2979: }
2980: /* if(sumnew <0.9){ */
2981: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2982: /* } */
2983: }
2984: k3=0.0; /* We put the last diagonal to 0 */
2985: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
2986: doldm[ii][ii]= k3;
2987: }
2988: /* End doldm, At the end doldm is diag[(w_i)] */
2989:
2990: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
2991: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
2992:
2993: /* Diag(Sum_i w^i_x p^ij_x */
2994: /* 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 2995: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 2996: sumnew=0.;
1.222 brouard 2997: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 2998: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 2999: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3000: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3001: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3002: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3003: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3004: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3005: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3006: /* }else */
1.268 brouard 3007: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3008: } /*End ii */
3009: } /* 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 */
3010:
3011: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
3012: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3013: /* end bmij */
1.266 brouard 3014: return ps; /*pointer is unchanged */
1.218 brouard 3015: }
1.217 brouard 3016: /*************** transition probabilities ***************/
3017:
1.218 brouard 3018: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3019: {
3020: /* According to parameters values stored in x and the covariate's values stored in cov,
3021: computes the probability to be observed in state j being in state i by appying the
3022: model to the ncovmodel covariates (including constant and age).
3023: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3024: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3025: ncth covariate in the global vector x is given by the formula:
3026: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3027: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3028: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3029: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3030: Outputs ps[i][j] the probability to be observed in j being in j according to
3031: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3032: */
3033: double s1, lnpijopii;
3034: /*double t34;*/
3035: int i,j, nc, ii, jj;
3036:
1.234 brouard 3037: for(i=1; i<= nlstate; i++){
3038: for(j=1; j<i;j++){
3039: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3040: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3041: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3042: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3043: }
3044: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3045: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3046: }
3047: for(j=i+1; j<=nlstate+ndeath;j++){
3048: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3049: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3050: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3051: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3052: }
3053: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3054: }
3055: }
3056:
3057: for(i=1; i<= nlstate; i++){
3058: s1=0;
3059: for(j=1; j<i; j++){
3060: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3061: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3062: }
3063: for(j=i+1; j<=nlstate+ndeath; j++){
3064: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3065: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3066: }
3067: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3068: ps[i][i]=1./(s1+1.);
3069: /* Computing other pijs */
3070: for(j=1; j<i; j++)
3071: ps[i][j]= exp(ps[i][j])*ps[i][i];
3072: for(j=i+1; j<=nlstate+ndeath; j++)
3073: ps[i][j]= exp(ps[i][j])*ps[i][i];
3074: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3075: } /* end i */
3076:
3077: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3078: for(jj=1; jj<= nlstate+ndeath; jj++){
3079: ps[ii][jj]=0;
3080: ps[ii][ii]=1;
3081: }
3082: }
3083: /* Added for backcast */ /* Transposed matrix too */
3084: for(jj=1; jj<= nlstate+ndeath; jj++){
3085: s1=0.;
3086: for(ii=1; ii<= nlstate+ndeath; ii++){
3087: s1+=ps[ii][jj];
3088: }
3089: for(ii=1; ii<= nlstate; ii++){
3090: ps[ii][jj]=ps[ii][jj]/s1;
3091: }
3092: }
3093: /* Transposition */
3094: for(jj=1; jj<= nlstate+ndeath; jj++){
3095: for(ii=jj; ii<= nlstate+ndeath; ii++){
3096: s1=ps[ii][jj];
3097: ps[ii][jj]=ps[jj][ii];
3098: ps[jj][ii]=s1;
3099: }
3100: }
3101: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3102: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3103: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3104: /* } */
3105: /* printf("\n "); */
3106: /* } */
3107: /* printf("\n ");printf("%lf ",cov[2]);*/
3108: /*
3109: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3110: goto end;*/
3111: return ps;
1.217 brouard 3112: }
3113:
3114:
1.126 brouard 3115: /**************** Product of 2 matrices ******************/
3116:
1.145 brouard 3117: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3118: {
3119: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3120: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3121: /* in, b, out are matrice of pointers which should have been initialized
3122: before: only the contents of out is modified. The function returns
3123: a pointer to pointers identical to out */
1.145 brouard 3124: int i, j, k;
1.126 brouard 3125: for(i=nrl; i<= nrh; i++)
1.145 brouard 3126: for(k=ncolol; k<=ncoloh; k++){
3127: out[i][k]=0.;
3128: for(j=ncl; j<=nch; j++)
3129: out[i][k] +=in[i][j]*b[j][k];
3130: }
1.126 brouard 3131: return out;
3132: }
3133:
3134:
3135: /************* Higher Matrix Product ***************/
3136:
1.235 brouard 3137: 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 3138: {
1.218 brouard 3139: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3140: 'nhstepm*hstepm*stepm' months (i.e. until
3141: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3142: nhstepm*hstepm matrices.
3143: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3144: (typically every 2 years instead of every month which is too big
3145: for the memory).
3146: Model is determined by parameters x and covariates have to be
3147: included manually here.
3148:
3149: */
3150:
3151: int i, j, d, h, k;
1.131 brouard 3152: double **out, cov[NCOVMAX+1];
1.126 brouard 3153: double **newm;
1.187 brouard 3154: double agexact;
1.214 brouard 3155: double agebegin, ageend;
1.126 brouard 3156:
3157: /* Hstepm could be zero and should return the unit matrix */
3158: for (i=1;i<=nlstate+ndeath;i++)
3159: for (j=1;j<=nlstate+ndeath;j++){
3160: oldm[i][j]=(i==j ? 1.0 : 0.0);
3161: po[i][j][0]=(i==j ? 1.0 : 0.0);
3162: }
3163: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3164: for(h=1; h <=nhstepm; h++){
3165: for(d=1; d <=hstepm; d++){
3166: newm=savm;
3167: /* Covariates have to be included here again */
3168: cov[1]=1.;
1.214 brouard 3169: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3170: cov[2]=agexact;
3171: if(nagesqr==1)
1.227 brouard 3172: cov[3]= agexact*agexact;
1.235 brouard 3173: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3174: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3175: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3176: /* 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)); */
3177: }
3178: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3179: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3180: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3181: /* 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]); */
3182: }
3183: for (k=1; k<=cptcovage;k++){
3184: if(Dummy[Tvar[Tage[k]]]){
3185: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3186: } else{
3187: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3188: }
3189: /* 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]); */
3190: }
3191: for (k=1; k<=cptcovprod;k++){ /* */
3192: /* 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]); */
3193: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3194: }
3195: /* for (k=1; k<=cptcovn;k++) */
3196: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3197: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3198: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3199: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3200: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3201:
3202:
1.126 brouard 3203: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3204: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3205: /* right multiplication of oldm by the current matrix */
1.126 brouard 3206: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3207: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3208: /* if((int)age == 70){ */
3209: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3210: /* for(i=1; i<=nlstate+ndeath; i++) { */
3211: /* printf("%d pmmij ",i); */
3212: /* for(j=1;j<=nlstate+ndeath;j++) { */
3213: /* printf("%f ",pmmij[i][j]); */
3214: /* } */
3215: /* printf(" oldm "); */
3216: /* for(j=1;j<=nlstate+ndeath;j++) { */
3217: /* printf("%f ",oldm[i][j]); */
3218: /* } */
3219: /* printf("\n"); */
3220: /* } */
3221: /* } */
1.126 brouard 3222: savm=oldm;
3223: oldm=newm;
3224: }
3225: for(i=1; i<=nlstate+ndeath; i++)
3226: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3227: po[i][j][h]=newm[i][j];
3228: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3229: }
1.128 brouard 3230: /*printf("h=%d ",h);*/
1.126 brouard 3231: } /* end h */
1.267 brouard 3232: /* printf("\n H=%d \n",h); */
1.126 brouard 3233: return po;
3234: }
3235:
1.217 brouard 3236: /************* Higher Back Matrix Product ***************/
1.218 brouard 3237: /* 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 3238: 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 3239: {
1.266 brouard 3240: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3241: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3242: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3243: nhstepm*hstepm matrices.
3244: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3245: (typically every 2 years instead of every month which is too big
1.217 brouard 3246: for the memory).
1.218 brouard 3247: Model is determined by parameters x and covariates have to be
1.266 brouard 3248: included manually here. Then we use a call to bmij(x and cov)
3249: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3250: */
1.217 brouard 3251:
3252: int i, j, d, h, k;
1.266 brouard 3253: double **out, cov[NCOVMAX+1], **bmij();
3254: double **newm, ***newmm;
1.217 brouard 3255: double agexact;
3256: double agebegin, ageend;
1.222 brouard 3257: double **oldm, **savm;
1.217 brouard 3258:
1.266 brouard 3259: newmm=po; /* To be saved */
3260: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3261: /* Hstepm could be zero and should return the unit matrix */
3262: for (i=1;i<=nlstate+ndeath;i++)
3263: for (j=1;j<=nlstate+ndeath;j++){
3264: oldm[i][j]=(i==j ? 1.0 : 0.0);
3265: po[i][j][0]=(i==j ? 1.0 : 0.0);
3266: }
3267: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3268: for(h=1; h <=nhstepm; h++){
3269: for(d=1; d <=hstepm; d++){
3270: newm=savm;
3271: /* Covariates have to be included here again */
3272: cov[1]=1.;
1.271 brouard 3273: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3274: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3275: cov[2]=agexact;
3276: if(nagesqr==1)
1.222 brouard 3277: cov[3]= agexact*agexact;
1.266 brouard 3278: for (k=1; k<=cptcovn;k++){
3279: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3280: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3281: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3282: /* 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)); */
3283: }
1.267 brouard 3284: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3285: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3286: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3287: /* 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]); */
3288: }
3289: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3290: if(Dummy[Tvar[Tage[k]]]){
3291: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3292: } else{
3293: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3294: }
3295: /* 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]); */
3296: }
3297: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3298: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3299: }
1.217 brouard 3300: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3301: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3302:
1.218 brouard 3303: /* Careful transposed matrix */
1.266 brouard 3304: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3305: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3306: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3307: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3308: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3309: /* if((int)age == 70){ */
3310: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3311: /* for(i=1; i<=nlstate+ndeath; i++) { */
3312: /* printf("%d pmmij ",i); */
3313: /* for(j=1;j<=nlstate+ndeath;j++) { */
3314: /* printf("%f ",pmmij[i][j]); */
3315: /* } */
3316: /* printf(" oldm "); */
3317: /* for(j=1;j<=nlstate+ndeath;j++) { */
3318: /* printf("%f ",oldm[i][j]); */
3319: /* } */
3320: /* printf("\n"); */
3321: /* } */
3322: /* } */
3323: savm=oldm;
3324: oldm=newm;
3325: }
3326: for(i=1; i<=nlstate+ndeath; i++)
3327: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3328: po[i][j][h]=newm[i][j];
1.268 brouard 3329: /* if(h==nhstepm) */
3330: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3331: }
1.268 brouard 3332: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3333: } /* end h */
1.268 brouard 3334: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3335: return po;
3336: }
3337:
3338:
1.162 brouard 3339: #ifdef NLOPT
3340: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3341: double fret;
3342: double *xt;
3343: int j;
3344: myfunc_data *d2 = (myfunc_data *) pd;
3345: /* xt = (p1-1); */
3346: xt=vector(1,n);
3347: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3348:
3349: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3350: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3351: printf("Function = %.12lf ",fret);
3352: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3353: printf("\n");
3354: free_vector(xt,1,n);
3355: return fret;
3356: }
3357: #endif
1.126 brouard 3358:
3359: /*************** log-likelihood *************/
3360: double func( double *x)
3361: {
1.226 brouard 3362: int i, ii, j, k, mi, d, kk;
3363: int ioffset=0;
3364: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3365: double **out;
3366: double lli; /* Individual log likelihood */
3367: int s1, s2;
1.228 brouard 3368: 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 3369: double bbh, survp;
3370: long ipmx;
3371: double agexact;
3372: /*extern weight */
3373: /* We are differentiating ll according to initial status */
3374: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3375: /*for(i=1;i<imx;i++)
3376: printf(" %d\n",s[4][i]);
3377: */
1.162 brouard 3378:
1.226 brouard 3379: ++countcallfunc;
1.162 brouard 3380:
1.226 brouard 3381: cov[1]=1.;
1.126 brouard 3382:
1.226 brouard 3383: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3384: ioffset=0;
1.226 brouard 3385: if(mle==1){
3386: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3387: /* Computes the values of the ncovmodel covariates of the model
3388: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3389: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3390: to be observed in j being in i according to the model.
3391: */
1.243 brouard 3392: ioffset=2+nagesqr ;
1.233 brouard 3393: /* Fixed */
1.234 brouard 3394: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3395: 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)*/
3396: }
1.226 brouard 3397: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3398: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3399: has been calculated etc */
3400: /* For an individual i, wav[i] gives the number of effective waves */
3401: /* We compute the contribution to Likelihood of each effective transition
3402: mw[mi][i] is real wave of the mi th effectve wave */
3403: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3404: s2=s[mw[mi+1][i]][i];
3405: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3406: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3407: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3408: */
3409: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3410: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3411: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3412: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3413: }
3414: for (ii=1;ii<=nlstate+ndeath;ii++)
3415: for (j=1;j<=nlstate+ndeath;j++){
3416: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3417: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3418: }
3419: for(d=0; d<dh[mi][i]; d++){
3420: newm=savm;
3421: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3422: cov[2]=agexact;
3423: if(nagesqr==1)
3424: cov[3]= agexact*agexact; /* Should be changed here */
3425: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3426: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3427: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3428: else
3429: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3430: }
3431: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3432: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3433: savm=oldm;
3434: oldm=newm;
3435: } /* end mult */
3436:
3437: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3438: /* But now since version 0.9 we anticipate for bias at large stepm.
3439: * If stepm is larger than one month (smallest stepm) and if the exact delay
3440: * (in months) between two waves is not a multiple of stepm, we rounded to
3441: * the nearest (and in case of equal distance, to the lowest) interval but now
3442: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3443: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3444: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3445: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3446: * -stepm/2 to stepm/2 .
3447: * For stepm=1 the results are the same as for previous versions of Imach.
3448: * For stepm > 1 the results are less biased than in previous versions.
3449: */
1.234 brouard 3450: s1=s[mw[mi][i]][i];
3451: s2=s[mw[mi+1][i]][i];
3452: bbh=(double)bh[mi][i]/(double)stepm;
3453: /* bias bh is positive if real duration
3454: * is higher than the multiple of stepm and negative otherwise.
3455: */
3456: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3457: if( s2 > nlstate){
3458: /* i.e. if s2 is a death state and if the date of death is known
3459: then the contribution to the likelihood is the probability to
3460: die between last step unit time and current step unit time,
3461: which is also equal to probability to die before dh
3462: minus probability to die before dh-stepm .
3463: In version up to 0.92 likelihood was computed
3464: as if date of death was unknown. Death was treated as any other
3465: health state: the date of the interview describes the actual state
3466: and not the date of a change in health state. The former idea was
3467: to consider that at each interview the state was recorded
3468: (healthy, disable or death) and IMaCh was corrected; but when we
3469: introduced the exact date of death then we should have modified
3470: the contribution of an exact death to the likelihood. This new
3471: contribution is smaller and very dependent of the step unit
3472: stepm. It is no more the probability to die between last interview
3473: and month of death but the probability to survive from last
3474: interview up to one month before death multiplied by the
3475: probability to die within a month. Thanks to Chris
3476: Jackson for correcting this bug. Former versions increased
3477: mortality artificially. The bad side is that we add another loop
3478: which slows down the processing. The difference can be up to 10%
3479: lower mortality.
3480: */
3481: /* If, at the beginning of the maximization mostly, the
3482: cumulative probability or probability to be dead is
3483: constant (ie = 1) over time d, the difference is equal to
3484: 0. out[s1][3] = savm[s1][3]: probability, being at state
3485: s1 at precedent wave, to be dead a month before current
3486: wave is equal to probability, being at state s1 at
3487: precedent wave, to be dead at mont of the current
3488: wave. Then the observed probability (that this person died)
3489: is null according to current estimated parameter. In fact,
3490: it should be very low but not zero otherwise the log go to
3491: infinity.
3492: */
1.183 brouard 3493: /* #ifdef INFINITYORIGINAL */
3494: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3495: /* #else */
3496: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3497: /* lli=log(mytinydouble); */
3498: /* else */
3499: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3500: /* #endif */
1.226 brouard 3501: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3502:
1.226 brouard 3503: } else if ( s2==-1 ) { /* alive */
3504: for (j=1,survp=0. ; j<=nlstate; j++)
3505: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3506: /*survp += out[s1][j]; */
3507: lli= log(survp);
3508: }
3509: else if (s2==-4) {
3510: for (j=3,survp=0. ; j<=nlstate; j++)
3511: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3512: lli= log(survp);
3513: }
3514: else if (s2==-5) {
3515: for (j=1,survp=0. ; j<=2; j++)
3516: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3517: lli= log(survp);
3518: }
3519: else{
3520: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3521: /* 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 */
3522: }
3523: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3524: /*if(lli ==000.0)*/
3525: /*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); */
3526: ipmx +=1;
3527: sw += weight[i];
3528: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3529: /* if (lli < log(mytinydouble)){ */
3530: /* 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); */
3531: /* 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]); */
3532: /* } */
3533: } /* end of wave */
3534: } /* end of individual */
3535: } else if(mle==2){
3536: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3537: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3538: for(mi=1; mi<= wav[i]-1; mi++){
3539: for (ii=1;ii<=nlstate+ndeath;ii++)
3540: for (j=1;j<=nlstate+ndeath;j++){
3541: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3542: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3543: }
3544: for(d=0; d<=dh[mi][i]; d++){
3545: newm=savm;
3546: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3547: cov[2]=agexact;
3548: if(nagesqr==1)
3549: cov[3]= agexact*agexact;
3550: for (kk=1; kk<=cptcovage;kk++) {
3551: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3552: }
3553: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3554: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3555: savm=oldm;
3556: oldm=newm;
3557: } /* end mult */
3558:
3559: s1=s[mw[mi][i]][i];
3560: s2=s[mw[mi+1][i]][i];
3561: bbh=(double)bh[mi][i]/(double)stepm;
3562: 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 */
3563: ipmx +=1;
3564: sw += weight[i];
3565: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3566: } /* end of wave */
3567: } /* end of individual */
3568: } else if(mle==3){ /* exponential inter-extrapolation */
3569: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3570: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3571: for(mi=1; mi<= wav[i]-1; mi++){
3572: for (ii=1;ii<=nlstate+ndeath;ii++)
3573: for (j=1;j<=nlstate+ndeath;j++){
3574: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3575: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3576: }
3577: for(d=0; d<dh[mi][i]; d++){
3578: newm=savm;
3579: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3580: cov[2]=agexact;
3581: if(nagesqr==1)
3582: cov[3]= agexact*agexact;
3583: for (kk=1; kk<=cptcovage;kk++) {
3584: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3585: }
3586: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3587: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3588: savm=oldm;
3589: oldm=newm;
3590: } /* end mult */
3591:
3592: s1=s[mw[mi][i]][i];
3593: s2=s[mw[mi+1][i]][i];
3594: bbh=(double)bh[mi][i]/(double)stepm;
3595: 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 */
3596: ipmx +=1;
3597: sw += weight[i];
3598: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3599: } /* end of wave */
3600: } /* end of individual */
3601: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3602: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3603: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3604: for(mi=1; mi<= wav[i]-1; mi++){
3605: for (ii=1;ii<=nlstate+ndeath;ii++)
3606: for (j=1;j<=nlstate+ndeath;j++){
3607: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3608: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3609: }
3610: for(d=0; d<dh[mi][i]; d++){
3611: newm=savm;
3612: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3613: cov[2]=agexact;
3614: if(nagesqr==1)
3615: cov[3]= agexact*agexact;
3616: for (kk=1; kk<=cptcovage;kk++) {
3617: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3618: }
1.126 brouard 3619:
1.226 brouard 3620: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3621: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3622: savm=oldm;
3623: oldm=newm;
3624: } /* end mult */
3625:
3626: s1=s[mw[mi][i]][i];
3627: s2=s[mw[mi+1][i]][i];
3628: if( s2 > nlstate){
3629: lli=log(out[s1][s2] - savm[s1][s2]);
3630: } else if ( s2==-1 ) { /* alive */
3631: for (j=1,survp=0. ; j<=nlstate; j++)
3632: survp += out[s1][j];
3633: lli= log(survp);
3634: }else{
3635: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3636: }
3637: ipmx +=1;
3638: sw += weight[i];
3639: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3640: /* 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 3641: } /* end of wave */
3642: } /* end of individual */
3643: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3644: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3645: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3646: for(mi=1; mi<= wav[i]-1; mi++){
3647: for (ii=1;ii<=nlstate+ndeath;ii++)
3648: for (j=1;j<=nlstate+ndeath;j++){
3649: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3650: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3651: }
3652: for(d=0; d<dh[mi][i]; d++){
3653: newm=savm;
3654: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3655: cov[2]=agexact;
3656: if(nagesqr==1)
3657: cov[3]= agexact*agexact;
3658: for (kk=1; kk<=cptcovage;kk++) {
3659: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3660: }
1.126 brouard 3661:
1.226 brouard 3662: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3663: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3664: savm=oldm;
3665: oldm=newm;
3666: } /* end mult */
3667:
3668: s1=s[mw[mi][i]][i];
3669: s2=s[mw[mi+1][i]][i];
3670: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3671: ipmx +=1;
3672: sw += weight[i];
3673: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3674: /*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]);*/
3675: } /* end of wave */
3676: } /* end of individual */
3677: } /* End of if */
3678: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3679: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3680: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3681: return -l;
1.126 brouard 3682: }
3683:
3684: /*************** log-likelihood *************/
3685: double funcone( double *x)
3686: {
1.228 brouard 3687: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3688: int i, ii, j, k, mi, d, kk;
1.228 brouard 3689: int ioffset=0;
1.131 brouard 3690: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3691: double **out;
3692: double lli; /* Individual log likelihood */
3693: double llt;
3694: int s1, s2;
1.228 brouard 3695: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3696:
1.126 brouard 3697: double bbh, survp;
1.187 brouard 3698: double agexact;
1.214 brouard 3699: double agebegin, ageend;
1.126 brouard 3700: /*extern weight */
3701: /* We are differentiating ll according to initial status */
3702: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3703: /*for(i=1;i<imx;i++)
3704: printf(" %d\n",s[4][i]);
3705: */
3706: cov[1]=1.;
3707:
3708: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3709: ioffset=0;
3710: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3711: /* ioffset=2+nagesqr+cptcovage; */
3712: ioffset=2+nagesqr;
1.232 brouard 3713: /* Fixed */
1.224 brouard 3714: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3715: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3716: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3717: 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)*/
3718: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3719: /* cov[2+6]=covar[Tvar[6]][i]; */
3720: /* cov[2+6]=covar[2][i]; V2 */
3721: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3722: /* cov[2+7]=covar[Tvar[7]][i]; */
3723: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3724: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3725: /* cov[2+9]=covar[Tvar[9]][i]; */
3726: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3727: }
1.232 brouard 3728: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3729: /* 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?)*\/ */
3730: /* } */
1.231 brouard 3731: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3732: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3733: /* } */
1.225 brouard 3734:
1.233 brouard 3735:
3736: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3737: /* Wave varying (but not age varying) */
3738: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3739: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3740: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3741: }
1.232 brouard 3742: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3743: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3744: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3745: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3746: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3747: /* 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 3748: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3749: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3750: /* /\* 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]); *\/ */
3751: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3752: /* } */
1.126 brouard 3753: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3754: for (j=1;j<=nlstate+ndeath;j++){
3755: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3756: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3757: }
1.214 brouard 3758:
3759: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3760: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3761: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3762: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3763: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3764: and mw[mi+1][i]. dh depends on stepm.*/
3765: newm=savm;
1.247 brouard 3766: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3767: cov[2]=agexact;
3768: if(nagesqr==1)
3769: cov[3]= agexact*agexact;
3770: for (kk=1; kk<=cptcovage;kk++) {
3771: if(!FixedV[Tvar[Tage[kk]]])
3772: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3773: else
3774: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3775: }
3776: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3777: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3778: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3779: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3780: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3781: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3782: savm=oldm;
3783: oldm=newm;
1.126 brouard 3784: } /* end mult */
3785:
3786: s1=s[mw[mi][i]][i];
3787: s2=s[mw[mi+1][i]][i];
1.217 brouard 3788: /* if(s2==-1){ */
1.268 brouard 3789: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3790: /* /\* exit(1); *\/ */
3791: /* } */
1.126 brouard 3792: bbh=(double)bh[mi][i]/(double)stepm;
3793: /* bias is positive if real duration
3794: * is higher than the multiple of stepm and negative otherwise.
3795: */
3796: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3797: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3798: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3799: for (j=1,survp=0. ; j<=nlstate; j++)
3800: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3801: lli= log(survp);
1.126 brouard 3802: }else if (mle==1){
1.242 brouard 3803: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3804: } else if(mle==2){
1.242 brouard 3805: 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 3806: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3807: 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 3808: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3809: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3810: } else{ /* mle=0 back to 1 */
1.242 brouard 3811: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3812: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3813: } /* End of if */
3814: ipmx +=1;
3815: sw += weight[i];
3816: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3817: /*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 3818: if(globpr){
1.246 brouard 3819: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3820: %11.6f %11.6f %11.6f ", \
1.242 brouard 3821: 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 3822: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3823: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3824: llt +=ll[k]*gipmx/gsw;
3825: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3826: }
3827: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3828: }
1.232 brouard 3829: } /* end of wave */
3830: } /* end of individual */
3831: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3832: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3833: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3834: if(globpr==0){ /* First time we count the contributions and weights */
3835: gipmx=ipmx;
3836: gsw=sw;
3837: }
3838: return -l;
1.126 brouard 3839: }
3840:
3841:
3842: /*************** function likelione ***********/
3843: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3844: {
3845: /* This routine should help understanding what is done with
3846: the selection of individuals/waves and
3847: to check the exact contribution to the likelihood.
3848: Plotting could be done.
3849: */
3850: int k;
3851:
3852: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3853: strcpy(fileresilk,"ILK_");
1.202 brouard 3854: strcat(fileresilk,fileresu);
1.126 brouard 3855: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3856: printf("Problem with resultfile: %s\n", fileresilk);
3857: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3858: }
1.214 brouard 3859: 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");
3860: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3861: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3862: for(k=1; k<=nlstate; k++)
3863: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3864: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3865: }
3866:
3867: *fretone=(*funcone)(p);
3868: if(*globpri !=0){
3869: fclose(ficresilk);
1.205 brouard 3870: if (mle ==0)
3871: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3872: else if(mle >=1)
3873: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3874: 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 3875: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3876:
3877: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3878: 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 3879: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3880: }
1.207 brouard 3881: 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 3882: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3883: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3884: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3885: fflush(fichtm);
1.205 brouard 3886: }
1.126 brouard 3887: return;
3888: }
3889:
3890:
3891: /*********** Maximum Likelihood Estimation ***************/
3892:
3893: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3894: {
1.165 brouard 3895: int i,j, iter=0;
1.126 brouard 3896: double **xi;
3897: double fret;
3898: double fretone; /* Only one call to likelihood */
3899: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3900:
3901: #ifdef NLOPT
3902: int creturn;
3903: nlopt_opt opt;
3904: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3905: double *lb;
3906: double minf; /* the minimum objective value, upon return */
3907: double * p1; /* Shifted parameters from 0 instead of 1 */
3908: myfunc_data dinst, *d = &dinst;
3909: #endif
3910:
3911:
1.126 brouard 3912: xi=matrix(1,npar,1,npar);
3913: for (i=1;i<=npar;i++)
3914: for (j=1;j<=npar;j++)
3915: xi[i][j]=(i==j ? 1.0 : 0.0);
3916: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3917: strcpy(filerespow,"POW_");
1.126 brouard 3918: strcat(filerespow,fileres);
3919: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3920: printf("Problem with resultfile: %s\n", filerespow);
3921: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3922: }
3923: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3924: for (i=1;i<=nlstate;i++)
3925: for(j=1;j<=nlstate+ndeath;j++)
3926: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3927: fprintf(ficrespow,"\n");
1.162 brouard 3928: #ifdef POWELL
1.126 brouard 3929: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3930: #endif
1.126 brouard 3931:
1.162 brouard 3932: #ifdef NLOPT
3933: #ifdef NEWUOA
3934: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3935: #else
3936: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3937: #endif
3938: lb=vector(0,npar-1);
3939: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3940: nlopt_set_lower_bounds(opt, lb);
3941: nlopt_set_initial_step1(opt, 0.1);
3942:
3943: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3944: d->function = func;
3945: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3946: nlopt_set_min_objective(opt, myfunc, d);
3947: nlopt_set_xtol_rel(opt, ftol);
3948: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3949: printf("nlopt failed! %d\n",creturn);
3950: }
3951: else {
3952: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3953: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3954: iter=1; /* not equal */
3955: }
3956: nlopt_destroy(opt);
3957: #endif
1.126 brouard 3958: free_matrix(xi,1,npar,1,npar);
3959: fclose(ficrespow);
1.203 brouard 3960: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3961: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3962: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3963:
3964: }
3965:
3966: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3967: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3968: {
3969: double **a,**y,*x,pd;
1.203 brouard 3970: /* double **hess; */
1.164 brouard 3971: int i, j;
1.126 brouard 3972: int *indx;
3973:
3974: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3975: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3976: void lubksb(double **a, int npar, int *indx, double b[]) ;
3977: void ludcmp(double **a, int npar, int *indx, double *d) ;
3978: double gompertz(double p[]);
1.203 brouard 3979: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3980:
3981: printf("\nCalculation of the hessian matrix. Wait...\n");
3982: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3983: for (i=1;i<=npar;i++){
1.203 brouard 3984: printf("%d-",i);fflush(stdout);
3985: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3986:
3987: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3988:
3989: /* printf(" %f ",p[i]);
3990: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3991: }
3992:
3993: for (i=1;i<=npar;i++) {
3994: for (j=1;j<=npar;j++) {
3995: if (j>i) {
1.203 brouard 3996: printf(".%d-%d",i,j);fflush(stdout);
3997: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3998: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3999:
4000: hess[j][i]=hess[i][j];
4001: /*printf(" %lf ",hess[i][j]);*/
4002: }
4003: }
4004: }
4005: printf("\n");
4006: fprintf(ficlog,"\n");
4007:
4008: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4009: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4010:
4011: a=matrix(1,npar,1,npar);
4012: y=matrix(1,npar,1,npar);
4013: x=vector(1,npar);
4014: indx=ivector(1,npar);
4015: for (i=1;i<=npar;i++)
4016: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4017: ludcmp(a,npar,indx,&pd);
4018:
4019: for (j=1;j<=npar;j++) {
4020: for (i=1;i<=npar;i++) x[i]=0;
4021: x[j]=1;
4022: lubksb(a,npar,indx,x);
4023: for (i=1;i<=npar;i++){
4024: matcov[i][j]=x[i];
4025: }
4026: }
4027:
4028: printf("\n#Hessian matrix#\n");
4029: fprintf(ficlog,"\n#Hessian matrix#\n");
4030: for (i=1;i<=npar;i++) {
4031: for (j=1;j<=npar;j++) {
1.203 brouard 4032: printf("%.6e ",hess[i][j]);
4033: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4034: }
4035: printf("\n");
4036: fprintf(ficlog,"\n");
4037: }
4038:
1.203 brouard 4039: /* printf("\n#Covariance matrix#\n"); */
4040: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4041: /* for (i=1;i<=npar;i++) { */
4042: /* for (j=1;j<=npar;j++) { */
4043: /* printf("%.6e ",matcov[i][j]); */
4044: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4045: /* } */
4046: /* printf("\n"); */
4047: /* fprintf(ficlog,"\n"); */
4048: /* } */
4049:
1.126 brouard 4050: /* Recompute Inverse */
1.203 brouard 4051: /* for (i=1;i<=npar;i++) */
4052: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4053: /* ludcmp(a,npar,indx,&pd); */
4054:
4055: /* printf("\n#Hessian matrix recomputed#\n"); */
4056:
4057: /* for (j=1;j<=npar;j++) { */
4058: /* for (i=1;i<=npar;i++) x[i]=0; */
4059: /* x[j]=1; */
4060: /* lubksb(a,npar,indx,x); */
4061: /* for (i=1;i<=npar;i++){ */
4062: /* y[i][j]=x[i]; */
4063: /* printf("%.3e ",y[i][j]); */
4064: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4065: /* } */
4066: /* printf("\n"); */
4067: /* fprintf(ficlog,"\n"); */
4068: /* } */
4069:
4070: /* Verifying the inverse matrix */
4071: #ifdef DEBUGHESS
4072: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4073:
1.203 brouard 4074: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4075: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4076:
4077: for (j=1;j<=npar;j++) {
4078: for (i=1;i<=npar;i++){
1.203 brouard 4079: printf("%.2f ",y[i][j]);
4080: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4081: }
4082: printf("\n");
4083: fprintf(ficlog,"\n");
4084: }
1.203 brouard 4085: #endif
1.126 brouard 4086:
4087: free_matrix(a,1,npar,1,npar);
4088: free_matrix(y,1,npar,1,npar);
4089: free_vector(x,1,npar);
4090: free_ivector(indx,1,npar);
1.203 brouard 4091: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4092:
4093:
4094: }
4095:
4096: /*************** hessian matrix ****************/
4097: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4098: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4099: int i;
4100: int l=1, lmax=20;
1.203 brouard 4101: double k1,k2, res, fx;
1.132 brouard 4102: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4103: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4104: int k=0,kmax=10;
4105: double l1;
4106:
4107: fx=func(x);
4108: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4109: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4110: l1=pow(10,l);
4111: delts=delt;
4112: for(k=1 ; k <kmax; k=k+1){
4113: delt = delta*(l1*k);
4114: p2[theta]=x[theta] +delt;
1.145 brouard 4115: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4116: p2[theta]=x[theta]-delt;
4117: k2=func(p2)-fx;
4118: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4119: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4120:
1.203 brouard 4121: #ifdef DEBUGHESSII
1.126 brouard 4122: 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);
4123: 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);
4124: #endif
4125: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4126: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4127: k=kmax;
4128: }
4129: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4130: k=kmax; l=lmax*10;
1.126 brouard 4131: }
4132: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4133: delts=delt;
4134: }
1.203 brouard 4135: } /* End loop k */
1.126 brouard 4136: }
4137: delti[theta]=delts;
4138: return res;
4139:
4140: }
4141:
1.203 brouard 4142: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4143: {
4144: int i;
1.164 brouard 4145: int l=1, lmax=20;
1.126 brouard 4146: double k1,k2,k3,k4,res,fx;
1.132 brouard 4147: double p2[MAXPARM+1];
1.203 brouard 4148: int k, kmax=1;
4149: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4150:
4151: int firstime=0;
1.203 brouard 4152:
1.126 brouard 4153: fx=func(x);
1.203 brouard 4154: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4155: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4156: p2[thetai]=x[thetai]+delti[thetai]*k;
4157: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4158: k1=func(p2)-fx;
4159:
1.203 brouard 4160: p2[thetai]=x[thetai]+delti[thetai]*k;
4161: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4162: k2=func(p2)-fx;
4163:
1.203 brouard 4164: p2[thetai]=x[thetai]-delti[thetai]*k;
4165: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4166: k3=func(p2)-fx;
4167:
1.203 brouard 4168: p2[thetai]=x[thetai]-delti[thetai]*k;
4169: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4170: k4=func(p2)-fx;
1.203 brouard 4171: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4172: if(k1*k2*k3*k4 <0.){
1.208 brouard 4173: firstime=1;
1.203 brouard 4174: kmax=kmax+10;
1.208 brouard 4175: }
4176: if(kmax >=10 || firstime ==1){
1.246 brouard 4177: 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);
4178: 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 4179: 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);
4180: 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);
4181: }
4182: #ifdef DEBUGHESSIJ
4183: v1=hess[thetai][thetai];
4184: v2=hess[thetaj][thetaj];
4185: cv12=res;
4186: /* Computing eigen value of Hessian matrix */
4187: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4188: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4189: if ((lc2 <0) || (lc1 <0) ){
4190: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4191: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4192: 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);
4193: 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);
4194: }
1.126 brouard 4195: #endif
4196: }
4197: return res;
4198: }
4199:
1.203 brouard 4200: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4201: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4202: /* { */
4203: /* int i; */
4204: /* int l=1, lmax=20; */
4205: /* double k1,k2,k3,k4,res,fx; */
4206: /* double p2[MAXPARM+1]; */
4207: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4208: /* int k=0,kmax=10; */
4209: /* double l1; */
4210:
4211: /* fx=func(x); */
4212: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4213: /* l1=pow(10,l); */
4214: /* delts=delt; */
4215: /* for(k=1 ; k <kmax; k=k+1){ */
4216: /* delt = delti*(l1*k); */
4217: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4218: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4219: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4220: /* k1=func(p2)-fx; */
4221:
4222: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4223: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4224: /* k2=func(p2)-fx; */
4225:
4226: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4227: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4228: /* k3=func(p2)-fx; */
4229:
4230: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4231: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4232: /* k4=func(p2)-fx; */
4233: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4234: /* #ifdef DEBUGHESSIJ */
4235: /* 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); */
4236: /* 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); */
4237: /* #endif */
4238: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4239: /* k=kmax; */
4240: /* } */
4241: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4242: /* k=kmax; l=lmax*10; */
4243: /* } */
4244: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4245: /* delts=delt; */
4246: /* } */
4247: /* } /\* End loop k *\/ */
4248: /* } */
4249: /* delti[theta]=delts; */
4250: /* return res; */
4251: /* } */
4252:
4253:
1.126 brouard 4254: /************** Inverse of matrix **************/
4255: void ludcmp(double **a, int n, int *indx, double *d)
4256: {
4257: int i,imax,j,k;
4258: double big,dum,sum,temp;
4259: double *vv;
4260:
4261: vv=vector(1,n);
4262: *d=1.0;
4263: for (i=1;i<=n;i++) {
4264: big=0.0;
4265: for (j=1;j<=n;j++)
4266: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4267: if (big == 0.0){
4268: printf(" Singular Hessian matrix at row %d:\n",i);
4269: for (j=1;j<=n;j++) {
4270: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4271: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4272: }
4273: fflush(ficlog);
4274: fclose(ficlog);
4275: nrerror("Singular matrix in routine ludcmp");
4276: }
1.126 brouard 4277: vv[i]=1.0/big;
4278: }
4279: for (j=1;j<=n;j++) {
4280: for (i=1;i<j;i++) {
4281: sum=a[i][j];
4282: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4283: a[i][j]=sum;
4284: }
4285: big=0.0;
4286: for (i=j;i<=n;i++) {
4287: sum=a[i][j];
4288: for (k=1;k<j;k++)
4289: sum -= a[i][k]*a[k][j];
4290: a[i][j]=sum;
4291: if ( (dum=vv[i]*fabs(sum)) >= big) {
4292: big=dum;
4293: imax=i;
4294: }
4295: }
4296: if (j != imax) {
4297: for (k=1;k<=n;k++) {
4298: dum=a[imax][k];
4299: a[imax][k]=a[j][k];
4300: a[j][k]=dum;
4301: }
4302: *d = -(*d);
4303: vv[imax]=vv[j];
4304: }
4305: indx[j]=imax;
4306: if (a[j][j] == 0.0) a[j][j]=TINY;
4307: if (j != n) {
4308: dum=1.0/(a[j][j]);
4309: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4310: }
4311: }
4312: free_vector(vv,1,n); /* Doesn't work */
4313: ;
4314: }
4315:
4316: void lubksb(double **a, int n, int *indx, double b[])
4317: {
4318: int i,ii=0,ip,j;
4319: double sum;
4320:
4321: for (i=1;i<=n;i++) {
4322: ip=indx[i];
4323: sum=b[ip];
4324: b[ip]=b[i];
4325: if (ii)
4326: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4327: else if (sum) ii=i;
4328: b[i]=sum;
4329: }
4330: for (i=n;i>=1;i--) {
4331: sum=b[i];
4332: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4333: b[i]=sum/a[i][i];
4334: }
4335: }
4336:
4337: void pstamp(FILE *fichier)
4338: {
1.196 brouard 4339: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4340: }
4341:
1.253 brouard 4342:
4343:
1.126 brouard 4344: /************ Frequencies ********************/
1.251 brouard 4345: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4346: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4347: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4348: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4349:
1.265 brouard 4350: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4351: int iind=0, iage=0;
4352: int mi; /* Effective wave */
4353: int first;
4354: double ***freq; /* Frequencies */
1.268 brouard 4355: 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 */
4356: 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);
1.226 brouard 4357: double *meanq;
4358: double **meanqt;
4359: double *pp, **prop, *posprop, *pospropt;
4360: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4361: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4362: double agebegin, ageend;
4363:
4364: pp=vector(1,nlstate);
1.251 brouard 4365: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4366: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4367: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4368: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4369: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4370: meanqt=matrix(1,lastpass,1,nqtveff);
4371: strcpy(fileresp,"P_");
4372: strcat(fileresp,fileresu);
4373: /*strcat(fileresphtm,fileresu);*/
4374: if((ficresp=fopen(fileresp,"w"))==NULL) {
4375: printf("Problem with prevalence resultfile: %s\n", fileresp);
4376: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4377: exit(0);
4378: }
1.240 brouard 4379:
1.226 brouard 4380: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4381: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4382: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4383: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4384: fflush(ficlog);
4385: exit(70);
4386: }
4387: else{
4388: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4389: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4390: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4391: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4392: }
1.237 brouard 4393: 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 4394:
1.226 brouard 4395: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4396: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4397: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4398: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4399: fflush(ficlog);
4400: exit(70);
1.240 brouard 4401: } else{
1.226 brouard 4402: 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 4403: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4404: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4405: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4406: }
1.240 brouard 4407: 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);
4408:
1.253 brouard 4409: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4410: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4411: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4412: j1=0;
1.126 brouard 4413:
1.227 brouard 4414: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4415: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4416: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4417:
4418:
1.226 brouard 4419: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4420: reference=low_education V1=0,V2=0
4421: med_educ V1=1 V2=0,
4422: high_educ V1=0 V2=1
4423: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4424: */
1.249 brouard 4425: dateintsum=0;
4426: k2cpt=0;
4427:
1.253 brouard 4428: if(cptcoveff == 0 )
1.265 brouard 4429: nl=1; /* Constant and age model only */
1.253 brouard 4430: else
4431: nl=2;
1.265 brouard 4432:
4433: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4434: /* Loop on nj=1 or 2 if dummy covariates j!=0
4435: * Loop on j1(1 to 2**cptcoveff) covariate combination
4436: * freq[s1][s2][iage] =0.
4437: * Loop on iind
4438: * ++freq[s1][s2][iage] weighted
4439: * end iind
4440: * if covariate and j!0
4441: * headers Variable on one line
4442: * endif cov j!=0
4443: * header of frequency table by age
4444: * Loop on age
4445: * pp[s1]+=freq[s1][s2][iage] weighted
4446: * pos+=freq[s1][s2][iage] weighted
4447: * Loop on s1 initial state
4448: * fprintf(ficresp
4449: * end s1
4450: * end age
4451: * if j!=0 computes starting values
4452: * end compute starting values
4453: * end j1
4454: * end nl
4455: */
1.253 brouard 4456: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4457: if(nj==1)
4458: j=0; /* First pass for the constant */
1.265 brouard 4459: else{
1.253 brouard 4460: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4461: }
1.251 brouard 4462: first=1;
1.265 brouard 4463: 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 4464: posproptt=0.;
4465: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4466: scanf("%d", i);*/
4467: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4468: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4469: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4470: freq[i][s2][m]=0;
1.251 brouard 4471:
4472: for (i=1; i<=nlstate; i++) {
1.240 brouard 4473: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4474: prop[i][m]=0;
4475: posprop[i]=0;
4476: pospropt[i]=0;
4477: }
4478: /* for (z1=1; z1<= nqfveff; z1++) { */
4479: /* meanq[z1]+=0.; */
4480: /* for(m=1;m<=lastpass;m++){ */
4481: /* meanqt[m][z1]=0.; */
4482: /* } */
4483: /* } */
4484:
4485: /* dateintsum=0; */
4486: /* k2cpt=0; */
4487:
1.265 brouard 4488: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4489: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4490: bool=1;
4491: if(j !=0){
4492: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4493: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4494: /* for (z1=1; z1<= nqfveff; z1++) { */
4495: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4496: /* } */
4497: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4498: /* if(Tvaraff[z1] ==-20){ */
4499: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4500: /* }else if(Tvaraff[z1] ==-10){ */
4501: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4502: /* }else */
4503: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4504: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4505: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4506: /* 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",
4507: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4508: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4509: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4510: } /* Onlyf fixed */
4511: } /* end z1 */
4512: } /* cptcovn > 0 */
4513: } /* end any */
4514: }/* end j==0 */
1.265 brouard 4515: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4516: /* for(m=firstpass; m<=lastpass; m++){ */
4517: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4518: m=mw[mi][iind];
4519: if(j!=0){
4520: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4521: for (z1=1; z1<=cptcoveff; z1++) {
4522: if( Fixed[Tmodelind[z1]]==1){
4523: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4524: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4525: value is -1, we don't select. It differs from the
4526: constant and age model which counts them. */
4527: bool=0; /* not selected */
4528: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4529: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4530: bool=0;
4531: }
4532: }
4533: }
4534: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4535: } /* end j==0 */
4536: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4537: if(bool==1){
4538: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4539: and mw[mi+1][iind]. dh depends on stepm. */
4540: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4541: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4542: if(m >=firstpass && m <=lastpass){
4543: k2=anint[m][iind]+(mint[m][iind]/12.);
4544: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4545: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4546: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4547: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4548: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4549: if (m<lastpass) {
4550: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4551: /* 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]); */
4552: if(s[m][iind]==-1)
4553: 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.));
4554: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4555: /* if((int)agev[m][iind] == 55) */
4556: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4557: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4558: 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 4559: }
1.251 brouard 4560: } /* end if between passes */
4561: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4562: dateintsum=dateintsum+k2; /* on all covariates ?*/
4563: k2cpt++;
4564: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4565: }
1.251 brouard 4566: }else{
4567: bool=1;
4568: }/* end bool 2 */
4569: } /* end m */
4570: } /* end bool */
4571: } /* end iind = 1 to imx */
4572: /* prop[s][age] is feeded for any initial and valid live state as well as
4573: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4574:
4575:
4576: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4577: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4578: pstamp(ficresp);
1.251 brouard 4579: if (cptcoveff>0 && j!=0){
1.265 brouard 4580: pstamp(ficresp);
1.251 brouard 4581: printf( "\n#********** Variable ");
4582: fprintf(ficresp, "\n#********** Variable ");
4583: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4584: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4585: fprintf(ficlog, "\n#********** Variable ");
4586: for (z1=1; z1<=cptcoveff; z1++){
4587: if(!FixedV[Tvaraff[z1]]){
4588: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4589: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4590: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4591: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4592: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4593: }else{
1.251 brouard 4594: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4595: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4596: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4597: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4598: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4599: }
4600: }
4601: printf( "**********\n#");
4602: fprintf(ficresp, "**********\n#");
4603: fprintf(ficresphtm, "**********</h3>\n");
4604: fprintf(ficresphtmfr, "**********</h3>\n");
4605: fprintf(ficlog, "**********\n");
4606: }
4607: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4608: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4609: fprintf(ficresp, " Age");
4610: 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 4611: for(i=1; i<=nlstate;i++) {
1.265 brouard 4612: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4613: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4614: }
1.265 brouard 4615: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4616: fprintf(ficresphtm, "\n");
4617:
4618: /* Header of frequency table by age */
4619: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4620: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4621: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4622: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4623: if(s2!=0 && m!=0)
4624: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4625: }
1.226 brouard 4626: }
1.251 brouard 4627: fprintf(ficresphtmfr, "\n");
4628:
4629: /* For each age */
4630: for(iage=iagemin; iage <= iagemax+3; iage++){
4631: fprintf(ficresphtm,"<tr>");
4632: if(iage==iagemax+1){
4633: fprintf(ficlog,"1");
4634: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4635: }else if(iage==iagemax+2){
4636: fprintf(ficlog,"0");
4637: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4638: }else if(iage==iagemax+3){
4639: fprintf(ficlog,"Total");
4640: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4641: }else{
1.240 brouard 4642: if(first==1){
1.251 brouard 4643: first=0;
4644: printf("See log file for details...\n");
4645: }
4646: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4647: fprintf(ficlog,"Age %d", iage);
4648: }
1.265 brouard 4649: for(s1=1; s1 <=nlstate ; s1++){
4650: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4651: pp[s1] += freq[s1][m][iage];
1.251 brouard 4652: }
1.265 brouard 4653: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4654: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4655: pos += freq[s1][m][iage];
4656: if(pp[s1]>=1.e-10){
1.251 brouard 4657: if(first==1){
1.265 brouard 4658: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4659: }
1.265 brouard 4660: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4661: }else{
4662: if(first==1)
1.265 brouard 4663: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4664: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4665: }
4666: }
4667:
1.265 brouard 4668: for(s1=1; s1 <=nlstate ; s1++){
4669: /* posprop[s1]=0; */
4670: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4671: pp[s1] += freq[s1][m][iage];
4672: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4673:
4674: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4675: pos += pp[s1]; /* pos is the total number of transitions until this age */
4676: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4677: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4678: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4679: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4680: }
4681:
4682: /* Writing ficresp */
4683: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4684: if( iage <= iagemax){
4685: fprintf(ficresp," %d",iage);
4686: }
4687: }else if( nj==2){
4688: if( iage <= iagemax){
4689: fprintf(ficresp," %d",iage);
4690: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4691: }
1.240 brouard 4692: }
1.265 brouard 4693: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4694: if(pos>=1.e-5){
1.251 brouard 4695: if(first==1)
1.265 brouard 4696: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4697: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4698: }else{
4699: if(first==1)
1.265 brouard 4700: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4701: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4702: }
4703: if( iage <= iagemax){
4704: if(pos>=1.e-5){
1.265 brouard 4705: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4706: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4707: }else if( nj==2){
4708: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4709: }
4710: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4711: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4712: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4713: } else{
4714: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4715: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4716: }
1.240 brouard 4717: }
1.265 brouard 4718: pospropt[s1] +=posprop[s1];
4719: } /* end loop s1 */
1.251 brouard 4720: /* pospropt=0.; */
1.265 brouard 4721: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4722: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4723: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4724: if(first==1){
1.265 brouard 4725: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4726: }
1.265 brouard 4727: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4728: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4729: }
1.265 brouard 4730: if(s1!=0 && m!=0)
4731: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4732: }
1.265 brouard 4733: } /* end loop s1 */
1.251 brouard 4734: posproptt=0.;
1.265 brouard 4735: for(s1=1; s1 <=nlstate; s1++){
4736: posproptt += pospropt[s1];
1.251 brouard 4737: }
4738: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4739: fprintf(ficresphtm,"</tr>\n");
4740: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4741: if(iage <= iagemax)
4742: fprintf(ficresp,"\n");
1.240 brouard 4743: }
1.251 brouard 4744: if(first==1)
4745: printf("Others in log...\n");
4746: fprintf(ficlog,"\n");
4747: } /* end loop age iage */
1.265 brouard 4748:
1.251 brouard 4749: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4750: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4751: if(posproptt < 1.e-5){
1.265 brouard 4752: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4753: }else{
1.265 brouard 4754: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4755: }
1.226 brouard 4756: }
1.251 brouard 4757: fprintf(ficresphtm,"</tr>\n");
4758: fprintf(ficresphtm,"</table>\n");
4759: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4760: if(posproptt < 1.e-5){
1.251 brouard 4761: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4762: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4763: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4764: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4765: invalidvarcomb[j1]=1;
1.226 brouard 4766: }else{
1.251 brouard 4767: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4768: invalidvarcomb[j1]=0;
1.226 brouard 4769: }
1.251 brouard 4770: fprintf(ficresphtmfr,"</table>\n");
4771: fprintf(ficlog,"\n");
4772: if(j!=0){
4773: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4774: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4775: for(k=1; k <=(nlstate+ndeath); k++){
4776: if (k != i) {
1.265 brouard 4777: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4778: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4779: if(j1==1){ /* All dummy covariates to zero */
4780: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4781: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4782: printf("%d%d ",i,k);
4783: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4784: 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]));
4785: 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]));
4786: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4787: }
1.253 brouard 4788: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4789: for(iage=iagemin; iage <= iagemax+3; iage++){
4790: x[iage]= (double)iage;
4791: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4792: /* 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 4793: }
1.268 brouard 4794: /* Some are not finite, but linreg will ignore these ages */
4795: no=0;
1.253 brouard 4796: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4797: pstart[s1]=b;
4798: pstart[s1-1]=a;
1.252 brouard 4799: }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 */
4800: 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]);
4801: 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 4802: 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 4803: printf("%d%d ",i,k);
4804: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4805: 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 4806: }else{ /* Other cases, like quantitative fixed or varying covariates */
4807: ;
4808: }
4809: /* printf("%12.7f )", param[i][jj][k]); */
4810: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4811: s1++;
1.251 brouard 4812: } /* end jj */
4813: } /* end k!= i */
4814: } /* end k */
1.265 brouard 4815: } /* end i, s1 */
1.251 brouard 4816: } /* end j !=0 */
4817: } /* end selected combination of covariate j1 */
4818: if(j==0){ /* We can estimate starting values from the occurences in each case */
4819: printf("#Freqsummary: Starting values for the constants:\n");
4820: fprintf(ficlog,"\n");
1.265 brouard 4821: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4822: for(k=1; k <=(nlstate+ndeath); k++){
4823: if (k != i) {
4824: printf("%d%d ",i,k);
4825: fprintf(ficlog,"%d%d ",i,k);
4826: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4827: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4828: if(jj==1){ /* Age has to be done */
1.265 brouard 4829: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4830: 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]));
4831: 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 4832: }
4833: /* printf("%12.7f )", param[i][jj][k]); */
4834: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4835: s1++;
1.250 brouard 4836: }
1.251 brouard 4837: printf("\n");
4838: fprintf(ficlog,"\n");
1.250 brouard 4839: }
4840: }
4841: }
1.251 brouard 4842: printf("#Freqsummary\n");
4843: fprintf(ficlog,"\n");
1.265 brouard 4844: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4845: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4846: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4847: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4848: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4849: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4850: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4851: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4852: /* } */
4853: }
1.265 brouard 4854: } /* end loop s1 */
1.251 brouard 4855:
4856: printf("\n");
4857: fprintf(ficlog,"\n");
4858: } /* end j=0 */
1.249 brouard 4859: } /* end j */
1.252 brouard 4860:
1.253 brouard 4861: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4862: for(i=1, jk=1; i <=nlstate; i++){
4863: for(j=1; j <=nlstate+ndeath; j++){
4864: if(j!=i){
4865: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4866: printf("%1d%1d",i,j);
4867: fprintf(ficparo,"%1d%1d",i,j);
4868: for(k=1; k<=ncovmodel;k++){
4869: /* printf(" %lf",param[i][j][k]); */
4870: /* fprintf(ficparo," %lf",param[i][j][k]); */
4871: p[jk]=pstart[jk];
4872: printf(" %f ",pstart[jk]);
4873: fprintf(ficparo," %f ",pstart[jk]);
4874: jk++;
4875: }
4876: printf("\n");
4877: fprintf(ficparo,"\n");
4878: }
4879: }
4880: }
4881: } /* end mle=-2 */
1.226 brouard 4882: dateintmean=dateintsum/k2cpt;
1.240 brouard 4883:
1.226 brouard 4884: fclose(ficresp);
4885: fclose(ficresphtm);
4886: fclose(ficresphtmfr);
4887: free_vector(meanq,1,nqfveff);
4888: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4889: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4890: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4891: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4892: free_vector(pospropt,1,nlstate);
4893: free_vector(posprop,1,nlstate);
1.251 brouard 4894: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4895: free_vector(pp,1,nlstate);
4896: /* End of freqsummary */
4897: }
1.126 brouard 4898:
1.268 brouard 4899: /* Simple linear regression */
4900: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4901:
4902: /* y=a+bx regression */
4903: double sumx = 0.0; /* sum of x */
4904: double sumx2 = 0.0; /* sum of x**2 */
4905: double sumxy = 0.0; /* sum of x * y */
4906: double sumy = 0.0; /* sum of y */
4907: double sumy2 = 0.0; /* sum of y**2 */
4908: double sume2 = 0.0; /* sum of square or residuals */
4909: double yhat;
4910:
4911: double denom=0;
4912: int i;
4913: int ne=*no;
4914:
4915: for ( i=ifi, ne=0;i<=ila;i++) {
4916: if(!isfinite(x[i]) || !isfinite(y[i])){
4917: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4918: continue;
4919: }
4920: ne=ne+1;
4921: sumx += x[i];
4922: sumx2 += x[i]*x[i];
4923: sumxy += x[i] * y[i];
4924: sumy += y[i];
4925: sumy2 += y[i]*y[i];
4926: denom = (ne * sumx2 - sumx*sumx);
4927: /* 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); */
4928: }
4929:
4930: denom = (ne * sumx2 - sumx*sumx);
4931: if (denom == 0) {
4932: // vertical, slope m is infinity
4933: *b = INFINITY;
4934: *a = 0;
4935: if (r) *r = 0;
4936: return 1;
4937: }
4938:
4939: *b = (ne * sumxy - sumx * sumy) / denom;
4940: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4941: if (r!=NULL) {
4942: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4943: sqrt((sumx2 - sumx*sumx/ne) *
4944: (sumy2 - sumy*sumy/ne));
4945: }
4946: *no=ne;
4947: for ( i=ifi, ne=0;i<=ila;i++) {
4948: if(!isfinite(x[i]) || !isfinite(y[i])){
4949: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4950: continue;
4951: }
4952: ne=ne+1;
4953: yhat = y[i] - *a -*b* x[i];
4954: sume2 += yhat * yhat ;
4955:
4956: denom = (ne * sumx2 - sumx*sumx);
4957: /* 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); */
4958: }
4959: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
4960: *sa= *sb * sqrt(sumx2/ne);
4961:
4962: return 0;
4963: }
4964:
1.126 brouard 4965: /************ Prevalence ********************/
1.227 brouard 4966: 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)
4967: {
4968: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4969: in each health status at the date of interview (if between dateprev1 and dateprev2).
4970: We still use firstpass and lastpass as another selection.
4971: */
1.126 brouard 4972:
1.227 brouard 4973: int i, m, jk, j1, bool, z1,j, iv;
4974: int mi; /* Effective wave */
4975: int iage;
4976: double agebegin, ageend;
4977:
4978: double **prop;
4979: double posprop;
4980: double y2; /* in fractional years */
4981: int iagemin, iagemax;
4982: int first; /** to stop verbosity which is redirected to log file */
4983:
4984: iagemin= (int) agemin;
4985: iagemax= (int) agemax;
4986: /*pp=vector(1,nlstate);*/
1.251 brouard 4987: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4988: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4989: j1=0;
1.222 brouard 4990:
1.227 brouard 4991: /*j=cptcoveff;*/
4992: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4993:
1.227 brouard 4994: first=1;
4995: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4996: for (i=1; i<=nlstate; i++)
1.251 brouard 4997: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4998: prop[i][iage]=0.0;
4999: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5000: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5001: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5002:
5003: for (i=1; i<=imx; i++) { /* Each individual */
5004: bool=1;
5005: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5006: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5007: m=mw[mi][i];
5008: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5009: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5010: for (z1=1; z1<=cptcoveff; z1++){
5011: if( Fixed[Tmodelind[z1]]==1){
5012: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5013: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5014: bool=0;
5015: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5016: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5017: bool=0;
5018: }
5019: }
5020: if(bool==1){ /* Otherwise we skip that wave/person */
5021: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5022: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5023: if(m >=firstpass && m <=lastpass){
5024: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5025: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5026: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5027: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5028: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5029: 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);
5030: exit(1);
5031: }
5032: if (s[m][i]>0 && s[m][i]<=nlstate) {
5033: /*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]]);*/
5034: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5035: prop[s[m][i]][iagemax+3] += weight[i];
5036: } /* end valid statuses */
5037: } /* end selection of dates */
5038: } /* end selection of waves */
5039: } /* end bool */
5040: } /* end wave */
5041: } /* end individual */
5042: for(i=iagemin; i <= iagemax+3; i++){
5043: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5044: posprop += prop[jk][i];
5045: }
5046:
5047: for(jk=1; jk <=nlstate ; jk++){
5048: if( i <= iagemax){
5049: if(posprop>=1.e-5){
5050: probs[i][jk][j1]= prop[jk][i]/posprop;
5051: } else{
5052: if(first==1){
5053: first=0;
1.266 brouard 5054: 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]);
5055: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
5056: }else{
5057: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
1.227 brouard 5058: }
5059: }
5060: }
5061: }/* end jk */
5062: }/* end i */
1.222 brouard 5063: /*} *//* end i1 */
1.227 brouard 5064: } /* end j1 */
1.222 brouard 5065:
1.227 brouard 5066: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5067: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5068: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5069: } /* End of prevalence */
1.126 brouard 5070:
5071: /************* Waves Concatenation ***************/
5072:
5073: 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)
5074: {
5075: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5076: Death is a valid wave (if date is known).
5077: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5078: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5079: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5080: */
1.126 brouard 5081:
1.224 brouard 5082: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5083: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5084: double sum=0., jmean=0.;*/
1.224 brouard 5085: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5086: int j, k=0,jk, ju, jl;
5087: double sum=0.;
5088: first=0;
1.214 brouard 5089: firstwo=0;
1.217 brouard 5090: firsthree=0;
1.218 brouard 5091: firstfour=0;
1.164 brouard 5092: jmin=100000;
1.126 brouard 5093: jmax=-1;
5094: jmean=0.;
1.224 brouard 5095:
5096: /* Treating live states */
1.214 brouard 5097: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5098: mi=0; /* First valid wave */
1.227 brouard 5099: mli=0; /* Last valid wave */
1.126 brouard 5100: m=firstpass;
1.214 brouard 5101: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5102: 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 */
5103: mli=m-1;/* mw[++mi][i]=m-1; */
5104: }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 */
5105: mw[++mi][i]=m;
5106: mli=m;
1.224 brouard 5107: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5108: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5109: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5110: }
1.227 brouard 5111: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5112: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5113: break;
1.224 brouard 5114: #else
1.227 brouard 5115: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5116: if(firsthree == 0){
1.262 brouard 5117: 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 5118: firsthree=1;
5119: }
1.262 brouard 5120: 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 5121: mw[++mi][i]=m;
5122: mli=m;
5123: }
5124: if(s[m][i]==-2){ /* Vital status is really unknown */
5125: nbwarn++;
5126: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5127: 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);
5128: 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);
5129: }
5130: break;
5131: }
5132: break;
1.224 brouard 5133: #endif
1.227 brouard 5134: }/* End m >= lastpass */
1.126 brouard 5135: }/* end while */
1.224 brouard 5136:
1.227 brouard 5137: /* 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 5138: /* After last pass */
1.224 brouard 5139: /* Treating death states */
1.214 brouard 5140: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5141: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5142: /* } */
1.126 brouard 5143: mi++; /* Death is another wave */
5144: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5145: /* Only death is a correct wave */
1.126 brouard 5146: mw[mi][i]=m;
1.257 brouard 5147: } /* else not in a death state */
1.224 brouard 5148: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5149: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5150: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5151: 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 */
5152: nbwarn++;
5153: if(firstfiv==0){
5154: 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 );
5155: firstfiv=1;
5156: }else{
5157: 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 );
5158: }
5159: }else{ /* Death occured afer last wave potential bias */
5160: nberr++;
5161: if(firstwo==0){
1.257 brouard 5162: 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 5163: firstwo=1;
5164: }
1.257 brouard 5165: 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 5166: }
1.257 brouard 5167: }else{ /* if date of interview is unknown */
1.227 brouard 5168: /* death is known but not confirmed by death status at any wave */
5169: if(firstfour==0){
5170: 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 );
5171: firstfour=1;
5172: }
5173: 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 5174: }
1.224 brouard 5175: } /* end if date of death is known */
5176: #endif
5177: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5178: /* wav[i]=mw[mi][i]; */
1.126 brouard 5179: if(mi==0){
5180: nbwarn++;
5181: if(first==0){
1.227 brouard 5182: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5183: first=1;
1.126 brouard 5184: }
5185: if(first==1){
1.227 brouard 5186: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5187: }
5188: } /* end mi==0 */
5189: } /* End individuals */
1.214 brouard 5190: /* wav and mw are no more changed */
1.223 brouard 5191:
1.214 brouard 5192:
1.126 brouard 5193: for(i=1; i<=imx; i++){
5194: for(mi=1; mi<wav[i];mi++){
5195: if (stepm <=0)
1.227 brouard 5196: dh[mi][i]=1;
1.126 brouard 5197: else{
1.260 brouard 5198: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5199: if (agedc[i] < 2*AGESUP) {
5200: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5201: if(j==0) j=1; /* Survives at least one month after exam */
5202: else if(j<0){
5203: nberr++;
5204: 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]);
5205: j=1; /* Temporary Dangerous patch */
5206: 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);
5207: 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]);
5208: 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);
5209: }
5210: k=k+1;
5211: if (j >= jmax){
5212: jmax=j;
5213: ijmax=i;
5214: }
5215: if (j <= jmin){
5216: jmin=j;
5217: ijmin=i;
5218: }
5219: sum=sum+j;
5220: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5221: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5222: }
5223: }
5224: else{
5225: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5226: /* 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 5227:
1.227 brouard 5228: k=k+1;
5229: if (j >= jmax) {
5230: jmax=j;
5231: ijmax=i;
5232: }
5233: else if (j <= jmin){
5234: jmin=j;
5235: ijmin=i;
5236: }
5237: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5238: /*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]);*/
5239: if(j<0){
5240: nberr++;
5241: 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]);
5242: 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]);
5243: }
5244: sum=sum+j;
5245: }
5246: jk= j/stepm;
5247: jl= j -jk*stepm;
5248: ju= j -(jk+1)*stepm;
5249: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5250: if(jl==0){
5251: dh[mi][i]=jk;
5252: bh[mi][i]=0;
5253: }else{ /* We want a negative bias in order to only have interpolation ie
5254: * to avoid the price of an extra matrix product in likelihood */
5255: dh[mi][i]=jk+1;
5256: bh[mi][i]=ju;
5257: }
5258: }else{
5259: if(jl <= -ju){
5260: dh[mi][i]=jk;
5261: bh[mi][i]=jl; /* bias is positive if real duration
5262: * is higher than the multiple of stepm and negative otherwise.
5263: */
5264: }
5265: else{
5266: dh[mi][i]=jk+1;
5267: bh[mi][i]=ju;
5268: }
5269: if(dh[mi][i]==0){
5270: dh[mi][i]=1; /* At least one step */
5271: bh[mi][i]=ju; /* At least one step */
5272: /* 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);*/
5273: }
5274: } /* end if mle */
1.126 brouard 5275: }
5276: } /* end wave */
5277: }
5278: jmean=sum/k;
5279: 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 5280: 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 5281: }
1.126 brouard 5282:
5283: /*********** Tricode ****************************/
1.220 brouard 5284: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5285: {
5286: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5287: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5288: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5289: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5290: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5291: */
1.130 brouard 5292:
1.242 brouard 5293: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5294: int modmaxcovj=0; /* Modality max of covariates j */
5295: int cptcode=0; /* Modality max of covariates j */
5296: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5297:
5298:
1.242 brouard 5299: /* cptcoveff=0; */
5300: /* *cptcov=0; */
1.126 brouard 5301:
1.242 brouard 5302: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5303:
1.242 brouard 5304: /* Loop on covariates without age and products and no quantitative variable */
5305: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5306: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5307: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5308: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5309: switch(Fixed[k]) {
5310: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5311: 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*/
5312: ij=(int)(covar[Tvar[k]][i]);
5313: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5314: * If product of Vn*Vm, still boolean *:
5315: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5316: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5317: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5318: modality of the nth covariate of individual i. */
5319: if (ij > modmaxcovj)
5320: modmaxcovj=ij;
5321: else if (ij < modmincovj)
5322: modmincovj=ij;
5323: if ((ij < -1) && (ij > NCOVMAX)){
5324: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5325: exit(1);
5326: }else
5327: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5328: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5329: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5330: /* getting the maximum value of the modality of the covariate
5331: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5332: female ies 1, then modmaxcovj=1.
5333: */
5334: } /* end for loop on individuals i */
5335: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5336: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5337: cptcode=modmaxcovj;
5338: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5339: /*for (i=0; i<=cptcode; i++) {*/
5340: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5341: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5342: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5343: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5344: if( j != -1){
5345: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5346: covariate for which somebody answered excluding
5347: undefined. Usually 2: 0 and 1. */
5348: }
5349: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5350: covariate for which somebody answered including
5351: undefined. Usually 3: -1, 0 and 1. */
5352: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5353: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5354: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5355:
1.242 brouard 5356: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5357: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5358: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5359: /* modmincovj=3; modmaxcovj = 7; */
5360: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5361: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5362: /* defining two dummy variables: variables V1_1 and V1_2.*/
5363: /* nbcode[Tvar[j]][ij]=k; */
5364: /* nbcode[Tvar[j]][1]=0; */
5365: /* nbcode[Tvar[j]][2]=1; */
5366: /* nbcode[Tvar[j]][3]=2; */
5367: /* To be continued (not working yet). */
5368: ij=0; /* ij is similar to i but can jump over null modalities */
5369: 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*/
5370: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5371: break;
5372: }
5373: ij++;
5374: 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*/
5375: cptcode = ij; /* New max modality for covar j */
5376: } /* end of loop on modality i=-1 to 1 or more */
5377: break;
5378: case 1: /* Testing on varying covariate, could be simple and
5379: * should look at waves or product of fixed *
5380: * varying. No time to test -1, assuming 0 and 1 only */
5381: ij=0;
5382: for(i=0; i<=1;i++){
5383: nbcode[Tvar[k]][++ij]=i;
5384: }
5385: break;
5386: default:
5387: break;
5388: } /* end switch */
5389: } /* end dummy test */
5390:
5391: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5392: /* /\*recode from 0 *\/ */
5393: /* k is a modality. If we have model=V1+V1*sex */
5394: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5395: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5396: /* } */
5397: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5398: /* if (ij > ncodemax[j]) { */
5399: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5400: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5401: /* break; */
5402: /* } */
5403: /* } /\* end of loop on modality k *\/ */
5404: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5405:
5406: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5407: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5408: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5409: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5410: 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 */
5411: 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 */
5412: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5413: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5414:
5415: ij=0;
5416: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5417: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5418: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5419: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5420: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5421: /* If product not in single variable we don't print results */
5422: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5423: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5424: 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*/
5425: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5426: 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 */
5427: if(Fixed[k]!=0)
5428: anyvaryingduminmodel=1;
5429: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5430: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5431: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5432: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5433: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5434: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5435: }
5436: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5437: /* ij--; */
5438: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5439: *cptcov=ij; /*Number of total real effective covariates: effective
5440: * because they can be excluded from the model and real
5441: * if in the model but excluded because missing values, but how to get k from ij?*/
5442: for(j=ij+1; j<= cptcovt; j++){
5443: Tvaraff[j]=0;
5444: Tmodelind[j]=0;
5445: }
5446: for(j=ntveff+1; j<= cptcovt; j++){
5447: TmodelInvind[j]=0;
5448: }
5449: /* To be sorted */
5450: ;
5451: }
1.126 brouard 5452:
1.145 brouard 5453:
1.126 brouard 5454: /*********** Health Expectancies ****************/
5455:
1.235 brouard 5456: 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 5457:
5458: {
5459: /* Health expectancies, no variances */
1.164 brouard 5460: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5461: int nhstepma, nstepma; /* Decreasing with age */
5462: double age, agelim, hf;
5463: double ***p3mat;
5464: double eip;
5465:
1.238 brouard 5466: /* pstamp(ficreseij); */
1.126 brouard 5467: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5468: fprintf(ficreseij,"# Age");
5469: for(i=1; i<=nlstate;i++){
5470: for(j=1; j<=nlstate;j++){
5471: fprintf(ficreseij," e%1d%1d ",i,j);
5472: }
5473: fprintf(ficreseij," e%1d. ",i);
5474: }
5475: fprintf(ficreseij,"\n");
5476:
5477:
5478: if(estepm < stepm){
5479: printf ("Problem %d lower than %d\n",estepm, stepm);
5480: }
5481: else hstepm=estepm;
5482: /* We compute the life expectancy from trapezoids spaced every estepm months
5483: * This is mainly to measure the difference between two models: for example
5484: * if stepm=24 months pijx are given only every 2 years and by summing them
5485: * we are calculating an estimate of the Life Expectancy assuming a linear
5486: * progression in between and thus overestimating or underestimating according
5487: * to the curvature of the survival function. If, for the same date, we
5488: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5489: * to compare the new estimate of Life expectancy with the same linear
5490: * hypothesis. A more precise result, taking into account a more precise
5491: * curvature will be obtained if estepm is as small as stepm. */
5492:
5493: /* For example we decided to compute the life expectancy with the smallest unit */
5494: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5495: nhstepm is the number of hstepm from age to agelim
5496: nstepm is the number of stepm from age to agelin.
1.270 brouard 5497: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5498: and note for a fixed period like estepm months */
5499: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5500: survival function given by stepm (the optimization length). Unfortunately it
5501: means that if the survival funtion is printed only each two years of age and if
5502: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5503: results. So we changed our mind and took the option of the best precision.
5504: */
5505: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5506:
5507: agelim=AGESUP;
5508: /* If stepm=6 months */
5509: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5510: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5511:
5512: /* nhstepm age range expressed in number of stepm */
5513: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5514: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5515: /* if (stepm >= YEARM) hstepm=1;*/
5516: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5517: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5518:
5519: for (age=bage; age<=fage; age ++){
5520: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5521: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5522: /* if (stepm >= YEARM) hstepm=1;*/
5523: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5524:
5525: /* If stepm=6 months */
5526: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5527: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5528:
1.235 brouard 5529: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5530:
5531: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5532:
5533: printf("%d|",(int)age);fflush(stdout);
5534: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5535:
5536: /* Computing expectancies */
5537: for(i=1; i<=nlstate;i++)
5538: for(j=1; j<=nlstate;j++)
5539: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5540: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5541:
5542: /* 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]);*/
5543:
5544: }
5545:
5546: fprintf(ficreseij,"%3.0f",age );
5547: for(i=1; i<=nlstate;i++){
5548: eip=0;
5549: for(j=1; j<=nlstate;j++){
5550: eip +=eij[i][j][(int)age];
5551: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5552: }
5553: fprintf(ficreseij,"%9.4f", eip );
5554: }
5555: fprintf(ficreseij,"\n");
5556:
5557: }
5558: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5559: printf("\n");
5560: fprintf(ficlog,"\n");
5561:
5562: }
5563:
1.235 brouard 5564: 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 5565:
5566: {
5567: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5568: to initial status i, ei. .
1.126 brouard 5569: */
5570: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5571: int nhstepma, nstepma; /* Decreasing with age */
5572: double age, agelim, hf;
5573: double ***p3matp, ***p3matm, ***varhe;
5574: double **dnewm,**doldm;
5575: double *xp, *xm;
5576: double **gp, **gm;
5577: double ***gradg, ***trgradg;
5578: int theta;
5579:
5580: double eip, vip;
5581:
5582: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5583: xp=vector(1,npar);
5584: xm=vector(1,npar);
5585: dnewm=matrix(1,nlstate*nlstate,1,npar);
5586: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5587:
5588: pstamp(ficresstdeij);
5589: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5590: fprintf(ficresstdeij,"# Age");
5591: for(i=1; i<=nlstate;i++){
5592: for(j=1; j<=nlstate;j++)
5593: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5594: fprintf(ficresstdeij," e%1d. ",i);
5595: }
5596: fprintf(ficresstdeij,"\n");
5597:
5598: pstamp(ficrescveij);
5599: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5600: fprintf(ficrescveij,"# Age");
5601: for(i=1; i<=nlstate;i++)
5602: for(j=1; j<=nlstate;j++){
5603: cptj= (j-1)*nlstate+i;
5604: for(i2=1; i2<=nlstate;i2++)
5605: for(j2=1; j2<=nlstate;j2++){
5606: cptj2= (j2-1)*nlstate+i2;
5607: if(cptj2 <= cptj)
5608: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5609: }
5610: }
5611: fprintf(ficrescveij,"\n");
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.
5632: Look at hpijx to understand the reason of that which relies in memory size
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: /* If stepm=6 months */
5643: /* nhstepm age range expressed in number of stepm */
5644: agelim=AGESUP;
5645: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5646: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5647: /* if (stepm >= YEARM) hstepm=1;*/
5648: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5649:
5650: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5651: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5652: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5653: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5654: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5655: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5656:
5657: for (age=bage; age<=fage; age ++){
5658: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5659: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5660: /* if (stepm >= YEARM) hstepm=1;*/
5661: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5662:
1.126 brouard 5663: /* If stepm=6 months */
5664: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5665: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5666:
5667: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5668:
1.126 brouard 5669: /* Computing Variances of health expectancies */
5670: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5671: decrease memory allocation */
5672: for(theta=1; theta <=npar; theta++){
5673: for(i=1; i<=npar; i++){
1.222 brouard 5674: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5675: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5676: }
1.235 brouard 5677: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5678: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5679:
1.126 brouard 5680: for(j=1; j<= nlstate; j++){
1.222 brouard 5681: for(i=1; i<=nlstate; i++){
5682: for(h=0; h<=nhstepm-1; h++){
5683: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5684: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5685: }
5686: }
1.126 brouard 5687: }
1.218 brouard 5688:
1.126 brouard 5689: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5690: for(h=0; h<=nhstepm-1; h++){
5691: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5692: }
1.126 brouard 5693: }/* End theta */
5694:
5695:
5696: for(h=0; h<=nhstepm-1; h++)
5697: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5698: for(theta=1; theta <=npar; theta++)
5699: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5700:
1.218 brouard 5701:
1.222 brouard 5702: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5703: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5704: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5705:
1.222 brouard 5706: printf("%d|",(int)age);fflush(stdout);
5707: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5708: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5709: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5710: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5711: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5712: for(ij=1;ij<=nlstate*nlstate;ij++)
5713: for(ji=1;ji<=nlstate*nlstate;ji++)
5714: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5715: }
5716: }
1.218 brouard 5717:
1.126 brouard 5718: /* Computing expectancies */
1.235 brouard 5719: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5720: for(i=1; i<=nlstate;i++)
5721: for(j=1; j<=nlstate;j++)
1.222 brouard 5722: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5723: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5724:
1.222 brouard 5725: /* 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 5726:
1.222 brouard 5727: }
1.269 brouard 5728:
5729: /* Standard deviation of expectancies ij */
1.126 brouard 5730: fprintf(ficresstdeij,"%3.0f",age );
5731: for(i=1; i<=nlstate;i++){
5732: eip=0.;
5733: vip=0.;
5734: for(j=1; j<=nlstate;j++){
1.222 brouard 5735: eip += eij[i][j][(int)age];
5736: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5737: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5738: 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 5739: }
5740: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5741: }
5742: fprintf(ficresstdeij,"\n");
1.218 brouard 5743:
1.269 brouard 5744: /* Variance of expectancies ij */
1.126 brouard 5745: fprintf(ficrescveij,"%3.0f",age );
5746: for(i=1; i<=nlstate;i++)
5747: for(j=1; j<=nlstate;j++){
1.222 brouard 5748: cptj= (j-1)*nlstate+i;
5749: for(i2=1; i2<=nlstate;i2++)
5750: for(j2=1; j2<=nlstate;j2++){
5751: cptj2= (j2-1)*nlstate+i2;
5752: if(cptj2 <= cptj)
5753: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5754: }
1.126 brouard 5755: }
5756: fprintf(ficrescveij,"\n");
1.218 brouard 5757:
1.126 brouard 5758: }
5759: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5760: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5761: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5762: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5763: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5764: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5765: printf("\n");
5766: fprintf(ficlog,"\n");
1.218 brouard 5767:
1.126 brouard 5768: free_vector(xm,1,npar);
5769: free_vector(xp,1,npar);
5770: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5771: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5772: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5773: }
1.218 brouard 5774:
1.126 brouard 5775: /************ Variance ******************/
1.235 brouard 5776: 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 5777: {
1.279 ! brouard 5778: /** Variance of health expectancies
! 5779: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
! 5780: * double **newm;
! 5781: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
! 5782: */
1.218 brouard 5783:
5784: /* int movingaverage(); */
5785: double **dnewm,**doldm;
5786: double **dnewmp,**doldmp;
5787: int i, j, nhstepm, hstepm, h, nstepm ;
5788: int k;
5789: double *xp;
1.279 ! brouard 5790: double **gp, **gm; /**< for var eij */
! 5791: double ***gradg, ***trgradg; /**< for var eij */
! 5792: double **gradgp, **trgradgp; /**< for var p point j */
! 5793: double *gpp, *gmp; /**< for var p point j */
! 5794: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5795: double ***p3mat;
5796: double age,agelim, hf;
5797: /* double ***mobaverage; */
5798: int theta;
5799: char digit[4];
5800: char digitp[25];
5801:
5802: char fileresprobmorprev[FILENAMELENGTH];
5803:
5804: if(popbased==1){
5805: if(mobilav!=0)
5806: strcpy(digitp,"-POPULBASED-MOBILAV_");
5807: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5808: }
5809: else
5810: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5811:
1.218 brouard 5812: /* if (mobilav!=0) { */
5813: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5814: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5815: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5816: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5817: /* } */
5818: /* } */
5819:
5820: strcpy(fileresprobmorprev,"PRMORPREV-");
5821: sprintf(digit,"%-d",ij);
5822: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5823: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5824: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5825: strcat(fileresprobmorprev,fileresu);
5826: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5827: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5828: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5829: }
5830: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5831: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5832: pstamp(ficresprobmorprev);
5833: 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 5834: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5835: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5836: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5837: }
5838: for(j=1;j<=cptcoveff;j++)
5839: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5840: fprintf(ficresprobmorprev,"\n");
5841:
1.218 brouard 5842: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5843: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5844: fprintf(ficresprobmorprev," p.%-d SE",j);
5845: for(i=1; i<=nlstate;i++)
5846: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5847: }
5848: fprintf(ficresprobmorprev,"\n");
5849:
5850: fprintf(ficgp,"\n# Routine varevsij");
5851: fprintf(ficgp,"\nunset title \n");
5852: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5853: 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");
5854: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 ! brouard 5855:
1.218 brouard 5856: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5857: pstamp(ficresvij);
5858: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5859: if(popbased==1)
5860: 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);
5861: else
5862: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5863: fprintf(ficresvij,"# Age");
5864: for(i=1; i<=nlstate;i++)
5865: for(j=1; j<=nlstate;j++)
5866: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5867: fprintf(ficresvij,"\n");
5868:
5869: xp=vector(1,npar);
5870: dnewm=matrix(1,nlstate,1,npar);
5871: doldm=matrix(1,nlstate,1,nlstate);
5872: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5873: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5874:
5875: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5876: gpp=vector(nlstate+1,nlstate+ndeath);
5877: gmp=vector(nlstate+1,nlstate+ndeath);
5878: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5879:
1.218 brouard 5880: if(estepm < stepm){
5881: printf ("Problem %d lower than %d\n",estepm, stepm);
5882: }
5883: else hstepm=estepm;
5884: /* For example we decided to compute the life expectancy with the smallest unit */
5885: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5886: nhstepm is the number of hstepm from age to agelim
5887: nstepm is the number of stepm from age to agelim.
5888: Look at function hpijx to understand why because of memory size limitations,
5889: we decided (b) to get a life expectancy respecting the most precise curvature of the
5890: survival function given by stepm (the optimization length). Unfortunately it
5891: means that if the survival funtion is printed every two years of age and if
5892: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5893: results. So we changed our mind and took the option of the best precision.
5894: */
5895: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5896: agelim = AGESUP;
5897: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5898: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5899: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5900: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5901: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5902: gp=matrix(0,nhstepm,1,nlstate);
5903: gm=matrix(0,nhstepm,1,nlstate);
5904:
5905:
5906: for(theta=1; theta <=npar; theta++){
5907: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5908: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5909: }
1.279 ! brouard 5910: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
! 5911: * returns into prlim .
! 5912: */
1.242 brouard 5913: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 ! brouard 5914:
! 5915: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 5916: if (popbased==1) {
5917: if(mobilav ==0){
5918: for(i=1; i<=nlstate;i++)
5919: prlim[i][i]=probs[(int)age][i][ij];
5920: }else{ /* mobilav */
5921: for(i=1; i<=nlstate;i++)
5922: prlim[i][i]=mobaverage[(int)age][i][ij];
5923: }
5924: }
1.279 ! brouard 5925: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}_x\f$ at horizon h.
! 5926: */
! 5927: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=0 to nhstepm */
! 5928: /**< And for each alive state j, sums over i \f$ w^i_x {}{h}_p^{ij}_x\f$, which are the probability
! 5929: * at horizon h in state j including mortality.
! 5930: */
1.218 brouard 5931: for(j=1; j<= nlstate; j++){
5932: for(h=0; h<=nhstepm; h++){
5933: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5934: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5935: }
5936: }
1.279 ! brouard 5937: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 5938: computed over hstepm matrices product = hstepm*stepm months)
1.279 ! brouard 5939: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 5940: */
5941: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5942: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5943: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 ! brouard 5944: }
! 5945:
! 5946: /* Again with minus shift */
1.218 brouard 5947:
5948: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5949: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5950:
1.242 brouard 5951: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5952:
5953: if (popbased==1) {
5954: if(mobilav ==0){
5955: for(i=1; i<=nlstate;i++)
5956: prlim[i][i]=probs[(int)age][i][ij];
5957: }else{ /* mobilav */
5958: for(i=1; i<=nlstate;i++)
5959: prlim[i][i]=mobaverage[(int)age][i][ij];
5960: }
5961: }
5962:
1.235 brouard 5963: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5964:
5965: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5966: for(h=0; h<=nhstepm; h++){
5967: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5968: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5969: }
5970: }
5971: /* This for computing probability of death (h=1 means
5972: computed over hstepm matrices product = hstepm*stepm months)
5973: as a weighted average of prlim.
5974: */
5975: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5976: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5977: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5978: }
1.279 ! brouard 5979: /* end shifting computations */
! 5980:
! 5981: /**< Computing gradient matrix at horizon h
! 5982: */
1.218 brouard 5983: for(j=1; j<= nlstate; j++) /* vareij */
5984: for(h=0; h<=nhstepm; h++){
5985: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5986: }
1.279 ! brouard 5987: /**< Gradient of overall mortality p.3 (or p.j)
! 5988: */
! 5989: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 5990: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5991: }
5992:
5993: } /* End theta */
1.279 ! brouard 5994:
! 5995: /* We got the gradient matrix for each theta and state j */
1.218 brouard 5996: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5997:
5998: for(h=0; h<=nhstepm; h++) /* veij */
5999: for(j=1; j<=nlstate;j++)
6000: for(theta=1; theta <=npar; theta++)
6001: trgradg[h][j][theta]=gradg[h][theta][j];
6002:
6003: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6004: for(theta=1; theta <=npar; theta++)
6005: trgradgp[j][theta]=gradgp[theta][j];
1.279 ! brouard 6006: /**< as well as its transposed matrix
! 6007: */
1.218 brouard 6008:
6009: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6010: for(i=1;i<=nlstate;i++)
6011: for(j=1;j<=nlstate;j++)
6012: vareij[i][j][(int)age] =0.;
1.279 ! brouard 6013:
! 6014: /* Computing trgradg by matcov by gradg at age and summing over h
! 6015: * and k (nhstepm) formula 15 of article
! 6016: * Lievre-Brouard-Heathcote
! 6017: */
! 6018:
1.218 brouard 6019: for(h=0;h<=nhstepm;h++){
6020: for(k=0;k<=nhstepm;k++){
6021: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6022: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6023: for(i=1;i<=nlstate;i++)
6024: for(j=1;j<=nlstate;j++)
6025: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6026: }
6027: }
6028:
1.279 ! brouard 6029: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
! 6030: * p.j overall mortality formula 49 but computed directly because
! 6031: * we compute the grad (wix pijx) instead of grad (pijx),even if
! 6032: * wix is independent of theta.
! 6033: */
1.218 brouard 6034: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6035: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6036: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6037: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6038: varppt[j][i]=doldmp[j][i];
6039: /* end ppptj */
6040: /* x centered again */
6041:
1.242 brouard 6042: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6043:
6044: if (popbased==1) {
6045: if(mobilav ==0){
6046: for(i=1; i<=nlstate;i++)
6047: prlim[i][i]=probs[(int)age][i][ij];
6048: }else{ /* mobilav */
6049: for(i=1; i<=nlstate;i++)
6050: prlim[i][i]=mobaverage[(int)age][i][ij];
6051: }
6052: }
6053:
6054: /* This for computing probability of death (h=1 means
6055: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6056: as a weighted average of prlim.
6057: */
1.235 brouard 6058: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6059: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6060: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6061: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6062: }
6063: /* end probability of death */
6064:
6065: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6066: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6067: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6068: for(i=1; i<=nlstate;i++){
6069: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6070: }
6071: }
6072: fprintf(ficresprobmorprev,"\n");
6073:
6074: fprintf(ficresvij,"%.0f ",age );
6075: for(i=1; i<=nlstate;i++)
6076: for(j=1; j<=nlstate;j++){
6077: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6078: }
6079: fprintf(ficresvij,"\n");
6080: free_matrix(gp,0,nhstepm,1,nlstate);
6081: free_matrix(gm,0,nhstepm,1,nlstate);
6082: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6083: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6084: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6085: } /* End age */
6086: free_vector(gpp,nlstate+1,nlstate+ndeath);
6087: free_vector(gmp,nlstate+1,nlstate+ndeath);
6088: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6089: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6090: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6091: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6092: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6093: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6094: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6095: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6096: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6097: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6098: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6099: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6100: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6101: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6102: 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);
6103: /* 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 6104: */
1.218 brouard 6105: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6106: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6107:
1.218 brouard 6108: free_vector(xp,1,npar);
6109: free_matrix(doldm,1,nlstate,1,nlstate);
6110: free_matrix(dnewm,1,nlstate,1,npar);
6111: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6112: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6113: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6114: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6115: fclose(ficresprobmorprev);
6116: fflush(ficgp);
6117: fflush(fichtm);
6118: } /* end varevsij */
1.126 brouard 6119:
6120: /************ Variance of prevlim ******************/
1.269 brouard 6121: 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 6122: {
1.205 brouard 6123: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6124: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6125:
1.268 brouard 6126: double **dnewmpar,**doldm;
1.126 brouard 6127: int i, j, nhstepm, hstepm;
6128: double *xp;
6129: double *gp, *gm;
6130: double **gradg, **trgradg;
1.208 brouard 6131: double **mgm, **mgp;
1.126 brouard 6132: double age,agelim;
6133: int theta;
6134:
6135: pstamp(ficresvpl);
6136: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6137: fprintf(ficresvpl,"# Age ");
6138: if(nresult >=1)
6139: fprintf(ficresvpl," Result# ");
1.126 brouard 6140: for(i=1; i<=nlstate;i++)
6141: fprintf(ficresvpl," %1d-%1d",i,i);
6142: fprintf(ficresvpl,"\n");
6143:
6144: xp=vector(1,npar);
1.268 brouard 6145: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6146: doldm=matrix(1,nlstate,1,nlstate);
6147:
6148: hstepm=1*YEARM; /* Every year of age */
6149: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6150: agelim = AGESUP;
6151: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6152: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6153: if (stepm >= YEARM) hstepm=1;
6154: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6155: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6156: mgp=matrix(1,npar,1,nlstate);
6157: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6158: gp=vector(1,nlstate);
6159: gm=vector(1,nlstate);
6160:
6161: for(theta=1; theta <=npar; theta++){
6162: for(i=1; i<=npar; i++){ /* Computes gradient */
6163: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6164: }
1.209 brouard 6165: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6166: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6167: else
1.235 brouard 6168: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6169: for(i=1;i<=nlstate;i++){
1.126 brouard 6170: gp[i] = prlim[i][i];
1.208 brouard 6171: mgp[theta][i] = prlim[i][i];
6172: }
1.126 brouard 6173: for(i=1; i<=npar; i++) /* Computes gradient */
6174: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6175: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6176: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6177: else
1.235 brouard 6178: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6179: for(i=1;i<=nlstate;i++){
1.126 brouard 6180: gm[i] = prlim[i][i];
1.208 brouard 6181: mgm[theta][i] = prlim[i][i];
6182: }
1.126 brouard 6183: for(i=1;i<=nlstate;i++)
6184: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6185: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6186: } /* End theta */
6187:
6188: trgradg =matrix(1,nlstate,1,npar);
6189:
6190: for(j=1; j<=nlstate;j++)
6191: for(theta=1; theta <=npar; theta++)
6192: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6193: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6194: /* printf("\nmgm mgp %d ",(int)age); */
6195: /* for(j=1; j<=nlstate;j++){ */
6196: /* printf(" %d ",j); */
6197: /* for(theta=1; theta <=npar; theta++) */
6198: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6199: /* printf("\n "); */
6200: /* } */
6201: /* } */
6202: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6203: /* printf("\n gradg %d ",(int)age); */
6204: /* for(j=1; j<=nlstate;j++){ */
6205: /* printf("%d ",j); */
6206: /* for(theta=1; theta <=npar; theta++) */
6207: /* printf("%d %lf ",theta,gradg[theta][j]); */
6208: /* printf("\n "); */
6209: /* } */
6210: /* } */
1.126 brouard 6211:
6212: for(i=1;i<=nlstate;i++)
6213: varpl[i][(int)age] =0.;
1.209 brouard 6214: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6215: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6216: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6217: }else{
1.268 brouard 6218: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6219: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6220: }
1.126 brouard 6221: for(i=1;i<=nlstate;i++)
6222: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6223:
6224: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6225: if(nresult >=1)
6226: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6227: for(i=1; i<=nlstate;i++)
6228: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6229: fprintf(ficresvpl,"\n");
6230: free_vector(gp,1,nlstate);
6231: free_vector(gm,1,nlstate);
1.208 brouard 6232: free_matrix(mgm,1,npar,1,nlstate);
6233: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6234: free_matrix(gradg,1,npar,1,nlstate);
6235: free_matrix(trgradg,1,nlstate,1,npar);
6236: } /* End age */
6237:
6238: free_vector(xp,1,npar);
6239: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6240: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6241:
6242: }
6243:
6244:
6245: /************ Variance of backprevalence limit ******************/
1.269 brouard 6246: 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 6247: {
6248: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6249: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6250:
6251: double **dnewmpar,**doldm;
6252: int i, j, nhstepm, hstepm;
6253: double *xp;
6254: double *gp, *gm;
6255: double **gradg, **trgradg;
6256: double **mgm, **mgp;
6257: double age,agelim;
6258: int theta;
6259:
6260: pstamp(ficresvbl);
6261: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6262: fprintf(ficresvbl,"# Age ");
6263: if(nresult >=1)
6264: fprintf(ficresvbl," Result# ");
6265: for(i=1; i<=nlstate;i++)
6266: fprintf(ficresvbl," %1d-%1d",i,i);
6267: fprintf(ficresvbl,"\n");
6268:
6269: xp=vector(1,npar);
6270: dnewmpar=matrix(1,nlstate,1,npar);
6271: doldm=matrix(1,nlstate,1,nlstate);
6272:
6273: hstepm=1*YEARM; /* Every year of age */
6274: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6275: agelim = AGEINF;
6276: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6277: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6278: if (stepm >= YEARM) hstepm=1;
6279: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6280: gradg=matrix(1,npar,1,nlstate);
6281: mgp=matrix(1,npar,1,nlstate);
6282: mgm=matrix(1,npar,1,nlstate);
6283: gp=vector(1,nlstate);
6284: gm=vector(1,nlstate);
6285:
6286: for(theta=1; theta <=npar; theta++){
6287: for(i=1; i<=npar; i++){ /* Computes gradient */
6288: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6289: }
6290: if(mobilavproj > 0 )
6291: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6292: else
6293: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6294: for(i=1;i<=nlstate;i++){
6295: gp[i] = bprlim[i][i];
6296: mgp[theta][i] = bprlim[i][i];
6297: }
6298: for(i=1; i<=npar; i++) /* Computes gradient */
6299: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6300: if(mobilavproj > 0 )
6301: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6302: else
6303: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6304: for(i=1;i<=nlstate;i++){
6305: gm[i] = bprlim[i][i];
6306: mgm[theta][i] = bprlim[i][i];
6307: }
6308: for(i=1;i<=nlstate;i++)
6309: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6310: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6311: } /* End theta */
6312:
6313: trgradg =matrix(1,nlstate,1,npar);
6314:
6315: for(j=1; j<=nlstate;j++)
6316: for(theta=1; theta <=npar; theta++)
6317: trgradg[j][theta]=gradg[theta][j];
6318: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6319: /* printf("\nmgm mgp %d ",(int)age); */
6320: /* for(j=1; j<=nlstate;j++){ */
6321: /* printf(" %d ",j); */
6322: /* for(theta=1; theta <=npar; theta++) */
6323: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6324: /* printf("\n "); */
6325: /* } */
6326: /* } */
6327: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6328: /* printf("\n gradg %d ",(int)age); */
6329: /* for(j=1; j<=nlstate;j++){ */
6330: /* printf("%d ",j); */
6331: /* for(theta=1; theta <=npar; theta++) */
6332: /* printf("%d %lf ",theta,gradg[theta][j]); */
6333: /* printf("\n "); */
6334: /* } */
6335: /* } */
6336:
6337: for(i=1;i<=nlstate;i++)
6338: varbpl[i][(int)age] =0.;
6339: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6340: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6341: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6342: }else{
6343: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6344: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6345: }
6346: for(i=1;i<=nlstate;i++)
6347: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6348:
6349: fprintf(ficresvbl,"%.0f ",age );
6350: if(nresult >=1)
6351: fprintf(ficresvbl,"%d ",nres );
6352: for(i=1; i<=nlstate;i++)
6353: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6354: fprintf(ficresvbl,"\n");
6355: free_vector(gp,1,nlstate);
6356: free_vector(gm,1,nlstate);
6357: free_matrix(mgm,1,npar,1,nlstate);
6358: free_matrix(mgp,1,npar,1,nlstate);
6359: free_matrix(gradg,1,npar,1,nlstate);
6360: free_matrix(trgradg,1,nlstate,1,npar);
6361: } /* End age */
6362:
6363: free_vector(xp,1,npar);
6364: free_matrix(doldm,1,nlstate,1,npar);
6365: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6366:
6367: }
6368:
6369: /************ Variance of one-step probabilities ******************/
6370: 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 6371: {
6372: int i, j=0, k1, l1, tj;
6373: int k2, l2, j1, z1;
6374: int k=0, l;
6375: int first=1, first1, first2;
6376: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6377: double **dnewm,**doldm;
6378: double *xp;
6379: double *gp, *gm;
6380: double **gradg, **trgradg;
6381: double **mu;
6382: double age, cov[NCOVMAX+1];
6383: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6384: int theta;
6385: char fileresprob[FILENAMELENGTH];
6386: char fileresprobcov[FILENAMELENGTH];
6387: char fileresprobcor[FILENAMELENGTH];
6388: double ***varpij;
6389:
6390: strcpy(fileresprob,"PROB_");
6391: strcat(fileresprob,fileres);
6392: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6393: printf("Problem with resultfile: %s\n", fileresprob);
6394: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6395: }
6396: strcpy(fileresprobcov,"PROBCOV_");
6397: strcat(fileresprobcov,fileresu);
6398: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6399: printf("Problem with resultfile: %s\n", fileresprobcov);
6400: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6401: }
6402: strcpy(fileresprobcor,"PROBCOR_");
6403: strcat(fileresprobcor,fileresu);
6404: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6405: printf("Problem with resultfile: %s\n", fileresprobcor);
6406: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6407: }
6408: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6409: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6410: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6411: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6412: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6413: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6414: pstamp(ficresprob);
6415: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6416: fprintf(ficresprob,"# Age");
6417: pstamp(ficresprobcov);
6418: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6419: fprintf(ficresprobcov,"# Age");
6420: pstamp(ficresprobcor);
6421: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6422: fprintf(ficresprobcor,"# Age");
1.126 brouard 6423:
6424:
1.222 brouard 6425: for(i=1; i<=nlstate;i++)
6426: for(j=1; j<=(nlstate+ndeath);j++){
6427: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6428: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6429: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6430: }
6431: /* fprintf(ficresprob,"\n");
6432: fprintf(ficresprobcov,"\n");
6433: fprintf(ficresprobcor,"\n");
6434: */
6435: xp=vector(1,npar);
6436: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6437: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6438: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6439: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6440: first=1;
6441: fprintf(ficgp,"\n# Routine varprob");
6442: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6443: fprintf(fichtm,"\n");
6444:
1.266 brouard 6445: fprintf(fichtm,"\n<li><h4> <a href=\"%s\">Matrix of variance-covariance of one-step probabilities (drawings)</a></h4> this page is important in order to visualize confidence intervals and especially correlation between disability and recovery, or more generally, way in and way back. %s</li>\n",optionfilehtmcov,optionfilehtmcov);
1.222 brouard 6446: 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);
6447: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6448: and drawn. It helps understanding how is the covariance between two incidences.\
6449: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6450: 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 6451: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6452: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6453: standard deviations wide on each axis. <br>\
6454: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6455: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6456: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6457:
1.222 brouard 6458: cov[1]=1;
6459: /* tj=cptcoveff; */
1.225 brouard 6460: tj = (int) pow(2,cptcoveff);
1.222 brouard 6461: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6462: j1=0;
1.224 brouard 6463: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6464: if (cptcovn>0) {
6465: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6466: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6467: fprintf(ficresprob, "**********\n#\n");
6468: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6469: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6470: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6471:
1.222 brouard 6472: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6473: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6474: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6475:
6476:
1.222 brouard 6477: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6478: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6479: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6480:
1.222 brouard 6481: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6482: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6483: fprintf(ficresprobcor, "**********\n#");
6484: if(invalidvarcomb[j1]){
6485: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6486: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6487: continue;
6488: }
6489: }
6490: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6491: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6492: gp=vector(1,(nlstate)*(nlstate+ndeath));
6493: gm=vector(1,(nlstate)*(nlstate+ndeath));
6494: for (age=bage; age<=fage; age ++){
6495: cov[2]=age;
6496: if(nagesqr==1)
6497: cov[3]= age*age;
6498: for (k=1; k<=cptcovn;k++) {
6499: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6500: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6501: * 1 1 1 1 1
6502: * 2 2 1 1 1
6503: * 3 1 2 1 1
6504: */
6505: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6506: }
6507: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6508: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6509: for (k=1; k<=cptcovprod;k++)
6510: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6511:
6512:
1.222 brouard 6513: for(theta=1; theta <=npar; theta++){
6514: for(i=1; i<=npar; i++)
6515: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6516:
1.222 brouard 6517: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6518:
1.222 brouard 6519: k=0;
6520: for(i=1; i<= (nlstate); i++){
6521: for(j=1; j<=(nlstate+ndeath);j++){
6522: k=k+1;
6523: gp[k]=pmmij[i][j];
6524: }
6525: }
1.220 brouard 6526:
1.222 brouard 6527: for(i=1; i<=npar; i++)
6528: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6529:
1.222 brouard 6530: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6531: k=0;
6532: for(i=1; i<=(nlstate); i++){
6533: for(j=1; j<=(nlstate+ndeath);j++){
6534: k=k+1;
6535: gm[k]=pmmij[i][j];
6536: }
6537: }
1.220 brouard 6538:
1.222 brouard 6539: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6540: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6541: }
1.126 brouard 6542:
1.222 brouard 6543: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6544: for(theta=1; theta <=npar; theta++)
6545: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6546:
1.222 brouard 6547: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6548: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6549:
1.222 brouard 6550: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6551:
1.222 brouard 6552: k=0;
6553: for(i=1; i<=(nlstate); i++){
6554: for(j=1; j<=(nlstate+ndeath);j++){
6555: k=k+1;
6556: mu[k][(int) age]=pmmij[i][j];
6557: }
6558: }
6559: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6560: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6561: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6562:
1.222 brouard 6563: /*printf("\n%d ",(int)age);
6564: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6565: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6566: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6567: }*/
1.220 brouard 6568:
1.222 brouard 6569: fprintf(ficresprob,"\n%d ",(int)age);
6570: fprintf(ficresprobcov,"\n%d ",(int)age);
6571: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6572:
1.222 brouard 6573: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6574: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6575: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6576: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6577: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6578: }
6579: i=0;
6580: for (k=1; k<=(nlstate);k++){
6581: for (l=1; l<=(nlstate+ndeath);l++){
6582: i++;
6583: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6584: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6585: for (j=1; j<=i;j++){
6586: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6587: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6588: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6589: }
6590: }
6591: }/* end of loop for state */
6592: } /* end of loop for age */
6593: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6594: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6595: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6596: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6597:
6598: /* Confidence intervalle of pij */
6599: /*
6600: fprintf(ficgp,"\nunset parametric;unset label");
6601: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6602: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6603: 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);
6604: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6605: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6606: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6607: */
6608:
6609: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6610: first1=1;first2=2;
6611: for (k2=1; k2<=(nlstate);k2++){
6612: for (l2=1; l2<=(nlstate+ndeath);l2++){
6613: if(l2==k2) continue;
6614: j=(k2-1)*(nlstate+ndeath)+l2;
6615: for (k1=1; k1<=(nlstate);k1++){
6616: for (l1=1; l1<=(nlstate+ndeath);l1++){
6617: if(l1==k1) continue;
6618: i=(k1-1)*(nlstate+ndeath)+l1;
6619: if(i<=j) continue;
6620: for (age=bage; age<=fage; age ++){
6621: if ((int)age %5==0){
6622: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6623: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6624: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6625: mu1=mu[i][(int) age]/stepm*YEARM ;
6626: mu2=mu[j][(int) age]/stepm*YEARM;
6627: c12=cv12/sqrt(v1*v2);
6628: /* Computing eigen value of matrix of covariance */
6629: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6630: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6631: if ((lc2 <0) || (lc1 <0) ){
6632: if(first2==1){
6633: first1=0;
6634: 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);
6635: }
6636: 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);
6637: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6638: /* lc2=fabs(lc2); */
6639: }
1.220 brouard 6640:
1.222 brouard 6641: /* Eigen vectors */
6642: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6643: /*v21=sqrt(1.-v11*v11); *//* error */
6644: v21=(lc1-v1)/cv12*v11;
6645: v12=-v21;
6646: v22=v11;
6647: tnalp=v21/v11;
6648: if(first1==1){
6649: first1=0;
6650: 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);
6651: }
6652: 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);
6653: /*printf(fignu*/
6654: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6655: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6656: if(first==1){
6657: first=0;
6658: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6659: fprintf(ficgp,"\nset parametric;unset label");
6660: 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);
6661: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6662: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6663: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6664: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6665: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6666: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6667: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6668: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6669: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6670: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6671: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6672: 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", \
1.266 brouard 6673: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6674: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6675: }else{
6676: first=0;
6677: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6678: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6679: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6680: 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 6681: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6682: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6683: }/* if first */
6684: } /* age mod 5 */
6685: } /* end loop age */
6686: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6687: first=1;
6688: } /*l12 */
6689: } /* k12 */
6690: } /*l1 */
6691: }/* k1 */
6692: } /* loop on combination of covariates j1 */
6693: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6694: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6695: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6696: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6697: free_vector(xp,1,npar);
6698: fclose(ficresprob);
6699: fclose(ficresprobcov);
6700: fclose(ficresprobcor);
6701: fflush(ficgp);
6702: fflush(fichtmcov);
6703: }
1.126 brouard 6704:
6705:
6706: /******************* Printing html file ***********/
1.201 brouard 6707: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6708: int lastpass, int stepm, int weightopt, char model[],\
6709: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6710: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6711: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6712: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6713: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6714:
6715: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6716: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6717: </ul>");
1.237 brouard 6718: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6719: </ul>", model);
1.214 brouard 6720: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6721: 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",
6722: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6723: 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 6724: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6725: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6726: fprintf(fichtm,"\
6727: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6728: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6729: fprintf(fichtm,"\
1.217 brouard 6730: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6731: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6732: fprintf(fichtm,"\
1.126 brouard 6733: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6734: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6735: fprintf(fichtm,"\
1.217 brouard 6736: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6737: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6738: fprintf(fichtm,"\
1.211 brouard 6739: - (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 6740: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6741: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6742: if(prevfcast==1){
6743: fprintf(fichtm,"\
6744: - Prevalence projections by age and states: \
1.201 brouard 6745: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6746: }
1.126 brouard 6747:
6748:
1.225 brouard 6749: m=pow(2,cptcoveff);
1.222 brouard 6750: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6751:
1.264 brouard 6752: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6753:
6754: jj1=0;
6755:
6756: fprintf(fichtm," \n<ul>");
6757: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6758: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6759: if(m != 1 && TKresult[nres]!= k1)
6760: continue;
6761: jj1++;
6762: if (cptcovn > 0) {
6763: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6764: for (cpt=1; cpt<=cptcoveff;cpt++){
6765: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6766: }
6767: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6768: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6769: }
6770: fprintf(fichtm,"\">");
6771:
6772: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6773: fprintf(fichtm,"************ Results for covariates");
6774: for (cpt=1; cpt<=cptcoveff;cpt++){
6775: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6776: }
6777: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6778: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6779: }
6780: if(invalidvarcomb[k1]){
6781: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6782: continue;
6783: }
6784: fprintf(fichtm,"</a></li>");
6785: } /* cptcovn >0 */
6786: }
6787: fprintf(fichtm," \n</ul>");
6788:
1.222 brouard 6789: jj1=0;
1.237 brouard 6790:
6791: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6792: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6793: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6794: continue;
1.220 brouard 6795:
1.222 brouard 6796: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6797: jj1++;
6798: if (cptcovn > 0) {
1.264 brouard 6799: fprintf(fichtm,"\n<p><a name=\"rescov");
6800: for (cpt=1; cpt<=cptcoveff;cpt++){
6801: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6802: }
6803: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6804: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6805: }
6806: fprintf(fichtm,"\"</a>");
6807:
1.222 brouard 6808: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6809: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6810: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6811: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6812: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6813: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6814: }
1.237 brouard 6815: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6816: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6817: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6818: }
6819:
1.230 brouard 6820: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6821: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6822: if(invalidvarcomb[k1]){
6823: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6824: printf("\nCombination (%d) ignored because no cases \n",k1);
6825: continue;
6826: }
6827: }
6828: /* aij, bij */
1.259 brouard 6829: 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 6830: <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 6831: /* Pij */
1.241 brouard 6832: 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> \
6833: <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 6834: /* Quasi-incidences */
6835: 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 6836: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6837: 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 6838: 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> \
6839: <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 6840: /* Survival functions (period) in state j */
6841: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6842: fprintf(fichtm,"<br>\n- Survival functions in state %d. Or probability to survive in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6843: <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 6844: }
6845: /* State specific survival functions (period) */
6846: for(cpt=1; cpt<=nlstate;cpt++){
6847: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6848: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6849: <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 6850: }
6851: /* Period (stable) prevalence in each health state */
6852: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6853: 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> \
6854: <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 6855: }
6856: if(backcast==1){
6857: /* Period (stable) back prevalence in each health state */
6858: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6859: 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 6860: <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 6861: }
1.217 brouard 6862: }
1.222 brouard 6863: if(prevfcast==1){
6864: /* Projection of prevalence up to period (stable) prevalence in each health state */
6865: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6866: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), from year %.1f up to year %.1f tending to period (stable) prevalence in state %d. Or probability to be in state %d being in an observed weighted state (from 1 to %d). <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6867: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateproj1, dateproj2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6868: }
6869: }
1.268 brouard 6870: if(backcast==1){
6871: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6872: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6873: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6874: 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 \
6875: 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) \
6876: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6877: <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 6878: }
6879: }
1.220 brouard 6880:
1.222 brouard 6881: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6882: 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> \
6883: <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 6884: }
6885: /* } /\* end i1 *\/ */
6886: }/* End k1 */
6887: fprintf(fichtm,"</ul>");
1.126 brouard 6888:
1.222 brouard 6889: fprintf(fichtm,"\
1.126 brouard 6890: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6891: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6892: - 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 6893: But because parameters are usually highly correlated (a higher incidence of disability \
6894: and a higher incidence of recovery can give very close observed transition) it might \
6895: be very useful to look not only at linear confidence intervals estimated from the \
6896: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6897: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6898: covariance matrix of the one-step probabilities. \
6899: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6900:
1.222 brouard 6901: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6902: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6903: fprintf(fichtm,"\
1.126 brouard 6904: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6905: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6906:
1.222 brouard 6907: fprintf(fichtm,"\
1.126 brouard 6908: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6909: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6910: fprintf(fichtm,"\
1.126 brouard 6911: - 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): \
6912: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6913: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6914: fprintf(fichtm,"\
1.126 brouard 6915: - (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): \
6916: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6917: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6918: fprintf(fichtm,"\
1.128 brouard 6919: - Variances and covariances of health expectancies by age. Status (i) based health expectancies (in state j), e<sup>ij</sup> are weighted by the period prevalences in each state i (if popbased=1, an additional computation is done using the cross-sectional prevalences, i.e population based) (estepm=%d months): <a href=\"%s\">%s</a><br>\n",
1.222 brouard 6920: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6921: fprintf(fichtm,"\
1.128 brouard 6922: - 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 6923: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6924: fprintf(fichtm,"\
1.126 brouard 6925: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6926: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6927:
6928: /* if(popforecast==1) fprintf(fichtm,"\n */
6929: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6930: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6931: /* <br>",fileres,fileres,fileres,fileres); */
6932: /* else */
6933: /* 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 6934: fflush(fichtm);
6935: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6936:
1.225 brouard 6937: m=pow(2,cptcoveff);
1.222 brouard 6938: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6939:
1.222 brouard 6940: jj1=0;
1.237 brouard 6941:
1.241 brouard 6942: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6943: for(k1=1; k1<=m;k1++){
1.253 brouard 6944: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6945: continue;
1.222 brouard 6946: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6947: jj1++;
1.126 brouard 6948: if (cptcovn > 0) {
6949: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6950: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6951: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6952: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6953: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6954: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6955: }
6956:
1.126 brouard 6957: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6958:
1.222 brouard 6959: if(invalidvarcomb[k1]){
6960: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6961: continue;
6962: }
1.126 brouard 6963: }
6964: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6965: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6966: 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 6967: <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 6968: }
6969: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6970: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6971: true period expectancies (those weighted with period prevalences are also\
6972: drawn in addition to the population based expectancies computed using\
1.241 brouard 6973: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6974: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6975: /* } /\* end i1 *\/ */
6976: }/* End k1 */
1.241 brouard 6977: }/* End nres */
1.222 brouard 6978: fprintf(fichtm,"</ul>");
6979: fflush(fichtm);
1.126 brouard 6980: }
6981:
6982: /******************* Gnuplot file **************/
1.270 brouard 6983: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double bage, double fage , int prevfcast, int backcast, char pathc[], double p[], int offyear, int offbyear){
1.126 brouard 6984:
6985: char dirfileres[132],optfileres[132];
1.264 brouard 6986: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6987: 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 6988: int lv=0, vlv=0, kl=0;
1.130 brouard 6989: int ng=0;
1.201 brouard 6990: int vpopbased;
1.223 brouard 6991: int ioffset; /* variable offset for columns */
1.270 brouard 6992: int iyearc=1; /* variable column for year of projection */
6993: int iagec=1; /* variable column for age of projection */
1.235 brouard 6994: int nres=0; /* Index of resultline */
1.266 brouard 6995: int istart=1; /* For starting graphs in projections */
1.219 brouard 6996:
1.126 brouard 6997: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6998: /* printf("Problem with file %s",optionfilegnuplot); */
6999: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7000: /* } */
7001:
7002: /*#ifdef windows */
7003: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7004: /*#endif */
1.225 brouard 7005: m=pow(2,cptcoveff);
1.126 brouard 7006:
1.274 brouard 7007: /* diagram of the model */
7008: fprintf(ficgp,"\n#Diagram of the model \n");
7009: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7010: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7011: 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);
7012:
7013: 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);
7014: fprintf(ficgp,"\n#show arrow\nunset label\n");
7015: 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);
7016: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7017: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7018: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7019: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7020:
1.202 brouard 7021: /* Contribution to likelihood */
7022: /* Plot the probability implied in the likelihood */
1.223 brouard 7023: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7024: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7025: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7026: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7027: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7028: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7029: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7030: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7031: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7032: 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));
7033: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7034: 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));
7035: for (i=1; i<= nlstate ; i ++) {
7036: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7037: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7038: 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);
7039: for (j=2; j<= nlstate+ndeath ; j ++) {
7040: 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);
7041: }
7042: fprintf(ficgp,";\nset out; unset ylabel;\n");
7043: }
7044: /* 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 */
7045: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7046: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7047: fprintf(ficgp,"\nset out;unset log\n");
7048: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7049:
1.126 brouard 7050: strcpy(dirfileres,optionfilefiname);
7051: strcpy(optfileres,"vpl");
1.223 brouard 7052: /* 1eme*/
1.238 brouard 7053: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7054: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7055: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7056: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7057: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7058: continue;
7059: /* We are interested in selected combination by the resultline */
1.246 brouard 7060: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7061: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7062: strcpy(gplotlabel,"(");
1.238 brouard 7063: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7064: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7065: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7066: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7067: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7068: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7069: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7070: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7071: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7072: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7073: }
7074: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7075: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7076: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7077: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7078: }
7079: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7080: /* printf("\n#\n"); */
1.238 brouard 7081: fprintf(ficgp,"\n#\n");
7082: if(invalidvarcomb[k1]){
1.260 brouard 7083: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7084: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7085: continue;
7086: }
1.235 brouard 7087:
1.241 brouard 7088: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7089: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7090: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7091: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7092: 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);
7093: /* 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); */
7094: /* k1-1 error should be nres-1*/
1.238 brouard 7095: for (i=1; i<= nlstate ; i ++) {
7096: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7097: else fprintf(ficgp," %%*lf (%%*lf)");
7098: }
1.260 brouard 7099: fprintf(ficgp,"\" t\"Period (stable) prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7100: for (i=1; i<= nlstate ; i ++) {
7101: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7102: else fprintf(ficgp," %%*lf (%%*lf)");
7103: }
1.260 brouard 7104: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7105: for (i=1; i<= nlstate ; i ++) {
7106: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7107: else fprintf(ficgp," %%*lf (%%*lf)");
7108: }
1.265 brouard 7109: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7110:
7111: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7112: if(cptcoveff ==0){
1.271 brouard 7113: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7114: }else{
7115: kl=0;
7116: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7117: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7118: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7119: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7120: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7121: vlv= nbcode[Tvaraff[k]][lv];
7122: kl++;
7123: /* 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 *\/ */
7124: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7125: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7126: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7127: if(k==cptcoveff){
7128: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7129: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7130: }else{
7131: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7132: kl++;
7133: }
7134: } /* end covariate */
7135: } /* end if no covariate */
7136:
1.238 brouard 7137: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7138: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7139: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7140: if(cptcoveff ==0){
1.245 brouard 7141: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7142: }else{
7143: kl=0;
7144: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7145: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7146: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7147: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7148: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7149: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7150: kl++;
1.238 brouard 7151: /* 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 *\/ */
7152: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7153: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7154: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7155: if(k==cptcoveff){
1.245 brouard 7156: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7157: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7158: }else{
7159: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7160: kl++;
7161: }
7162: } /* end covariate */
7163: } /* end if no covariate */
1.268 brouard 7164: if(backcast == 1){
7165: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7166: /* k1-1 error should be nres-1*/
7167: for (i=1; i<= nlstate ; i ++) {
7168: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7169: else fprintf(ficgp," %%*lf (%%*lf)");
7170: }
1.271 brouard 7171: 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 7172: for (i=1; i<= nlstate ; i ++) {
7173: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7174: else fprintf(ficgp," %%*lf (%%*lf)");
7175: }
1.276 brouard 7176: 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 7177: for (i=1; i<= nlstate ; i ++) {
7178: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7179: else fprintf(ficgp," %%*lf (%%*lf)");
7180: }
1.274 brouard 7181: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7182: } /* end if backprojcast */
1.238 brouard 7183: } /* end if backcast */
1.276 brouard 7184: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7185: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7186: } /* nres */
1.201 brouard 7187: } /* k1 */
7188: } /* cpt */
1.235 brouard 7189:
7190:
1.126 brouard 7191: /*2 eme*/
1.238 brouard 7192: for (k1=1; k1<= m ; k1 ++){
7193: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7194: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7195: continue;
7196: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7197: strcpy(gplotlabel,"(");
1.238 brouard 7198: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7199: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7200: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7201: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7202: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7203: vlv= nbcode[Tvaraff[k]][lv];
7204: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7205: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7206: }
1.237 brouard 7207: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7208: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7209: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7210: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7211: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7212: }
1.264 brouard 7213: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7214: fprintf(ficgp,"\n#\n");
1.223 brouard 7215: if(invalidvarcomb[k1]){
7216: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7217: continue;
7218: }
1.219 brouard 7219:
1.241 brouard 7220: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7221: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7222: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7223: if(vpopbased==0){
1.238 brouard 7224: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7225: }else
1.238 brouard 7226: fprintf(ficgp,"\nreplot ");
7227: for (i=1; i<= nlstate+1 ; i ++) {
7228: k=2*i;
1.261 brouard 7229: 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 7230: for (j=1; j<= nlstate+1 ; j ++) {
7231: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7232: else fprintf(ficgp," %%*lf (%%*lf)");
7233: }
7234: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7235: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7236: 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 7237: for (j=1; j<= nlstate+1 ; j ++) {
7238: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7239: else fprintf(ficgp," %%*lf (%%*lf)");
7240: }
7241: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7242: 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 7243: for (j=1; j<= nlstate+1 ; j ++) {
7244: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7245: else fprintf(ficgp," %%*lf (%%*lf)");
7246: }
7247: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7248: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7249: } /* state */
7250: } /* vpopbased */
1.264 brouard 7251: 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 7252: } /* end nres */
7253: } /* k1 end 2 eme*/
7254:
7255:
7256: /*3eme*/
7257: for (k1=1; k1<= m ; k1 ++){
7258: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7259: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7260: continue;
7261:
7262: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7263: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7264: strcpy(gplotlabel,"(");
1.238 brouard 7265: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7266: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7267: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7268: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7269: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7270: vlv= nbcode[Tvaraff[k]][lv];
7271: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7272: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7273: }
7274: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7275: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7276: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7277: }
1.264 brouard 7278: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7279: fprintf(ficgp,"\n#\n");
7280: if(invalidvarcomb[k1]){
7281: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7282: continue;
7283: }
7284:
7285: /* k=2+nlstate*(2*cpt-2); */
7286: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7287: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7288: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7289: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7290: 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 7291: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7292: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7293: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7294: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7295: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7296: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7297:
1.238 brouard 7298: */
7299: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7300: 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 7301: /* 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 7302:
1.238 brouard 7303: }
1.261 brouard 7304: 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 7305: }
1.264 brouard 7306: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7307: } /* end nres */
7308: } /* end kl 3eme */
1.126 brouard 7309:
1.223 brouard 7310: /* 4eme */
1.201 brouard 7311: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7312: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7313: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7314: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7315: continue;
1.238 brouard 7316: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7317: strcpy(gplotlabel,"(");
1.238 brouard 7318: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7319: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7320: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7321: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7322: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7323: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7324: vlv= nbcode[Tvaraff[k]][lv];
7325: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7326: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7327: }
7328: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7329: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7330: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7331: }
1.264 brouard 7332: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7333: fprintf(ficgp,"\n#\n");
7334: if(invalidvarcomb[k1]){
7335: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7336: continue;
1.223 brouard 7337: }
1.238 brouard 7338:
1.241 brouard 7339: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7340: 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 7341: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7342: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7343: k=3;
7344: for (i=1; i<= nlstate ; i ++){
7345: if(i==1){
7346: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7347: }else{
7348: fprintf(ficgp,", '' ");
7349: }
7350: l=(nlstate+ndeath)*(i-1)+1;
7351: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7352: for (j=2; j<= nlstate+ndeath ; j ++)
7353: fprintf(ficgp,"+$%d",k+l+j-1);
7354: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7355: } /* nlstate */
1.264 brouard 7356: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7357: } /* end cpt state*/
7358: } /* end nres */
7359: } /* end covariate k1 */
7360:
1.220 brouard 7361: /* 5eme */
1.201 brouard 7362: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7363: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7364: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7365: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7366: continue;
1.238 brouard 7367: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7368: strcpy(gplotlabel,"(");
1.238 brouard 7369: 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);
7370: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7371: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7372: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7373: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7374: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7375: vlv= nbcode[Tvaraff[k]][lv];
7376: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7377: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7378: }
7379: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7380: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7381: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7382: }
1.264 brouard 7383: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7384: fprintf(ficgp,"\n#\n");
7385: if(invalidvarcomb[k1]){
7386: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7387: continue;
7388: }
1.227 brouard 7389:
1.241 brouard 7390: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7391: 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 7392: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7393: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7394: k=3;
7395: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7396: if(j==1)
7397: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7398: else
7399: fprintf(ficgp,", '' ");
7400: l=(nlstate+ndeath)*(cpt-1) +j;
7401: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7402: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7403: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7404: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7405: } /* nlstate */
7406: fprintf(ficgp,", '' ");
7407: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7408: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7409: l=(nlstate+ndeath)*(cpt-1) +j;
7410: if(j < nlstate)
7411: fprintf(ficgp,"$%d +",k+l);
7412: else
7413: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7414: }
1.264 brouard 7415: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7416: } /* end cpt state*/
7417: } /* end covariate */
7418: } /* end nres */
1.227 brouard 7419:
1.220 brouard 7420: /* 6eme */
1.202 brouard 7421: /* CV preval stable (period) for each covariate */
1.237 brouard 7422: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7423: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7424: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7425: continue;
1.255 brouard 7426: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7427: strcpy(gplotlabel,"(");
1.211 brouard 7428: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7429: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7430: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7431: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7432: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7433: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7434: vlv= nbcode[Tvaraff[k]][lv];
7435: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7436: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7437: }
1.237 brouard 7438: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7439: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7440: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7441: }
1.264 brouard 7442: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7443: fprintf(ficgp,"\n#\n");
1.223 brouard 7444: if(invalidvarcomb[k1]){
1.227 brouard 7445: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7446: continue;
1.223 brouard 7447: }
1.227 brouard 7448:
1.241 brouard 7449: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7450: 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 7451: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7452: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7453: k=3; /* Offset */
1.255 brouard 7454: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7455: if(i==1)
7456: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7457: else
7458: fprintf(ficgp,", '' ");
1.255 brouard 7459: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7460: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7461: for (j=2; j<= nlstate ; j ++)
7462: fprintf(ficgp,"+$%d",k+l+j-1);
7463: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7464: } /* nlstate */
1.264 brouard 7465: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7466: } /* end cpt state*/
7467: } /* end covariate */
1.227 brouard 7468:
7469:
1.220 brouard 7470: /* 7eme */
1.218 brouard 7471: if(backcast == 1){
1.217 brouard 7472: /* CV back preval stable (period) for each covariate */
1.237 brouard 7473: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7474: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7475: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7476: continue;
1.268 brouard 7477: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7478: strcpy(gplotlabel,"(");
7479: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7480: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7481: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7482: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7483: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7484: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7485: vlv= nbcode[Tvaraff[k]][lv];
7486: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7487: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7488: }
1.237 brouard 7489: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7490: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7491: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7492: }
1.264 brouard 7493: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7494: fprintf(ficgp,"\n#\n");
7495: if(invalidvarcomb[k1]){
7496: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7497: continue;
7498: }
7499:
1.241 brouard 7500: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7501: 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 7502: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7503: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7504: k=3; /* Offset */
1.268 brouard 7505: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7506: if(i==1)
7507: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7508: else
7509: fprintf(ficgp,", '' ");
7510: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7511: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7512: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7513: /* 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 7514: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7515: /* for (j=2; j<= nlstate ; j ++) */
7516: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7517: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7518: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7519: } /* nlstate */
1.264 brouard 7520: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7521: } /* end cpt state*/
7522: } /* end covariate */
7523: } /* End if backcast */
7524:
1.223 brouard 7525: /* 8eme */
1.218 brouard 7526: if(prevfcast==1){
7527: /* Projection from cross-sectional to stable (period) for each covariate */
7528:
1.237 brouard 7529: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7530: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7531: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7532: continue;
1.211 brouard 7533: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7534: strcpy(gplotlabel,"(");
1.227 brouard 7535: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7536: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7537: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7538: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7539: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7540: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7541: vlv= nbcode[Tvaraff[k]][lv];
7542: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7543: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7544: }
1.237 brouard 7545: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7546: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7547: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7548: }
1.264 brouard 7549: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7550: fprintf(ficgp,"\n#\n");
7551: if(invalidvarcomb[k1]){
7552: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7553: continue;
7554: }
7555:
7556: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7557: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7558: 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 7559: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7560: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7561:
7562: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7563: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7564: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7565: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7566: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7567: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7568: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7569: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7570: if(i==istart){
1.227 brouard 7571: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7572: }else{
7573: fprintf(ficgp,",\\\n '' ");
7574: }
7575: if(cptcoveff ==0){ /* No covariate */
7576: ioffset=2; /* Age is in 2 */
7577: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7578: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7579: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7580: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7581: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7582: if(i==nlstate+1){
1.270 brouard 7583: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7584: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7585: fprintf(ficgp,",\\\n '' ");
7586: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7587: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7588: offyear, \
1.268 brouard 7589: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7590: }else
1.227 brouard 7591: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7592: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7593: }else{ /* more than 2 covariates */
1.270 brouard 7594: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7595: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7596: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7597: iyearc=ioffset-1;
7598: iagec=ioffset;
1.227 brouard 7599: fprintf(ficgp," u %d:(",ioffset);
7600: kl=0;
7601: strcpy(gplotcondition,"(");
7602: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7603: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7604: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7605: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7606: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7607: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7608: kl++;
7609: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7610: kl++;
7611: if(k <cptcoveff && cptcoveff>1)
7612: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7613: }
7614: strcpy(gplotcondition+strlen(gplotcondition),")");
7615: /* 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 *\/ */
7616: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7617: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7618: /* '' 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*/
7619: if(i==nlstate+1){
1.270 brouard 7620: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7621: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7622: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7623: fprintf(ficgp," u %d:(",iagec);
7624: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7625: iyearc, iagec, offyear, \
7626: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7627: /* '' 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 7628: }else{
7629: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7630: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7631: }
7632: } /* end if covariate */
7633: } /* nlstate */
1.264 brouard 7634: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7635: } /* end cpt state*/
7636: } /* end covariate */
7637: } /* End if prevfcast */
1.227 brouard 7638:
1.268 brouard 7639: if(backcast==1){
7640: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7641:
7642: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7643: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7644: if(m != 1 && TKresult[nres]!= k1)
7645: continue;
7646: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7647: strcpy(gplotlabel,"(");
7648: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7649: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7650: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7651: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7652: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7653: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7654: vlv= nbcode[Tvaraff[k]][lv];
7655: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7656: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7657: }
7658: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7659: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7660: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7661: }
7662: strcpy(gplotlabel+strlen(gplotlabel),")");
7663: fprintf(ficgp,"\n#\n");
7664: if(invalidvarcomb[k1]){
7665: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7666: continue;
7667: }
7668:
7669: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7670: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7671: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7672: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7673: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7674:
7675: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7676: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7677: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7678: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7679: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7680: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7681: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7682: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7683: if(i==istart){
7684: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7685: }else{
7686: fprintf(ficgp,",\\\n '' ");
7687: }
7688: if(cptcoveff ==0){ /* No covariate */
7689: ioffset=2; /* Age is in 2 */
7690: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7691: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7692: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7693: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7694: fprintf(ficgp," u %d:(", ioffset);
7695: if(i==nlstate+1){
1.270 brouard 7696: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7697: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7698: fprintf(ficgp,",\\\n '' ");
7699: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7700: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7701: offbyear, \
7702: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7703: }else
7704: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7705: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7706: }else{ /* more than 2 covariates */
1.270 brouard 7707: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7708: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7709: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7710: iyearc=ioffset-1;
7711: iagec=ioffset;
1.268 brouard 7712: fprintf(ficgp," u %d:(",ioffset);
7713: kl=0;
7714: strcpy(gplotcondition,"(");
7715: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7716: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7717: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7718: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7719: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7720: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7721: kl++;
7722: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7723: kl++;
7724: if(k <cptcoveff && cptcoveff>1)
7725: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7726: }
7727: strcpy(gplotcondition+strlen(gplotcondition),")");
7728: /* 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 *\/ */
7729: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7730: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7731: /* '' 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*/
7732: if(i==nlstate+1){
1.270 brouard 7733: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7734: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7735: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7736: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7737: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7738: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7739: iyearc,iagec,offbyear, \
7740: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7741: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7742: }else{
7743: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7744: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7745: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7746: }
7747: } /* end if covariate */
7748: } /* nlstate */
7749: fprintf(ficgp,"\nset out; unset label;\n");
7750: } /* end cpt state*/
7751: } /* end covariate */
7752: } /* End if backcast */
7753:
1.227 brouard 7754:
1.238 brouard 7755: /* 9eme writing MLE parameters */
7756: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7757: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7758: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7759: for(k=1; k <=(nlstate+ndeath); k++){
7760: if (k != i) {
1.227 brouard 7761: fprintf(ficgp,"# current state %d\n",k);
7762: for(j=1; j <=ncovmodel; j++){
7763: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7764: jk++;
7765: }
7766: fprintf(ficgp,"\n");
1.126 brouard 7767: }
7768: }
1.223 brouard 7769: }
1.187 brouard 7770: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7771:
1.145 brouard 7772: /*goto avoid;*/
1.238 brouard 7773: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7774: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7775: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7776: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7777: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7778: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7779: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7780: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7781: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7782: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7783: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7784: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7785: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7786: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7787: fprintf(ficgp,"#\n");
1.223 brouard 7788: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7789: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7790: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7791: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7792: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7793: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7794: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7795: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7796: continue;
1.264 brouard 7797: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7798: strcpy(gplotlabel,"(");
1.276 brouard 7799: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7800: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7801: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7802: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7803: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7804: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7805: vlv= nbcode[Tvaraff[k]][lv];
7806: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7807: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7808: }
1.237 brouard 7809: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7810: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7811: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7812: }
1.264 brouard 7813: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7814: fprintf(ficgp,"\n#\n");
1.264 brouard 7815: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7816: fprintf(ficgp,"\nset key outside ");
7817: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7818: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7819: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7820: if (ng==1){
7821: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7822: fprintf(ficgp,"\nunset log y");
7823: }else if (ng==2){
7824: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7825: fprintf(ficgp,"\nset log y");
7826: }else if (ng==3){
7827: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7828: fprintf(ficgp,"\nset log y");
7829: }else
7830: fprintf(ficgp,"\nunset title ");
7831: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7832: i=1;
7833: for(k2=1; k2<=nlstate; k2++) {
7834: k3=i;
7835: for(k=1; k<=(nlstate+ndeath); k++) {
7836: if (k != k2){
7837: switch( ng) {
7838: case 1:
7839: if(nagesqr==0)
7840: fprintf(ficgp," p%d+p%d*x",i,i+1);
7841: else /* nagesqr =1 */
7842: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7843: break;
7844: case 2: /* ng=2 */
7845: if(nagesqr==0)
7846: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7847: else /* nagesqr =1 */
7848: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7849: break;
7850: case 3:
7851: if(nagesqr==0)
7852: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7853: else /* nagesqr =1 */
7854: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7855: break;
7856: }
7857: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7858: ijp=1; /* product no age */
7859: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7860: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7861: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7862: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7863: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7864: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7865: if(DummyV[j]==0){
7866: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7867: }else{ /* quantitative */
7868: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7869: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7870: }
7871: ij++;
1.237 brouard 7872: }
1.268 brouard 7873: }
7874: }else if(cptcovprod >0){
7875: if(j==Tprod[ijp]) { /* */
7876: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7877: if(ijp <=cptcovprod) { /* Product */
7878: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7879: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7880: /* 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)]); */
7881: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7882: }else{ /* Vn is dummy and Vm is quanti */
7883: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7884: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7885: }
7886: }else{ /* Vn*Vm Vn is quanti */
7887: if(DummyV[Tvard[ijp][2]]==0){
7888: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7889: }else{ /* Both quanti */
7890: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7891: }
1.237 brouard 7892: }
1.268 brouard 7893: ijp++;
1.237 brouard 7894: }
1.268 brouard 7895: } /* end Tprod */
1.237 brouard 7896: } else{ /* simple covariate */
1.264 brouard 7897: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7898: if(Dummy[j]==0){
7899: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7900: }else{ /* quantitative */
7901: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7902: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7903: }
1.237 brouard 7904: } /* end simple */
7905: } /* end j */
1.223 brouard 7906: }else{
7907: i=i-ncovmodel;
7908: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7909: fprintf(ficgp," (1.");
7910: }
1.227 brouard 7911:
1.223 brouard 7912: if(ng != 1){
7913: fprintf(ficgp,")/(1");
1.227 brouard 7914:
1.264 brouard 7915: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7916: if(nagesqr==0)
1.264 brouard 7917: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7918: else /* nagesqr =1 */
1.264 brouard 7919: 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 7920:
1.223 brouard 7921: ij=1;
7922: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7923: if(cptcovage >0){
7924: if((j-2)==Tage[ij]) { /* Bug valgrind */
7925: if(ij <=cptcovage) { /* Bug valgrind */
7926: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7927: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7928: ij++;
7929: }
7930: }
7931: }else
7932: 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 7933: }
7934: fprintf(ficgp,")");
7935: }
7936: fprintf(ficgp,")");
7937: if(ng ==2)
1.276 brouard 7938: 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 7939: else /* ng= 3 */
1.276 brouard 7940: 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 7941: }else{ /* end ng <> 1 */
7942: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 7943: 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 7944: }
7945: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7946: fprintf(ficgp,",");
7947: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7948: fprintf(ficgp,",");
7949: i=i+ncovmodel;
7950: } /* end k */
7951: } /* end k2 */
1.276 brouard 7952: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
7953: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 7954: } /* end k1 */
1.223 brouard 7955: } /* end ng */
7956: /* avoid: */
7957: fflush(ficgp);
1.126 brouard 7958: } /* end gnuplot */
7959:
7960:
7961: /*************** Moving average **************/
1.219 brouard 7962: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7963: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7964:
1.222 brouard 7965: int i, cpt, cptcod;
7966: int modcovmax =1;
7967: int mobilavrange, mob;
7968: int iage=0;
7969:
1.266 brouard 7970: double sum=0., sumr=0.;
1.222 brouard 7971: double age;
1.266 brouard 7972: double *sumnewp, *sumnewm, *sumnewmr;
7973: double *agemingood, *agemaxgood;
7974: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7975:
7976:
1.278 brouard 7977: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
7978: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 7979:
7980: sumnewp = vector(1,ncovcombmax);
7981: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7982: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7983: agemingood = vector(1,ncovcombmax);
1.266 brouard 7984: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7985: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7986: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7987:
7988: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7989: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7990: sumnewp[cptcod]=0.;
1.266 brouard 7991: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
7992: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7993: }
7994: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7995:
1.266 brouard 7996: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7997: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7998: else mobilavrange=mobilav;
7999: for (age=bage; age<=fage; age++)
8000: for (i=1; i<=nlstate;i++)
8001: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8002: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8003: /* We keep the original values on the extreme ages bage, fage and for
8004: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8005: we use a 5 terms etc. until the borders are no more concerned.
8006: */
8007: for (mob=3;mob <=mobilavrange;mob=mob+2){
8008: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8009: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8010: sumnewm[cptcod]=0.;
8011: for (i=1; i<=nlstate;i++){
1.222 brouard 8012: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8013: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8014: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8015: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8016: }
8017: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8018: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8019: } /* end i */
8020: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8021: } /* end cptcod */
1.222 brouard 8022: }/* end age */
8023: }/* end mob */
1.266 brouard 8024: }else{
8025: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8026: return -1;
1.266 brouard 8027: }
8028:
8029: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8030: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8031: if(invalidvarcomb[cptcod]){
8032: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8033: continue;
8034: }
1.219 brouard 8035:
1.266 brouard 8036: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8037: sumnewm[cptcod]=0.;
8038: sumnewmr[cptcod]=0.;
8039: for (i=1; i<=nlstate;i++){
8040: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8041: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8042: }
8043: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8044: agemingoodr[cptcod]=age;
8045: }
8046: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8047: agemingood[cptcod]=age;
8048: }
8049: } /* age */
8050: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8051: sumnewm[cptcod]=0.;
1.266 brouard 8052: sumnewmr[cptcod]=0.;
1.222 brouard 8053: for (i=1; i<=nlstate;i++){
8054: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8055: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8056: }
8057: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8058: agemaxgoodr[cptcod]=age;
1.222 brouard 8059: }
8060: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8061: agemaxgood[cptcod]=age;
8062: }
8063: } /* age */
8064: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8065: /* but they will change */
8066: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8067: sumnewm[cptcod]=0.;
8068: sumnewmr[cptcod]=0.;
8069: for (i=1; i<=nlstate;i++){
8070: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8071: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8072: }
8073: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8074: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8075: agemaxgoodr[cptcod]=age; /* age min */
8076: for (i=1; i<=nlstate;i++)
8077: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8078: }else{ /* bad we change the value with the values of good ages */
8079: for (i=1; i<=nlstate;i++){
8080: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8081: } /* i */
8082: } /* end bad */
8083: }else{
8084: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8085: agemaxgood[cptcod]=age;
8086: }else{ /* bad we change the value with the values of good ages */
8087: for (i=1; i<=nlstate;i++){
8088: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8089: } /* i */
8090: } /* end bad */
8091: }/* end else */
8092: sum=0.;sumr=0.;
8093: for (i=1; i<=nlstate;i++){
8094: sum+=mobaverage[(int)age][i][cptcod];
8095: sumr+=probs[(int)age][i][cptcod];
8096: }
8097: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8098: printf("Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.266 brouard 8099: } /* end bad */
8100: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8101: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8102: printf("Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.222 brouard 8103: } /* end bad */
8104: }/* age */
1.266 brouard 8105:
8106: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8107: sumnewm[cptcod]=0.;
1.266 brouard 8108: sumnewmr[cptcod]=0.;
1.222 brouard 8109: for (i=1; i<=nlstate;i++){
8110: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8111: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8112: }
8113: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8114: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8115: agemingoodr[cptcod]=age;
8116: for (i=1; i<=nlstate;i++)
8117: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8118: }else{ /* bad we change the value with the values of good ages */
8119: for (i=1; i<=nlstate;i++){
8120: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8121: } /* i */
8122: } /* end bad */
8123: }else{
8124: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8125: agemingood[cptcod]=age;
8126: }else{ /* bad */
8127: for (i=1; i<=nlstate;i++){
8128: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8129: } /* i */
8130: } /* end bad */
8131: }/* end else */
8132: sum=0.;sumr=0.;
8133: for (i=1; i<=nlstate;i++){
8134: sum+=mobaverage[(int)age][i][cptcod];
8135: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8136: }
1.266 brouard 8137: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8138: 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 8139: } /* end bad */
8140: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8141: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8142: 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 8143: } /* end bad */
8144: }/* age */
1.266 brouard 8145:
1.222 brouard 8146:
8147: for (age=bage; age<=fage; age++){
1.235 brouard 8148: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8149: sumnewp[cptcod]=0.;
8150: sumnewm[cptcod]=0.;
8151: for (i=1; i<=nlstate;i++){
8152: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8153: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8154: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8155: }
8156: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8157: }
8158: /* printf("\n"); */
8159: /* } */
1.266 brouard 8160:
1.222 brouard 8161: /* brutal averaging */
1.266 brouard 8162: /* for (i=1; i<=nlstate;i++){ */
8163: /* for (age=1; age<=bage; age++){ */
8164: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8165: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8166: /* } */
8167: /* for (age=fage; age<=AGESUP; age++){ */
8168: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8169: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8170: /* } */
8171: /* } /\* end i status *\/ */
8172: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8173: /* for (age=1; age<=AGESUP; age++){ */
8174: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8175: /* mobaverage[(int)age][i][cptcod]=0.; */
8176: /* } */
8177: /* } */
1.222 brouard 8178: }/* end cptcod */
1.266 brouard 8179: free_vector(agemaxgoodr,1, ncovcombmax);
8180: free_vector(agemaxgood,1, ncovcombmax);
8181: free_vector(agemingood,1, ncovcombmax);
8182: free_vector(agemingoodr,1, ncovcombmax);
8183: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8184: free_vector(sumnewm,1, ncovcombmax);
8185: free_vector(sumnewp,1, ncovcombmax);
8186: return 0;
8187: }/* End movingaverage */
1.218 brouard 8188:
1.126 brouard 8189:
8190: /************** Forecasting ******************/
1.269 brouard 8191: void prevforecast(char fileres[], double anproj1, double mproj1, double jproj1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double anproj2, double p[], int cptcoveff){
1.126 brouard 8192: /* proj1, year, month, day of starting projection
8193: agemin, agemax range of age
8194: dateprev1 dateprev2 range of dates during which prevalence is computed
8195: anproj2 year of en of projection (same day and month as proj1).
8196: */
1.267 brouard 8197: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8198: double agec; /* generic age */
8199: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8200: double *popeffectif,*popcount;
8201: double ***p3mat;
1.218 brouard 8202: /* double ***mobaverage; */
1.126 brouard 8203: char fileresf[FILENAMELENGTH];
8204:
8205: agelim=AGESUP;
1.211 brouard 8206: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8207: in each health status at the date of interview (if between dateprev1 and dateprev2).
8208: We still use firstpass and lastpass as another selection.
8209: */
1.214 brouard 8210: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8211: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8212:
1.201 brouard 8213: strcpy(fileresf,"F_");
8214: strcat(fileresf,fileresu);
1.126 brouard 8215: if((ficresf=fopen(fileresf,"w"))==NULL) {
8216: printf("Problem with forecast resultfile: %s\n", fileresf);
8217: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8218: }
1.235 brouard 8219: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8220: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8221:
1.225 brouard 8222: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8223:
8224:
8225: stepsize=(int) (stepm+YEARM-1)/YEARM;
8226: if (stepm<=12) stepsize=1;
8227: if(estepm < stepm){
8228: printf ("Problem %d lower than %d\n",estepm, stepm);
8229: }
1.270 brouard 8230: else{
8231: hstepm=estepm;
8232: }
8233: if(estepm > stepm){ /* Yes every two year */
8234: stepsize=2;
8235: }
1.126 brouard 8236:
8237: hstepm=hstepm/stepm;
8238: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8239: fractional in yp1 */
8240: anprojmean=yp;
8241: yp2=modf((yp1*12),&yp);
8242: mprojmean=yp;
8243: yp1=modf((yp2*30.5),&yp);
8244: jprojmean=yp;
8245: if(jprojmean==0) jprojmean=1;
8246: if(mprojmean==0) jprojmean=1;
8247:
1.227 brouard 8248: i1=pow(2,cptcoveff);
1.126 brouard 8249: if (cptcovn < 1){i1=1;}
8250:
8251: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8252:
8253: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8254:
1.126 brouard 8255: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8256: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8257: for(k=1; k<=i1;k++){
1.253 brouard 8258: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8259: continue;
1.227 brouard 8260: if(invalidvarcomb[k]){
8261: printf("\nCombination (%d) projection ignored because no cases \n",k);
8262: continue;
8263: }
8264: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8265: for(j=1;j<=cptcoveff;j++) {
8266: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8267: }
1.235 brouard 8268: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8269: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8270: }
1.227 brouard 8271: fprintf(ficresf," yearproj age");
8272: for(j=1; j<=nlstate+ndeath;j++){
8273: for(i=1; i<=nlstate;i++)
8274: fprintf(ficresf," p%d%d",i,j);
8275: fprintf(ficresf," wp.%d",j);
8276: }
8277: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8278: fprintf(ficresf,"\n");
8279: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8280: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8281: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8282: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8283: nhstepm = nhstepm/hstepm;
8284: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8285: oldm=oldms;savm=savms;
1.268 brouard 8286: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8287: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8288: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8289: for (h=0; h<=nhstepm; h++){
8290: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8291: break;
8292: }
8293: }
8294: fprintf(ficresf,"\n");
8295: for(j=1;j<=cptcoveff;j++)
8296: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8297: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8298:
8299: for(j=1; j<=nlstate+ndeath;j++) {
8300: ppij=0.;
8301: for(i=1; i<=nlstate;i++) {
1.278 brouard 8302: if (mobilav>=1)
8303: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8304: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8305: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8306: }
1.268 brouard 8307: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8308: } /* end i */
8309: fprintf(ficresf," %.3f", ppij);
8310: }/* end j */
1.227 brouard 8311: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8312: } /* end agec */
1.266 brouard 8313: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8314: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8315: } /* end yearp */
8316: } /* end k */
1.219 brouard 8317:
1.126 brouard 8318: fclose(ficresf);
1.215 brouard 8319: printf("End of Computing forecasting \n");
8320: fprintf(ficlog,"End of Computing forecasting\n");
8321:
1.126 brouard 8322: }
8323:
1.269 brouard 8324: /************** Back Forecasting ******************/
8325: void prevbackforecast(char fileres[], double ***prevacurrent, double anback1, double mback1, double jback1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anback2, double p[], int cptcoveff){
1.267 brouard 8326: /* back1, year, month, day of starting backection
8327: agemin, agemax range of age
8328: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8329: anback2 year of end of backprojection (same day and month as back1).
8330: prevacurrent and prev are prevalences.
1.267 brouard 8331: */
8332: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8333: double agec; /* generic age */
1.268 brouard 8334: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8335: double *popeffectif,*popcount;
8336: double ***p3mat;
8337: /* double ***mobaverage; */
8338: char fileresfb[FILENAMELENGTH];
8339:
1.268 brouard 8340: agelim=AGEINF;
1.267 brouard 8341: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8342: in each health status at the date of interview (if between dateprev1 and dateprev2).
8343: We still use firstpass and lastpass as another selection.
8344: */
8345: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8346: /* firstpass, lastpass, stepm, weightopt, model); */
8347:
8348: /*Do we need to compute prevalence again?*/
8349:
8350: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8351:
8352: strcpy(fileresfb,"FB_");
8353: strcat(fileresfb,fileresu);
8354: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8355: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8356: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8357: }
8358: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8359: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8360:
8361: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8362:
8363:
8364: stepsize=(int) (stepm+YEARM-1)/YEARM;
8365: if (stepm<=12) stepsize=1;
8366: if(estepm < stepm){
8367: printf ("Problem %d lower than %d\n",estepm, stepm);
8368: }
1.270 brouard 8369: else{
8370: hstepm=estepm;
8371: }
8372: if(estepm >= stepm){ /* Yes every two year */
8373: stepsize=2;
8374: }
1.267 brouard 8375:
8376: hstepm=hstepm/stepm;
8377: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8378: fractional in yp1 */
8379: anprojmean=yp;
8380: yp2=modf((yp1*12),&yp);
8381: mprojmean=yp;
8382: yp1=modf((yp2*30.5),&yp);
8383: jprojmean=yp;
8384: if(jprojmean==0) jprojmean=1;
8385: if(mprojmean==0) jprojmean=1;
8386:
8387: i1=pow(2,cptcoveff);
8388: if (cptcovn < 1){i1=1;}
8389:
8390: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8391: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8392:
8393: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8394:
8395: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8396: for(k=1; k<=i1;k++){
8397: if(i1 != 1 && TKresult[nres]!= k)
8398: continue;
8399: if(invalidvarcomb[k]){
8400: printf("\nCombination (%d) projection ignored because no cases \n",k);
8401: continue;
8402: }
1.268 brouard 8403: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8404: for(j=1;j<=cptcoveff;j++) {
8405: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8406: }
8407: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8408: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8409: }
8410: fprintf(ficresfb," yearbproj age");
8411: for(j=1; j<=nlstate+ndeath;j++){
8412: for(i=1; i<=nlstate;i++)
1.268 brouard 8413: fprintf(ficresfb," b%d%d",i,j);
8414: fprintf(ficresfb," b.%d",j);
1.267 brouard 8415: }
8416: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8417: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8418: fprintf(ficresfb,"\n");
8419: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8420: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8421: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8422: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8423: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8424: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8425: nhstepm = nhstepm/hstepm;
8426: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8427: oldm=oldms;savm=savms;
1.268 brouard 8428: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8429: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8430: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8431: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8432: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8433: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8434: for (h=0; h<=nhstepm; h++){
1.268 brouard 8435: if (h*hstepm/YEARM*stepm ==-yearp) {
8436: break;
8437: }
8438: }
8439: fprintf(ficresfb,"\n");
8440: for(j=1;j<=cptcoveff;j++)
8441: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8442: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8443: for(i=1; i<=nlstate+ndeath;i++) {
8444: ppij=0.;ppi=0.;
8445: for(j=1; j<=nlstate;j++) {
8446: /* if (mobilav==1) */
1.269 brouard 8447: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8448: ppi=ppi+prevacurrent[(int)agec][j][k];
8449: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8450: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8451: /* else { */
8452: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8453: /* } */
1.268 brouard 8454: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8455: } /* end j */
8456: if(ppi <0.99){
8457: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8458: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8459: }
8460: fprintf(ficresfb," %.3f", ppij);
8461: }/* end j */
1.267 brouard 8462: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8463: } /* end agec */
8464: } /* end yearp */
8465: } /* end k */
1.217 brouard 8466:
1.267 brouard 8467: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8468:
1.267 brouard 8469: fclose(ficresfb);
8470: printf("End of Computing Back forecasting \n");
8471: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8472:
1.267 brouard 8473: }
1.217 brouard 8474:
1.269 brouard 8475: /* Variance of prevalence limit: varprlim */
8476: 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){
8477: /*------- Variance of period (stable) prevalence------*/
8478:
8479: char fileresvpl[FILENAMELENGTH];
8480: FILE *ficresvpl;
8481: double **oldm, **savm;
8482: double **varpl; /* Variances of prevalence limits by age */
8483: int i1, k, nres, j ;
8484:
8485: strcpy(fileresvpl,"VPL_");
8486: strcat(fileresvpl,fileresu);
8487: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8488: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8489: exit(0);
8490: }
8491: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8492: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8493:
8494: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8495: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8496:
8497: i1=pow(2,cptcoveff);
8498: if (cptcovn < 1){i1=1;}
8499:
8500: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8501: for(k=1; k<=i1;k++){
8502: if(i1 != 1 && TKresult[nres]!= k)
8503: continue;
8504: fprintf(ficresvpl,"\n#****** ");
8505: printf("\n#****** ");
8506: fprintf(ficlog,"\n#****** ");
8507: for(j=1;j<=cptcoveff;j++) {
8508: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8509: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8510: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8511: }
8512: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8513: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8514: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8515: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8516: }
8517: fprintf(ficresvpl,"******\n");
8518: printf("******\n");
8519: fprintf(ficlog,"******\n");
8520:
8521: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8522: oldm=oldms;savm=savms;
8523: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8524: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8525: /*}*/
8526: }
8527:
8528: fclose(ficresvpl);
8529: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8530: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8531:
8532: }
8533: /* Variance of back prevalence: varbprlim */
8534: 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){
8535: /*------- Variance of back (stable) prevalence------*/
8536:
8537: char fileresvbl[FILENAMELENGTH];
8538: FILE *ficresvbl;
8539:
8540: double **oldm, **savm;
8541: double **varbpl; /* Variances of back prevalence limits by age */
8542: int i1, k, nres, j ;
8543:
8544: strcpy(fileresvbl,"VBL_");
8545: strcat(fileresvbl,fileresu);
8546: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8547: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8548: exit(0);
8549: }
8550: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8551: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8552:
8553:
8554: i1=pow(2,cptcoveff);
8555: if (cptcovn < 1){i1=1;}
8556:
8557: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8558: for(k=1; k<=i1;k++){
8559: if(i1 != 1 && TKresult[nres]!= k)
8560: continue;
8561: fprintf(ficresvbl,"\n#****** ");
8562: printf("\n#****** ");
8563: fprintf(ficlog,"\n#****** ");
8564: for(j=1;j<=cptcoveff;j++) {
8565: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8566: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8567: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8568: }
8569: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8570: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8571: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8572: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8573: }
8574: fprintf(ficresvbl,"******\n");
8575: printf("******\n");
8576: fprintf(ficlog,"******\n");
8577:
8578: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8579: oldm=oldms;savm=savms;
8580:
8581: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8582: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8583: /*}*/
8584: }
8585:
8586: fclose(ficresvbl);
8587: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8588: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8589:
8590: } /* End of varbprlim */
8591:
1.126 brouard 8592: /************** Forecasting *****not tested NB*************/
1.227 brouard 8593: /* 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 8594:
1.227 brouard 8595: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8596: /* int *popage; */
8597: /* double calagedatem, agelim, kk1, kk2; */
8598: /* double *popeffectif,*popcount; */
8599: /* double ***p3mat,***tabpop,***tabpopprev; */
8600: /* /\* double ***mobaverage; *\/ */
8601: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8602:
1.227 brouard 8603: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8604: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8605: /* agelim=AGESUP; */
8606: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8607:
1.227 brouard 8608: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8609:
8610:
1.227 brouard 8611: /* strcpy(filerespop,"POP_"); */
8612: /* strcat(filerespop,fileresu); */
8613: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8614: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8615: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8616: /* } */
8617: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8618: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8619:
1.227 brouard 8620: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8621:
1.227 brouard 8622: /* /\* if (mobilav!=0) { *\/ */
8623: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8624: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8625: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8626: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8627: /* /\* } *\/ */
8628: /* /\* } *\/ */
1.126 brouard 8629:
1.227 brouard 8630: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8631: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8632:
1.227 brouard 8633: /* agelim=AGESUP; */
1.126 brouard 8634:
1.227 brouard 8635: /* hstepm=1; */
8636: /* hstepm=hstepm/stepm; */
1.218 brouard 8637:
1.227 brouard 8638: /* if (popforecast==1) { */
8639: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8640: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8641: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8642: /* } */
8643: /* popage=ivector(0,AGESUP); */
8644: /* popeffectif=vector(0,AGESUP); */
8645: /* popcount=vector(0,AGESUP); */
1.126 brouard 8646:
1.227 brouard 8647: /* i=1; */
8648: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8649:
1.227 brouard 8650: /* imx=i; */
8651: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8652: /* } */
1.218 brouard 8653:
1.227 brouard 8654: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8655: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8656: /* k=k+1; */
8657: /* fprintf(ficrespop,"\n#******"); */
8658: /* for(j=1;j<=cptcoveff;j++) { */
8659: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8660: /* } */
8661: /* fprintf(ficrespop,"******\n"); */
8662: /* fprintf(ficrespop,"# Age"); */
8663: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8664: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8665:
1.227 brouard 8666: /* for (cpt=0; cpt<=0;cpt++) { */
8667: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8668:
1.227 brouard 8669: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8670: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8671: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8672:
1.227 brouard 8673: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8674: /* oldm=oldms;savm=savms; */
8675: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8676:
1.227 brouard 8677: /* for (h=0; h<=nhstepm; h++){ */
8678: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8679: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8680: /* } */
8681: /* for(j=1; j<=nlstate+ndeath;j++) { */
8682: /* kk1=0.;kk2=0; */
8683: /* for(i=1; i<=nlstate;i++) { */
8684: /* if (mobilav==1) */
8685: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8686: /* else { */
8687: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8688: /* } */
8689: /* } */
8690: /* if (h==(int)(calagedatem+12*cpt)){ */
8691: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8692: /* /\*fprintf(ficrespop," %.3f", kk1); */
8693: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8694: /* } */
8695: /* } */
8696: /* for(i=1; i<=nlstate;i++){ */
8697: /* kk1=0.; */
8698: /* for(j=1; j<=nlstate;j++){ */
8699: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8700: /* } */
8701: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8702: /* } */
1.218 brouard 8703:
1.227 brouard 8704: /* if (h==(int)(calagedatem+12*cpt)) */
8705: /* for(j=1; j<=nlstate;j++) */
8706: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8707: /* } */
8708: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8709: /* } */
8710: /* } */
1.218 brouard 8711:
1.227 brouard 8712: /* /\******\/ */
1.218 brouard 8713:
1.227 brouard 8714: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8715: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8716: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8717: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8718: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8719:
1.227 brouard 8720: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8721: /* oldm=oldms;savm=savms; */
8722: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8723: /* for (h=0; h<=nhstepm; h++){ */
8724: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8725: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8726: /* } */
8727: /* for(j=1; j<=nlstate+ndeath;j++) { */
8728: /* kk1=0.;kk2=0; */
8729: /* for(i=1; i<=nlstate;i++) { */
8730: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8731: /* } */
8732: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8733: /* } */
8734: /* } */
8735: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8736: /* } */
8737: /* } */
8738: /* } */
8739: /* } */
1.218 brouard 8740:
1.227 brouard 8741: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8742:
1.227 brouard 8743: /* if (popforecast==1) { */
8744: /* free_ivector(popage,0,AGESUP); */
8745: /* free_vector(popeffectif,0,AGESUP); */
8746: /* free_vector(popcount,0,AGESUP); */
8747: /* } */
8748: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8749: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8750: /* fclose(ficrespop); */
8751: /* } /\* End of popforecast *\/ */
1.218 brouard 8752:
1.126 brouard 8753: int fileappend(FILE *fichier, char *optionfich)
8754: {
8755: if((fichier=fopen(optionfich,"a"))==NULL) {
8756: printf("Problem with file: %s\n", optionfich);
8757: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8758: return (0);
8759: }
8760: fflush(fichier);
8761: return (1);
8762: }
8763:
8764:
8765: /**************** function prwizard **********************/
8766: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8767: {
8768:
8769: /* Wizard to print covariance matrix template */
8770:
1.164 brouard 8771: char ca[32], cb[32];
8772: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8773: int numlinepar;
8774:
8775: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8776: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8777: for(i=1; i <=nlstate; i++){
8778: jj=0;
8779: for(j=1; j <=nlstate+ndeath; j++){
8780: if(j==i) continue;
8781: jj++;
8782: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8783: printf("%1d%1d",i,j);
8784: fprintf(ficparo,"%1d%1d",i,j);
8785: for(k=1; k<=ncovmodel;k++){
8786: /* printf(" %lf",param[i][j][k]); */
8787: /* fprintf(ficparo," %lf",param[i][j][k]); */
8788: printf(" 0.");
8789: fprintf(ficparo," 0.");
8790: }
8791: printf("\n");
8792: fprintf(ficparo,"\n");
8793: }
8794: }
8795: printf("# Scales (for hessian or gradient estimation)\n");
8796: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8797: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8798: for(i=1; i <=nlstate; i++){
8799: jj=0;
8800: for(j=1; j <=nlstate+ndeath; j++){
8801: if(j==i) continue;
8802: jj++;
8803: fprintf(ficparo,"%1d%1d",i,j);
8804: printf("%1d%1d",i,j);
8805: fflush(stdout);
8806: for(k=1; k<=ncovmodel;k++){
8807: /* printf(" %le",delti3[i][j][k]); */
8808: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8809: printf(" 0.");
8810: fprintf(ficparo," 0.");
8811: }
8812: numlinepar++;
8813: printf("\n");
8814: fprintf(ficparo,"\n");
8815: }
8816: }
8817: printf("# Covariance matrix\n");
8818: /* # 121 Var(a12)\n\ */
8819: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8820: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8821: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8822: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8823: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8824: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8825: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8826: fflush(stdout);
8827: fprintf(ficparo,"# Covariance matrix\n");
8828: /* # 121 Var(a12)\n\ */
8829: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8830: /* # ...\n\ */
8831: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8832:
8833: for(itimes=1;itimes<=2;itimes++){
8834: jj=0;
8835: for(i=1; i <=nlstate; i++){
8836: for(j=1; j <=nlstate+ndeath; j++){
8837: if(j==i) continue;
8838: for(k=1; k<=ncovmodel;k++){
8839: jj++;
8840: ca[0]= k+'a'-1;ca[1]='\0';
8841: if(itimes==1){
8842: printf("#%1d%1d%d",i,j,k);
8843: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8844: }else{
8845: printf("%1d%1d%d",i,j,k);
8846: fprintf(ficparo,"%1d%1d%d",i,j,k);
8847: /* printf(" %.5le",matcov[i][j]); */
8848: }
8849: ll=0;
8850: for(li=1;li <=nlstate; li++){
8851: for(lj=1;lj <=nlstate+ndeath; lj++){
8852: if(lj==li) continue;
8853: for(lk=1;lk<=ncovmodel;lk++){
8854: ll++;
8855: if(ll<=jj){
8856: cb[0]= lk +'a'-1;cb[1]='\0';
8857: if(ll<jj){
8858: if(itimes==1){
8859: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8860: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8861: }else{
8862: printf(" 0.");
8863: fprintf(ficparo," 0.");
8864: }
8865: }else{
8866: if(itimes==1){
8867: printf(" Var(%s%1d%1d)",ca,i,j);
8868: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8869: }else{
8870: printf(" 0.");
8871: fprintf(ficparo," 0.");
8872: }
8873: }
8874: }
8875: } /* end lk */
8876: } /* end lj */
8877: } /* end li */
8878: printf("\n");
8879: fprintf(ficparo,"\n");
8880: numlinepar++;
8881: } /* end k*/
8882: } /*end j */
8883: } /* end i */
8884: } /* end itimes */
8885:
8886: } /* end of prwizard */
8887: /******************* Gompertz Likelihood ******************************/
8888: double gompertz(double x[])
8889: {
8890: double A,B,L=0.0,sump=0.,num=0.;
8891: int i,n=0; /* n is the size of the sample */
8892:
1.220 brouard 8893: for (i=1;i<=imx ; i++) {
1.126 brouard 8894: sump=sump+weight[i];
8895: /* sump=sump+1;*/
8896: num=num+1;
8897: }
8898:
8899:
8900: /* for (i=0; i<=imx; i++)
8901: 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]);*/
8902:
8903: for (i=1;i<=imx ; i++)
8904: {
8905: if (cens[i] == 1 && wav[i]>1)
8906: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8907:
8908: if (cens[i] == 0 && wav[i]>1)
8909: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8910: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8911:
8912: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8913: if (wav[i] > 1 ) { /* ??? */
8914: L=L+A*weight[i];
8915: /* 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]);*/
8916: }
8917: }
8918:
8919: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8920:
8921: return -2*L*num/sump;
8922: }
8923:
1.136 brouard 8924: #ifdef GSL
8925: /******************* Gompertz_f Likelihood ******************************/
8926: double gompertz_f(const gsl_vector *v, void *params)
8927: {
8928: double A,B,LL=0.0,sump=0.,num=0.;
8929: double *x= (double *) v->data;
8930: int i,n=0; /* n is the size of the sample */
8931:
8932: for (i=0;i<=imx-1 ; i++) {
8933: sump=sump+weight[i];
8934: /* sump=sump+1;*/
8935: num=num+1;
8936: }
8937:
8938:
8939: /* for (i=0; i<=imx; i++)
8940: 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]);*/
8941: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8942: for (i=1;i<=imx ; i++)
8943: {
8944: if (cens[i] == 1 && wav[i]>1)
8945: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8946:
8947: if (cens[i] == 0 && wav[i]>1)
8948: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8949: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8950:
8951: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8952: if (wav[i] > 1 ) { /* ??? */
8953: LL=LL+A*weight[i];
8954: /* 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]);*/
8955: }
8956: }
8957:
8958: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8959: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8960:
8961: return -2*LL*num/sump;
8962: }
8963: #endif
8964:
1.126 brouard 8965: /******************* Printing html file ***********/
1.201 brouard 8966: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8967: int lastpass, int stepm, int weightopt, char model[],\
8968: int imx, double p[],double **matcov,double agemortsup){
8969: int i,k;
8970:
8971: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8972: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8973: for (i=1;i<=2;i++)
8974: 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 8975: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8976: fprintf(fichtm,"</ul>");
8977:
8978: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8979:
8980: 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>");
8981:
8982: for (k=agegomp;k<(agemortsup-2);k++)
8983: 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]);
8984:
8985:
8986: fflush(fichtm);
8987: }
8988:
8989: /******************* Gnuplot file **************/
1.201 brouard 8990: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8991:
8992: char dirfileres[132],optfileres[132];
1.164 brouard 8993:
1.126 brouard 8994: int ng;
8995:
8996:
8997: /*#ifdef windows */
8998: fprintf(ficgp,"cd \"%s\" \n",pathc);
8999: /*#endif */
9000:
9001:
9002: strcpy(dirfileres,optionfilefiname);
9003: strcpy(optfileres,"vpl");
1.199 brouard 9004: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9005: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9006: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9007: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9008: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9009:
9010: }
9011:
1.136 brouard 9012: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9013: {
1.126 brouard 9014:
1.136 brouard 9015: /*-------- data file ----------*/
9016: FILE *fic;
9017: char dummy[]=" ";
1.240 brouard 9018: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9019: int lstra;
1.136 brouard 9020: int linei, month, year,iout;
9021: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9022: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9023: char *stratrunc;
1.223 brouard 9024:
1.240 brouard 9025: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9026: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9027:
1.240 brouard 9028: for(v=1; v <=ncovcol;v++){
9029: DummyV[v]=0;
9030: FixedV[v]=0;
9031: }
9032: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9033: DummyV[v]=1;
9034: FixedV[v]=0;
9035: }
9036: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9037: DummyV[v]=0;
9038: FixedV[v]=1;
9039: }
9040: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9041: DummyV[v]=1;
9042: FixedV[v]=1;
9043: }
9044: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9045: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9046: 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]);
9047: }
1.126 brouard 9048:
1.136 brouard 9049: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9050: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9051: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9052: }
1.126 brouard 9053:
1.136 brouard 9054: i=1;
9055: linei=0;
9056: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9057: linei=linei+1;
9058: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9059: if(line[j] == '\t')
9060: line[j] = ' ';
9061: }
9062: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9063: ;
9064: };
9065: line[j+1]=0; /* Trims blanks at end of line */
9066: if(line[0]=='#'){
9067: fprintf(ficlog,"Comment line\n%s\n",line);
9068: printf("Comment line\n%s\n",line);
9069: continue;
9070: }
9071: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9072: strcpy(line, linetmp);
1.223 brouard 9073:
9074: /* Loops on waves */
9075: for (j=maxwav;j>=1;j--){
9076: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9077: cutv(stra, strb, line, ' ');
9078: if(strb[0]=='.') { /* Missing value */
9079: lval=-1;
9080: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9081: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9082: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9083: 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);
9084: 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);
9085: return 1;
9086: }
9087: }else{
9088: errno=0;
9089: /* what_kind_of_number(strb); */
9090: dval=strtod(strb,&endptr);
9091: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9092: /* if(strb != endptr && *endptr == '\0') */
9093: /* dval=dlval; */
9094: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9095: if( strb[0]=='\0' || (*endptr != '\0')){
9096: 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);
9097: 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);
9098: return 1;
9099: }
9100: cotqvar[j][iv][i]=dval;
9101: cotvar[j][ntv+iv][i]=dval;
9102: }
9103: strcpy(line,stra);
1.223 brouard 9104: }/* end loop ntqv */
1.225 brouard 9105:
1.223 brouard 9106: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9107: cutv(stra, strb, line, ' ');
9108: if(strb[0]=='.') { /* Missing value */
9109: lval=-1;
9110: }else{
9111: errno=0;
9112: lval=strtol(strb,&endptr,10);
9113: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9114: if( strb[0]=='\0' || (*endptr != '\0')){
9115: 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);
9116: 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);
9117: return 1;
9118: }
9119: }
9120: if(lval <-1 || lval >1){
9121: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9122: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9123: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9124: For example, for multinomial values like 1, 2 and 3,\n \
9125: build V1=0 V2=0 for the reference value (1),\n \
9126: V1=1 V2=0 for (2) \n \
1.223 brouard 9127: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9128: output of IMaCh is often meaningless.\n \
1.223 brouard 9129: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9130: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9131: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9132: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9133: For example, for multinomial values like 1, 2 and 3,\n \
9134: build V1=0 V2=0 for the reference value (1),\n \
9135: V1=1 V2=0 for (2) \n \
1.223 brouard 9136: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9137: output of IMaCh is often meaningless.\n \
1.223 brouard 9138: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9139: return 1;
9140: }
9141: cotvar[j][iv][i]=(double)(lval);
9142: strcpy(line,stra);
1.223 brouard 9143: }/* end loop ntv */
1.225 brouard 9144:
1.223 brouard 9145: /* Statuses at wave */
1.137 brouard 9146: cutv(stra, strb, line, ' ');
1.223 brouard 9147: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9148: lval=-1;
1.136 brouard 9149: }else{
1.238 brouard 9150: errno=0;
9151: lval=strtol(strb,&endptr,10);
9152: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9153: if( strb[0]=='\0' || (*endptr != '\0')){
9154: 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);
9155: 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);
9156: return 1;
9157: }
1.136 brouard 9158: }
1.225 brouard 9159:
1.136 brouard 9160: s[j][i]=lval;
1.225 brouard 9161:
1.223 brouard 9162: /* Date of Interview */
1.136 brouard 9163: strcpy(line,stra);
9164: cutv(stra, strb,line,' ');
1.169 brouard 9165: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9166: }
1.169 brouard 9167: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9168: month=99;
9169: year=9999;
1.136 brouard 9170: }else{
1.225 brouard 9171: 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);
9172: 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);
9173: return 1;
1.136 brouard 9174: }
9175: anint[j][i]= (double) year;
9176: mint[j][i]= (double)month;
9177: strcpy(line,stra);
1.223 brouard 9178: } /* End loop on waves */
1.225 brouard 9179:
1.223 brouard 9180: /* Date of death */
1.136 brouard 9181: cutv(stra, strb,line,' ');
1.169 brouard 9182: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9183: }
1.169 brouard 9184: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9185: month=99;
9186: year=9999;
9187: }else{
1.141 brouard 9188: 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 9189: 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);
9190: return 1;
1.136 brouard 9191: }
9192: andc[i]=(double) year;
9193: moisdc[i]=(double) month;
9194: strcpy(line,stra);
9195:
1.223 brouard 9196: /* Date of birth */
1.136 brouard 9197: cutv(stra, strb,line,' ');
1.169 brouard 9198: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9199: }
1.169 brouard 9200: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9201: month=99;
9202: year=9999;
9203: }else{
1.141 brouard 9204: 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);
9205: 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 9206: return 1;
1.136 brouard 9207: }
9208: if (year==9999) {
1.141 brouard 9209: 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);
9210: 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 9211: return 1;
9212:
1.136 brouard 9213: }
9214: annais[i]=(double)(year);
9215: moisnais[i]=(double)(month);
9216: strcpy(line,stra);
1.225 brouard 9217:
1.223 brouard 9218: /* Sample weight */
1.136 brouard 9219: cutv(stra, strb,line,' ');
9220: errno=0;
9221: dval=strtod(strb,&endptr);
9222: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9223: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9224: 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 9225: fflush(ficlog);
9226: return 1;
9227: }
9228: weight[i]=dval;
9229: strcpy(line,stra);
1.225 brouard 9230:
1.223 brouard 9231: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9232: cutv(stra, strb, line, ' ');
9233: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9234: lval=-1;
1.223 brouard 9235: }else{
1.225 brouard 9236: errno=0;
9237: /* what_kind_of_number(strb); */
9238: dval=strtod(strb,&endptr);
9239: /* if(strb != endptr && *endptr == '\0') */
9240: /* dval=dlval; */
9241: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9242: if( strb[0]=='\0' || (*endptr != '\0')){
9243: 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);
9244: 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);
9245: return 1;
9246: }
9247: coqvar[iv][i]=dval;
1.226 brouard 9248: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9249: }
9250: strcpy(line,stra);
9251: }/* end loop nqv */
1.136 brouard 9252:
1.223 brouard 9253: /* Covariate values */
1.136 brouard 9254: for (j=ncovcol;j>=1;j--){
9255: cutv(stra, strb,line,' ');
1.223 brouard 9256: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9257: lval=-1;
1.136 brouard 9258: }else{
1.225 brouard 9259: errno=0;
9260: lval=strtol(strb,&endptr,10);
9261: if( strb[0]=='\0' || (*endptr != '\0')){
9262: 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);
9263: 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);
9264: return 1;
9265: }
1.136 brouard 9266: }
9267: if(lval <-1 || lval >1){
1.225 brouard 9268: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9269: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9270: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9271: For example, for multinomial values like 1, 2 and 3,\n \
9272: build V1=0 V2=0 for the reference value (1),\n \
9273: V1=1 V2=0 for (2) \n \
1.136 brouard 9274: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9275: output of IMaCh is often meaningless.\n \
1.136 brouard 9276: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9277: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9278: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9279: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9280: For example, for multinomial values like 1, 2 and 3,\n \
9281: build V1=0 V2=0 for the reference value (1),\n \
9282: V1=1 V2=0 for (2) \n \
1.136 brouard 9283: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9284: output of IMaCh is often meaningless.\n \
1.136 brouard 9285: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9286: return 1;
1.136 brouard 9287: }
9288: covar[j][i]=(double)(lval);
9289: strcpy(line,stra);
9290: }
9291: lstra=strlen(stra);
1.225 brouard 9292:
1.136 brouard 9293: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9294: stratrunc = &(stra[lstra-9]);
9295: num[i]=atol(stratrunc);
9296: }
9297: else
9298: num[i]=atol(stra);
9299: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9300: 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;}*/
9301:
9302: i=i+1;
9303: } /* End loop reading data */
1.225 brouard 9304:
1.136 brouard 9305: *imax=i-1; /* Number of individuals */
9306: fclose(fic);
1.225 brouard 9307:
1.136 brouard 9308: return (0);
1.164 brouard 9309: /* endread: */
1.225 brouard 9310: printf("Exiting readdata: ");
9311: fclose(fic);
9312: return (1);
1.223 brouard 9313: }
1.126 brouard 9314:
1.234 brouard 9315: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9316: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9317: while (*p2 == ' ')
1.234 brouard 9318: p2++;
9319: /* while ((*p1++ = *p2++) !=0) */
9320: /* ; */
9321: /* do */
9322: /* while (*p2 == ' ') */
9323: /* p2++; */
9324: /* while (*p1++ == *p2++); */
9325: *stri=p2;
1.145 brouard 9326: }
9327:
1.235 brouard 9328: int decoderesult ( char resultline[], int nres)
1.230 brouard 9329: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9330: {
1.235 brouard 9331: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9332: char resultsav[MAXLINE];
1.234 brouard 9333: int resultmodel[MAXLINE];
9334: int modelresult[MAXLINE];
1.230 brouard 9335: char stra[80], strb[80], strc[80], strd[80],stre[80];
9336:
1.234 brouard 9337: removefirstspace(&resultline);
1.233 brouard 9338: printf("decoderesult:%s\n",resultline);
1.230 brouard 9339:
9340: if (strstr(resultline,"v") !=0){
9341: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9342: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9343: return 1;
9344: }
9345: trimbb(resultsav, resultline);
9346: if (strlen(resultsav) >1){
9347: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9348: }
1.253 brouard 9349: if(j == 0){ /* Resultline but no = */
9350: TKresult[nres]=0; /* Combination for the nresult and the model */
9351: return (0);
9352: }
9353:
1.234 brouard 9354: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9355: 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);
9356: 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);
9357: }
9358: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9359: if(nbocc(resultsav,'=') >1){
9360: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9361: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9362: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9363: }else
9364: cutl(strc,strd,resultsav,'=');
1.230 brouard 9365: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9366:
1.230 brouard 9367: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9368: Tvarsel[k]=atoi(strc);
9369: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9370: /* cptcovsel++; */
9371: if (nbocc(stra,'=') >0)
9372: strcpy(resultsav,stra); /* and analyzes it */
9373: }
1.235 brouard 9374: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9375: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9376: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9377: match=0;
1.236 brouard 9378: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9379: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9380: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9381: match=1;
9382: break;
9383: }
9384: }
9385: if(match == 0){
9386: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9387: }
9388: }
9389: }
1.235 brouard 9390: /* Checking for missing or useless values in comparison of current model needs */
9391: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9392: match=0;
1.235 brouard 9393: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9394: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9395: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9396: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9397: ++match;
9398: }
9399: }
9400: }
9401: if(match == 0){
9402: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9403: }else if(match > 1){
9404: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9405: }
9406: }
1.235 brouard 9407:
1.234 brouard 9408: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9409: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9410: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9411: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9412: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9413: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9414: /* 1 0 0 0 */
9415: /* 2 1 0 0 */
9416: /* 3 0 1 0 */
9417: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9418: /* 5 0 0 1 */
9419: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9420: /* 7 0 1 1 */
9421: /* 8 1 1 1 */
1.237 brouard 9422: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9423: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9424: /* V5*age V5 known which value for nres? */
9425: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9426: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9427: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9428: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9429: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9430: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9431: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9432: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9433: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9434: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9435: k4++;;
9436: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9437: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9438: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9439: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9440: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9441: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9442: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9443: k4q++;;
9444: }
9445: }
1.234 brouard 9446:
1.235 brouard 9447: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9448: return (0);
9449: }
1.235 brouard 9450:
1.230 brouard 9451: int decodemodel( char model[], int lastobs)
9452: /**< This routine decodes the model and returns:
1.224 brouard 9453: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9454: * - nagesqr = 1 if age*age in the model, otherwise 0.
9455: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9456: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9457: * - cptcovage number of covariates with age*products =2
9458: * - cptcovs number of simple covariates
9459: * - 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
9460: * which is a new column after the 9 (ncovcol) variables.
9461: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9462: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9463: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9464: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9465: */
1.136 brouard 9466: {
1.238 brouard 9467: int i, j, k, ks, v;
1.227 brouard 9468: int j1, k1, k2, k3, k4;
1.136 brouard 9469: char modelsav[80];
1.145 brouard 9470: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9471: char *strpt;
1.136 brouard 9472:
1.145 brouard 9473: /*removespace(model);*/
1.136 brouard 9474: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9475: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9476: if (strstr(model,"AGE") !=0){
1.192 brouard 9477: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9478: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9479: return 1;
9480: }
1.141 brouard 9481: if (strstr(model,"v") !=0){
9482: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9483: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9484: return 1;
9485: }
1.187 brouard 9486: strcpy(modelsav,model);
9487: if ((strpt=strstr(model,"age*age")) !=0){
9488: printf(" strpt=%s, model=%s\n",strpt, model);
9489: if(strpt != model){
1.234 brouard 9490: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9491: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9492: corresponding column of parameters.\n",model);
1.234 brouard 9493: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9494: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9495: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9496: return 1;
1.225 brouard 9497: }
1.187 brouard 9498: nagesqr=1;
9499: if (strstr(model,"+age*age") !=0)
1.234 brouard 9500: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9501: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9502: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9503: else
1.234 brouard 9504: substrchaine(modelsav, model, "age*age");
1.187 brouard 9505: }else
9506: nagesqr=0;
9507: if (strlen(modelsav) >1){
9508: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9509: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9510: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9511: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9512: * cst, age and age*age
9513: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9514: /* including age products which are counted in cptcovage.
9515: * but the covariates which are products must be treated
9516: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9517: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9518: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9519:
9520:
1.187 brouard 9521: /* Design
9522: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9523: * < ncovcol=8 >
9524: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9525: * k= 1 2 3 4 5 6 7 8
9526: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9527: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9528: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9529: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9530: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9531: * Tage[++cptcovage]=k
9532: * if products, new covar are created after ncovcol with k1
9533: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9534: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9535: * 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
9536: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9537: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9538: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9539: * < ncovcol=8 >
9540: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9541: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9542: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9543: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9544: * p Tprod[1]@2={ 6, 5}
9545: *p Tvard[1][1]@4= {7, 8, 5, 6}
9546: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9547: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9548: *How to reorganize?
9549: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9550: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9551: * {2, 1, 4, 8, 5, 6, 3, 7}
9552: * Struct []
9553: */
1.225 brouard 9554:
1.187 brouard 9555: /* This loop fills the array Tvar from the string 'model'.*/
9556: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9557: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9558: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9559: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9560: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9561: /* k=1 Tvar[1]=2 (from V2) */
9562: /* k=5 Tvar[5] */
9563: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9564: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9565: /* } */
1.198 brouard 9566: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9567: /*
9568: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9569: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9570: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9571: }
1.187 brouard 9572: cptcovage=0;
9573: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9574: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9575: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9576: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9577: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9578: /*scanf("%d",i);*/
9579: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9580: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9581: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9582: /* covar is not filled and then is empty */
9583: cptcovprod--;
9584: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9585: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9586: Typevar[k]=1; /* 1 for age product */
9587: cptcovage++; /* Sums the number of covariates which include age as a product */
9588: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9589: /*printf("stre=%s ", stre);*/
9590: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9591: cptcovprod--;
9592: cutl(stre,strb,strc,'V');
9593: Tvar[k]=atoi(stre);
9594: Typevar[k]=1; /* 1 for age product */
9595: cptcovage++;
9596: Tage[cptcovage]=k;
9597: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9598: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9599: cptcovn++;
9600: cptcovprodnoage++;k1++;
9601: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9602: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9603: because this model-covariate is a construction we invent a new column
9604: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9605: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9606: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9607: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9608: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9609: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9610: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9611: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9612: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9613: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9614: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9615: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9616: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9617: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9618: for (i=1; i<=lastobs;i++){
9619: /* Computes the new covariate which is a product of
9620: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9621: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9622: }
9623: } /* End age is not in the model */
9624: } /* End if model includes a product */
9625: else { /* no more sum */
9626: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9627: /* scanf("%d",i);*/
9628: cutl(strd,strc,strb,'V');
9629: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9630: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9631: Tvar[k]=atoi(strd);
9632: Typevar[k]=0; /* 0 for simple covariates */
9633: }
9634: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9635: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9636: scanf("%d",i);*/
1.187 brouard 9637: } /* end of loop + on total covariates */
9638: } /* end if strlen(modelsave == 0) age*age might exist */
9639: } /* end if strlen(model == 0) */
1.136 brouard 9640:
9641: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9642: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9643:
1.136 brouard 9644: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9645: printf("cptcovprod=%d ", cptcovprod);
9646: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9647: scanf("%d ",i);*/
9648:
9649:
1.230 brouard 9650: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9651: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9652: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9653: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9654: k = 1 2 3 4 5 6 7 8 9
9655: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9656: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9657: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9658: Dummy[k] 1 0 0 0 3 1 1 2 3
9659: Tmodelind[combination of covar]=k;
1.225 brouard 9660: */
9661: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9662: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9663: /* 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 9664: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9665: printf("Model=%s\n\
9666: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9667: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9668: 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);
9669: fprintf(ficlog,"Model=%s\n\
9670: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9671: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9672: 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);
1.240 brouard 9673: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9674: 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 */
9675: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9676: Fixed[k]= 0;
9677: Dummy[k]= 0;
1.225 brouard 9678: ncoveff++;
1.232 brouard 9679: ncovf++;
1.234 brouard 9680: nsd++;
9681: modell[k].maintype= FTYPE;
9682: TvarsD[nsd]=Tvar[k];
9683: TvarsDind[nsd]=k;
9684: TvarF[ncovf]=Tvar[k];
9685: TvarFind[ncovf]=k;
9686: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9687: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9688: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9689: Fixed[k]= 0;
9690: Dummy[k]= 0;
9691: ncoveff++;
9692: ncovf++;
9693: modell[k].maintype= FTYPE;
9694: TvarF[ncovf]=Tvar[k];
9695: TvarFind[ncovf]=k;
1.230 brouard 9696: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9697: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9698: }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 9699: Fixed[k]= 0;
9700: Dummy[k]= 1;
1.230 brouard 9701: nqfveff++;
1.234 brouard 9702: modell[k].maintype= FTYPE;
9703: modell[k].subtype= FQ;
9704: nsq++;
9705: TvarsQ[nsq]=Tvar[k];
9706: TvarsQind[nsq]=k;
1.232 brouard 9707: ncovf++;
1.234 brouard 9708: TvarF[ncovf]=Tvar[k];
9709: TvarFind[ncovf]=k;
1.231 brouard 9710: 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 9711: 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 9712: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9713: Fixed[k]= 1;
9714: Dummy[k]= 0;
1.225 brouard 9715: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9716: modell[k].maintype= VTYPE;
9717: modell[k].subtype= VD;
9718: nsd++;
9719: TvarsD[nsd]=Tvar[k];
9720: TvarsDind[nsd]=k;
9721: ncovv++; /* Only simple time varying variables */
9722: TvarV[ncovv]=Tvar[k];
1.242 brouard 9723: 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 9724: 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 */
9725: 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 9726: 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);
9727: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9728: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9729: Fixed[k]= 1;
9730: Dummy[k]= 1;
9731: nqtveff++;
9732: modell[k].maintype= VTYPE;
9733: modell[k].subtype= VQ;
9734: ncovv++; /* Only simple time varying variables */
9735: nsq++;
9736: TvarsQ[nsq]=Tvar[k];
9737: TvarsQind[nsq]=k;
9738: TvarV[ncovv]=Tvar[k];
1.242 brouard 9739: 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 9740: 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 */
9741: 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 9742: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9743: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9744: 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 9745: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9746: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9747: ncova++;
9748: TvarA[ncova]=Tvar[k];
9749: TvarAind[ncova]=k;
1.231 brouard 9750: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9751: Fixed[k]= 2;
9752: Dummy[k]= 2;
9753: modell[k].maintype= ATYPE;
9754: modell[k].subtype= APFD;
9755: /* ncoveff++; */
1.227 brouard 9756: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9757: Fixed[k]= 2;
9758: Dummy[k]= 3;
9759: modell[k].maintype= ATYPE;
9760: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9761: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9762: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9763: Fixed[k]= 3;
9764: Dummy[k]= 2;
9765: modell[k].maintype= ATYPE;
9766: modell[k].subtype= APVD; /* Product age * varying dummy */
9767: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9768: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9769: Fixed[k]= 3;
9770: Dummy[k]= 3;
9771: modell[k].maintype= ATYPE;
9772: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9773: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9774: }
9775: }else if (Typevar[k] == 2) { /* product without age */
9776: k1=Tposprod[k];
9777: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9778: if(Tvard[k1][2] <=ncovcol){
9779: Fixed[k]= 1;
9780: Dummy[k]= 0;
9781: modell[k].maintype= FTYPE;
9782: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9783: ncovf++; /* Fixed variables without age */
9784: TvarF[ncovf]=Tvar[k];
9785: TvarFind[ncovf]=k;
9786: }else if(Tvard[k1][2] <=ncovcol+nqv){
9787: Fixed[k]= 0; /* or 2 ?*/
9788: Dummy[k]= 1;
9789: modell[k].maintype= FTYPE;
9790: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9791: ncovf++; /* Varying variables without age */
9792: TvarF[ncovf]=Tvar[k];
9793: TvarFind[ncovf]=k;
9794: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9795: Fixed[k]= 1;
9796: Dummy[k]= 0;
9797: modell[k].maintype= VTYPE;
9798: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9799: ncovv++; /* Varying variables without age */
9800: TvarV[ncovv]=Tvar[k];
9801: TvarVind[ncovv]=k;
9802: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9803: Fixed[k]= 1;
9804: Dummy[k]= 1;
9805: modell[k].maintype= VTYPE;
9806: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9807: ncovv++; /* Varying variables without age */
9808: TvarV[ncovv]=Tvar[k];
9809: TvarVind[ncovv]=k;
9810: }
1.227 brouard 9811: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9812: if(Tvard[k1][2] <=ncovcol){
9813: Fixed[k]= 0; /* or 2 ?*/
9814: Dummy[k]= 1;
9815: modell[k].maintype= FTYPE;
9816: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9817: ncovf++; /* Fixed variables without age */
9818: TvarF[ncovf]=Tvar[k];
9819: TvarFind[ncovf]=k;
9820: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9821: Fixed[k]= 1;
9822: Dummy[k]= 1;
9823: modell[k].maintype= VTYPE;
9824: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9825: ncovv++; /* Varying variables without age */
9826: TvarV[ncovv]=Tvar[k];
9827: TvarVind[ncovv]=k;
9828: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9829: Fixed[k]= 1;
9830: Dummy[k]= 1;
9831: modell[k].maintype= VTYPE;
9832: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9833: ncovv++; /* Varying variables without age */
9834: TvarV[ncovv]=Tvar[k];
9835: TvarVind[ncovv]=k;
9836: ncovv++; /* Varying variables without age */
9837: TvarV[ncovv]=Tvar[k];
9838: TvarVind[ncovv]=k;
9839: }
1.227 brouard 9840: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9841: if(Tvard[k1][2] <=ncovcol){
9842: Fixed[k]= 1;
9843: Dummy[k]= 1;
9844: modell[k].maintype= VTYPE;
9845: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9846: ncovv++; /* Varying variables without age */
9847: TvarV[ncovv]=Tvar[k];
9848: TvarVind[ncovv]=k;
9849: }else if(Tvard[k1][2] <=ncovcol+nqv){
9850: Fixed[k]= 1;
9851: Dummy[k]= 1;
9852: modell[k].maintype= VTYPE;
9853: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9854: ncovv++; /* Varying variables without age */
9855: TvarV[ncovv]=Tvar[k];
9856: TvarVind[ncovv]=k;
9857: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9858: Fixed[k]= 1;
9859: Dummy[k]= 0;
9860: modell[k].maintype= VTYPE;
9861: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9862: ncovv++; /* Varying variables without age */
9863: TvarV[ncovv]=Tvar[k];
9864: TvarVind[ncovv]=k;
9865: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9866: Fixed[k]= 1;
9867: Dummy[k]= 1;
9868: modell[k].maintype= VTYPE;
9869: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9870: ncovv++; /* Varying variables without age */
9871: TvarV[ncovv]=Tvar[k];
9872: TvarVind[ncovv]=k;
9873: }
1.227 brouard 9874: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9875: if(Tvard[k1][2] <=ncovcol){
9876: Fixed[k]= 1;
9877: Dummy[k]= 1;
9878: modell[k].maintype= VTYPE;
9879: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9880: ncovv++; /* Varying variables without age */
9881: TvarV[ncovv]=Tvar[k];
9882: TvarVind[ncovv]=k;
9883: }else if(Tvard[k1][2] <=ncovcol+nqv){
9884: Fixed[k]= 1;
9885: Dummy[k]= 1;
9886: modell[k].maintype= VTYPE;
9887: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9888: ncovv++; /* Varying variables without age */
9889: TvarV[ncovv]=Tvar[k];
9890: TvarVind[ncovv]=k;
9891: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9892: Fixed[k]= 1;
9893: Dummy[k]= 1;
9894: modell[k].maintype= VTYPE;
9895: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9896: ncovv++; /* Varying variables without age */
9897: TvarV[ncovv]=Tvar[k];
9898: TvarVind[ncovv]=k;
9899: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9900: Fixed[k]= 1;
9901: Dummy[k]= 1;
9902: modell[k].maintype= VTYPE;
9903: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9904: ncovv++; /* Varying variables without age */
9905: TvarV[ncovv]=Tvar[k];
9906: TvarVind[ncovv]=k;
9907: }
1.227 brouard 9908: }else{
1.240 brouard 9909: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9910: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9911: } /*end k1*/
1.225 brouard 9912: }else{
1.226 brouard 9913: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9914: 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 9915: }
1.227 brouard 9916: 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 9917: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9918: 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]);
9919: }
9920: /* Searching for doublons in the model */
9921: for(k1=1; k1<= cptcovt;k1++){
9922: for(k2=1; k2 <k1;k2++){
9923: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9924: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9925: if(Tvar[k1]==Tvar[k2]){
9926: 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[Tvar[k1]],Dummy[Tvar[k1]]);
9927: 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[Tvar[k1]],Dummy[Tvar[k1]]); fflush(ficlog);
9928: return(1);
9929: }
9930: }else if (Typevar[k1] ==2){
9931: k3=Tposprod[k1];
9932: k4=Tposprod[k2];
9933: 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])) ){
9934: 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]]);
9935: 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);
9936: return(1);
9937: }
9938: }
1.227 brouard 9939: }
9940: }
1.225 brouard 9941: }
9942: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9943: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9944: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9945: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9946: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9947: /*endread:*/
1.225 brouard 9948: printf("Exiting decodemodel: ");
9949: return (1);
1.136 brouard 9950: }
9951:
1.169 brouard 9952: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9953: {/* Check ages at death */
1.136 brouard 9954: int i, m;
1.218 brouard 9955: int firstone=0;
9956:
1.136 brouard 9957: for (i=1; i<=imx; i++) {
9958: for(m=2; (m<= maxwav); m++) {
9959: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9960: anint[m][i]=9999;
1.216 brouard 9961: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9962: s[m][i]=-1;
1.136 brouard 9963: }
9964: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9965: *nberr = *nberr + 1;
1.218 brouard 9966: if(firstone == 0){
9967: firstone=1;
1.260 brouard 9968: 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 9969: }
1.262 brouard 9970: 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 9971: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9972: }
9973: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9974: (*nberr)++;
1.259 brouard 9975: 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 9976: 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 9977: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9978: }
9979: }
9980: }
9981:
9982: for (i=1; i<=imx; i++) {
9983: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9984: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9985: 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 9986: if (s[m][i] >= nlstate+1) {
1.169 brouard 9987: if(agedc[i]>0){
9988: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9989: agev[m][i]=agedc[i];
1.214 brouard 9990: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9991: }else {
1.136 brouard 9992: if ((int)andc[i]!=9999){
9993: nbwarn++;
9994: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9995: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9996: agev[m][i]=-1;
9997: }
9998: }
1.169 brouard 9999: } /* agedc > 0 */
1.214 brouard 10000: } /* end if */
1.136 brouard 10001: else if(s[m][i] !=9){ /* Standard case, age in fractional
10002: years but with the precision of a month */
10003: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10004: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10005: agev[m][i]=1;
10006: else if(agev[m][i] < *agemin){
10007: *agemin=agev[m][i];
10008: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10009: }
10010: else if(agev[m][i] >*agemax){
10011: *agemax=agev[m][i];
1.156 brouard 10012: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10013: }
10014: /*agev[m][i]=anint[m][i]-annais[i];*/
10015: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10016: } /* en if 9*/
1.136 brouard 10017: else { /* =9 */
1.214 brouard 10018: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10019: agev[m][i]=1;
10020: s[m][i]=-1;
10021: }
10022: }
1.214 brouard 10023: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10024: agev[m][i]=1;
1.214 brouard 10025: else{
10026: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10027: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10028: agev[m][i]=0;
10029: }
10030: } /* End for lastpass */
10031: }
1.136 brouard 10032:
10033: for (i=1; i<=imx; i++) {
10034: for(m=firstpass; (m<=lastpass); m++){
10035: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10036: (*nberr)++;
1.136 brouard 10037: 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);
10038: 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);
10039: return 1;
10040: }
10041: }
10042: }
10043:
10044: /*for (i=1; i<=imx; i++){
10045: for (m=firstpass; (m<lastpass); m++){
10046: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10047: }
10048:
10049: }*/
10050:
10051:
1.139 brouard 10052: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10053: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10054:
10055: return (0);
1.164 brouard 10056: /* endread:*/
1.136 brouard 10057: printf("Exiting calandcheckages: ");
10058: return (1);
10059: }
10060:
1.172 brouard 10061: #if defined(_MSC_VER)
10062: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10063: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10064: //#include "stdafx.h"
10065: //#include <stdio.h>
10066: //#include <tchar.h>
10067: //#include <windows.h>
10068: //#include <iostream>
10069: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10070:
10071: LPFN_ISWOW64PROCESS fnIsWow64Process;
10072:
10073: BOOL IsWow64()
10074: {
10075: BOOL bIsWow64 = FALSE;
10076:
10077: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10078: // (HANDLE, PBOOL);
10079:
10080: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10081:
10082: HMODULE module = GetModuleHandle(_T("kernel32"));
10083: const char funcName[] = "IsWow64Process";
10084: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10085: GetProcAddress(module, funcName);
10086:
10087: if (NULL != fnIsWow64Process)
10088: {
10089: if (!fnIsWow64Process(GetCurrentProcess(),
10090: &bIsWow64))
10091: //throw std::exception("Unknown error");
10092: printf("Unknown error\n");
10093: }
10094: return bIsWow64 != FALSE;
10095: }
10096: #endif
1.177 brouard 10097:
1.191 brouard 10098: void syscompilerinfo(int logged)
1.167 brouard 10099: {
10100: /* #include "syscompilerinfo.h"*/
1.185 brouard 10101: /* command line Intel compiler 32bit windows, XP compatible:*/
10102: /* /GS /W3 /Gy
10103: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10104: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10105: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10106: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10107: */
10108: /* 64 bits */
1.185 brouard 10109: /*
10110: /GS /W3 /Gy
10111: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10112: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10113: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10114: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10115: /* Optimization are useless and O3 is slower than O2 */
10116: /*
10117: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10118: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10119: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10120: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10121: */
1.186 brouard 10122: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10123: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10124: /PDB:"visual studio
10125: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10126: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10127: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10128: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10129: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10130: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10131: uiAccess='false'"
10132: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10133: /NOLOGO /TLBID:1
10134: */
1.177 brouard 10135: #if defined __INTEL_COMPILER
1.178 brouard 10136: #if defined(__GNUC__)
10137: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10138: #endif
1.177 brouard 10139: #elif defined(__GNUC__)
1.179 brouard 10140: #ifndef __APPLE__
1.174 brouard 10141: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10142: #endif
1.177 brouard 10143: struct utsname sysInfo;
1.178 brouard 10144: int cross = CROSS;
10145: if (cross){
10146: printf("Cross-");
1.191 brouard 10147: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10148: }
1.174 brouard 10149: #endif
10150:
1.171 brouard 10151: #include <stdint.h>
1.178 brouard 10152:
1.191 brouard 10153: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10154: #if defined(__clang__)
1.191 brouard 10155: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10156: #endif
10157: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10158: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10159: #endif
10160: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10161: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10162: #endif
10163: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10164: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10165: #endif
10166: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10167: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10168: #endif
10169: #if defined(_MSC_VER)
1.191 brouard 10170: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10171: #endif
10172: #if defined(__PGI)
1.191 brouard 10173: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10174: #endif
10175: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10176: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10177: #endif
1.191 brouard 10178: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10179:
1.167 brouard 10180: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10181: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10182: // Windows (x64 and x86)
1.191 brouard 10183: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10184: #elif __unix__ // all unices, not all compilers
10185: // Unix
1.191 brouard 10186: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10187: #elif __linux__
10188: // linux
1.191 brouard 10189: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10190: #elif __APPLE__
1.174 brouard 10191: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10192: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10193: #endif
10194:
10195: /* __MINGW32__ */
10196: /* __CYGWIN__ */
10197: /* __MINGW64__ */
10198: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10199: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10200: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10201: /* _WIN64 // Defined for applications for Win64. */
10202: /* _M_X64 // Defined for compilations that target x64 processors. */
10203: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10204:
1.167 brouard 10205: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10206: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10207: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10208: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10209: #else
1.191 brouard 10210: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10211: #endif
10212:
1.169 brouard 10213: #if defined(__GNUC__)
10214: # if defined(__GNUC_PATCHLEVEL__)
10215: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10216: + __GNUC_MINOR__ * 100 \
10217: + __GNUC_PATCHLEVEL__)
10218: # else
10219: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10220: + __GNUC_MINOR__ * 100)
10221: # endif
1.174 brouard 10222: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10223: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10224:
10225: if (uname(&sysInfo) != -1) {
10226: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10227: 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 10228: }
10229: else
10230: perror("uname() error");
1.179 brouard 10231: //#ifndef __INTEL_COMPILER
10232: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10233: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10234: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10235: #endif
1.169 brouard 10236: #endif
1.172 brouard 10237:
10238: // void main()
10239: // {
1.169 brouard 10240: #if defined(_MSC_VER)
1.174 brouard 10241: if (IsWow64()){
1.191 brouard 10242: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10243: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10244: }
10245: else{
1.191 brouard 10246: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10247: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10248: }
1.172 brouard 10249: // printf("\nPress Enter to continue...");
10250: // getchar();
10251: // }
10252:
1.169 brouard 10253: #endif
10254:
1.167 brouard 10255:
1.219 brouard 10256: }
1.136 brouard 10257:
1.219 brouard 10258: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10259: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10260: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10261: /* double ftolpl = 1.e-10; */
1.180 brouard 10262: double age, agebase, agelim;
1.203 brouard 10263: double tot;
1.180 brouard 10264:
1.202 brouard 10265: strcpy(filerespl,"PL_");
10266: strcat(filerespl,fileresu);
10267: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10268: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10269: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10270: }
1.227 brouard 10271: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10272: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10273: pstamp(ficrespl);
1.203 brouard 10274: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10275: fprintf(ficrespl,"#Age ");
10276: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10277: fprintf(ficrespl,"\n");
1.180 brouard 10278:
1.219 brouard 10279: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10280:
1.219 brouard 10281: agebase=ageminpar;
10282: agelim=agemaxpar;
1.180 brouard 10283:
1.227 brouard 10284: /* i1=pow(2,ncoveff); */
1.234 brouard 10285: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10286: if (cptcovn < 1){i1=1;}
1.180 brouard 10287:
1.238 brouard 10288: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10289: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10290: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10291: continue;
1.235 brouard 10292:
1.238 brouard 10293: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10294: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10295: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10296: /* k=k+1; */
10297: /* to clean */
10298: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10299: fprintf(ficrespl,"#******");
10300: printf("#******");
10301: fprintf(ficlog,"#******");
10302: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10303: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10304: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10305: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10306: }
10307: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10308: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10309: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10310: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10311: }
10312: fprintf(ficrespl,"******\n");
10313: printf("******\n");
10314: fprintf(ficlog,"******\n");
10315: if(invalidvarcomb[k]){
10316: printf("\nCombination (%d) ignored because no case \n",k);
10317: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10318: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10319: continue;
10320: }
1.219 brouard 10321:
1.238 brouard 10322: fprintf(ficrespl,"#Age ");
10323: for(j=1;j<=cptcoveff;j++) {
10324: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10325: }
10326: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10327: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10328:
1.238 brouard 10329: for (age=agebase; age<=agelim; age++){
10330: /* for (age=agebase; age<=agebase; age++){ */
10331: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10332: fprintf(ficrespl,"%.0f ",age );
10333: for(j=1;j<=cptcoveff;j++)
10334: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10335: tot=0.;
10336: for(i=1; i<=nlstate;i++){
10337: tot += prlim[i][i];
10338: fprintf(ficrespl," %.5f", prlim[i][i]);
10339: }
10340: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10341: } /* Age */
10342: /* was end of cptcod */
10343: } /* cptcov */
10344: } /* nres */
1.219 brouard 10345: return 0;
1.180 brouard 10346: }
10347:
1.218 brouard 10348: 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){
10349: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10350:
10351: /* Computes the back prevalence limit for any combination of covariate values
10352: * at any age between ageminpar and agemaxpar
10353: */
1.235 brouard 10354: int i, j, k, i1, nres=0 ;
1.217 brouard 10355: /* double ftolpl = 1.e-10; */
10356: double age, agebase, agelim;
10357: double tot;
1.218 brouard 10358: /* double ***mobaverage; */
10359: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10360:
10361: strcpy(fileresplb,"PLB_");
10362: strcat(fileresplb,fileresu);
10363: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10364: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10365: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10366: }
10367: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10368: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10369: pstamp(ficresplb);
10370: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10371: fprintf(ficresplb,"#Age ");
10372: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10373: fprintf(ficresplb,"\n");
10374:
1.218 brouard 10375:
10376: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10377:
10378: agebase=ageminpar;
10379: agelim=agemaxpar;
10380:
10381:
1.227 brouard 10382: i1=pow(2,cptcoveff);
1.218 brouard 10383: if (cptcovn < 1){i1=1;}
1.227 brouard 10384:
1.238 brouard 10385: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10386: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10387: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10388: continue;
10389: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10390: fprintf(ficresplb,"#******");
10391: printf("#******");
10392: fprintf(ficlog,"#******");
10393: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10394: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10395: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10396: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10397: }
10398: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10399: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10400: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10401: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10402: }
10403: fprintf(ficresplb,"******\n");
10404: printf("******\n");
10405: fprintf(ficlog,"******\n");
10406: if(invalidvarcomb[k]){
10407: printf("\nCombination (%d) ignored because no cases \n",k);
10408: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10409: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10410: continue;
10411: }
1.218 brouard 10412:
1.238 brouard 10413: fprintf(ficresplb,"#Age ");
10414: for(j=1;j<=cptcoveff;j++) {
10415: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10416: }
10417: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10418: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10419:
10420:
1.238 brouard 10421: for (age=agebase; age<=agelim; age++){
10422: /* for (age=agebase; age<=agebase; age++){ */
10423: if(mobilavproj > 0){
10424: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10425: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10426: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10427: }else if (mobilavproj == 0){
10428: 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);
10429: 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);
10430: exit(1);
10431: }else{
10432: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10433: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10434: /* printf("TOTOT\n"); */
10435: /* exit(1); */
1.238 brouard 10436: }
10437: fprintf(ficresplb,"%.0f ",age );
10438: for(j=1;j<=cptcoveff;j++)
10439: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10440: tot=0.;
10441: for(i=1; i<=nlstate;i++){
10442: tot += bprlim[i][i];
10443: fprintf(ficresplb," %.5f", bprlim[i][i]);
10444: }
10445: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10446: } /* Age */
10447: /* was end of cptcod */
1.255 brouard 10448: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10449: } /* end of any combination */
10450: } /* end of nres */
1.218 brouard 10451: /* hBijx(p, bage, fage); */
10452: /* fclose(ficrespijb); */
10453:
10454: return 0;
1.217 brouard 10455: }
1.218 brouard 10456:
1.180 brouard 10457: int hPijx(double *p, int bage, int fage){
10458: /*------------- h Pij x at various ages ------------*/
10459:
10460: int stepsize;
10461: int agelim;
10462: int hstepm;
10463: int nhstepm;
1.235 brouard 10464: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10465:
10466: double agedeb;
10467: double ***p3mat;
10468:
1.201 brouard 10469: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10470: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10471: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10472: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10473: }
10474: printf("Computing pij: result on file '%s' \n", filerespij);
10475: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10476:
10477: stepsize=(int) (stepm+YEARM-1)/YEARM;
10478: /*if (stepm<=24) stepsize=2;*/
10479:
10480: agelim=AGESUP;
10481: hstepm=stepsize*YEARM; /* Every year of age */
10482: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10483:
1.180 brouard 10484: /* hstepm=1; aff par mois*/
10485: pstamp(ficrespij);
10486: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10487: i1= pow(2,cptcoveff);
1.218 brouard 10488: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10489: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10490: /* k=k+1; */
1.235 brouard 10491: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10492: for(k=1; k<=i1;k++){
1.253 brouard 10493: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10494: continue;
1.183 brouard 10495: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10496: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10497: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10498: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10499: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10500: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10501: }
1.183 brouard 10502: fprintf(ficrespij,"******\n");
10503:
10504: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10505: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10506: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10507:
10508: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10509:
1.183 brouard 10510: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10511: oldm=oldms;savm=savms;
1.235 brouard 10512: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10513: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10514: for(i=1; i<=nlstate;i++)
10515: for(j=1; j<=nlstate+ndeath;j++)
10516: fprintf(ficrespij," %1d-%1d",i,j);
10517: fprintf(ficrespij,"\n");
10518: for (h=0; h<=nhstepm; h++){
10519: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10520: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10521: for(i=1; i<=nlstate;i++)
10522: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10523: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10524: fprintf(ficrespij,"\n");
10525: }
1.183 brouard 10526: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10527: fprintf(ficrespij,"\n");
10528: }
1.180 brouard 10529: /*}*/
10530: }
1.218 brouard 10531: return 0;
1.180 brouard 10532: }
1.218 brouard 10533:
10534: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10535: /*------------- h Bij x at various ages ------------*/
10536:
10537: int stepsize;
1.218 brouard 10538: /* int agelim; */
10539: int ageminl;
1.217 brouard 10540: int hstepm;
10541: int nhstepm;
1.238 brouard 10542: int h, i, i1, j, k, nres;
1.218 brouard 10543:
1.217 brouard 10544: double agedeb;
10545: double ***p3mat;
1.218 brouard 10546:
10547: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10548: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10549: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10550: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10551: }
10552: printf("Computing pij back: result on file '%s' \n", filerespijb);
10553: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10554:
10555: stepsize=(int) (stepm+YEARM-1)/YEARM;
10556: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10557:
1.218 brouard 10558: /* agelim=AGESUP; */
10559: ageminl=30;
10560: hstepm=stepsize*YEARM; /* Every year of age */
10561: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10562:
10563: /* hstepm=1; aff par mois*/
10564: pstamp(ficrespijb);
1.255 brouard 10565: 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 10566: i1= pow(2,cptcoveff);
1.218 brouard 10567: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10568: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10569: /* k=k+1; */
1.238 brouard 10570: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10571: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10572: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10573: continue;
10574: fprintf(ficrespijb,"\n#****** ");
10575: for(j=1;j<=cptcoveff;j++)
10576: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10577: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10578: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10579: }
10580: fprintf(ficrespijb,"******\n");
1.264 brouard 10581: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10582: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10583: continue;
10584: }
10585:
10586: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10587: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10588: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10589: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10590: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10591:
10592: /* nhstepm=nhstepm*YEARM; aff par mois*/
10593:
1.266 brouard 10594: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10595: /* and memory limitations if stepm is small */
10596:
1.238 brouard 10597: /* oldm=oldms;savm=savms; */
10598: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10599: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10600: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10601: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10602: for(i=1; i<=nlstate;i++)
10603: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10604: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10605: fprintf(ficrespijb,"\n");
1.238 brouard 10606: for (h=0; h<=nhstepm; h++){
10607: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10608: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10609: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10610: for(i=1; i<=nlstate;i++)
10611: for(j=1; j<=nlstate+ndeath;j++)
10612: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10613: fprintf(ficrespijb,"\n");
10614: }
10615: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10616: fprintf(ficrespijb,"\n");
10617: } /* end age deb */
10618: } /* end combination */
10619: } /* end nres */
1.218 brouard 10620: return 0;
10621: } /* hBijx */
1.217 brouard 10622:
1.180 brouard 10623:
1.136 brouard 10624: /***********************************************/
10625: /**************** Main Program *****************/
10626: /***********************************************/
10627:
10628: int main(int argc, char *argv[])
10629: {
10630: #ifdef GSL
10631: const gsl_multimin_fminimizer_type *T;
10632: size_t iteri = 0, it;
10633: int rval = GSL_CONTINUE;
10634: int status = GSL_SUCCESS;
10635: double ssval;
10636: #endif
10637: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10638: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10639: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10640: int jj, ll, li, lj, lk;
1.136 brouard 10641: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10642: int num_filled;
1.136 brouard 10643: int itimes;
10644: int NDIM=2;
10645: int vpopbased=0;
1.235 brouard 10646: int nres=0;
1.258 brouard 10647: int endishere=0;
1.277 brouard 10648: int noffset=0;
1.274 brouard 10649: int ncurrv=0; /* Temporary variable */
10650:
1.164 brouard 10651: char ca[32], cb[32];
1.136 brouard 10652: /* FILE *fichtm; *//* Html File */
10653: /* FILE *ficgp;*/ /*Gnuplot File */
10654: struct stat info;
1.191 brouard 10655: double agedeb=0.;
1.194 brouard 10656:
10657: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10658: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10659:
1.165 brouard 10660: double fret;
1.191 brouard 10661: double dum=0.; /* Dummy variable */
1.136 brouard 10662: double ***p3mat;
1.218 brouard 10663: /* double ***mobaverage; */
1.164 brouard 10664:
10665: char line[MAXLINE];
1.197 brouard 10666: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10667:
1.234 brouard 10668: char modeltemp[MAXLINE];
1.230 brouard 10669: char resultline[MAXLINE];
10670:
1.136 brouard 10671: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10672: char *tok, *val; /* pathtot */
1.136 brouard 10673: int firstobs=1, lastobs=10;
1.195 brouard 10674: int c, h , cpt, c2;
1.191 brouard 10675: int jl=0;
10676: int i1, j1, jk, stepsize=0;
1.194 brouard 10677: int count=0;
10678:
1.164 brouard 10679: int *tab;
1.136 brouard 10680: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10681: int backcast=0;
1.136 brouard 10682: int mobilav=0,popforecast=0;
1.191 brouard 10683: int hstepm=0, nhstepm=0;
1.136 brouard 10684: int agemortsup;
10685: float sumlpop=0.;
10686: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10687: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10688:
1.191 brouard 10689: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10690: double ftolpl=FTOL;
10691: double **prlim;
1.217 brouard 10692: double **bprlim;
1.136 brouard 10693: double ***param; /* Matrix of parameters */
1.251 brouard 10694: double ***paramstart; /* Matrix of starting parameter values */
10695: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10696: double **matcov; /* Matrix of covariance */
1.203 brouard 10697: double **hess; /* Hessian matrix */
1.136 brouard 10698: double ***delti3; /* Scale */
10699: double *delti; /* Scale */
10700: double ***eij, ***vareij;
10701: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10702:
1.136 brouard 10703: double *epj, vepp;
1.164 brouard 10704:
1.273 brouard 10705: double dateprev1, dateprev2;
10706: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10707: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10708:
1.136 brouard 10709: double **ximort;
1.145 brouard 10710: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10711: int *dcwave;
10712:
1.164 brouard 10713: char z[1]="c";
1.136 brouard 10714:
10715: /*char *strt;*/
10716: char strtend[80];
1.126 brouard 10717:
1.164 brouard 10718:
1.126 brouard 10719: /* setlocale (LC_ALL, ""); */
10720: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10721: /* textdomain (PACKAGE); */
10722: /* setlocale (LC_CTYPE, ""); */
10723: /* setlocale (LC_MESSAGES, ""); */
10724:
10725: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10726: rstart_time = time(NULL);
10727: /* (void) gettimeofday(&start_time,&tzp);*/
10728: start_time = *localtime(&rstart_time);
1.126 brouard 10729: curr_time=start_time;
1.157 brouard 10730: /*tml = *localtime(&start_time.tm_sec);*/
10731: /* strcpy(strstart,asctime(&tml)); */
10732: strcpy(strstart,asctime(&start_time));
1.126 brouard 10733:
10734: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10735: /* tp.tm_sec = tp.tm_sec +86400; */
10736: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10737: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10738: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10739: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10740: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10741: /* strt=asctime(&tmg); */
10742: /* printf("Time(after) =%s",strstart); */
10743: /* (void) time (&time_value);
10744: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10745: * tm = *localtime(&time_value);
10746: * strstart=asctime(&tm);
10747: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10748: */
10749:
10750: nberr=0; /* Number of errors and warnings */
10751: nbwarn=0;
1.184 brouard 10752: #ifdef WIN32
10753: _getcwd(pathcd, size);
10754: #else
1.126 brouard 10755: getcwd(pathcd, size);
1.184 brouard 10756: #endif
1.191 brouard 10757: syscompilerinfo(0);
1.196 brouard 10758: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10759: if(argc <=1){
10760: printf("\nEnter the parameter file name: ");
1.205 brouard 10761: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10762: printf("ERROR Empty parameter file name\n");
10763: goto end;
10764: }
1.126 brouard 10765: i=strlen(pathr);
10766: if(pathr[i-1]=='\n')
10767: pathr[i-1]='\0';
1.156 brouard 10768: i=strlen(pathr);
1.205 brouard 10769: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10770: pathr[i-1]='\0';
1.205 brouard 10771: }
10772: i=strlen(pathr);
10773: if( i==0 ){
10774: printf("ERROR Empty parameter file name\n");
10775: goto end;
10776: }
10777: for (tok = pathr; tok != NULL; ){
1.126 brouard 10778: printf("Pathr |%s|\n",pathr);
10779: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10780: printf("val= |%s| pathr=%s\n",val,pathr);
10781: strcpy (pathtot, val);
10782: if(pathr[0] == '\0') break; /* Dirty */
10783: }
10784: }
10785: else{
10786: strcpy(pathtot,argv[1]);
10787: }
10788: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10789: /*cygwin_split_path(pathtot,path,optionfile);
10790: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10791: /* cutv(path,optionfile,pathtot,'\\');*/
10792:
10793: /* Split argv[0], imach program to get pathimach */
10794: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10795: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10796: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10797: /* strcpy(pathimach,argv[0]); */
10798: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10799: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10800: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10801: #ifdef WIN32
10802: _chdir(path); /* Can be a relative path */
10803: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10804: #else
1.126 brouard 10805: chdir(path); /* Can be a relative path */
1.184 brouard 10806: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10807: #endif
10808: printf("Current directory %s!\n",pathcd);
1.126 brouard 10809: strcpy(command,"mkdir ");
10810: strcat(command,optionfilefiname);
10811: if((outcmd=system(command)) != 0){
1.169 brouard 10812: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10813: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10814: /* fclose(ficlog); */
10815: /* exit(1); */
10816: }
10817: /* if((imk=mkdir(optionfilefiname))<0){ */
10818: /* perror("mkdir"); */
10819: /* } */
10820:
10821: /*-------- arguments in the command line --------*/
10822:
1.186 brouard 10823: /* Main Log file */
1.126 brouard 10824: strcat(filelog, optionfilefiname);
10825: strcat(filelog,".log"); /* */
10826: if((ficlog=fopen(filelog,"w"))==NULL) {
10827: printf("Problem with logfile %s\n",filelog);
10828: goto end;
10829: }
10830: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10831: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10832: fprintf(ficlog,"\nEnter the parameter file name: \n");
10833: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10834: path=%s \n\
10835: optionfile=%s\n\
10836: optionfilext=%s\n\
1.156 brouard 10837: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10838:
1.197 brouard 10839: syscompilerinfo(1);
1.167 brouard 10840:
1.126 brouard 10841: printf("Local time (at start):%s",strstart);
10842: fprintf(ficlog,"Local time (at start): %s",strstart);
10843: fflush(ficlog);
10844: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10845: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10846:
10847: /* */
10848: strcpy(fileres,"r");
10849: strcat(fileres, optionfilefiname);
1.201 brouard 10850: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10851: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10852: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10853:
1.186 brouard 10854: /* Main ---------arguments file --------*/
1.126 brouard 10855:
10856: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10857: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10858: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10859: fflush(ficlog);
1.149 brouard 10860: /* goto end; */
10861: exit(70);
1.126 brouard 10862: }
10863:
10864:
10865:
10866: strcpy(filereso,"o");
1.201 brouard 10867: strcat(filereso,fileresu);
1.126 brouard 10868: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10869: printf("Problem with Output resultfile: %s\n", filereso);
10870: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10871: fflush(ficlog);
10872: goto end;
10873: }
1.278 brouard 10874: /*-------- Rewriting parameter file ----------*/
10875: strcpy(rfileres,"r"); /* "Rparameterfile */
10876: strcat(rfileres,optionfilefiname); /* Parameter file first name */
10877: strcat(rfileres,"."); /* */
10878: strcat(rfileres,optionfilext); /* Other files have txt extension */
10879: if((ficres =fopen(rfileres,"w"))==NULL) {
10880: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10881: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
10882: fflush(ficlog);
10883: goto end;
10884: }
10885: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 10886:
1.278 brouard 10887:
1.126 brouard 10888: /* Reads comments: lines beginning with '#' */
10889: numlinepar=0;
1.277 brouard 10890: /* Is it a BOM UTF-8 Windows file? */
10891: /* First parameter line */
1.197 brouard 10892: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 10893: noffset=0;
10894: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10895: {
10896: noffset=noffset+3;
10897: printf("# File is an UTF8 Bom.\n"); // 0xBF
10898: }
10899: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
10900: {
10901: noffset=noffset+2;
10902: printf("# File is an UTF16BE BOM file\n");
10903: }
10904: else if( line[0] == 0 && line[1] == 0)
10905: {
10906: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10907: noffset=noffset+4;
10908: printf("# File is an UTF16BE BOM file\n");
10909: }
10910: } else{
10911: ;/*printf(" Not a BOM file\n");*/
10912: }
10913:
1.197 brouard 10914: /* If line starts with a # it is a comment */
1.277 brouard 10915: if (line[noffset] == '#') {
1.197 brouard 10916: numlinepar++;
10917: fputs(line,stdout);
10918: fputs(line,ficparo);
1.278 brouard 10919: fputs(line,ficres);
1.197 brouard 10920: fputs(line,ficlog);
10921: continue;
10922: }else
10923: break;
10924: }
10925: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10926: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10927: if (num_filled != 5) {
10928: printf("Should be 5 parameters\n");
10929: }
1.126 brouard 10930: numlinepar++;
1.197 brouard 10931: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10932: }
10933: /* Second parameter line */
10934: while(fgets(line, MAXLINE, ficpar)) {
10935: /* If line starts with a # it is a comment */
10936: if (line[0] == '#') {
10937: numlinepar++;
10938: fputs(line,stdout);
10939: fputs(line,ficparo);
1.278 brouard 10940: fputs(line,ficres);
1.197 brouard 10941: fputs(line,ficlog);
10942: continue;
10943: }else
10944: break;
10945: }
1.223 brouard 10946: 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", \
10947: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10948: if (num_filled != 11) {
10949: 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 10950: printf("but line=%s\n",line);
1.197 brouard 10951: }
1.223 brouard 10952: 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);
1.126 brouard 10953: }
1.203 brouard 10954: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10955: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10956: /* Third parameter line */
10957: while(fgets(line, MAXLINE, ficpar)) {
10958: /* If line starts with a # it is a comment */
10959: if (line[0] == '#') {
10960: numlinepar++;
10961: fputs(line,stdout);
10962: fputs(line,ficparo);
1.278 brouard 10963: fputs(line,ficres);
1.197 brouard 10964: fputs(line,ficlog);
10965: continue;
10966: }else
10967: break;
10968: }
1.201 brouard 10969: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 ! brouard 10970: if (num_filled != 1){
! 10971: printf("ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
! 10972: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
1.197 brouard 10973: model[0]='\0';
10974: goto end;
10975: }
10976: else{
10977: if (model[0]=='+'){
10978: for(i=1; i<=strlen(model);i++)
10979: modeltemp[i-1]=model[i];
1.201 brouard 10980: strcpy(model,modeltemp);
1.197 brouard 10981: }
10982: }
1.199 brouard 10983: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10984: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10985: }
10986: /* 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); */
10987: /* numlinepar=numlinepar+3; /\* In general *\/ */
10988: /* 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); */
1.223 brouard 10989: 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);
10990: 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 10991: fflush(ficlog);
1.190 brouard 10992: /* if(model[0]=='#'|| model[0]== '\0'){ */
10993: if(model[0]=='#'){
1.279 ! brouard 10994: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
! 10995: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
! 10996: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 10997: if(mle != -1){
1.279 ! brouard 10998: 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 10999: exit(1);
11000: }
11001: }
1.126 brouard 11002: while((c=getc(ficpar))=='#' && c!= EOF){
11003: ungetc(c,ficpar);
11004: fgets(line, MAXLINE, ficpar);
11005: numlinepar++;
1.195 brouard 11006: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11007: z[0]=line[1];
11008: }
11009: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11010: fputs(line, stdout);
11011: //puts(line);
1.126 brouard 11012: fputs(line,ficparo);
11013: fputs(line,ficlog);
11014: }
11015: ungetc(c,ficpar);
11016:
11017:
1.145 brouard 11018: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 11019: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
11020: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
11021: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11022: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11023: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11024: v1+v2*age+v2*v3 makes cptcovn = 3
11025: */
11026: if (strlen(model)>1)
1.187 brouard 11027: 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 11028: else
1.187 brouard 11029: ncovmodel=2; /* Constant and age */
1.133 brouard 11030: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11031: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11032: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11033: 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);
11034: 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);
11035: fflush(stdout);
11036: fclose (ficlog);
11037: goto end;
11038: }
1.126 brouard 11039: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11040: delti=delti3[1][1];
11041: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11042: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11043: /* We could also provide initial parameters values giving by simple logistic regression
11044: * only one way, that is without matrix product. We will have nlstate maximizations */
11045: /* for(i=1;i<nlstate;i++){ */
11046: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11047: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11048: /* } */
1.126 brouard 11049: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11050: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11051: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11052: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11053: fclose (ficparo);
11054: fclose (ficlog);
11055: goto end;
11056: exit(0);
1.220 brouard 11057: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11058: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11059: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11060: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11061: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11062: matcov=matrix(1,npar,1,npar);
1.203 brouard 11063: hess=matrix(1,npar,1,npar);
1.220 brouard 11064: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11065: /* Read guessed parameters */
1.126 brouard 11066: /* Reads comments: lines beginning with '#' */
11067: while((c=getc(ficpar))=='#' && c!= EOF){
11068: ungetc(c,ficpar);
11069: fgets(line, MAXLINE, ficpar);
11070: numlinepar++;
1.141 brouard 11071: fputs(line,stdout);
1.126 brouard 11072: fputs(line,ficparo);
11073: fputs(line,ficlog);
11074: }
11075: ungetc(c,ficpar);
11076:
11077: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11078: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11079: for(i=1; i <=nlstate; i++){
1.234 brouard 11080: j=0;
1.126 brouard 11081: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11082: if(jj==i) continue;
11083: j++;
11084: fscanf(ficpar,"%1d%1d",&i1,&j1);
11085: if ((i1 != i) || (j1 != jj)){
11086: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11087: It might be a problem of design; if ncovcol and the model are correct\n \
11088: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11089: exit(1);
11090: }
11091: fprintf(ficparo,"%1d%1d",i1,j1);
11092: if(mle==1)
11093: printf("%1d%1d",i,jj);
11094: fprintf(ficlog,"%1d%1d",i,jj);
11095: for(k=1; k<=ncovmodel;k++){
11096: fscanf(ficpar," %lf",¶m[i][j][k]);
11097: if(mle==1){
11098: printf(" %lf",param[i][j][k]);
11099: fprintf(ficlog," %lf",param[i][j][k]);
11100: }
11101: else
11102: fprintf(ficlog," %lf",param[i][j][k]);
11103: fprintf(ficparo," %lf",param[i][j][k]);
11104: }
11105: fscanf(ficpar,"\n");
11106: numlinepar++;
11107: if(mle==1)
11108: printf("\n");
11109: fprintf(ficlog,"\n");
11110: fprintf(ficparo,"\n");
1.126 brouard 11111: }
11112: }
11113: fflush(ficlog);
1.234 brouard 11114:
1.251 brouard 11115: /* Reads parameters values */
1.126 brouard 11116: p=param[1][1];
1.251 brouard 11117: pstart=paramstart[1][1];
1.126 brouard 11118:
11119: /* Reads comments: lines beginning with '#' */
11120: while((c=getc(ficpar))=='#' && c!= EOF){
11121: ungetc(c,ficpar);
11122: fgets(line, MAXLINE, ficpar);
11123: numlinepar++;
1.141 brouard 11124: fputs(line,stdout);
1.126 brouard 11125: fputs(line,ficparo);
11126: fputs(line,ficlog);
11127: }
11128: ungetc(c,ficpar);
11129:
11130: for(i=1; i <=nlstate; i++){
11131: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11132: fscanf(ficpar,"%1d%1d",&i1,&j1);
11133: if ( (i1-i) * (j1-j) != 0){
11134: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11135: exit(1);
11136: }
11137: printf("%1d%1d",i,j);
11138: fprintf(ficparo,"%1d%1d",i1,j1);
11139: fprintf(ficlog,"%1d%1d",i1,j1);
11140: for(k=1; k<=ncovmodel;k++){
11141: fscanf(ficpar,"%le",&delti3[i][j][k]);
11142: printf(" %le",delti3[i][j][k]);
11143: fprintf(ficparo," %le",delti3[i][j][k]);
11144: fprintf(ficlog," %le",delti3[i][j][k]);
11145: }
11146: fscanf(ficpar,"\n");
11147: numlinepar++;
11148: printf("\n");
11149: fprintf(ficparo,"\n");
11150: fprintf(ficlog,"\n");
1.126 brouard 11151: }
11152: }
11153: fflush(ficlog);
1.234 brouard 11154:
1.145 brouard 11155: /* Reads covariance matrix */
1.126 brouard 11156: delti=delti3[1][1];
1.220 brouard 11157:
11158:
1.126 brouard 11159: /* 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 11160:
1.126 brouard 11161: /* Reads comments: lines beginning with '#' */
11162: while((c=getc(ficpar))=='#' && c!= EOF){
11163: ungetc(c,ficpar);
11164: fgets(line, MAXLINE, ficpar);
11165: numlinepar++;
1.141 brouard 11166: fputs(line,stdout);
1.126 brouard 11167: fputs(line,ficparo);
11168: fputs(line,ficlog);
11169: }
11170: ungetc(c,ficpar);
1.220 brouard 11171:
1.126 brouard 11172: matcov=matrix(1,npar,1,npar);
1.203 brouard 11173: hess=matrix(1,npar,1,npar);
1.131 brouard 11174: for(i=1; i <=npar; i++)
11175: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11176:
1.194 brouard 11177: /* Scans npar lines */
1.126 brouard 11178: for(i=1; i <=npar; i++){
1.226 brouard 11179: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11180: if(count != 3){
1.226 brouard 11181: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11182: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11183: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11184: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11185: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11186: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11187: exit(1);
1.220 brouard 11188: }else{
1.226 brouard 11189: if(mle==1)
11190: printf("%1d%1d%d",i1,j1,jk);
11191: }
11192: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11193: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11194: for(j=1; j <=i; j++){
1.226 brouard 11195: fscanf(ficpar," %le",&matcov[i][j]);
11196: if(mle==1){
11197: printf(" %.5le",matcov[i][j]);
11198: }
11199: fprintf(ficlog," %.5le",matcov[i][j]);
11200: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11201: }
11202: fscanf(ficpar,"\n");
11203: numlinepar++;
11204: if(mle==1)
1.220 brouard 11205: printf("\n");
1.126 brouard 11206: fprintf(ficlog,"\n");
11207: fprintf(ficparo,"\n");
11208: }
1.194 brouard 11209: /* End of read covariance matrix npar lines */
1.126 brouard 11210: for(i=1; i <=npar; i++)
11211: for(j=i+1;j<=npar;j++)
1.226 brouard 11212: matcov[i][j]=matcov[j][i];
1.126 brouard 11213:
11214: if(mle==1)
11215: printf("\n");
11216: fprintf(ficlog,"\n");
11217:
11218: fflush(ficlog);
11219:
11220: } /* End of mle != -3 */
1.218 brouard 11221:
1.186 brouard 11222: /* Main data
11223: */
1.126 brouard 11224: n= lastobs;
11225: num=lvector(1,n);
11226: moisnais=vector(1,n);
11227: annais=vector(1,n);
11228: moisdc=vector(1,n);
11229: andc=vector(1,n);
1.220 brouard 11230: weight=vector(1,n);
1.126 brouard 11231: agedc=vector(1,n);
11232: cod=ivector(1,n);
1.220 brouard 11233: for(i=1;i<=n;i++){
1.234 brouard 11234: num[i]=0;
11235: moisnais[i]=0;
11236: annais[i]=0;
11237: moisdc[i]=0;
11238: andc[i]=0;
11239: agedc[i]=0;
11240: cod[i]=0;
11241: weight[i]=1.0; /* Equal weights, 1 by default */
11242: }
1.126 brouard 11243: mint=matrix(1,maxwav,1,n);
11244: anint=matrix(1,maxwav,1,n);
1.131 brouard 11245: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11246: tab=ivector(1,NCOVMAX);
1.144 brouard 11247: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11248: 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 11249:
1.136 brouard 11250: /* Reads data from file datafile */
11251: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11252: goto end;
11253:
11254: /* Calculation of the number of parameters from char model */
1.234 brouard 11255: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11256: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11257: k=3 V4 Tvar[k=3]= 4 (from V4)
11258: k=2 V1 Tvar[k=2]= 1 (from V1)
11259: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11260: */
11261:
11262: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11263: TvarsDind=ivector(1,NCOVMAX); /* */
11264: TvarsD=ivector(1,NCOVMAX); /* */
11265: TvarsQind=ivector(1,NCOVMAX); /* */
11266: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11267: TvarF=ivector(1,NCOVMAX); /* */
11268: TvarFind=ivector(1,NCOVMAX); /* */
11269: TvarV=ivector(1,NCOVMAX); /* */
11270: TvarVind=ivector(1,NCOVMAX); /* */
11271: TvarA=ivector(1,NCOVMAX); /* */
11272: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11273: TvarFD=ivector(1,NCOVMAX); /* */
11274: TvarFDind=ivector(1,NCOVMAX); /* */
11275: TvarFQ=ivector(1,NCOVMAX); /* */
11276: TvarFQind=ivector(1,NCOVMAX); /* */
11277: TvarVD=ivector(1,NCOVMAX); /* */
11278: TvarVDind=ivector(1,NCOVMAX); /* */
11279: TvarVQ=ivector(1,NCOVMAX); /* */
11280: TvarVQind=ivector(1,NCOVMAX); /* */
11281:
1.230 brouard 11282: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11283: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11284: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11285: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11286: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11287: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11288: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11289: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11290: */
11291: /* For model-covariate k tells which data-covariate to use but
11292: because this model-covariate is a construction we invent a new column
11293: ncovcol + k1
11294: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11295: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11296: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11297: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11298: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11299: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11300: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11301: */
1.145 brouard 11302: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11303: 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 11304: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11305: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11306: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11307: 4 covariates (3 plus signs)
11308: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11309: */
1.230 brouard 11310: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11311: * individual dummy, fixed or varying:
11312: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11313: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11314: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11315: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11316: * Tmodelind[1]@9={9,0,3,2,}*/
11317: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11318: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11319: * individual quantitative, fixed or varying:
11320: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11321: * 3, 1, 0, 0, 0, 0, 0, 0},
11322: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11323: /* Main decodemodel */
11324:
1.187 brouard 11325:
1.223 brouard 11326: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11327: goto end;
11328:
1.137 brouard 11329: if((double)(lastobs-imx)/(double)imx > 1.10){
11330: nbwarn++;
11331: 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);
11332: 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);
11333: }
1.136 brouard 11334: /* if(mle==1){*/
1.137 brouard 11335: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11336: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11337: }
11338:
11339: /*-calculation of age at interview from date of interview and age at death -*/
11340: agev=matrix(1,maxwav,1,imx);
11341:
11342: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11343: goto end;
11344:
1.126 brouard 11345:
1.136 brouard 11346: agegomp=(int)agemin;
11347: free_vector(moisnais,1,n);
11348: free_vector(annais,1,n);
1.126 brouard 11349: /* free_matrix(mint,1,maxwav,1,n);
11350: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11351: /* free_vector(moisdc,1,n); */
11352: /* free_vector(andc,1,n); */
1.145 brouard 11353: /* */
11354:
1.126 brouard 11355: wav=ivector(1,imx);
1.214 brouard 11356: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11357: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11358: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11359: 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.*/
11360: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11361: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11362:
11363: /* Concatenates waves */
1.214 brouard 11364: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11365: Death is a valid wave (if date is known).
11366: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11367: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11368: and mw[mi+1][i]. dh depends on stepm.
11369: */
11370:
1.126 brouard 11371: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11372: /* Concatenates waves */
1.145 brouard 11373:
1.215 brouard 11374: free_vector(moisdc,1,n);
11375: free_vector(andc,1,n);
11376:
1.126 brouard 11377: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11378: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11379: ncodemax[1]=1;
1.145 brouard 11380: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11381: cptcoveff=0;
1.220 brouard 11382: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11383: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11384: }
11385:
11386: ncovcombmax=pow(2,cptcoveff);
11387: invalidvarcomb=ivector(1, ncovcombmax);
11388: for(i=1;i<ncovcombmax;i++)
11389: invalidvarcomb[i]=0;
11390:
1.211 brouard 11391: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11392: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11393: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11394:
1.200 brouard 11395: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11396: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11397: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11398: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11399: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11400: * (currently 0 or 1) in the data.
11401: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11402: * corresponding modality (h,j).
11403: */
11404:
1.145 brouard 11405: h=0;
11406: /*if (cptcovn > 0) */
1.126 brouard 11407: m=pow(2,cptcoveff);
11408:
1.144 brouard 11409: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11410: * For k=4 covariates, h goes from 1 to m=2**k
11411: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11412: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11413: * h\k 1 2 3 4
1.143 brouard 11414: *______________________________
11415: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11416: * 2 2 1 1 1
11417: * 3 i=2 1 2 1 1
11418: * 4 2 2 1 1
11419: * 5 i=3 1 i=2 1 2 1
11420: * 6 2 1 2 1
11421: * 7 i=4 1 2 2 1
11422: * 8 2 2 2 1
1.197 brouard 11423: * 9 i=5 1 i=3 1 i=2 1 2
11424: * 10 2 1 1 2
11425: * 11 i=6 1 2 1 2
11426: * 12 2 2 1 2
11427: * 13 i=7 1 i=4 1 2 2
11428: * 14 2 1 2 2
11429: * 15 i=8 1 2 2 2
11430: * 16 2 2 2 2
1.143 brouard 11431: */
1.212 brouard 11432: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11433: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11434: * and the value of each covariate?
11435: * V1=1, V2=1, V3=2, V4=1 ?
11436: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11437: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11438: * In order to get the real value in the data, we use nbcode
11439: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11440: * We are keeping this crazy system in order to be able (in the future?)
11441: * to have more than 2 values (0 or 1) for a covariate.
11442: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11443: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11444: * bbbbbbbb
11445: * 76543210
11446: * h-1 00000101 (6-1=5)
1.219 brouard 11447: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11448: * &
11449: * 1 00000001 (1)
1.219 brouard 11450: * 00000000 = 1 & ((h-1) >> (k-1))
11451: * +1= 00000001 =1
1.211 brouard 11452: *
11453: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11454: * h' 1101 =2^3+2^2+0x2^1+2^0
11455: * >>k' 11
11456: * & 00000001
11457: * = 00000001
11458: * +1 = 00000010=2 = codtabm(14,3)
11459: * Reverse h=6 and m=16?
11460: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11461: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11462: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11463: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11464: * V3=decodtabm(14,3,2**4)=2
11465: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11466: *(h-1) >> (j-1) 0011 =13 >> 2
11467: * &1 000000001
11468: * = 000000001
11469: * +1= 000000010 =2
11470: * 2211
11471: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11472: * V3=2
1.220 brouard 11473: * codtabm and decodtabm are identical
1.211 brouard 11474: */
11475:
1.145 brouard 11476:
11477: free_ivector(Ndum,-1,NCOVMAX);
11478:
11479:
1.126 brouard 11480:
1.186 brouard 11481: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11482: strcpy(optionfilegnuplot,optionfilefiname);
11483: if(mle==-3)
1.201 brouard 11484: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11485: strcat(optionfilegnuplot,".gp");
11486:
11487: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11488: printf("Problem with file %s",optionfilegnuplot);
11489: }
11490: else{
1.204 brouard 11491: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11492: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11493: //fprintf(ficgp,"set missing 'NaNq'\n");
11494: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11495: }
11496: /* fclose(ficgp);*/
1.186 brouard 11497:
11498:
11499: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11500:
11501: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11502: if(mle==-3)
1.201 brouard 11503: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11504: strcat(optionfilehtm,".htm");
11505: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11506: printf("Problem with %s \n",optionfilehtm);
11507: exit(0);
1.126 brouard 11508: }
11509:
11510: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11511: strcat(optionfilehtmcov,"-cov.htm");
11512: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11513: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11514: }
11515: else{
11516: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11517: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11518: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11519: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11520: }
11521:
1.213 brouard 11522: 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 11523: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11524: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11525: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11526: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11527: \n\
11528: <hr size=\"2\" color=\"#EC5E5E\">\
11529: <ul><li><h4>Parameter files</h4>\n\
11530: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11531: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11532: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11533: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11534: - Date and time at start: %s</ul>\n",\
11535: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11536: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11537: fileres,fileres,\
11538: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11539: fflush(fichtm);
11540:
11541: strcpy(pathr,path);
11542: strcat(pathr,optionfilefiname);
1.184 brouard 11543: #ifdef WIN32
11544: _chdir(optionfilefiname); /* Move to directory named optionfile */
11545: #else
1.126 brouard 11546: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11547: #endif
11548:
1.126 brouard 11549:
1.220 brouard 11550: /* Calculates basic frequencies. Computes observed prevalence at single age
11551: and for any valid combination of covariates
1.126 brouard 11552: and prints on file fileres'p'. */
1.251 brouard 11553: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11554: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11555:
11556: fprintf(fichtm,"\n");
1.274 brouard 11557: fprintf(fichtm,"<h4>Parameter line 2</h4><ul><li>Tolerance for the convergence of the likelihood: ftol=%f \n<li>Interval for the elementary matrix (in month): stepm=%d",\
11558: ftol, stepm);
11559: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11560: ncurrv=1;
11561: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11562: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11563: ncurrv=i;
11564: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
11565: fprintf(fichtm,"\n<li> Number of time varying (wave varying) covariates: ntv=%d ", ntv);
11566: ncurrv=i;
11567: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
11568: fprintf(fichtm,"\n<li>Number of quantitative time varying covariates: nqtv=%d ", nqtv);
11569: ncurrv=i;
11570: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11571: 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", \
11572: nlstate, ndeath, maxwav, mle, weightopt);
11573:
11574: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11575: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11576:
11577:
11578: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11579: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11580: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11581: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11582: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11583: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11584: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11585: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11586: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11587:
1.126 brouard 11588: /* For Powell, parameters are in a vector p[] starting at p[1]
11589: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11590: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11591:
11592: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11593: /* For mortality only */
1.126 brouard 11594: if (mle==-3){
1.136 brouard 11595: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11596: for(i=1;i<=NDIM;i++)
11597: for(j=1;j<=NDIM;j++)
11598: ximort[i][j]=0.;
1.186 brouard 11599: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11600: cens=ivector(1,n);
11601: ageexmed=vector(1,n);
11602: agecens=vector(1,n);
11603: dcwave=ivector(1,n);
1.223 brouard 11604:
1.126 brouard 11605: for (i=1; i<=imx; i++){
11606: dcwave[i]=-1;
11607: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11608: if (s[m][i]>nlstate) {
11609: dcwave[i]=m;
11610: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11611: break;
11612: }
1.126 brouard 11613: }
1.226 brouard 11614:
1.126 brouard 11615: for (i=1; i<=imx; i++) {
11616: if (wav[i]>0){
1.226 brouard 11617: ageexmed[i]=agev[mw[1][i]][i];
11618: j=wav[i];
11619: agecens[i]=1.;
11620:
11621: if (ageexmed[i]> 1 && wav[i] > 0){
11622: agecens[i]=agev[mw[j][i]][i];
11623: cens[i]= 1;
11624: }else if (ageexmed[i]< 1)
11625: cens[i]= -1;
11626: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11627: cens[i]=0 ;
1.126 brouard 11628: }
11629: else cens[i]=-1;
11630: }
11631:
11632: for (i=1;i<=NDIM;i++) {
11633: for (j=1;j<=NDIM;j++)
1.226 brouard 11634: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11635: }
11636:
1.145 brouard 11637: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11638: /*printf("%lf %lf", p[1], p[2]);*/
11639:
11640:
1.136 brouard 11641: #ifdef GSL
11642: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11643: #else
1.126 brouard 11644: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11645: #endif
1.201 brouard 11646: strcpy(filerespow,"POW-MORT_");
11647: strcat(filerespow,fileresu);
1.126 brouard 11648: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11649: printf("Problem with resultfile: %s\n", filerespow);
11650: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11651: }
1.136 brouard 11652: #ifdef GSL
11653: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11654: #else
1.126 brouard 11655: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11656: #endif
1.126 brouard 11657: /* for (i=1;i<=nlstate;i++)
11658: for(j=1;j<=nlstate+ndeath;j++)
11659: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11660: */
11661: fprintf(ficrespow,"\n");
1.136 brouard 11662: #ifdef GSL
11663: /* gsl starts here */
11664: T = gsl_multimin_fminimizer_nmsimplex;
11665: gsl_multimin_fminimizer *sfm = NULL;
11666: gsl_vector *ss, *x;
11667: gsl_multimin_function minex_func;
11668:
11669: /* Initial vertex size vector */
11670: ss = gsl_vector_alloc (NDIM);
11671:
11672: if (ss == NULL){
11673: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11674: }
11675: /* Set all step sizes to 1 */
11676: gsl_vector_set_all (ss, 0.001);
11677:
11678: /* Starting point */
1.126 brouard 11679:
1.136 brouard 11680: x = gsl_vector_alloc (NDIM);
11681:
11682: if (x == NULL){
11683: gsl_vector_free(ss);
11684: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11685: }
11686:
11687: /* Initialize method and iterate */
11688: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11689: /* gsl_vector_set(x, 0, 0.0268); */
11690: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11691: gsl_vector_set(x, 0, p[1]);
11692: gsl_vector_set(x, 1, p[2]);
11693:
11694: minex_func.f = &gompertz_f;
11695: minex_func.n = NDIM;
11696: minex_func.params = (void *)&p; /* ??? */
11697:
11698: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11699: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11700:
11701: printf("Iterations beginning .....\n\n");
11702: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11703:
11704: iteri=0;
11705: while (rval == GSL_CONTINUE){
11706: iteri++;
11707: status = gsl_multimin_fminimizer_iterate(sfm);
11708:
11709: if (status) printf("error: %s\n", gsl_strerror (status));
11710: fflush(0);
11711:
11712: if (status)
11713: break;
11714:
11715: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11716: ssval = gsl_multimin_fminimizer_size (sfm);
11717:
11718: if (rval == GSL_SUCCESS)
11719: printf ("converged to a local maximum at\n");
11720:
11721: printf("%5d ", iteri);
11722: for (it = 0; it < NDIM; it++){
11723: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11724: }
11725: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11726: }
11727:
11728: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11729:
11730: gsl_vector_free(x); /* initial values */
11731: gsl_vector_free(ss); /* inital step size */
11732: for (it=0; it<NDIM; it++){
11733: p[it+1]=gsl_vector_get(sfm->x,it);
11734: fprintf(ficrespow," %.12lf", p[it]);
11735: }
11736: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11737: #endif
11738: #ifdef POWELL
11739: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11740: #endif
1.126 brouard 11741: fclose(ficrespow);
11742:
1.203 brouard 11743: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11744:
11745: for(i=1; i <=NDIM; i++)
11746: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11747: matcov[i][j]=matcov[j][i];
1.126 brouard 11748:
11749: printf("\nCovariance matrix\n ");
1.203 brouard 11750: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11751: for(i=1; i <=NDIM; i++) {
11752: for(j=1;j<=NDIM;j++){
1.220 brouard 11753: printf("%f ",matcov[i][j]);
11754: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11755: }
1.203 brouard 11756: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11757: }
11758:
11759: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11760: for (i=1;i<=NDIM;i++) {
1.126 brouard 11761: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11762: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11763: }
1.126 brouard 11764: lsurv=vector(1,AGESUP);
11765: lpop=vector(1,AGESUP);
11766: tpop=vector(1,AGESUP);
11767: lsurv[agegomp]=100000;
11768:
11769: for (k=agegomp;k<=AGESUP;k++) {
11770: agemortsup=k;
11771: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11772: }
11773:
11774: for (k=agegomp;k<agemortsup;k++)
11775: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11776:
11777: for (k=agegomp;k<agemortsup;k++){
11778: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11779: sumlpop=sumlpop+lpop[k];
11780: }
11781:
11782: tpop[agegomp]=sumlpop;
11783: for (k=agegomp;k<(agemortsup-3);k++){
11784: /* tpop[k+1]=2;*/
11785: tpop[k+1]=tpop[k]-lpop[k];
11786: }
11787:
11788:
11789: printf("\nAge lx qx dx Lx Tx e(x)\n");
11790: for (k=agegomp;k<(agemortsup-2);k++)
11791: 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]);
11792:
11793:
11794: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11795: ageminpar=50;
11796: agemaxpar=100;
1.194 brouard 11797: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11798: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11799: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11800: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11801: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11802: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11803: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11804: }else{
11805: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11806: 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 11807: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11808: }
1.201 brouard 11809: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11810: stepm, weightopt,\
11811: model,imx,p,matcov,agemortsup);
11812:
11813: free_vector(lsurv,1,AGESUP);
11814: free_vector(lpop,1,AGESUP);
11815: free_vector(tpop,1,AGESUP);
1.220 brouard 11816: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11817: free_ivector(cens,1,n);
11818: free_vector(agecens,1,n);
11819: free_ivector(dcwave,1,n);
1.220 brouard 11820: #ifdef GSL
1.136 brouard 11821: #endif
1.186 brouard 11822: } /* Endof if mle==-3 mortality only */
1.205 brouard 11823: /* Standard */
11824: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11825: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11826: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11827: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11828: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11829: for (k=1; k<=npar;k++)
11830: printf(" %d %8.5f",k,p[k]);
11831: printf("\n");
1.205 brouard 11832: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11833: /* mlikeli uses func not funcone */
1.247 brouard 11834: /* for(i=1;i<nlstate;i++){ */
11835: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11836: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11837: /* } */
1.205 brouard 11838: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11839: }
11840: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11841: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11842: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11843: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11844: }
11845: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11846: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11847: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11848: for (k=1; k<=npar;k++)
11849: printf(" %d %8.5f",k,p[k]);
11850: printf("\n");
11851:
11852: /*--------- results files --------------*/
1.224 brouard 11853: 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 11854:
11855:
11856: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11857: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11858: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11859: for(i=1,jk=1; i <=nlstate; i++){
11860: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11861: if (k != i) {
11862: printf("%d%d ",i,k);
11863: fprintf(ficlog,"%d%d ",i,k);
11864: fprintf(ficres,"%1d%1d ",i,k);
11865: for(j=1; j <=ncovmodel; j++){
11866: printf("%12.7f ",p[jk]);
11867: fprintf(ficlog,"%12.7f ",p[jk]);
11868: fprintf(ficres,"%12.7f ",p[jk]);
11869: jk++;
11870: }
11871: printf("\n");
11872: fprintf(ficlog,"\n");
11873: fprintf(ficres,"\n");
11874: }
1.126 brouard 11875: }
11876: }
1.203 brouard 11877: if(mle != 0){
11878: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11879: ftolhess=ftol; /* Usually correct */
1.203 brouard 11880: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11881: 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");
11882: 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");
11883: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11884: for(k=1; k <=(nlstate+ndeath); k++){
11885: if (k != i) {
11886: printf("%d%d ",i,k);
11887: fprintf(ficlog,"%d%d ",i,k);
11888: for(j=1; j <=ncovmodel; j++){
11889: 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]));
11890: 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]));
11891: jk++;
11892: }
11893: printf("\n");
11894: fprintf(ficlog,"\n");
11895: }
11896: }
1.193 brouard 11897: }
1.203 brouard 11898: } /* end of hesscov and Wald tests */
1.225 brouard 11899:
1.203 brouard 11900: /* */
1.126 brouard 11901: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11902: printf("# Scales (for hessian or gradient estimation)\n");
11903: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11904: for(i=1,jk=1; i <=nlstate; i++){
11905: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11906: if (j!=i) {
11907: fprintf(ficres,"%1d%1d",i,j);
11908: printf("%1d%1d",i,j);
11909: fprintf(ficlog,"%1d%1d",i,j);
11910: for(k=1; k<=ncovmodel;k++){
11911: printf(" %.5e",delti[jk]);
11912: fprintf(ficlog," %.5e",delti[jk]);
11913: fprintf(ficres," %.5e",delti[jk]);
11914: jk++;
11915: }
11916: printf("\n");
11917: fprintf(ficlog,"\n");
11918: fprintf(ficres,"\n");
11919: }
1.126 brouard 11920: }
11921: }
11922:
11923: 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 11924: if(mle >= 1) /* To big for the screen */
1.126 brouard 11925: 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");
11926: 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");
11927: /* # 121 Var(a12)\n\ */
11928: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11929: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11930: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11931: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11932: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11933: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11934: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11935:
11936:
11937: /* Just to have a covariance matrix which will be more understandable
11938: even is we still don't want to manage dictionary of variables
11939: */
11940: for(itimes=1;itimes<=2;itimes++){
11941: jj=0;
11942: for(i=1; i <=nlstate; i++){
1.225 brouard 11943: for(j=1; j <=nlstate+ndeath; j++){
11944: if(j==i) continue;
11945: for(k=1; k<=ncovmodel;k++){
11946: jj++;
11947: ca[0]= k+'a'-1;ca[1]='\0';
11948: if(itimes==1){
11949: if(mle>=1)
11950: printf("#%1d%1d%d",i,j,k);
11951: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11952: fprintf(ficres,"#%1d%1d%d",i,j,k);
11953: }else{
11954: if(mle>=1)
11955: printf("%1d%1d%d",i,j,k);
11956: fprintf(ficlog,"%1d%1d%d",i,j,k);
11957: fprintf(ficres,"%1d%1d%d",i,j,k);
11958: }
11959: ll=0;
11960: for(li=1;li <=nlstate; li++){
11961: for(lj=1;lj <=nlstate+ndeath; lj++){
11962: if(lj==li) continue;
11963: for(lk=1;lk<=ncovmodel;lk++){
11964: ll++;
11965: if(ll<=jj){
11966: cb[0]= lk +'a'-1;cb[1]='\0';
11967: if(ll<jj){
11968: if(itimes==1){
11969: if(mle>=1)
11970: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11971: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11972: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11973: }else{
11974: if(mle>=1)
11975: printf(" %.5e",matcov[jj][ll]);
11976: fprintf(ficlog," %.5e",matcov[jj][ll]);
11977: fprintf(ficres," %.5e",matcov[jj][ll]);
11978: }
11979: }else{
11980: if(itimes==1){
11981: if(mle>=1)
11982: printf(" Var(%s%1d%1d)",ca,i,j);
11983: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11984: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11985: }else{
11986: if(mle>=1)
11987: printf(" %.7e",matcov[jj][ll]);
11988: fprintf(ficlog," %.7e",matcov[jj][ll]);
11989: fprintf(ficres," %.7e",matcov[jj][ll]);
11990: }
11991: }
11992: }
11993: } /* end lk */
11994: } /* end lj */
11995: } /* end li */
11996: if(mle>=1)
11997: printf("\n");
11998: fprintf(ficlog,"\n");
11999: fprintf(ficres,"\n");
12000: numlinepar++;
12001: } /* end k*/
12002: } /*end j */
1.126 brouard 12003: } /* end i */
12004: } /* end itimes */
12005:
12006: fflush(ficlog);
12007: fflush(ficres);
1.225 brouard 12008: while(fgets(line, MAXLINE, ficpar)) {
12009: /* If line starts with a # it is a comment */
12010: if (line[0] == '#') {
12011: numlinepar++;
12012: fputs(line,stdout);
12013: fputs(line,ficparo);
12014: fputs(line,ficlog);
12015: continue;
12016: }else
12017: break;
12018: }
12019:
1.209 brouard 12020: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12021: /* ungetc(c,ficpar); */
12022: /* fgets(line, MAXLINE, ficpar); */
12023: /* fputs(line,stdout); */
12024: /* fputs(line,ficparo); */
12025: /* } */
12026: /* ungetc(c,ficpar); */
1.126 brouard 12027:
12028: estepm=0;
1.209 brouard 12029: 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 12030:
12031: if (num_filled != 6) {
12032: 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);
12033: 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);
12034: goto end;
12035: }
12036: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12037: }
12038: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12039: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12040:
1.209 brouard 12041: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12042: if (estepm==0 || estepm < stepm) estepm=stepm;
12043: if (fage <= 2) {
12044: bage = ageminpar;
12045: fage = agemaxpar;
12046: }
12047:
12048: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12049: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12050: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12051:
1.186 brouard 12052: /* Other stuffs, more or less useful */
1.254 brouard 12053: while(fgets(line, MAXLINE, ficpar)) {
12054: /* If line starts with a # it is a comment */
12055: if (line[0] == '#') {
12056: numlinepar++;
12057: fputs(line,stdout);
12058: fputs(line,ficparo);
12059: fputs(line,ficlog);
12060: continue;
12061: }else
12062: break;
12063: }
12064:
12065: 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){
12066:
12067: if (num_filled != 7) {
12068: 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);
12069: 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);
12070: goto end;
12071: }
12072: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12073: 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);
12074: 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);
12075: 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 12076: }
1.254 brouard 12077:
12078: while(fgets(line, MAXLINE, ficpar)) {
12079: /* If line starts with a # it is a comment */
12080: if (line[0] == '#') {
12081: numlinepar++;
12082: fputs(line,stdout);
12083: fputs(line,ficparo);
12084: fputs(line,ficlog);
12085: continue;
12086: }else
12087: break;
1.126 brouard 12088: }
12089:
12090:
12091: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12092: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12093:
1.254 brouard 12094: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12095: if (num_filled != 1) {
12096: 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);
12097: 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);
12098: goto end;
12099: }
12100: printf("pop_based=%d\n",popbased);
12101: fprintf(ficlog,"pop_based=%d\n",popbased);
12102: fprintf(ficparo,"pop_based=%d\n",popbased);
12103: fprintf(ficres,"pop_based=%d\n",popbased);
12104: }
12105:
1.258 brouard 12106: /* Results */
12107: nresult=0;
12108: do{
12109: if(!fgets(line, MAXLINE, ficpar)){
12110: endishere=1;
12111: parameterline=14;
12112: }else if (line[0] == '#') {
12113: /* If line starts with a # it is a comment */
1.254 brouard 12114: numlinepar++;
12115: fputs(line,stdout);
12116: fputs(line,ficparo);
12117: fputs(line,ficlog);
12118: continue;
1.258 brouard 12119: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12120: parameterline=11;
12121: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12122: parameterline=12;
12123: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12124: parameterline=13;
12125: else{
12126: parameterline=14;
1.254 brouard 12127: }
1.258 brouard 12128: switch (parameterline){
12129: case 11:
12130: 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){
12131: if (num_filled != 8) {
12132: printf("Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
12133: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
12134: goto end;
12135: }
12136: 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);
12137: 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);
12138: 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);
12139: 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);
12140: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12141: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12142: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12143:
1.258 brouard 12144: }
1.254 brouard 12145: break;
1.258 brouard 12146: case 12:
12147: /*fscanf(ficpar,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj);*/
12148: if((num_filled=sscanf(line,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj)) !=EOF){
12149: if (num_filled != 8) {
1.262 brouard 12150: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
12151: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
1.258 brouard 12152: goto end;
12153: }
12154: printf("backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12155: fprintf(ficparo,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12156: fprintf(ficlog,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12157: fprintf(ficres,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
12158: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12159: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12160: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12161: }
1.230 brouard 12162: break;
1.258 brouard 12163: case 13:
12164: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12165: if (num_filled == 0){
12166: resultline[0]='\0';
12167: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12168: 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);
12169: break;
12170: } else if (num_filled != 1){
12171: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12172: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12173: }
12174: nresult++; /* Sum of resultlines */
12175: printf("Result %d: result=%s\n",nresult, resultline);
12176: if(nresult > MAXRESULTLINES){
12177: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12178: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12179: goto end;
12180: }
12181: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12182: fprintf(ficparo,"result: %s\n",resultline);
12183: fprintf(ficres,"result: %s\n",resultline);
12184: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12185: break;
1.258 brouard 12186: case 14:
1.259 brouard 12187: if(ncovmodel >2 && nresult==0 ){
12188: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12189: goto end;
12190: }
1.259 brouard 12191: break;
1.258 brouard 12192: default:
12193: nresult=1;
12194: decoderesult(".",nresult ); /* No covariate */
12195: }
12196: } /* End switch parameterline */
12197: }while(endishere==0); /* End do */
1.126 brouard 12198:
1.230 brouard 12199: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12200: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12201:
12202: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12203: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12204: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12205: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12206: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12207: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12208: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12209: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12210: }else{
1.270 brouard 12211: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12212: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12213: }
12214: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12215: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12216: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12217:
1.225 brouard 12218: /*------------ free_vector -------------*/
12219: /* chdir(path); */
1.220 brouard 12220:
1.215 brouard 12221: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12222: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12223: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12224: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12225: free_lvector(num,1,n);
12226: free_vector(agedc,1,n);
12227: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12228: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12229: fclose(ficparo);
12230: fclose(ficres);
1.220 brouard 12231:
12232:
1.186 brouard 12233: /* Other results (useful)*/
1.220 brouard 12234:
12235:
1.126 brouard 12236: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12237: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12238: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12239: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12240: fclose(ficrespl);
12241:
12242: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12243: /*#include "hpijx.h"*/
12244: hPijx(p, bage, fage);
1.145 brouard 12245: fclose(ficrespij);
1.227 brouard 12246:
1.220 brouard 12247: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12248: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12249: k=1;
1.126 brouard 12250: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12251:
1.269 brouard 12252: /* Prevalence for each covariate combination in probs[age][status][cov] */
12253: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12254: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12255: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12256: for(k=1;k<=ncovcombmax;k++)
12257: probs[i][j][k]=0.;
1.269 brouard 12258: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12259: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12260: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12261: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12262: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12263: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12264: for(k=1;k<=ncovcombmax;k++)
12265: mobaverages[i][j][k]=0.;
1.219 brouard 12266: mobaverage=mobaverages;
12267: if (mobilav!=0) {
1.235 brouard 12268: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12269: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12270: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12271: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12272: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12273: }
1.269 brouard 12274: } else if (mobilavproj !=0) {
1.235 brouard 12275: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12276: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12277: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12278: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12279: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12280: }
1.269 brouard 12281: }else{
12282: printf("Internal error moving average\n");
12283: fflush(stdout);
12284: exit(1);
1.219 brouard 12285: }
12286: }/* end if moving average */
1.227 brouard 12287:
1.126 brouard 12288: /*---------- Forecasting ------------------*/
12289: if(prevfcast==1){
12290: /* if(stepm ==1){*/
1.269 brouard 12291: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12292: }
1.269 brouard 12293:
12294: /* Backcasting */
1.217 brouard 12295: if(backcast==1){
1.219 brouard 12296: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12297: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12298: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12299:
12300: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12301:
12302: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12303:
1.219 brouard 12304: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12305: fclose(ficresplb);
12306:
1.222 brouard 12307: hBijx(p, bage, fage, mobaverage);
12308: fclose(ficrespijb);
1.219 brouard 12309:
1.269 brouard 12310: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12311: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12312: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12313:
12314:
1.269 brouard 12315: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12316: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12317: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12318: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12319: } /* end Backcasting */
1.268 brouard 12320:
1.186 brouard 12321:
12322: /* ------ Other prevalence ratios------------ */
1.126 brouard 12323:
1.215 brouard 12324: free_ivector(wav,1,imx);
12325: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12326: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12327: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12328:
12329:
1.127 brouard 12330: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12331:
1.201 brouard 12332: strcpy(filerese,"E_");
12333: strcat(filerese,fileresu);
1.126 brouard 12334: if((ficreseij=fopen(filerese,"w"))==NULL) {
12335: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12336: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12337: }
1.208 brouard 12338: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12339: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12340:
12341: pstamp(ficreseij);
1.219 brouard 12342:
1.235 brouard 12343: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12344: if (cptcovn < 1){i1=1;}
12345:
12346: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12347: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12348: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12349: continue;
1.219 brouard 12350: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12351: printf("\n#****** ");
1.225 brouard 12352: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12353: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12354: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12355: }
12356: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12357: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12358: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12359: }
12360: fprintf(ficreseij,"******\n");
1.235 brouard 12361: printf("******\n");
1.219 brouard 12362:
12363: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12364: oldm=oldms;savm=savms;
1.235 brouard 12365: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12366:
1.219 brouard 12367: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12368: }
12369: fclose(ficreseij);
1.208 brouard 12370: printf("done evsij\n");fflush(stdout);
12371: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12372:
1.218 brouard 12373:
1.227 brouard 12374: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12375:
1.201 brouard 12376: strcpy(filerest,"T_");
12377: strcat(filerest,fileresu);
1.127 brouard 12378: if((ficrest=fopen(filerest,"w"))==NULL) {
12379: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12380: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12381: }
1.208 brouard 12382: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12383: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12384: strcpy(fileresstde,"STDE_");
12385: strcat(fileresstde,fileresu);
1.126 brouard 12386: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12387: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12388: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12389: }
1.227 brouard 12390: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12391: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12392:
1.201 brouard 12393: strcpy(filerescve,"CVE_");
12394: strcat(filerescve,fileresu);
1.126 brouard 12395: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12396: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12397: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12398: }
1.227 brouard 12399: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12400: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12401:
1.201 brouard 12402: strcpy(fileresv,"V_");
12403: strcat(fileresv,fileresu);
1.126 brouard 12404: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12405: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12406: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12407: }
1.227 brouard 12408: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12409: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12410:
1.235 brouard 12411: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12412: if (cptcovn < 1){i1=1;}
12413:
12414: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12415: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12416: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12417: continue;
1.242 brouard 12418: printf("\n#****** Result for:");
12419: fprintf(ficrest,"\n#****** Result for:");
12420: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12421: for(j=1;j<=cptcoveff;j++){
12422: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12423: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12424: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12425: }
1.235 brouard 12426: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12427: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12428: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12429: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12430: }
1.208 brouard 12431: fprintf(ficrest,"******\n");
1.227 brouard 12432: fprintf(ficlog,"******\n");
12433: printf("******\n");
1.208 brouard 12434:
12435: fprintf(ficresstdeij,"\n#****** ");
12436: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12437: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12438: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12439: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12440: }
1.235 brouard 12441: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12442: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12443: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12444: }
1.208 brouard 12445: fprintf(ficresstdeij,"******\n");
12446: fprintf(ficrescveij,"******\n");
12447:
12448: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12449: /* pstamp(ficresvij); */
1.225 brouard 12450: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12451: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12452: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12453: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12454: }
1.208 brouard 12455: fprintf(ficresvij,"******\n");
12456:
12457: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12458: oldm=oldms;savm=savms;
1.235 brouard 12459: printf(" cvevsij ");
12460: fprintf(ficlog, " cvevsij ");
12461: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12462: printf(" end cvevsij \n ");
12463: fprintf(ficlog, " end cvevsij \n ");
12464:
12465: /*
12466: */
12467: /* goto endfree; */
12468:
12469: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12470: pstamp(ficrest);
12471:
1.269 brouard 12472: epj=vector(1,nlstate+1);
1.208 brouard 12473: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12474: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12475: cptcod= 0; /* To be deleted */
12476: printf("varevsij vpopbased=%d \n",vpopbased);
12477: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12478: 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 12479: 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 ");
12480: if(vpopbased==1)
12481: 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);
12482: else
12483: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12484: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12485: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12486: fprintf(ficrest,"\n");
12487: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12488: printf("Computing age specific period (stable) prevalences in each health state \n");
12489: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12490: for(age=bage; age <=fage ;age++){
1.235 brouard 12491: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12492: if (vpopbased==1) {
12493: if(mobilav ==0){
12494: for(i=1; i<=nlstate;i++)
12495: prlim[i][i]=probs[(int)age][i][k];
12496: }else{ /* mobilav */
12497: for(i=1; i<=nlstate;i++)
12498: prlim[i][i]=mobaverage[(int)age][i][k];
12499: }
12500: }
1.219 brouard 12501:
1.227 brouard 12502: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12503: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12504: /* printf(" age %4.0f ",age); */
12505: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12506: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12507: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12508: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12509: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12510: }
12511: epj[nlstate+1] +=epj[j];
12512: }
12513: /* printf(" age %4.0f \n",age); */
1.219 brouard 12514:
1.227 brouard 12515: for(i=1, vepp=0.;i <=nlstate;i++)
12516: for(j=1;j <=nlstate;j++)
12517: vepp += vareij[i][j][(int)age];
12518: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12519: for(j=1;j <=nlstate;j++){
12520: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12521: }
12522: fprintf(ficrest,"\n");
12523: }
1.208 brouard 12524: } /* End vpopbased */
1.269 brouard 12525: free_vector(epj,1,nlstate+1);
1.208 brouard 12526: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12527: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12528: printf("done selection\n");fflush(stdout);
12529: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12530:
1.235 brouard 12531: } /* End k selection */
1.227 brouard 12532:
12533: printf("done State-specific expectancies\n");fflush(stdout);
12534: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12535:
1.269 brouard 12536: /* variance-covariance of period prevalence*/
12537: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12538:
1.227 brouard 12539:
12540: free_vector(weight,1,n);
12541: free_imatrix(Tvard,1,NCOVMAX,1,2);
12542: free_imatrix(s,1,maxwav+1,1,n);
12543: free_matrix(anint,1,maxwav,1,n);
12544: free_matrix(mint,1,maxwav,1,n);
12545: free_ivector(cod,1,n);
12546: free_ivector(tab,1,NCOVMAX);
12547: fclose(ficresstdeij);
12548: fclose(ficrescveij);
12549: fclose(ficresvij);
12550: fclose(ficrest);
12551: fclose(ficpar);
12552:
12553:
1.126 brouard 12554: /*---------- End : free ----------------*/
1.219 brouard 12555: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12556: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12557: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12558: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12559: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12560: } /* mle==-3 arrives here for freeing */
1.227 brouard 12561: /* endfree:*/
12562: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12563: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12564: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12565: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12566: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12567: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12568: free_matrix(covar,0,NCOVMAX,1,n);
12569: free_matrix(matcov,1,npar,1,npar);
12570: free_matrix(hess,1,npar,1,npar);
12571: /*free_vector(delti,1,npar);*/
12572: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12573: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12574: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12575: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12576:
12577: free_ivector(ncodemax,1,NCOVMAX);
12578: free_ivector(ncodemaxwundef,1,NCOVMAX);
12579: free_ivector(Dummy,-1,NCOVMAX);
12580: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12581: free_ivector(DummyV,1,NCOVMAX);
12582: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12583: free_ivector(Typevar,-1,NCOVMAX);
12584: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12585: free_ivector(TvarsQ,1,NCOVMAX);
12586: free_ivector(TvarsQind,1,NCOVMAX);
12587: free_ivector(TvarsD,1,NCOVMAX);
12588: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12589: free_ivector(TvarFD,1,NCOVMAX);
12590: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12591: free_ivector(TvarF,1,NCOVMAX);
12592: free_ivector(TvarFind,1,NCOVMAX);
12593: free_ivector(TvarV,1,NCOVMAX);
12594: free_ivector(TvarVind,1,NCOVMAX);
12595: free_ivector(TvarA,1,NCOVMAX);
12596: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12597: free_ivector(TvarFQ,1,NCOVMAX);
12598: free_ivector(TvarFQind,1,NCOVMAX);
12599: free_ivector(TvarVD,1,NCOVMAX);
12600: free_ivector(TvarVDind,1,NCOVMAX);
12601: free_ivector(TvarVQ,1,NCOVMAX);
12602: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12603: free_ivector(Tvarsel,1,NCOVMAX);
12604: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12605: free_ivector(Tposprod,1,NCOVMAX);
12606: free_ivector(Tprod,1,NCOVMAX);
12607: free_ivector(Tvaraff,1,NCOVMAX);
12608: free_ivector(invalidvarcomb,1,ncovcombmax);
12609: free_ivector(Tage,1,NCOVMAX);
12610: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12611: free_ivector(TmodelInvind,1,NCOVMAX);
12612: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12613:
12614: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12615: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12616: fflush(fichtm);
12617: fflush(ficgp);
12618:
1.227 brouard 12619:
1.126 brouard 12620: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12621: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12622: 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 12623: }else{
12624: printf("End of Imach\n");
12625: fprintf(ficlog,"End of Imach\n");
12626: }
12627: printf("See log file on %s\n",filelog);
12628: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12629: /*(void) gettimeofday(&end_time,&tzp);*/
12630: rend_time = time(NULL);
12631: end_time = *localtime(&rend_time);
12632: /* tml = *localtime(&end_time.tm_sec); */
12633: strcpy(strtend,asctime(&end_time));
1.126 brouard 12634: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12635: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12636: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12637:
1.157 brouard 12638: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12639: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12640: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12641: /* printf("Total time was %d uSec.\n", total_usecs);*/
12642: /* if(fileappend(fichtm,optionfilehtm)){ */
12643: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12644: fclose(fichtm);
12645: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12646: fclose(fichtmcov);
12647: fclose(ficgp);
12648: fclose(ficlog);
12649: /*------ End -----------*/
1.227 brouard 12650:
12651:
12652: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12653: #ifdef WIN32
1.227 brouard 12654: if (_chdir(pathcd) != 0)
12655: printf("Can't move to directory %s!\n",path);
12656: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12657: #else
1.227 brouard 12658: if(chdir(pathcd) != 0)
12659: printf("Can't move to directory %s!\n", path);
12660: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12661: #endif
1.126 brouard 12662: printf("Current directory %s!\n",pathcd);
12663: /*strcat(plotcmd,CHARSEPARATOR);*/
12664: sprintf(plotcmd,"gnuplot");
1.157 brouard 12665: #ifdef _WIN32
1.126 brouard 12666: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12667: #endif
12668: if(!stat(plotcmd,&info)){
1.158 brouard 12669: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12670: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12671: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12672: }else
12673: strcpy(pplotcmd,plotcmd);
1.157 brouard 12674: #ifdef __unix
1.126 brouard 12675: strcpy(plotcmd,GNUPLOTPROGRAM);
12676: if(!stat(plotcmd,&info)){
1.158 brouard 12677: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12678: }else
12679: strcpy(pplotcmd,plotcmd);
12680: #endif
12681: }else
12682: strcpy(pplotcmd,plotcmd);
12683:
12684: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12685: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12686:
1.126 brouard 12687: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12688: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12689: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12690: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12691: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12692: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12693: }
1.158 brouard 12694: printf(" Successful, please wait...");
1.126 brouard 12695: while (z[0] != 'q') {
12696: /* chdir(path); */
1.154 brouard 12697: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12698: scanf("%s",z);
12699: /* if (z[0] == 'c') system("./imach"); */
12700: if (z[0] == 'e') {
1.158 brouard 12701: #ifdef __APPLE__
1.152 brouard 12702: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12703: #elif __linux
12704: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12705: #else
1.152 brouard 12706: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12707: #endif
12708: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12709: system(pplotcmd);
1.126 brouard 12710: }
12711: else if (z[0] == 'g') system(plotcmd);
12712: else if (z[0] == 'q') exit(0);
12713: }
1.227 brouard 12714: end:
1.126 brouard 12715: while (z[0] != 'q') {
1.195 brouard 12716: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12717: scanf("%s",z);
12718: }
12719: }
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