Annotation of imach/src/imach.c, revision 1.280
1.280 ! brouard 1: /* $Id: imach.c,v 1.279 2017/07/20 13:35:01 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.280 ! brouard 4: Revision 1.279 2017/07/20 13:35:01 brouard
! 5: Summary: temporary working
! 6:
1.279 brouard 7: Revision 1.278 2017/07/19 14:09:02 brouard
8: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
9:
1.278 brouard 10: Revision 1.277 2017/07/17 08:53:49 brouard
11: Summary: BOM files can be read now
12:
1.277 brouard 13: Revision 1.276 2017/06/30 15:48:31 brouard
14: Summary: Graphs improvements
15:
1.276 brouard 16: Revision 1.275 2017/06/30 13:39:33 brouard
17: Summary: Saito's color
18:
1.275 brouard 19: Revision 1.274 2017/06/29 09:47:08 brouard
20: Summary: Version 0.99r14
21:
1.274 brouard 22: Revision 1.273 2017/06/27 11:06:02 brouard
23: Summary: More documentation on projections
24:
1.273 brouard 25: Revision 1.272 2017/06/27 10:22:40 brouard
26: Summary: Color of backprojection changed from 6 to 5(yellow)
27:
1.272 brouard 28: Revision 1.271 2017/06/27 10:17:50 brouard
29: Summary: Some bug with rint
30:
1.271 brouard 31: Revision 1.270 2017/05/24 05:45:29 brouard
32: *** empty log message ***
33:
1.270 brouard 34: Revision 1.269 2017/05/23 08:39:25 brouard
35: Summary: Code into subroutine, cleanings
36:
1.269 brouard 37: Revision 1.268 2017/05/18 20:09:32 brouard
38: Summary: backprojection and confidence intervals of backprevalence
39:
1.268 brouard 40: Revision 1.267 2017/05/13 10:25:05 brouard
41: Summary: temporary save for backprojection
42:
1.267 brouard 43: Revision 1.266 2017/05/13 07:26:12 brouard
44: Summary: Version 0.99r13 (improvements and bugs fixed)
45:
1.266 brouard 46: Revision 1.265 2017/04/26 16:22:11 brouard
47: Summary: imach 0.99r13 Some bugs fixed
48:
1.265 brouard 49: Revision 1.264 2017/04/26 06:01:29 brouard
50: Summary: Labels in graphs
51:
1.264 brouard 52: Revision 1.263 2017/04/24 15:23:15 brouard
53: Summary: to save
54:
1.263 brouard 55: Revision 1.262 2017/04/18 16:48:12 brouard
56: *** empty log message ***
57:
1.262 brouard 58: Revision 1.261 2017/04/05 10:14:09 brouard
59: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
60:
1.261 brouard 61: Revision 1.260 2017/04/04 17:46:59 brouard
62: Summary: Gnuplot indexations fixed (humm)
63:
1.260 brouard 64: Revision 1.259 2017/04/04 13:01:16 brouard
65: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
66:
1.259 brouard 67: Revision 1.258 2017/04/03 10:17:47 brouard
68: Summary: Version 0.99r12
69:
70: Some cleanings, conformed with updated documentation.
71:
1.258 brouard 72: Revision 1.257 2017/03/29 16:53:30 brouard
73: Summary: Temp
74:
1.257 brouard 75: Revision 1.256 2017/03/27 05:50:23 brouard
76: Summary: Temporary
77:
1.256 brouard 78: Revision 1.255 2017/03/08 16:02:28 brouard
79: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
80:
1.255 brouard 81: Revision 1.254 2017/03/08 07:13:00 brouard
82: Summary: Fixing data parameter line
83:
1.254 brouard 84: Revision 1.253 2016/12/15 11:59:41 brouard
85: Summary: 0.99 in progress
86:
1.253 brouard 87: Revision 1.252 2016/09/15 21:15:37 brouard
88: *** empty log message ***
89:
1.252 brouard 90: Revision 1.251 2016/09/15 15:01:13 brouard
91: Summary: not working
92:
1.251 brouard 93: Revision 1.250 2016/09/08 16:07:27 brouard
94: Summary: continue
95:
1.250 brouard 96: Revision 1.249 2016/09/07 17:14:18 brouard
97: Summary: Starting values from frequencies
98:
1.249 brouard 99: Revision 1.248 2016/09/07 14:10:18 brouard
100: *** empty log message ***
101:
1.248 brouard 102: Revision 1.247 2016/09/02 11:11:21 brouard
103: *** empty log message ***
104:
1.247 brouard 105: Revision 1.246 2016/09/02 08:49:22 brouard
106: *** empty log message ***
107:
1.246 brouard 108: Revision 1.245 2016/09/02 07:25:01 brouard
109: *** empty log message ***
110:
1.245 brouard 111: Revision 1.244 2016/09/02 07:17:34 brouard
112: *** empty log message ***
113:
1.244 brouard 114: Revision 1.243 2016/09/02 06:45:35 brouard
115: *** empty log message ***
116:
1.243 brouard 117: Revision 1.242 2016/08/30 15:01:20 brouard
118: Summary: Fixing a lots
119:
1.242 brouard 120: Revision 1.241 2016/08/29 17:17:25 brouard
121: Summary: gnuplot problem in Back projection to fix
122:
1.241 brouard 123: Revision 1.240 2016/08/29 07:53:18 brouard
124: Summary: Better
125:
1.240 brouard 126: Revision 1.239 2016/08/26 15:51:03 brouard
127: Summary: Improvement in Powell output in order to copy and paste
128:
129: Author:
130:
1.239 brouard 131: Revision 1.238 2016/08/26 14:23:35 brouard
132: Summary: Starting tests of 0.99
133:
1.238 brouard 134: Revision 1.237 2016/08/26 09:20:19 brouard
135: Summary: to valgrind
136:
1.237 brouard 137: Revision 1.236 2016/08/25 10:50:18 brouard
138: *** empty log message ***
139:
1.236 brouard 140: Revision 1.235 2016/08/25 06:59:23 brouard
141: *** empty log message ***
142:
1.235 brouard 143: Revision 1.234 2016/08/23 16:51:20 brouard
144: *** empty log message ***
145:
1.234 brouard 146: Revision 1.233 2016/08/23 07:40:50 brouard
147: Summary: not working
148:
1.233 brouard 149: Revision 1.232 2016/08/22 14:20:21 brouard
150: Summary: not working
151:
1.232 brouard 152: Revision 1.231 2016/08/22 07:17:15 brouard
153: Summary: not working
154:
1.231 brouard 155: Revision 1.230 2016/08/22 06:55:53 brouard
156: Summary: Not working
157:
1.230 brouard 158: Revision 1.229 2016/07/23 09:45:53 brouard
159: Summary: Completing for func too
160:
1.229 brouard 161: Revision 1.228 2016/07/22 17:45:30 brouard
162: Summary: Fixing some arrays, still debugging
163:
1.227 brouard 164: Revision 1.226 2016/07/12 18:42:34 brouard
165: Summary: temp
166:
1.226 brouard 167: Revision 1.225 2016/07/12 08:40:03 brouard
168: Summary: saving but not running
169:
1.225 brouard 170: Revision 1.224 2016/07/01 13:16:01 brouard
171: Summary: Fixes
172:
1.224 brouard 173: Revision 1.223 2016/02/19 09:23:35 brouard
174: Summary: temporary
175:
1.223 brouard 176: Revision 1.222 2016/02/17 08:14:50 brouard
177: Summary: Probably last 0.98 stable version 0.98r6
178:
1.222 brouard 179: Revision 1.221 2016/02/15 23:35:36 brouard
180: Summary: minor bug
181:
1.220 brouard 182: Revision 1.219 2016/02/15 00:48:12 brouard
183: *** empty log message ***
184:
1.219 brouard 185: Revision 1.218 2016/02/12 11:29:23 brouard
186: Summary: 0.99 Back projections
187:
1.218 brouard 188: Revision 1.217 2015/12/23 17:18:31 brouard
189: Summary: Experimental backcast
190:
1.217 brouard 191: Revision 1.216 2015/12/18 17:32:11 brouard
192: Summary: 0.98r4 Warning and status=-2
193:
194: Version 0.98r4 is now:
195: - displaying an error when status is -1, date of interview unknown and date of death known;
196: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
197: Older changes concerning s=-2, dating from 2005 have been supersed.
198:
1.216 brouard 199: Revision 1.215 2015/12/16 08:52:24 brouard
200: Summary: 0.98r4 working
201:
1.215 brouard 202: Revision 1.214 2015/12/16 06:57:54 brouard
203: Summary: temporary not working
204:
1.214 brouard 205: Revision 1.213 2015/12/11 18:22:17 brouard
206: Summary: 0.98r4
207:
1.213 brouard 208: Revision 1.212 2015/11/21 12:47:24 brouard
209: Summary: minor typo
210:
1.212 brouard 211: Revision 1.211 2015/11/21 12:41:11 brouard
212: Summary: 0.98r3 with some graph of projected cross-sectional
213:
214: Author: Nicolas Brouard
215:
1.211 brouard 216: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 217: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 218: Summary: Adding ftolpl parameter
219: Author: N Brouard
220:
221: We had difficulties to get smoothed confidence intervals. It was due
222: to the period prevalence which wasn't computed accurately. The inner
223: parameter ftolpl is now an outer parameter of the .imach parameter
224: file after estepm. If ftolpl is small 1.e-4 and estepm too,
225: computation are long.
226:
1.209 brouard 227: Revision 1.208 2015/11/17 14:31:57 brouard
228: Summary: temporary
229:
1.208 brouard 230: Revision 1.207 2015/10/27 17:36:57 brouard
231: *** empty log message ***
232:
1.207 brouard 233: Revision 1.206 2015/10/24 07:14:11 brouard
234: *** empty log message ***
235:
1.206 brouard 236: Revision 1.205 2015/10/23 15:50:53 brouard
237: Summary: 0.98r3 some clarification for graphs on likelihood contributions
238:
1.205 brouard 239: Revision 1.204 2015/10/01 16:20:26 brouard
240: Summary: Some new graphs of contribution to likelihood
241:
1.204 brouard 242: Revision 1.203 2015/09/30 17:45:14 brouard
243: Summary: looking at better estimation of the hessian
244:
245: Also a better criteria for convergence to the period prevalence And
246: therefore adding the number of years needed to converge. (The
247: prevalence in any alive state shold sum to one
248:
1.203 brouard 249: Revision 1.202 2015/09/22 19:45:16 brouard
250: Summary: Adding some overall graph on contribution to likelihood. Might change
251:
1.202 brouard 252: Revision 1.201 2015/09/15 17:34:58 brouard
253: Summary: 0.98r0
254:
255: - Some new graphs like suvival functions
256: - Some bugs fixed like model=1+age+V2.
257:
1.201 brouard 258: Revision 1.200 2015/09/09 16:53:55 brouard
259: Summary: Big bug thanks to Flavia
260:
261: Even model=1+age+V2. did not work anymore
262:
1.200 brouard 263: Revision 1.199 2015/09/07 14:09:23 brouard
264: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
265:
1.199 brouard 266: Revision 1.198 2015/09/03 07:14:39 brouard
267: Summary: 0.98q5 Flavia
268:
1.198 brouard 269: Revision 1.197 2015/09/01 18:24:39 brouard
270: *** empty log message ***
271:
1.197 brouard 272: Revision 1.196 2015/08/18 23:17:52 brouard
273: Summary: 0.98q5
274:
1.196 brouard 275: Revision 1.195 2015/08/18 16:28:39 brouard
276: Summary: Adding a hack for testing purpose
277:
278: After reading the title, ftol and model lines, if the comment line has
279: a q, starting with #q, the answer at the end of the run is quit. It
280: permits to run test files in batch with ctest. The former workaround was
281: $ echo q | imach foo.imach
282:
1.195 brouard 283: Revision 1.194 2015/08/18 13:32:00 brouard
284: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
285:
1.194 brouard 286: Revision 1.193 2015/08/04 07:17:42 brouard
287: Summary: 0.98q4
288:
1.193 brouard 289: Revision 1.192 2015/07/16 16:49:02 brouard
290: Summary: Fixing some outputs
291:
1.192 brouard 292: Revision 1.191 2015/07/14 10:00:33 brouard
293: Summary: Some fixes
294:
1.191 brouard 295: Revision 1.190 2015/05/05 08:51:13 brouard
296: Summary: Adding digits in output parameters (7 digits instead of 6)
297:
298: Fix 1+age+.
299:
1.190 brouard 300: Revision 1.189 2015/04/30 14:45:16 brouard
301: Summary: 0.98q2
302:
1.189 brouard 303: Revision 1.188 2015/04/30 08:27:53 brouard
304: *** empty log message ***
305:
1.188 brouard 306: Revision 1.187 2015/04/29 09:11:15 brouard
307: *** empty log message ***
308:
1.187 brouard 309: Revision 1.186 2015/04/23 12:01:52 brouard
310: Summary: V1*age is working now, version 0.98q1
311:
312: Some codes had been disabled in order to simplify and Vn*age was
313: working in the optimization phase, ie, giving correct MLE parameters,
314: but, as usual, outputs were not correct and program core dumped.
315:
1.186 brouard 316: Revision 1.185 2015/03/11 13:26:42 brouard
317: Summary: Inclusion of compile and links command line for Intel Compiler
318:
1.185 brouard 319: Revision 1.184 2015/03/11 11:52:39 brouard
320: Summary: Back from Windows 8. Intel Compiler
321:
1.184 brouard 322: Revision 1.183 2015/03/10 20:34:32 brouard
323: Summary: 0.98q0, trying with directest, mnbrak fixed
324:
325: We use directest instead of original Powell test; probably no
326: incidence on the results, but better justifications;
327: We fixed Numerical Recipes mnbrak routine which was wrong and gave
328: wrong results.
329:
1.183 brouard 330: Revision 1.182 2015/02/12 08:19:57 brouard
331: Summary: Trying to keep directest which seems simpler and more general
332: Author: Nicolas Brouard
333:
1.182 brouard 334: Revision 1.181 2015/02/11 23:22:24 brouard
335: Summary: Comments on Powell added
336:
337: Author:
338:
1.181 brouard 339: Revision 1.180 2015/02/11 17:33:45 brouard
340: Summary: Finishing move from main to function (hpijx and prevalence_limit)
341:
1.180 brouard 342: Revision 1.179 2015/01/04 09:57:06 brouard
343: Summary: back to OS/X
344:
1.179 brouard 345: Revision 1.178 2015/01/04 09:35:48 brouard
346: *** empty log message ***
347:
1.178 brouard 348: Revision 1.177 2015/01/03 18:40:56 brouard
349: Summary: Still testing ilc32 on OSX
350:
1.177 brouard 351: Revision 1.176 2015/01/03 16:45:04 brouard
352: *** empty log message ***
353:
1.176 brouard 354: Revision 1.175 2015/01/03 16:33:42 brouard
355: *** empty log message ***
356:
1.175 brouard 357: Revision 1.174 2015/01/03 16:15:49 brouard
358: Summary: Still in cross-compilation
359:
1.174 brouard 360: Revision 1.173 2015/01/03 12:06:26 brouard
361: Summary: trying to detect cross-compilation
362:
1.173 brouard 363: Revision 1.172 2014/12/27 12:07:47 brouard
364: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
365:
1.172 brouard 366: Revision 1.171 2014/12/23 13:26:59 brouard
367: Summary: Back from Visual C
368:
369: Still problem with utsname.h on Windows
370:
1.171 brouard 371: Revision 1.170 2014/12/23 11:17:12 brouard
372: Summary: Cleaning some \%% back to %%
373:
374: The escape was mandatory for a specific compiler (which one?), but too many warnings.
375:
1.170 brouard 376: Revision 1.169 2014/12/22 23:08:31 brouard
377: Summary: 0.98p
378:
379: Outputs some informations on compiler used, OS etc. Testing on different platforms.
380:
1.169 brouard 381: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 382: Summary: update
1.169 brouard 383:
1.168 brouard 384: Revision 1.167 2014/12/22 13:50:56 brouard
385: Summary: Testing uname and compiler version and if compiled 32 or 64
386:
387: Testing on Linux 64
388:
1.167 brouard 389: Revision 1.166 2014/12/22 11:40:47 brouard
390: *** empty log message ***
391:
1.166 brouard 392: Revision 1.165 2014/12/16 11:20:36 brouard
393: Summary: After compiling on Visual C
394:
395: * imach.c (Module): Merging 1.61 to 1.162
396:
1.165 brouard 397: Revision 1.164 2014/12/16 10:52:11 brouard
398: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
399:
400: * imach.c (Module): Merging 1.61 to 1.162
401:
1.164 brouard 402: Revision 1.163 2014/12/16 10:30:11 brouard
403: * imach.c (Module): Merging 1.61 to 1.162
404:
1.163 brouard 405: Revision 1.162 2014/09/25 11:43:39 brouard
406: Summary: temporary backup 0.99!
407:
1.162 brouard 408: Revision 1.1 2014/09/16 11:06:58 brouard
409: Summary: With some code (wrong) for nlopt
410:
411: Author:
412:
413: Revision 1.161 2014/09/15 20:41:41 brouard
414: Summary: Problem with macro SQR on Intel compiler
415:
1.161 brouard 416: Revision 1.160 2014/09/02 09:24:05 brouard
417: *** empty log message ***
418:
1.160 brouard 419: Revision 1.159 2014/09/01 10:34:10 brouard
420: Summary: WIN32
421: Author: Brouard
422:
1.159 brouard 423: Revision 1.158 2014/08/27 17:11:51 brouard
424: *** empty log message ***
425:
1.158 brouard 426: Revision 1.157 2014/08/27 16:26:55 brouard
427: Summary: Preparing windows Visual studio version
428: Author: Brouard
429:
430: In order to compile on Visual studio, time.h is now correct and time_t
431: and tm struct should be used. difftime should be used but sometimes I
432: just make the differences in raw time format (time(&now).
433: Trying to suppress #ifdef LINUX
434: Add xdg-open for __linux in order to open default browser.
435:
1.157 brouard 436: Revision 1.156 2014/08/25 20:10:10 brouard
437: *** empty log message ***
438:
1.156 brouard 439: Revision 1.155 2014/08/25 18:32:34 brouard
440: Summary: New compile, minor changes
441: Author: Brouard
442:
1.155 brouard 443: Revision 1.154 2014/06/20 17:32:08 brouard
444: Summary: Outputs now all graphs of convergence to period prevalence
445:
1.154 brouard 446: Revision 1.153 2014/06/20 16:45:46 brouard
447: Summary: If 3 live state, convergence to period prevalence on same graph
448: Author: Brouard
449:
1.153 brouard 450: Revision 1.152 2014/06/18 17:54:09 brouard
451: Summary: open browser, use gnuplot on same dir than imach if not found in the path
452:
1.152 brouard 453: Revision 1.151 2014/06/18 16:43:30 brouard
454: *** empty log message ***
455:
1.151 brouard 456: Revision 1.150 2014/06/18 16:42:35 brouard
457: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
458: Author: brouard
459:
1.150 brouard 460: Revision 1.149 2014/06/18 15:51:14 brouard
461: Summary: Some fixes in parameter files errors
462: Author: Nicolas Brouard
463:
1.149 brouard 464: Revision 1.148 2014/06/17 17:38:48 brouard
465: Summary: Nothing new
466: Author: Brouard
467:
468: Just a new packaging for OS/X version 0.98nS
469:
1.148 brouard 470: Revision 1.147 2014/06/16 10:33:11 brouard
471: *** empty log message ***
472:
1.147 brouard 473: Revision 1.146 2014/06/16 10:20:28 brouard
474: Summary: Merge
475: Author: Brouard
476:
477: Merge, before building revised version.
478:
1.146 brouard 479: Revision 1.145 2014/06/10 21:23:15 brouard
480: Summary: Debugging with valgrind
481: Author: Nicolas Brouard
482:
483: Lot of changes in order to output the results with some covariates
484: After the Edimburgh REVES conference 2014, it seems mandatory to
485: improve the code.
486: No more memory valgrind error but a lot has to be done in order to
487: continue the work of splitting the code into subroutines.
488: Also, decodemodel has been improved. Tricode is still not
489: optimal. nbcode should be improved. Documentation has been added in
490: the source code.
491:
1.144 brouard 492: Revision 1.143 2014/01/26 09:45:38 brouard
493: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
494:
495: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
496: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
497:
1.143 brouard 498: Revision 1.142 2014/01/26 03:57:36 brouard
499: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
500:
501: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
502:
1.142 brouard 503: Revision 1.141 2014/01/26 02:42:01 brouard
504: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
505:
1.141 brouard 506: Revision 1.140 2011/09/02 10:37:54 brouard
507: Summary: times.h is ok with mingw32 now.
508:
1.140 brouard 509: Revision 1.139 2010/06/14 07:50:17 brouard
510: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
511: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
512:
1.139 brouard 513: Revision 1.138 2010/04/30 18:19:40 brouard
514: *** empty log message ***
515:
1.138 brouard 516: Revision 1.137 2010/04/29 18:11:38 brouard
517: (Module): Checking covariates for more complex models
518: than V1+V2. A lot of change to be done. Unstable.
519:
1.137 brouard 520: Revision 1.136 2010/04/26 20:30:53 brouard
521: (Module): merging some libgsl code. Fixing computation
522: of likelione (using inter/intrapolation if mle = 0) in order to
523: get same likelihood as if mle=1.
524: Some cleaning of code and comments added.
525:
1.136 brouard 526: Revision 1.135 2009/10/29 15:33:14 brouard
527: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
528:
1.135 brouard 529: Revision 1.134 2009/10/29 13:18:53 brouard
530: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
531:
1.134 brouard 532: Revision 1.133 2009/07/06 10:21:25 brouard
533: just nforces
534:
1.133 brouard 535: Revision 1.132 2009/07/06 08:22:05 brouard
536: Many tings
537:
1.132 brouard 538: Revision 1.131 2009/06/20 16:22:47 brouard
539: Some dimensions resccaled
540:
1.131 brouard 541: Revision 1.130 2009/05/26 06:44:34 brouard
542: (Module): Max Covariate is now set to 20 instead of 8. A
543: lot of cleaning with variables initialized to 0. Trying to make
544: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
545:
1.130 brouard 546: Revision 1.129 2007/08/31 13:49:27 lievre
547: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
548:
1.129 lievre 549: Revision 1.128 2006/06/30 13:02:05 brouard
550: (Module): Clarifications on computing e.j
551:
1.128 brouard 552: Revision 1.127 2006/04/28 18:11:50 brouard
553: (Module): Yes the sum of survivors was wrong since
554: imach-114 because nhstepm was no more computed in the age
555: loop. Now we define nhstepma in the age loop.
556: (Module): In order to speed up (in case of numerous covariates) we
557: compute health expectancies (without variances) in a first step
558: and then all the health expectancies with variances or standard
559: deviation (needs data from the Hessian matrices) which slows the
560: computation.
561: In the future we should be able to stop the program is only health
562: expectancies and graph are needed without standard deviations.
563:
1.127 brouard 564: Revision 1.126 2006/04/28 17:23:28 brouard
565: (Module): Yes the sum of survivors was wrong since
566: imach-114 because nhstepm was no more computed in the age
567: loop. Now we define nhstepma in the age loop.
568: Version 0.98h
569:
1.126 brouard 570: Revision 1.125 2006/04/04 15:20:31 lievre
571: Errors in calculation of health expectancies. Age was not initialized.
572: Forecasting file added.
573:
574: Revision 1.124 2006/03/22 17:13:53 lievre
575: Parameters are printed with %lf instead of %f (more numbers after the comma).
576: The log-likelihood is printed in the log file
577:
578: Revision 1.123 2006/03/20 10:52:43 brouard
579: * imach.c (Module): <title> changed, corresponds to .htm file
580: name. <head> headers where missing.
581:
582: * imach.c (Module): Weights can have a decimal point as for
583: English (a comma might work with a correct LC_NUMERIC environment,
584: otherwise the weight is truncated).
585: Modification of warning when the covariates values are not 0 or
586: 1.
587: Version 0.98g
588:
589: Revision 1.122 2006/03/20 09:45:41 brouard
590: (Module): Weights can have a decimal point as for
591: English (a comma might work with a correct LC_NUMERIC environment,
592: otherwise the weight is truncated).
593: Modification of warning when the covariates values are not 0 or
594: 1.
595: Version 0.98g
596:
597: Revision 1.121 2006/03/16 17:45:01 lievre
598: * imach.c (Module): Comments concerning covariates added
599:
600: * imach.c (Module): refinements in the computation of lli if
601: status=-2 in order to have more reliable computation if stepm is
602: not 1 month. Version 0.98f
603:
604: Revision 1.120 2006/03/16 15:10:38 lievre
605: (Module): refinements in the computation of lli if
606: status=-2 in order to have more reliable computation if stepm is
607: not 1 month. Version 0.98f
608:
609: Revision 1.119 2006/03/15 17:42:26 brouard
610: (Module): Bug if status = -2, the loglikelihood was
611: computed as likelihood omitting the logarithm. Version O.98e
612:
613: Revision 1.118 2006/03/14 18:20:07 brouard
614: (Module): varevsij Comments added explaining the second
615: table of variances if popbased=1 .
616: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
617: (Module): Function pstamp added
618: (Module): Version 0.98d
619:
620: Revision 1.117 2006/03/14 17:16:22 brouard
621: (Module): varevsij Comments added explaining the second
622: table of variances if popbased=1 .
623: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
624: (Module): Function pstamp added
625: (Module): Version 0.98d
626:
627: Revision 1.116 2006/03/06 10:29:27 brouard
628: (Module): Variance-covariance wrong links and
629: varian-covariance of ej. is needed (Saito).
630:
631: Revision 1.115 2006/02/27 12:17:45 brouard
632: (Module): One freematrix added in mlikeli! 0.98c
633:
634: Revision 1.114 2006/02/26 12:57:58 brouard
635: (Module): Some improvements in processing parameter
636: filename with strsep.
637:
638: Revision 1.113 2006/02/24 14:20:24 brouard
639: (Module): Memory leaks checks with valgrind and:
640: datafile was not closed, some imatrix were not freed and on matrix
641: allocation too.
642:
643: Revision 1.112 2006/01/30 09:55:26 brouard
644: (Module): Back to gnuplot.exe instead of wgnuplot.exe
645:
646: Revision 1.111 2006/01/25 20:38:18 brouard
647: (Module): Lots of cleaning and bugs added (Gompertz)
648: (Module): Comments can be added in data file. Missing date values
649: can be a simple dot '.'.
650:
651: Revision 1.110 2006/01/25 00:51:50 brouard
652: (Module): Lots of cleaning and bugs added (Gompertz)
653:
654: Revision 1.109 2006/01/24 19:37:15 brouard
655: (Module): Comments (lines starting with a #) are allowed in data.
656:
657: Revision 1.108 2006/01/19 18:05:42 lievre
658: Gnuplot problem appeared...
659: To be fixed
660:
661: Revision 1.107 2006/01/19 16:20:37 brouard
662: Test existence of gnuplot in imach path
663:
664: Revision 1.106 2006/01/19 13:24:36 brouard
665: Some cleaning and links added in html output
666:
667: Revision 1.105 2006/01/05 20:23:19 lievre
668: *** empty log message ***
669:
670: Revision 1.104 2005/09/30 16:11:43 lievre
671: (Module): sump fixed, loop imx fixed, and simplifications.
672: (Module): If the status is missing at the last wave but we know
673: that the person is alive, then we can code his/her status as -2
674: (instead of missing=-1 in earlier versions) and his/her
675: contributions to the likelihood is 1 - Prob of dying from last
676: health status (= 1-p13= p11+p12 in the easiest case of somebody in
677: the healthy state at last known wave). Version is 0.98
678:
679: Revision 1.103 2005/09/30 15:54:49 lievre
680: (Module): sump fixed, loop imx fixed, and simplifications.
681:
682: Revision 1.102 2004/09/15 17:31:30 brouard
683: Add the possibility to read data file including tab characters.
684:
685: Revision 1.101 2004/09/15 10:38:38 brouard
686: Fix on curr_time
687:
688: Revision 1.100 2004/07/12 18:29:06 brouard
689: Add version for Mac OS X. Just define UNIX in Makefile
690:
691: Revision 1.99 2004/06/05 08:57:40 brouard
692: *** empty log message ***
693:
694: Revision 1.98 2004/05/16 15:05:56 brouard
695: New version 0.97 . First attempt to estimate force of mortality
696: directly from the data i.e. without the need of knowing the health
697: state at each age, but using a Gompertz model: log u =a + b*age .
698: This is the basic analysis of mortality and should be done before any
699: other analysis, in order to test if the mortality estimated from the
700: cross-longitudinal survey is different from the mortality estimated
701: from other sources like vital statistic data.
702:
703: The same imach parameter file can be used but the option for mle should be -3.
704:
1.133 brouard 705: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 706: former routines in order to include the new code within the former code.
707:
708: The output is very simple: only an estimate of the intercept and of
709: the slope with 95% confident intervals.
710:
711: Current limitations:
712: A) Even if you enter covariates, i.e. with the
713: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
714: B) There is no computation of Life Expectancy nor Life Table.
715:
716: Revision 1.97 2004/02/20 13:25:42 lievre
717: Version 0.96d. Population forecasting command line is (temporarily)
718: suppressed.
719:
720: Revision 1.96 2003/07/15 15:38:55 brouard
721: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
722: rewritten within the same printf. Workaround: many printfs.
723:
724: Revision 1.95 2003/07/08 07:54:34 brouard
725: * imach.c (Repository):
726: (Repository): Using imachwizard code to output a more meaningful covariance
727: matrix (cov(a12,c31) instead of numbers.
728:
729: Revision 1.94 2003/06/27 13:00:02 brouard
730: Just cleaning
731:
732: Revision 1.93 2003/06/25 16:33:55 brouard
733: (Module): On windows (cygwin) function asctime_r doesn't
734: exist so I changed back to asctime which exists.
735: (Module): Version 0.96b
736:
737: Revision 1.92 2003/06/25 16:30:45 brouard
738: (Module): On windows (cygwin) function asctime_r doesn't
739: exist so I changed back to asctime which exists.
740:
741: Revision 1.91 2003/06/25 15:30:29 brouard
742: * imach.c (Repository): Duplicated warning errors corrected.
743: (Repository): Elapsed time after each iteration is now output. It
744: helps to forecast when convergence will be reached. Elapsed time
745: is stamped in powell. We created a new html file for the graphs
746: concerning matrix of covariance. It has extension -cov.htm.
747:
748: Revision 1.90 2003/06/24 12:34:15 brouard
749: (Module): Some bugs corrected for windows. Also, when
750: mle=-1 a template is output in file "or"mypar.txt with the design
751: of the covariance matrix to be input.
752:
753: Revision 1.89 2003/06/24 12:30:52 brouard
754: (Module): Some bugs corrected for windows. Also, when
755: mle=-1 a template is output in file "or"mypar.txt with the design
756: of the covariance matrix to be input.
757:
758: Revision 1.88 2003/06/23 17:54:56 brouard
759: * 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.
760:
761: Revision 1.87 2003/06/18 12:26:01 brouard
762: Version 0.96
763:
764: Revision 1.86 2003/06/17 20:04:08 brouard
765: (Module): Change position of html and gnuplot routines and added
766: routine fileappend.
767:
768: Revision 1.85 2003/06/17 13:12:43 brouard
769: * imach.c (Repository): Check when date of death was earlier that
770: current date of interview. It may happen when the death was just
771: prior to the death. In this case, dh was negative and likelihood
772: was wrong (infinity). We still send an "Error" but patch by
773: assuming that the date of death was just one stepm after the
774: interview.
775: (Repository): Because some people have very long ID (first column)
776: we changed int to long in num[] and we added a new lvector for
777: memory allocation. But we also truncated to 8 characters (left
778: truncation)
779: (Repository): No more line truncation errors.
780:
781: Revision 1.84 2003/06/13 21:44:43 brouard
782: * imach.c (Repository): Replace "freqsummary" at a correct
783: place. It differs from routine "prevalence" which may be called
784: many times. Probs is memory consuming and must be used with
785: parcimony.
786: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
787:
788: Revision 1.83 2003/06/10 13:39:11 lievre
789: *** empty log message ***
790:
791: Revision 1.82 2003/06/05 15:57:20 brouard
792: Add log in imach.c and fullversion number is now printed.
793:
794: */
795: /*
796: Interpolated Markov Chain
797:
798: Short summary of the programme:
799:
1.227 brouard 800: This program computes Healthy Life Expectancies or State-specific
801: (if states aren't health statuses) Expectancies from
802: cross-longitudinal data. Cross-longitudinal data consist in:
803:
804: -1- a first survey ("cross") where individuals from different ages
805: are interviewed on their health status or degree of disability (in
806: the case of a health survey which is our main interest)
807:
808: -2- at least a second wave of interviews ("longitudinal") which
809: measure each change (if any) in individual health status. Health
810: expectancies are computed from the time spent in each health state
811: according to a model. More health states you consider, more time is
812: necessary to reach the Maximum Likelihood of the parameters involved
813: in the model. The simplest model is the multinomial logistic model
814: where pij is the probability to be observed in state j at the second
815: wave conditional to be observed in state i at the first
816: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
817: etc , where 'age' is age and 'sex' is a covariate. If you want to
818: have a more complex model than "constant and age", you should modify
819: the program where the markup *Covariates have to be included here
820: again* invites you to do it. More covariates you add, slower the
1.126 brouard 821: convergence.
822:
823: The advantage of this computer programme, compared to a simple
824: multinomial logistic model, is clear when the delay between waves is not
825: identical for each individual. Also, if a individual missed an
826: intermediate interview, the information is lost, but taken into
827: account using an interpolation or extrapolation.
828:
829: hPijx is the probability to be observed in state i at age x+h
830: conditional to the observed state i at age x. The delay 'h' can be
831: split into an exact number (nh*stepm) of unobserved intermediate
832: states. This elementary transition (by month, quarter,
833: semester or year) is modelled as a multinomial logistic. The hPx
834: matrix is simply the matrix product of nh*stepm elementary matrices
835: and the contribution of each individual to the likelihood is simply
836: hPijx.
837:
838: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 839: of the life expectancies. It also computes the period (stable) prevalence.
840:
841: Back prevalence and projections:
1.227 brouard 842:
843: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
844: double agemaxpar, double ftolpl, int *ncvyearp, double
845: dateprev1,double dateprev2, int firstpass, int lastpass, int
846: mobilavproj)
847:
848: Computes the back prevalence limit for any combination of
849: covariate values k at any age between ageminpar and agemaxpar and
850: returns it in **bprlim. In the loops,
851:
852: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
853: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
854:
855: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 856: Computes for any combination of covariates k and any age between bage and fage
857: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
858: oldm=oldms;savm=savms;
1.227 brouard 859:
1.267 brouard 860: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 861: Computes the transition matrix starting at age 'age' over
862: 'nhstepm*hstepm*stepm' months (i.e. until
863: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 864: nhstepm*hstepm matrices.
865:
866: Returns p3mat[i][j][h] after calling
867: p3mat[i][j][h]=matprod2(newm,
868: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
869: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
870: oldm);
1.226 brouard 871:
872: Important routines
873:
874: - func (or funcone), computes logit (pij) distinguishing
875: o fixed variables (single or product dummies or quantitative);
876: o varying variables by:
877: (1) wave (single, product dummies, quantitative),
878: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
879: % fixed dummy (treated) or quantitative (not done because time-consuming);
880: % varying dummy (not done) or quantitative (not done);
881: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
882: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
883: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
884: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
885: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 886:
1.226 brouard 887:
888:
1.133 brouard 889: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
890: Institut national d'études démographiques, Paris.
1.126 brouard 891: This software have been partly granted by Euro-REVES, a concerted action
892: from the European Union.
893: It is copyrighted identically to a GNU software product, ie programme and
894: software can be distributed freely for non commercial use. Latest version
895: can be accessed at http://euroreves.ined.fr/imach .
896:
897: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
898: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
899:
900: **********************************************************************/
901: /*
902: main
903: read parameterfile
904: read datafile
905: concatwav
906: freqsummary
907: if (mle >= 1)
908: mlikeli
909: print results files
910: if mle==1
911: computes hessian
912: read end of parameter file: agemin, agemax, bage, fage, estepm
913: begin-prev-date,...
914: open gnuplot file
915: open html file
1.145 brouard 916: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
917: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
918: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
919: freexexit2 possible for memory heap.
920:
921: h Pij x | pij_nom ficrestpij
922: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
923: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
924: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
925:
926: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
927: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
928: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
929: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
930: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
931:
1.126 brouard 932: forecasting if prevfcast==1 prevforecast call prevalence()
933: health expectancies
934: Variance-covariance of DFLE
935: prevalence()
936: movingaverage()
937: varevsij()
938: if popbased==1 varevsij(,popbased)
939: total life expectancies
940: Variance of period (stable) prevalence
941: end
942: */
943:
1.187 brouard 944: /* #define DEBUG */
945: /* #define DEBUGBRENT */
1.203 brouard 946: /* #define DEBUGLINMIN */
947: /* #define DEBUGHESS */
948: #define DEBUGHESSIJ
1.224 brouard 949: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 950: #define POWELL /* Instead of NLOPT */
1.224 brouard 951: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 952: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
953: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 954:
955: #include <math.h>
956: #include <stdio.h>
957: #include <stdlib.h>
958: #include <string.h>
1.226 brouard 959: #include <ctype.h>
1.159 brouard 960:
961: #ifdef _WIN32
962: #include <io.h>
1.172 brouard 963: #include <windows.h>
964: #include <tchar.h>
1.159 brouard 965: #else
1.126 brouard 966: #include <unistd.h>
1.159 brouard 967: #endif
1.126 brouard 968:
969: #include <limits.h>
970: #include <sys/types.h>
1.171 brouard 971:
972: #if defined(__GNUC__)
973: #include <sys/utsname.h> /* Doesn't work on Windows */
974: #endif
975:
1.126 brouard 976: #include <sys/stat.h>
977: #include <errno.h>
1.159 brouard 978: /* extern int errno; */
1.126 brouard 979:
1.157 brouard 980: /* #ifdef LINUX */
981: /* #include <time.h> */
982: /* #include "timeval.h" */
983: /* #else */
984: /* #include <sys/time.h> */
985: /* #endif */
986:
1.126 brouard 987: #include <time.h>
988:
1.136 brouard 989: #ifdef GSL
990: #include <gsl/gsl_errno.h>
991: #include <gsl/gsl_multimin.h>
992: #endif
993:
1.167 brouard 994:
1.162 brouard 995: #ifdef NLOPT
996: #include <nlopt.h>
997: typedef struct {
998: double (* function)(double [] );
999: } myfunc_data ;
1000: #endif
1001:
1.126 brouard 1002: /* #include <libintl.h> */
1003: /* #define _(String) gettext (String) */
1004:
1.251 brouard 1005: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1006:
1007: #define GNUPLOTPROGRAM "gnuplot"
1008: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1009: #define FILENAMELENGTH 132
1010:
1011: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1012: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1013:
1.144 brouard 1014: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1015: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1016:
1017: #define NINTERVMAX 8
1.144 brouard 1018: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1019: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1020: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1021: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1022: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1023: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1024: #define MAXN 20000
1.144 brouard 1025: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1026: /* #define AGESUP 130 */
1027: #define AGESUP 150
1.268 brouard 1028: #define AGEINF 0
1.218 brouard 1029: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1030: #define AGEBASE 40
1.194 brouard 1031: #define AGEOVERFLOW 1.e20
1.164 brouard 1032: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1033: #ifdef _WIN32
1034: #define DIRSEPARATOR '\\'
1035: #define CHARSEPARATOR "\\"
1036: #define ODIRSEPARATOR '/'
1037: #else
1.126 brouard 1038: #define DIRSEPARATOR '/'
1039: #define CHARSEPARATOR "/"
1040: #define ODIRSEPARATOR '\\'
1041: #endif
1042:
1.280 ! brouard 1043: /* $Id: imach.c,v 1.279 2017/07/20 13:35:01 brouard Exp $ */
1.126 brouard 1044: /* $State: Exp $ */
1.196 brouard 1045: #include "version.h"
1046: char version[]=__IMACH_VERSION__;
1.224 brouard 1047: 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";
1.280 ! brouard 1048: char fullversion[]="$Revision: 1.279 $ $Date: 2017/07/20 13:35:01 $";
1.126 brouard 1049: char strstart[80];
1050: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1051: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1052: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1053: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1054: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1055: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1056: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1057: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1058: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1059: int cptcovprodnoage=0; /**< Number of covariate products without age */
1060: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1061: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1062: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1063: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1064: int nsd=0; /**< Total number of single dummy variables (output) */
1065: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1066: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1067: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1068: int ntveff=0; /**< ntveff number of effective time varying variables */
1069: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1070: int cptcov=0; /* Working variable */
1.218 brouard 1071: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1072: int npar=NPARMAX;
1073: int nlstate=2; /* Number of live states */
1074: int ndeath=1; /* Number of dead states */
1.130 brouard 1075: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1076: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1077: int popbased=0;
1078:
1079: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1080: int maxwav=0; /* Maxim number of waves */
1081: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1082: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1083: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1084: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1085: int mle=1, weightopt=0;
1.126 brouard 1086: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1087: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1088: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1089: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1090: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1091: int selected(int kvar); /* Is covariate kvar selected for printing results */
1092:
1.130 brouard 1093: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1094: double **matprod2(); /* test */
1.126 brouard 1095: double **oldm, **newm, **savm; /* Working pointers to matrices */
1096: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1097: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1098:
1.136 brouard 1099: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1100: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1101: FILE *ficlog, *ficrespow;
1.130 brouard 1102: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1103: double fretone; /* Only one call to likelihood */
1.130 brouard 1104: long ipmx=0; /* Number of contributions */
1.126 brouard 1105: double sw; /* Sum of weights */
1106: char filerespow[FILENAMELENGTH];
1107: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1108: FILE *ficresilk;
1109: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1110: FILE *ficresprobmorprev;
1111: FILE *fichtm, *fichtmcov; /* Html File */
1112: FILE *ficreseij;
1113: char filerese[FILENAMELENGTH];
1114: FILE *ficresstdeij;
1115: char fileresstde[FILENAMELENGTH];
1116: FILE *ficrescveij;
1117: char filerescve[FILENAMELENGTH];
1118: FILE *ficresvij;
1119: char fileresv[FILENAMELENGTH];
1.269 brouard 1120:
1.126 brouard 1121: char title[MAXLINE];
1.234 brouard 1122: char model[MAXLINE]; /**< The model line */
1.217 brouard 1123: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1124: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1125: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1126: char command[FILENAMELENGTH];
1127: int outcmd=0;
1128:
1.217 brouard 1129: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1130: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1131: char filelog[FILENAMELENGTH]; /* Log file */
1132: char filerest[FILENAMELENGTH];
1133: char fileregp[FILENAMELENGTH];
1134: char popfile[FILENAMELENGTH];
1135:
1136: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1137:
1.157 brouard 1138: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1139: /* struct timezone tzp; */
1140: /* extern int gettimeofday(); */
1141: struct tm tml, *gmtime(), *localtime();
1142:
1143: extern time_t time();
1144:
1145: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1146: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1147: struct tm tm;
1148:
1.126 brouard 1149: char strcurr[80], strfor[80];
1150:
1151: char *endptr;
1152: long lval;
1153: double dval;
1154:
1155: #define NR_END 1
1156: #define FREE_ARG char*
1157: #define FTOL 1.0e-10
1158:
1159: #define NRANSI
1.240 brouard 1160: #define ITMAX 200
1161: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1162:
1163: #define TOL 2.0e-4
1164:
1165: #define CGOLD 0.3819660
1166: #define ZEPS 1.0e-10
1167: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1168:
1169: #define GOLD 1.618034
1170: #define GLIMIT 100.0
1171: #define TINY 1.0e-20
1172:
1173: static double maxarg1,maxarg2;
1174: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1175: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1176:
1177: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1178: #define rint(a) floor(a+0.5)
1.166 brouard 1179: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1180: #define mytinydouble 1.0e-16
1.166 brouard 1181: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1182: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1183: /* static double dsqrarg; */
1184: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1185: static double sqrarg;
1186: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1187: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1188: int agegomp= AGEGOMP;
1189:
1190: int imx;
1191: int stepm=1;
1192: /* Stepm, step in month: minimum step interpolation*/
1193:
1194: int estepm;
1195: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1196:
1197: int m,nb;
1198: long *num;
1.197 brouard 1199: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1200: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1201: covariate for which somebody answered excluding
1202: undefined. Usually 2: 0 and 1. */
1203: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1204: covariate for which somebody answered including
1205: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1206: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1207: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1208: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1209: double *ageexmed,*agecens;
1210: double dateintmean=0;
1211:
1212: double *weight;
1213: int **s; /* Status */
1.141 brouard 1214: double *agedc;
1.145 brouard 1215: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1216: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1217: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1218: double **coqvar; /* Fixed quantitative covariate nqv */
1219: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1220: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1221: double idx;
1222: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1223: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1224: /*k 1 2 3 4 5 6 7 8 9 */
1225: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1226: /* Tndvar[k] 1 2 3 4 5 */
1227: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1228: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1229: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1230: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1231: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1232: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1233: /* Tprod[i]=k 4 7 */
1234: /* Tage[i]=k 5 8 */
1235: /* */
1236: /* Type */
1237: /* V 1 2 3 4 5 */
1238: /* F F V V V */
1239: /* D Q D D Q */
1240: /* */
1241: int *TvarsD;
1242: int *TvarsDind;
1243: int *TvarsQ;
1244: int *TvarsQind;
1245:
1.235 brouard 1246: #define MAXRESULTLINES 10
1247: int nresult=0;
1.258 brouard 1248: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1249: int TKresult[MAXRESULTLINES];
1.237 brouard 1250: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1251: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1252: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1253: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1254: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1255: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1256:
1.234 brouard 1257: /* 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 1258: 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 */
1259: 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 */
1260: 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 */
1261: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1262: 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 */
1263: 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 1264: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1265: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1266: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1267: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1268: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1269: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1270: 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 */
1271: 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 */
1272:
1.230 brouard 1273: int *Tvarsel; /**< Selected covariates for output */
1274: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1275: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1276: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1277: 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 1278: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1279: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1280: int *Tage;
1.227 brouard 1281: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1282: 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 1283: 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*/
1284: 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 1285: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1286: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1287: int **Tvard;
1288: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1289: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1290: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1291: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1292: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1293: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1294: double *lsurv, *lpop, *tpop;
1295:
1.231 brouard 1296: #define FD 1; /* Fixed dummy covariate */
1297: #define FQ 2; /* Fixed quantitative covariate */
1298: #define FP 3; /* Fixed product covariate */
1299: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1300: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1301: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1302: #define VD 10; /* Varying dummy covariate */
1303: #define VQ 11; /* Varying quantitative covariate */
1304: #define VP 12; /* Varying product covariate */
1305: #define VPDD 13; /* Varying product dummy*dummy covariate */
1306: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1307: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1308: #define APFD 16; /* Age product * fixed dummy covariate */
1309: #define APFQ 17; /* Age product * fixed quantitative covariate */
1310: #define APVD 18; /* Age product * varying dummy covariate */
1311: #define APVQ 19; /* Age product * varying quantitative covariate */
1312:
1313: #define FTYPE 1; /* Fixed covariate */
1314: #define VTYPE 2; /* Varying covariate (loop in wave) */
1315: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1316:
1317: struct kmodel{
1318: int maintype; /* main type */
1319: int subtype; /* subtype */
1320: };
1321: struct kmodel modell[NCOVMAX];
1322:
1.143 brouard 1323: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1324: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1325:
1326: /**************** split *************************/
1327: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1328: {
1329: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1330: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1331: */
1332: char *ss; /* pointer */
1.186 brouard 1333: int l1=0, l2=0; /* length counters */
1.126 brouard 1334:
1335: l1 = strlen(path ); /* length of path */
1336: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1337: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1338: if ( ss == NULL ) { /* no directory, so determine current directory */
1339: strcpy( name, path ); /* we got the fullname name because no directory */
1340: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1341: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1342: /* get current working directory */
1343: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1344: #ifdef WIN32
1345: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1346: #else
1347: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1348: #endif
1.126 brouard 1349: return( GLOCK_ERROR_GETCWD );
1350: }
1351: /* got dirc from getcwd*/
1352: printf(" DIRC = %s \n",dirc);
1.205 brouard 1353: } else { /* strip directory from path */
1.126 brouard 1354: ss++; /* after this, the filename */
1355: l2 = strlen( ss ); /* length of filename */
1356: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1357: strcpy( name, ss ); /* save file name */
1358: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1359: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1360: printf(" DIRC2 = %s \n",dirc);
1361: }
1362: /* We add a separator at the end of dirc if not exists */
1363: l1 = strlen( dirc ); /* length of directory */
1364: if( dirc[l1-1] != DIRSEPARATOR ){
1365: dirc[l1] = DIRSEPARATOR;
1366: dirc[l1+1] = 0;
1367: printf(" DIRC3 = %s \n",dirc);
1368: }
1369: ss = strrchr( name, '.' ); /* find last / */
1370: if (ss >0){
1371: ss++;
1372: strcpy(ext,ss); /* save extension */
1373: l1= strlen( name);
1374: l2= strlen(ss)+1;
1375: strncpy( finame, name, l1-l2);
1376: finame[l1-l2]= 0;
1377: }
1378:
1379: return( 0 ); /* we're done */
1380: }
1381:
1382:
1383: /******************************************/
1384:
1385: void replace_back_to_slash(char *s, char*t)
1386: {
1387: int i;
1388: int lg=0;
1389: i=0;
1390: lg=strlen(t);
1391: for(i=0; i<= lg; i++) {
1392: (s[i] = t[i]);
1393: if (t[i]== '\\') s[i]='/';
1394: }
1395: }
1396:
1.132 brouard 1397: char *trimbb(char *out, char *in)
1.137 brouard 1398: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1399: char *s;
1400: s=out;
1401: while (*in != '\0'){
1.137 brouard 1402: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1403: in++;
1404: }
1405: *out++ = *in++;
1406: }
1407: *out='\0';
1408: return s;
1409: }
1410:
1.187 brouard 1411: /* char *substrchaine(char *out, char *in, char *chain) */
1412: /* { */
1413: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1414: /* char *s, *t; */
1415: /* t=in;s=out; */
1416: /* while ((*in != *chain) && (*in != '\0')){ */
1417: /* *out++ = *in++; */
1418: /* } */
1419:
1420: /* /\* *in matches *chain *\/ */
1421: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1422: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1423: /* } */
1424: /* in--; chain--; */
1425: /* while ( (*in != '\0')){ */
1426: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1427: /* *out++ = *in++; */
1428: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1429: /* } */
1430: /* *out='\0'; */
1431: /* out=s; */
1432: /* return out; */
1433: /* } */
1434: char *substrchaine(char *out, char *in, char *chain)
1435: {
1436: /* Substract chain 'chain' from 'in', return and output 'out' */
1437: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1438:
1439: char *strloc;
1440:
1441: strcpy (out, in);
1442: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1443: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1444: if(strloc != NULL){
1445: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1446: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1447: /* strcpy (strloc, strloc +strlen(chain));*/
1448: }
1449: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1450: return out;
1451: }
1452:
1453:
1.145 brouard 1454: char *cutl(char *blocc, char *alocc, char *in, char occ)
1455: {
1.187 brouard 1456: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1457: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1458: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1459: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1460: */
1.160 brouard 1461: char *s, *t;
1.145 brouard 1462: t=in;s=in;
1463: while ((*in != occ) && (*in != '\0')){
1464: *alocc++ = *in++;
1465: }
1466: if( *in == occ){
1467: *(alocc)='\0';
1468: s=++in;
1469: }
1470:
1471: if (s == t) {/* occ not found */
1472: *(alocc-(in-s))='\0';
1473: in=s;
1474: }
1475: while ( *in != '\0'){
1476: *blocc++ = *in++;
1477: }
1478:
1479: *blocc='\0';
1480: return t;
1481: }
1.137 brouard 1482: char *cutv(char *blocc, char *alocc, char *in, char occ)
1483: {
1.187 brouard 1484: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1485: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1486: gives blocc="abcdef2ghi" and alocc="j".
1487: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1488: */
1489: char *s, *t;
1490: t=in;s=in;
1491: while (*in != '\0'){
1492: while( *in == occ){
1493: *blocc++ = *in++;
1494: s=in;
1495: }
1496: *blocc++ = *in++;
1497: }
1498: if (s == t) /* occ not found */
1499: *(blocc-(in-s))='\0';
1500: else
1501: *(blocc-(in-s)-1)='\0';
1502: in=s;
1503: while ( *in != '\0'){
1504: *alocc++ = *in++;
1505: }
1506:
1507: *alocc='\0';
1508: return s;
1509: }
1510:
1.126 brouard 1511: int nbocc(char *s, char occ)
1512: {
1513: int i,j=0;
1514: int lg=20;
1515: i=0;
1516: lg=strlen(s);
1517: for(i=0; i<= lg; i++) {
1.234 brouard 1518: if (s[i] == occ ) j++;
1.126 brouard 1519: }
1520: return j;
1521: }
1522:
1.137 brouard 1523: /* void cutv(char *u,char *v, char*t, char occ) */
1524: /* { */
1525: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1526: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1527: /* gives u="abcdef2ghi" and v="j" *\/ */
1528: /* int i,lg,j,p=0; */
1529: /* i=0; */
1530: /* lg=strlen(t); */
1531: /* for(j=0; j<=lg-1; j++) { */
1532: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1533: /* } */
1.126 brouard 1534:
1.137 brouard 1535: /* for(j=0; j<p; j++) { */
1536: /* (u[j] = t[j]); */
1537: /* } */
1538: /* u[p]='\0'; */
1.126 brouard 1539:
1.137 brouard 1540: /* for(j=0; j<= lg; j++) { */
1541: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1542: /* } */
1543: /* } */
1.126 brouard 1544:
1.160 brouard 1545: #ifdef _WIN32
1546: char * strsep(char **pp, const char *delim)
1547: {
1548: char *p, *q;
1549:
1550: if ((p = *pp) == NULL)
1551: return 0;
1552: if ((q = strpbrk (p, delim)) != NULL)
1553: {
1554: *pp = q + 1;
1555: *q = '\0';
1556: }
1557: else
1558: *pp = 0;
1559: return p;
1560: }
1561: #endif
1562:
1.126 brouard 1563: /********************** nrerror ********************/
1564:
1565: void nrerror(char error_text[])
1566: {
1567: fprintf(stderr,"ERREUR ...\n");
1568: fprintf(stderr,"%s\n",error_text);
1569: exit(EXIT_FAILURE);
1570: }
1571: /*********************** vector *******************/
1572: double *vector(int nl, int nh)
1573: {
1574: double *v;
1575: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1576: if (!v) nrerror("allocation failure in vector");
1577: return v-nl+NR_END;
1578: }
1579:
1580: /************************ free vector ******************/
1581: void free_vector(double*v, int nl, int nh)
1582: {
1583: free((FREE_ARG)(v+nl-NR_END));
1584: }
1585:
1586: /************************ivector *******************************/
1587: int *ivector(long nl,long nh)
1588: {
1589: int *v;
1590: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1591: if (!v) nrerror("allocation failure in ivector");
1592: return v-nl+NR_END;
1593: }
1594:
1595: /******************free ivector **************************/
1596: void free_ivector(int *v, long nl, long nh)
1597: {
1598: free((FREE_ARG)(v+nl-NR_END));
1599: }
1600:
1601: /************************lvector *******************************/
1602: long *lvector(long nl,long nh)
1603: {
1604: long *v;
1605: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1606: if (!v) nrerror("allocation failure in ivector");
1607: return v-nl+NR_END;
1608: }
1609:
1610: /******************free lvector **************************/
1611: void free_lvector(long *v, long nl, long nh)
1612: {
1613: free((FREE_ARG)(v+nl-NR_END));
1614: }
1615:
1616: /******************* imatrix *******************************/
1617: int **imatrix(long nrl, long nrh, long ncl, long nch)
1618: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1619: {
1620: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1621: int **m;
1622:
1623: /* allocate pointers to rows */
1624: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1625: if (!m) nrerror("allocation failure 1 in matrix()");
1626: m += NR_END;
1627: m -= nrl;
1628:
1629:
1630: /* allocate rows and set pointers to them */
1631: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1632: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1633: m[nrl] += NR_END;
1634: m[nrl] -= ncl;
1635:
1636: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1637:
1638: /* return pointer to array of pointers to rows */
1639: return m;
1640: }
1641:
1642: /****************** free_imatrix *************************/
1643: void free_imatrix(m,nrl,nrh,ncl,nch)
1644: int **m;
1645: long nch,ncl,nrh,nrl;
1646: /* free an int matrix allocated by imatrix() */
1647: {
1648: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1649: free((FREE_ARG) (m+nrl-NR_END));
1650: }
1651:
1652: /******************* matrix *******************************/
1653: double **matrix(long nrl, long nrh, long ncl, long nch)
1654: {
1655: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1656: double **m;
1657:
1658: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1659: if (!m) nrerror("allocation failure 1 in matrix()");
1660: m += NR_END;
1661: m -= nrl;
1662:
1663: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1664: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1665: m[nrl] += NR_END;
1666: m[nrl] -= ncl;
1667:
1668: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1669: return m;
1.145 brouard 1670: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1671: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1672: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1673: */
1674: }
1675:
1676: /*************************free matrix ************************/
1677: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1678: {
1679: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1680: free((FREE_ARG)(m+nrl-NR_END));
1681: }
1682:
1683: /******************* ma3x *******************************/
1684: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1685: {
1686: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1687: double ***m;
1688:
1689: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1690: if (!m) nrerror("allocation failure 1 in matrix()");
1691: m += NR_END;
1692: m -= nrl;
1693:
1694: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1695: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1696: m[nrl] += NR_END;
1697: m[nrl] -= ncl;
1698:
1699: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1700:
1701: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1702: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1703: m[nrl][ncl] += NR_END;
1704: m[nrl][ncl] -= nll;
1705: for (j=ncl+1; j<=nch; j++)
1706: m[nrl][j]=m[nrl][j-1]+nlay;
1707:
1708: for (i=nrl+1; i<=nrh; i++) {
1709: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1710: for (j=ncl+1; j<=nch; j++)
1711: m[i][j]=m[i][j-1]+nlay;
1712: }
1713: return m;
1714: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1715: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1716: */
1717: }
1718:
1719: /*************************free ma3x ************************/
1720: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1721: {
1722: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1723: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1724: free((FREE_ARG)(m+nrl-NR_END));
1725: }
1726:
1727: /*************** function subdirf ***********/
1728: char *subdirf(char fileres[])
1729: {
1730: /* Caution optionfilefiname is hidden */
1731: strcpy(tmpout,optionfilefiname);
1732: strcat(tmpout,"/"); /* Add to the right */
1733: strcat(tmpout,fileres);
1734: return tmpout;
1735: }
1736:
1737: /*************** function subdirf2 ***********/
1738: char *subdirf2(char fileres[], char *preop)
1739: {
1740:
1741: /* Caution optionfilefiname is hidden */
1742: strcpy(tmpout,optionfilefiname);
1743: strcat(tmpout,"/");
1744: strcat(tmpout,preop);
1745: strcat(tmpout,fileres);
1746: return tmpout;
1747: }
1748:
1749: /*************** function subdirf3 ***********/
1750: char *subdirf3(char fileres[], char *preop, char *preop2)
1751: {
1752:
1753: /* Caution optionfilefiname is hidden */
1754: strcpy(tmpout,optionfilefiname);
1755: strcat(tmpout,"/");
1756: strcat(tmpout,preop);
1757: strcat(tmpout,preop2);
1758: strcat(tmpout,fileres);
1759: return tmpout;
1760: }
1.213 brouard 1761:
1762: /*************** function subdirfext ***********/
1763: char *subdirfext(char fileres[], char *preop, char *postop)
1764: {
1765:
1766: strcpy(tmpout,preop);
1767: strcat(tmpout,fileres);
1768: strcat(tmpout,postop);
1769: return tmpout;
1770: }
1.126 brouard 1771:
1.213 brouard 1772: /*************** function subdirfext3 ***********/
1773: char *subdirfext3(char fileres[], char *preop, char *postop)
1774: {
1775:
1776: /* Caution optionfilefiname is hidden */
1777: strcpy(tmpout,optionfilefiname);
1778: strcat(tmpout,"/");
1779: strcat(tmpout,preop);
1780: strcat(tmpout,fileres);
1781: strcat(tmpout,postop);
1782: return tmpout;
1783: }
1784:
1.162 brouard 1785: char *asc_diff_time(long time_sec, char ascdiff[])
1786: {
1787: long sec_left, days, hours, minutes;
1788: days = (time_sec) / (60*60*24);
1789: sec_left = (time_sec) % (60*60*24);
1790: hours = (sec_left) / (60*60) ;
1791: sec_left = (sec_left) %(60*60);
1792: minutes = (sec_left) /60;
1793: sec_left = (sec_left) % (60);
1794: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1795: return ascdiff;
1796: }
1797:
1.126 brouard 1798: /***************** f1dim *************************/
1799: extern int ncom;
1800: extern double *pcom,*xicom;
1801: extern double (*nrfunc)(double []);
1802:
1803: double f1dim(double x)
1804: {
1805: int j;
1806: double f;
1807: double *xt;
1808:
1809: xt=vector(1,ncom);
1810: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1811: f=(*nrfunc)(xt);
1812: free_vector(xt,1,ncom);
1813: return f;
1814: }
1815:
1816: /*****************brent *************************/
1817: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1818: {
1819: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1820: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1821: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1822: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1823: * returned function value.
1824: */
1.126 brouard 1825: int iter;
1826: double a,b,d,etemp;
1.159 brouard 1827: double fu=0,fv,fw,fx;
1.164 brouard 1828: double ftemp=0.;
1.126 brouard 1829: double p,q,r,tol1,tol2,u,v,w,x,xm;
1830: double e=0.0;
1831:
1832: a=(ax < cx ? ax : cx);
1833: b=(ax > cx ? ax : cx);
1834: x=w=v=bx;
1835: fw=fv=fx=(*f)(x);
1836: for (iter=1;iter<=ITMAX;iter++) {
1837: xm=0.5*(a+b);
1838: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1839: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1840: printf(".");fflush(stdout);
1841: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1842: #ifdef DEBUGBRENT
1.126 brouard 1843: 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);
1844: 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);
1845: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1846: #endif
1847: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1848: *xmin=x;
1849: return fx;
1850: }
1851: ftemp=fu;
1852: if (fabs(e) > tol1) {
1853: r=(x-w)*(fx-fv);
1854: q=(x-v)*(fx-fw);
1855: p=(x-v)*q-(x-w)*r;
1856: q=2.0*(q-r);
1857: if (q > 0.0) p = -p;
1858: q=fabs(q);
1859: etemp=e;
1860: e=d;
1861: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1862: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1863: else {
1.224 brouard 1864: d=p/q;
1865: u=x+d;
1866: if (u-a < tol2 || b-u < tol2)
1867: d=SIGN(tol1,xm-x);
1.126 brouard 1868: }
1869: } else {
1870: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1871: }
1872: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1873: fu=(*f)(u);
1874: if (fu <= fx) {
1875: if (u >= x) a=x; else b=x;
1876: SHFT(v,w,x,u)
1.183 brouard 1877: SHFT(fv,fw,fx,fu)
1878: } else {
1879: if (u < x) a=u; else b=u;
1880: if (fu <= fw || w == x) {
1.224 brouard 1881: v=w;
1882: w=u;
1883: fv=fw;
1884: fw=fu;
1.183 brouard 1885: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1886: v=u;
1887: fv=fu;
1.183 brouard 1888: }
1889: }
1.126 brouard 1890: }
1891: nrerror("Too many iterations in brent");
1892: *xmin=x;
1893: return fx;
1894: }
1895:
1896: /****************** mnbrak ***********************/
1897:
1898: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1899: double (*func)(double))
1.183 brouard 1900: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1901: the downhill direction (defined by the function as evaluated at the initial points) and returns
1902: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1903: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1904: */
1.126 brouard 1905: double ulim,u,r,q, dum;
1906: double fu;
1.187 brouard 1907:
1908: double scale=10.;
1909: int iterscale=0;
1910:
1911: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1912: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1913:
1914:
1915: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1916: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1917: /* *bx = *ax - (*ax - *bx)/scale; */
1918: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1919: /* } */
1920:
1.126 brouard 1921: if (*fb > *fa) {
1922: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1923: SHFT(dum,*fb,*fa,dum)
1924: }
1.126 brouard 1925: *cx=(*bx)+GOLD*(*bx-*ax);
1926: *fc=(*func)(*cx);
1.183 brouard 1927: #ifdef DEBUG
1.224 brouard 1928: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1929: 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 1930: #endif
1.224 brouard 1931: 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 1932: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1933: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1934: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1935: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1936: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1937: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1938: fu=(*func)(u);
1.163 brouard 1939: #ifdef DEBUG
1940: /* f(x)=A(x-u)**2+f(u) */
1941: double A, fparabu;
1942: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1943: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1944: 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);
1945: 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 1946: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1947: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1948: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1949: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1950: #endif
1.184 brouard 1951: #ifdef MNBRAKORIGINAL
1.183 brouard 1952: #else
1.191 brouard 1953: /* if (fu > *fc) { */
1954: /* #ifdef DEBUG */
1955: /* printf("mnbrak4 fu > fc \n"); */
1956: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1957: /* #endif */
1958: /* /\* 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 *\\/ *\/ */
1959: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1960: /* dum=u; /\* Shifting c and u *\/ */
1961: /* u = *cx; */
1962: /* *cx = dum; */
1963: /* dum = fu; */
1964: /* fu = *fc; */
1965: /* *fc =dum; */
1966: /* } else { /\* end *\/ */
1967: /* #ifdef DEBUG */
1968: /* printf("mnbrak3 fu < fc \n"); */
1969: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1970: /* #endif */
1971: /* dum=u; /\* Shifting c and u *\/ */
1972: /* u = *cx; */
1973: /* *cx = dum; */
1974: /* dum = fu; */
1975: /* fu = *fc; */
1976: /* *fc =dum; */
1977: /* } */
1.224 brouard 1978: #ifdef DEBUGMNBRAK
1979: double A, fparabu;
1980: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1981: fparabu= *fa - A*(*ax-u)*(*ax-u);
1982: 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);
1983: 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 1984: #endif
1.191 brouard 1985: dum=u; /* Shifting c and u */
1986: u = *cx;
1987: *cx = dum;
1988: dum = fu;
1989: fu = *fc;
1990: *fc =dum;
1.183 brouard 1991: #endif
1.162 brouard 1992: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1993: #ifdef DEBUG
1.224 brouard 1994: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1995: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1996: #endif
1.126 brouard 1997: fu=(*func)(u);
1998: if (fu < *fc) {
1.183 brouard 1999: #ifdef DEBUG
1.224 brouard 2000: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2001: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2002: #endif
2003: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2004: SHFT(*fb,*fc,fu,(*func)(u))
2005: #ifdef DEBUG
2006: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2007: #endif
2008: }
1.162 brouard 2009: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2010: #ifdef DEBUG
1.224 brouard 2011: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2012: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2013: #endif
1.126 brouard 2014: u=ulim;
2015: fu=(*func)(u);
1.183 brouard 2016: } else { /* u could be left to b (if r > q parabola has a maximum) */
2017: #ifdef DEBUG
1.224 brouard 2018: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2019: 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 2020: #endif
1.126 brouard 2021: u=(*cx)+GOLD*(*cx-*bx);
2022: fu=(*func)(u);
1.224 brouard 2023: #ifdef DEBUG
2024: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2025: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2026: #endif
1.183 brouard 2027: } /* end tests */
1.126 brouard 2028: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2029: SHFT(*fa,*fb,*fc,fu)
2030: #ifdef DEBUG
1.224 brouard 2031: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2032: 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 2033: #endif
2034: } /* 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 2035: }
2036:
2037: /*************** linmin ************************/
1.162 brouard 2038: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2039: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2040: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2041: the value of func at the returned location p . This is actually all accomplished by calling the
2042: routines mnbrak and brent .*/
1.126 brouard 2043: int ncom;
2044: double *pcom,*xicom;
2045: double (*nrfunc)(double []);
2046:
1.224 brouard 2047: #ifdef LINMINORIGINAL
1.126 brouard 2048: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2049: #else
2050: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2051: #endif
1.126 brouard 2052: {
2053: double brent(double ax, double bx, double cx,
2054: double (*f)(double), double tol, double *xmin);
2055: double f1dim(double x);
2056: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2057: double *fc, double (*func)(double));
2058: int j;
2059: double xx,xmin,bx,ax;
2060: double fx,fb,fa;
1.187 brouard 2061:
1.203 brouard 2062: #ifdef LINMINORIGINAL
2063: #else
2064: double scale=10., axs, xxs; /* Scale added for infinity */
2065: #endif
2066:
1.126 brouard 2067: ncom=n;
2068: pcom=vector(1,n);
2069: xicom=vector(1,n);
2070: nrfunc=func;
2071: for (j=1;j<=n;j++) {
2072: pcom[j]=p[j];
1.202 brouard 2073: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2074: }
1.187 brouard 2075:
1.203 brouard 2076: #ifdef LINMINORIGINAL
2077: xx=1.;
2078: #else
2079: axs=0.0;
2080: xxs=1.;
2081: do{
2082: xx= xxs;
2083: #endif
1.187 brouard 2084: ax=0.;
2085: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2086: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2087: /* 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)) */
2088: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2089: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2090: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2091: /* 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 2092: #ifdef LINMINORIGINAL
2093: #else
2094: if (fx != fx){
1.224 brouard 2095: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2096: printf("|");
2097: fprintf(ficlog,"|");
1.203 brouard 2098: #ifdef DEBUGLINMIN
1.224 brouard 2099: 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 2100: #endif
2101: }
1.224 brouard 2102: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2103: #endif
2104:
1.191 brouard 2105: #ifdef DEBUGLINMIN
2106: 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 2107: 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 2108: #endif
1.224 brouard 2109: #ifdef LINMINORIGINAL
2110: #else
2111: if(fb == fx){ /* Flat function in the direction */
2112: xmin=xx;
2113: *flat=1;
2114: }else{
2115: *flat=0;
2116: #endif
2117: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2118: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2119: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2120: /* fmin = f(p[j] + xmin * xi[j]) */
2121: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2122: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2123: #ifdef DEBUG
1.224 brouard 2124: 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);
2125: 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);
2126: #endif
2127: #ifdef LINMINORIGINAL
2128: #else
2129: }
1.126 brouard 2130: #endif
1.191 brouard 2131: #ifdef DEBUGLINMIN
2132: printf("linmin end ");
1.202 brouard 2133: fprintf(ficlog,"linmin end ");
1.191 brouard 2134: #endif
1.126 brouard 2135: for (j=1;j<=n;j++) {
1.203 brouard 2136: #ifdef LINMINORIGINAL
2137: xi[j] *= xmin;
2138: #else
2139: #ifdef DEBUGLINMIN
2140: if(xxs <1.0)
2141: printf(" before xi[%d]=%12.8f", j,xi[j]);
2142: #endif
2143: 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) */
2144: #ifdef DEBUGLINMIN
2145: if(xxs <1.0)
2146: 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 );
2147: #endif
2148: #endif
1.187 brouard 2149: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2150: }
1.191 brouard 2151: #ifdef DEBUGLINMIN
1.203 brouard 2152: printf("\n");
1.191 brouard 2153: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2154: 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 2155: for (j=1;j<=n;j++) {
1.202 brouard 2156: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2157: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2158: if(j % ncovmodel == 0){
1.191 brouard 2159: printf("\n");
1.202 brouard 2160: fprintf(ficlog,"\n");
2161: }
1.191 brouard 2162: }
1.203 brouard 2163: #else
1.191 brouard 2164: #endif
1.126 brouard 2165: free_vector(xicom,1,n);
2166: free_vector(pcom,1,n);
2167: }
2168:
2169:
2170: /*************** powell ************************/
1.162 brouard 2171: /*
2172: Minimization of a function func of n variables. Input consists of an initial starting point
2173: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2174: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2175: such that failure to decrease by more than this amount on one iteration signals doneness. On
2176: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2177: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2178: */
1.224 brouard 2179: #ifdef LINMINORIGINAL
2180: #else
2181: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2182: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2183: #endif
1.126 brouard 2184: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2185: double (*func)(double []))
2186: {
1.224 brouard 2187: #ifdef LINMINORIGINAL
2188: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2189: double (*func)(double []));
1.224 brouard 2190: #else
1.241 brouard 2191: void linmin(double p[], double xi[], int n, double *fret,
2192: double (*func)(double []),int *flat);
1.224 brouard 2193: #endif
1.239 brouard 2194: int i,ibig,j,jk,k;
1.126 brouard 2195: double del,t,*pt,*ptt,*xit;
1.181 brouard 2196: double directest;
1.126 brouard 2197: double fp,fptt;
2198: double *xits;
2199: int niterf, itmp;
1.224 brouard 2200: #ifdef LINMINORIGINAL
2201: #else
2202:
2203: flatdir=ivector(1,n);
2204: for (j=1;j<=n;j++) flatdir[j]=0;
2205: #endif
1.126 brouard 2206:
2207: pt=vector(1,n);
2208: ptt=vector(1,n);
2209: xit=vector(1,n);
2210: xits=vector(1,n);
2211: *fret=(*func)(p);
2212: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2213: rcurr_time = time(NULL);
1.126 brouard 2214: for (*iter=1;;++(*iter)) {
1.187 brouard 2215: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2216: ibig=0;
2217: del=0.0;
1.157 brouard 2218: rlast_time=rcurr_time;
2219: /* (void) gettimeofday(&curr_time,&tzp); */
2220: rcurr_time = time(NULL);
2221: curr_time = *localtime(&rcurr_time);
2222: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2223: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2224: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2225: for (i=1;i<=n;i++) {
1.126 brouard 2226: fprintf(ficrespow," %.12lf", p[i]);
2227: }
1.239 brouard 2228: fprintf(ficrespow,"\n");fflush(ficrespow);
2229: printf("\n#model= 1 + age ");
2230: fprintf(ficlog,"\n#model= 1 + age ");
2231: if(nagesqr==1){
1.241 brouard 2232: printf(" + age*age ");
2233: fprintf(ficlog," + age*age ");
1.239 brouard 2234: }
2235: for(j=1;j <=ncovmodel-2;j++){
2236: if(Typevar[j]==0) {
2237: printf(" + V%d ",Tvar[j]);
2238: fprintf(ficlog," + V%d ",Tvar[j]);
2239: }else if(Typevar[j]==1) {
2240: printf(" + V%d*age ",Tvar[j]);
2241: fprintf(ficlog," + V%d*age ",Tvar[j]);
2242: }else if(Typevar[j]==2) {
2243: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2244: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2245: }
2246: }
1.126 brouard 2247: printf("\n");
1.239 brouard 2248: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2249: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2250: fprintf(ficlog,"\n");
1.239 brouard 2251: for(i=1,jk=1; i <=nlstate; i++){
2252: for(k=1; k <=(nlstate+ndeath); k++){
2253: if (k != i) {
2254: printf("%d%d ",i,k);
2255: fprintf(ficlog,"%d%d ",i,k);
2256: for(j=1; j <=ncovmodel; j++){
2257: printf("%12.7f ",p[jk]);
2258: fprintf(ficlog,"%12.7f ",p[jk]);
2259: jk++;
2260: }
2261: printf("\n");
2262: fprintf(ficlog,"\n");
2263: }
2264: }
2265: }
1.241 brouard 2266: if(*iter <=3 && *iter >1){
1.157 brouard 2267: tml = *localtime(&rcurr_time);
2268: strcpy(strcurr,asctime(&tml));
2269: rforecast_time=rcurr_time;
1.126 brouard 2270: itmp = strlen(strcurr);
2271: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2272: strcurr[itmp-1]='\0';
1.162 brouard 2273: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2274: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2275: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2276: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2277: forecast_time = *localtime(&rforecast_time);
2278: strcpy(strfor,asctime(&forecast_time));
2279: itmp = strlen(strfor);
2280: if(strfor[itmp-1]=='\n')
2281: strfor[itmp-1]='\0';
2282: 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);
2283: 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 2284: }
2285: }
1.187 brouard 2286: for (i=1;i<=n;i++) { /* For each direction i */
2287: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2288: fptt=(*fret);
2289: #ifdef DEBUG
1.203 brouard 2290: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2291: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2292: #endif
1.203 brouard 2293: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2294: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2295: #ifdef LINMINORIGINAL
1.188 brouard 2296: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2297: #else
2298: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2299: flatdir[i]=flat; /* Function is vanishing in that direction i */
2300: #endif
2301: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2302: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2303: /* because that direction will be replaced unless the gain del is small */
2304: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2305: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2306: /* with the new direction. */
2307: del=fabs(fptt-(*fret));
2308: ibig=i;
1.126 brouard 2309: }
2310: #ifdef DEBUG
2311: printf("%d %.12e",i,(*fret));
2312: fprintf(ficlog,"%d %.12e",i,(*fret));
2313: for (j=1;j<=n;j++) {
1.224 brouard 2314: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2315: printf(" x(%d)=%.12e",j,xit[j]);
2316: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2317: }
2318: for(j=1;j<=n;j++) {
1.225 brouard 2319: printf(" p(%d)=%.12e",j,p[j]);
2320: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2321: }
2322: printf("\n");
2323: fprintf(ficlog,"\n");
2324: #endif
1.187 brouard 2325: } /* end loop on each direction i */
2326: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2327: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2328: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2329: for(j=1;j<=n;j++) {
1.225 brouard 2330: if(flatdir[j] >0){
2331: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2332: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2333: }
2334: /* printf("\n"); */
2335: /* fprintf(ficlog,"\n"); */
2336: }
1.243 brouard 2337: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2338: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2339: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2340: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2341: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2342: /* decreased of more than 3.84 */
2343: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2344: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2345: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2346:
1.188 brouard 2347: /* Starting the program with initial values given by a former maximization will simply change */
2348: /* the scales of the directions and the directions, because the are reset to canonical directions */
2349: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2350: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2351: #ifdef DEBUG
2352: int k[2],l;
2353: k[0]=1;
2354: k[1]=-1;
2355: printf("Max: %.12e",(*func)(p));
2356: fprintf(ficlog,"Max: %.12e",(*func)(p));
2357: for (j=1;j<=n;j++) {
2358: printf(" %.12e",p[j]);
2359: fprintf(ficlog," %.12e",p[j]);
2360: }
2361: printf("\n");
2362: fprintf(ficlog,"\n");
2363: for(l=0;l<=1;l++) {
2364: for (j=1;j<=n;j++) {
2365: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2366: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2367: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2368: }
2369: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2370: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2371: }
2372: #endif
2373:
1.224 brouard 2374: #ifdef LINMINORIGINAL
2375: #else
2376: free_ivector(flatdir,1,n);
2377: #endif
1.126 brouard 2378: free_vector(xit,1,n);
2379: free_vector(xits,1,n);
2380: free_vector(ptt,1,n);
2381: free_vector(pt,1,n);
2382: return;
1.192 brouard 2383: } /* enough precision */
1.240 brouard 2384: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2385: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2386: ptt[j]=2.0*p[j]-pt[j];
2387: xit[j]=p[j]-pt[j];
2388: pt[j]=p[j];
2389: }
1.181 brouard 2390: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2391: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2392: if (*iter <=4) {
1.225 brouard 2393: #else
2394: #endif
1.224 brouard 2395: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2396: #else
1.161 brouard 2397: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2398: #endif
1.162 brouard 2399: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2400: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2401: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2402: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2403: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2404: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2405: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2406: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2407: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2408: /* Even if f3 <f1, directest can be negative and t >0 */
2409: /* mu² and del² are equal when f3=f1 */
2410: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2411: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2412: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2413: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2414: #ifdef NRCORIGINAL
2415: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2416: #else
2417: 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 2418: t= t- del*SQR(fp-fptt);
1.183 brouard 2419: #endif
1.202 brouard 2420: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2421: #ifdef DEBUG
1.181 brouard 2422: 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);
2423: 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 2424: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2425: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2426: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2427: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2428: 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);
2429: 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);
2430: #endif
1.183 brouard 2431: #ifdef POWELLORIGINAL
2432: if (t < 0.0) { /* Then we use it for new direction */
2433: #else
1.182 brouard 2434: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2435: 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 2436: 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 2437: 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 2438: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2439: }
1.181 brouard 2440: if (directest < 0.0) { /* Then we use it for new direction */
2441: #endif
1.191 brouard 2442: #ifdef DEBUGLINMIN
1.234 brouard 2443: printf("Before linmin in direction P%d-P0\n",n);
2444: for (j=1;j<=n;j++) {
2445: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2446: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2447: if(j % ncovmodel == 0){
2448: printf("\n");
2449: fprintf(ficlog,"\n");
2450: }
2451: }
1.224 brouard 2452: #endif
2453: #ifdef LINMINORIGINAL
1.234 brouard 2454: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2455: #else
1.234 brouard 2456: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2457: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2458: #endif
1.234 brouard 2459:
1.191 brouard 2460: #ifdef DEBUGLINMIN
1.234 brouard 2461: for (j=1;j<=n;j++) {
2462: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2463: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2464: if(j % ncovmodel == 0){
2465: printf("\n");
2466: fprintf(ficlog,"\n");
2467: }
2468: }
1.224 brouard 2469: #endif
1.234 brouard 2470: for (j=1;j<=n;j++) {
2471: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2472: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2473: }
1.224 brouard 2474: #ifdef LINMINORIGINAL
2475: #else
1.234 brouard 2476: for (j=1, flatd=0;j<=n;j++) {
2477: if(flatdir[j]>0)
2478: flatd++;
2479: }
2480: if(flatd >0){
1.255 brouard 2481: printf("%d flat directions: ",flatd);
2482: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2483: for (j=1;j<=n;j++) {
2484: if(flatdir[j]>0){
2485: printf("%d ",j);
2486: fprintf(ficlog,"%d ",j);
2487: }
2488: }
2489: printf("\n");
2490: fprintf(ficlog,"\n");
2491: }
1.191 brouard 2492: #endif
1.234 brouard 2493: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2494: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2495:
1.126 brouard 2496: #ifdef DEBUG
1.234 brouard 2497: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2498: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2499: for(j=1;j<=n;j++){
2500: printf(" %lf",xit[j]);
2501: fprintf(ficlog," %lf",xit[j]);
2502: }
2503: printf("\n");
2504: fprintf(ficlog,"\n");
1.126 brouard 2505: #endif
1.192 brouard 2506: } /* end of t or directest negative */
1.224 brouard 2507: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2508: #else
1.234 brouard 2509: } /* end if (fptt < fp) */
1.192 brouard 2510: #endif
1.225 brouard 2511: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2512: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2513: #else
1.224 brouard 2514: #endif
1.234 brouard 2515: } /* loop iteration */
1.126 brouard 2516: }
1.234 brouard 2517:
1.126 brouard 2518: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2519:
1.235 brouard 2520: 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 2521: {
1.279 brouard 2522: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2523: * (and selected quantitative values in nres)
2524: * by left multiplying the unit
2525: * matrix by transitions matrix until convergence is reached with precision ftolpl
2526: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2527: * Wx is row vector: population in state 1, population in state 2, population dead
2528: * or prevalence in state 1, prevalence in state 2, 0
2529: * newm is the matrix after multiplications, its rows are identical at a factor.
2530: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2531: * Output is prlim.
2532: * Initial matrix pimij
2533: */
1.206 brouard 2534: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2535: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2536: /* 0, 0 , 1} */
2537: /*
2538: * and after some iteration: */
2539: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2540: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2541: /* 0, 0 , 1} */
2542: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2543: /* {0.51571254859325999, 0.4842874514067399, */
2544: /* 0.51326036147820708, 0.48673963852179264} */
2545: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2546:
1.126 brouard 2547: int i, ii,j,k;
1.209 brouard 2548: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2549: /* double **matprod2(); */ /* test */
1.218 brouard 2550: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2551: double **newm;
1.209 brouard 2552: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2553: int ncvloop=0;
1.169 brouard 2554:
1.209 brouard 2555: min=vector(1,nlstate);
2556: max=vector(1,nlstate);
2557: meandiff=vector(1,nlstate);
2558:
1.218 brouard 2559: /* Starting with matrix unity */
1.126 brouard 2560: for (ii=1;ii<=nlstate+ndeath;ii++)
2561: for (j=1;j<=nlstate+ndeath;j++){
2562: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2563: }
1.169 brouard 2564:
2565: cov[1]=1.;
2566:
2567: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2568: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2569: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2570: ncvloop++;
1.126 brouard 2571: newm=savm;
2572: /* Covariates have to be included here again */
1.138 brouard 2573: cov[2]=agefin;
1.187 brouard 2574: if(nagesqr==1)
2575: cov[3]= agefin*agefin;;
1.234 brouard 2576: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2577: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2578: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2579: /* 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 2580: }
2581: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2582: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2583: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2584: /* 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 2585: }
1.237 brouard 2586: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2587: if(Dummy[Tvar[Tage[k]]]){
2588: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2589: } else{
1.235 brouard 2590: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2591: }
1.235 brouard 2592: /* 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 2593: }
1.237 brouard 2594: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2595: /* 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 2596: if(Dummy[Tvard[k][1]==0]){
2597: if(Dummy[Tvard[k][2]==0]){
2598: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2599: }else{
2600: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2601: }
2602: }else{
2603: if(Dummy[Tvard[k][2]==0]){
2604: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2605: }else{
2606: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2607: }
2608: }
1.234 brouard 2609: }
1.138 brouard 2610: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2611: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2612: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2613: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2614: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2615: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2616: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2617:
1.126 brouard 2618: savm=oldm;
2619: oldm=newm;
1.209 brouard 2620:
2621: for(j=1; j<=nlstate; j++){
2622: max[j]=0.;
2623: min[j]=1.;
2624: }
2625: for(i=1;i<=nlstate;i++){
2626: sumnew=0;
2627: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2628: for(j=1; j<=nlstate; j++){
2629: prlim[i][j]= newm[i][j]/(1-sumnew);
2630: max[j]=FMAX(max[j],prlim[i][j]);
2631: min[j]=FMIN(min[j],prlim[i][j]);
2632: }
2633: }
2634:
1.126 brouard 2635: maxmax=0.;
1.209 brouard 2636: for(j=1; j<=nlstate; j++){
2637: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2638: maxmax=FMAX(maxmax,meandiff[j]);
2639: /* 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 2640: } /* j loop */
1.203 brouard 2641: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2642: /* 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 2643: if(maxmax < ftolpl){
1.209 brouard 2644: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2645: free_vector(min,1,nlstate);
2646: free_vector(max,1,nlstate);
2647: free_vector(meandiff,1,nlstate);
1.126 brouard 2648: return prlim;
2649: }
1.169 brouard 2650: } /* age loop */
1.208 brouard 2651: /* After some age loop it doesn't converge */
1.209 brouard 2652: 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 2653: 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 2654: /* 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); */
2655: free_vector(min,1,nlstate);
2656: free_vector(max,1,nlstate);
2657: free_vector(meandiff,1,nlstate);
1.208 brouard 2658:
1.169 brouard 2659: return prlim; /* should not reach here */
1.126 brouard 2660: }
2661:
1.217 brouard 2662:
2663: /**** Back Prevalence limit (stable or period prevalence) ****************/
2664:
1.218 brouard 2665: /* 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) */
2666: /* 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 2667: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2668: {
1.264 brouard 2669: /* 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 2670: matrix by transitions matrix until convergence is reached with precision ftolpl */
2671: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2672: /* Wx is row vector: population in state 1, population in state 2, population dead */
2673: /* or prevalence in state 1, prevalence in state 2, 0 */
2674: /* newm is the matrix after multiplications, its rows are identical at a factor */
2675: /* Initial matrix pimij */
2676: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2677: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2678: /* 0, 0 , 1} */
2679: /*
2680: * and after some iteration: */
2681: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2682: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2683: /* 0, 0 , 1} */
2684: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2685: /* {0.51571254859325999, 0.4842874514067399, */
2686: /* 0.51326036147820708, 0.48673963852179264} */
2687: /* If we start from prlim again, prlim tends to a constant matrix */
2688:
2689: int i, ii,j,k;
1.247 brouard 2690: int first=0;
1.217 brouard 2691: double *min, *max, *meandiff, maxmax,sumnew=0.;
2692: /* double **matprod2(); */ /* test */
2693: double **out, cov[NCOVMAX+1], **bmij();
2694: double **newm;
1.218 brouard 2695: double **dnewm, **doldm, **dsavm; /* for use */
2696: double **oldm, **savm; /* for use */
2697:
1.217 brouard 2698: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2699: int ncvloop=0;
2700:
2701: min=vector(1,nlstate);
2702: max=vector(1,nlstate);
2703: meandiff=vector(1,nlstate);
2704:
1.266 brouard 2705: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2706: oldm=oldms; savm=savms;
2707:
2708: /* Starting with matrix unity */
2709: for (ii=1;ii<=nlstate+ndeath;ii++)
2710: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2711: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2712: }
2713:
2714: cov[1]=1.;
2715:
2716: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2717: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2718: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2719: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2720: ncvloop++;
1.218 brouard 2721: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2722: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2723: /* Covariates have to be included here again */
2724: cov[2]=agefin;
2725: if(nagesqr==1)
2726: cov[3]= agefin*agefin;;
1.242 brouard 2727: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2728: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2729: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2730: /* 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 2731: }
2732: /* for (k=1; k<=cptcovn;k++) { */
2733: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2734: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2735: /* /\* 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])]); *\/ */
2736: /* } */
2737: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2738: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2739: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2740: /* 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]); */
2741: }
2742: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2743: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2744: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2745: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2746: for (k=1; k<=cptcovage;k++){ /* For product with age */
2747: if(Dummy[Tvar[Tage[k]]]){
2748: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2749: } else{
2750: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2751: }
2752: /* 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]); */
2753: }
2754: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2755: /* 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]); */
2756: if(Dummy[Tvard[k][1]==0]){
2757: if(Dummy[Tvard[k][2]==0]){
2758: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2759: }else{
2760: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2761: }
2762: }else{
2763: if(Dummy[Tvard[k][2]==0]){
2764: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2765: }else{
2766: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2767: }
2768: }
1.217 brouard 2769: }
2770:
2771: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2772: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2773: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2774: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2775: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2776: /* ij should be linked to the correct index of cov */
2777: /* age and covariate values ij are in 'cov', but we need to pass
2778: * ij for the observed prevalence at age and status and covariate
2779: * number: prevacurrent[(int)agefin][ii][ij]
2780: */
2781: /* 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 *\/ */
2782: /* 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 *\/ */
2783: 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 2784: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2785: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2786: /* for(i=1; i<=nlstate+ndeath; i++) { */
2787: /* printf("%d newm= ",i); */
2788: /* for(j=1;j<=nlstate+ndeath;j++) { */
2789: /* printf("%f ",newm[i][j]); */
2790: /* } */
2791: /* printf("oldm * "); */
2792: /* for(j=1;j<=nlstate+ndeath;j++) { */
2793: /* printf("%f ",oldm[i][j]); */
2794: /* } */
1.268 brouard 2795: /* printf(" bmmij "); */
1.266 brouard 2796: /* for(j=1;j<=nlstate+ndeath;j++) { */
2797: /* printf("%f ",pmmij[i][j]); */
2798: /* } */
2799: /* printf("\n"); */
2800: /* } */
2801: /* } */
1.217 brouard 2802: savm=oldm;
2803: oldm=newm;
1.266 brouard 2804:
1.217 brouard 2805: for(j=1; j<=nlstate; j++){
2806: max[j]=0.;
2807: min[j]=1.;
2808: }
2809: for(j=1; j<=nlstate; j++){
2810: for(i=1;i<=nlstate;i++){
1.234 brouard 2811: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2812: bprlim[i][j]= newm[i][j];
2813: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2814: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2815: }
2816: }
1.218 brouard 2817:
1.217 brouard 2818: maxmax=0.;
2819: for(i=1; i<=nlstate; i++){
2820: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2821: maxmax=FMAX(maxmax,meandiff[i]);
2822: /* 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 2823: } /* i loop */
1.217 brouard 2824: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2825: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2826: if(maxmax < ftolpl){
1.220 brouard 2827: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2828: free_vector(min,1,nlstate);
2829: free_vector(max,1,nlstate);
2830: free_vector(meandiff,1,nlstate);
2831: return bprlim;
2832: }
2833: } /* age loop */
2834: /* After some age loop it doesn't converge */
1.247 brouard 2835: if(first){
2836: first=1;
2837: 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\
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: }
2840: 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 2841: 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);
2842: /* 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); */
2843: free_vector(min,1,nlstate);
2844: free_vector(max,1,nlstate);
2845: free_vector(meandiff,1,nlstate);
2846:
2847: return bprlim; /* should not reach here */
2848: }
2849:
1.126 brouard 2850: /*************** transition probabilities ***************/
2851:
2852: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2853: {
1.138 brouard 2854: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2855: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2856: model to the ncovmodel covariates (including constant and age).
2857: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2858: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2859: ncth covariate in the global vector x is given by the formula:
2860: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2861: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2862: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2863: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2864: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2865: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2866: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2867: */
2868: double s1, lnpijopii;
1.126 brouard 2869: /*double t34;*/
1.164 brouard 2870: int i,j, nc, ii, jj;
1.126 brouard 2871:
1.223 brouard 2872: for(i=1; i<= nlstate; i++){
2873: for(j=1; j<i;j++){
2874: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2875: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2876: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2877: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2878: }
2879: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2880: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2881: }
2882: for(j=i+1; j<=nlstate+ndeath;j++){
2883: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2884: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2885: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2886: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2887: }
2888: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2889: }
2890: }
1.218 brouard 2891:
1.223 brouard 2892: for(i=1; i<= nlstate; i++){
2893: s1=0;
2894: for(j=1; j<i; j++){
2895: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2896: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2897: }
2898: for(j=i+1; j<=nlstate+ndeath; j++){
2899: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2900: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2901: }
2902: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2903: ps[i][i]=1./(s1+1.);
2904: /* Computing other pijs */
2905: for(j=1; j<i; j++)
2906: ps[i][j]= exp(ps[i][j])*ps[i][i];
2907: for(j=i+1; j<=nlstate+ndeath; j++)
2908: ps[i][j]= exp(ps[i][j])*ps[i][i];
2909: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2910: } /* end i */
1.218 brouard 2911:
1.223 brouard 2912: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2913: for(jj=1; jj<= nlstate+ndeath; jj++){
2914: ps[ii][jj]=0;
2915: ps[ii][ii]=1;
2916: }
2917: }
1.218 brouard 2918:
2919:
1.223 brouard 2920: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2921: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2922: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2923: /* } */
2924: /* printf("\n "); */
2925: /* } */
2926: /* printf("\n ");printf("%lf ",cov[2]);*/
2927: /*
2928: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2929: goto end;*/
1.266 brouard 2930: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2931: }
2932:
1.218 brouard 2933: /*************** backward transition probabilities ***************/
2934:
2935: /* 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 ) */
2936: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2937: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2938: {
1.266 brouard 2939: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2940: * 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 2941: */
1.218 brouard 2942: int i, ii, j,k;
1.222 brouard 2943:
2944: double **out, **pmij();
2945: double sumnew=0.;
1.218 brouard 2946: double agefin;
1.268 brouard 2947: 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 2948: double **dnewm, **dsavm, **doldm;
2949: double **bbmij;
2950:
1.218 brouard 2951: doldm=ddoldms; /* global pointers */
1.222 brouard 2952: dnewm=ddnewms;
2953: dsavm=ddsavms;
2954:
2955: agefin=cov[2];
1.268 brouard 2956: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2957: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2958: the observed prevalence (with this covariate ij) at beginning of transition */
2959: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2960:
2961: /* P_x */
1.266 brouard 2962: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2963: /* outputs pmmij which is a stochastic matrix in row */
2964:
2965: /* Diag(w_x) */
2966: /* Problem with prevacurrent which can be zero */
2967: sumnew=0.;
1.269 brouard 2968: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2969: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2970: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2971: sumnew+=prevacurrent[(int)agefin][ii][ij];
2972: }
2973: if(sumnew >0.01){ /* At least some value in the prevalence */
2974: for (ii=1;ii<=nlstate+ndeath;ii++){
2975: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2976: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2977: }
2978: }else{
2979: for (ii=1;ii<=nlstate+ndeath;ii++){
2980: for (j=1;j<=nlstate+ndeath;j++)
2981: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
2982: }
2983: /* if(sumnew <0.9){ */
2984: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2985: /* } */
2986: }
2987: k3=0.0; /* We put the last diagonal to 0 */
2988: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
2989: doldm[ii][ii]= k3;
2990: }
2991: /* End doldm, At the end doldm is diag[(w_i)] */
2992:
2993: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
2994: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
2995:
2996: /* Diag(Sum_i w^i_x p^ij_x */
2997: /* 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 2998: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 2999: sumnew=0.;
1.222 brouard 3000: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3001: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3002: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3003: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3004: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3005: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3006: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3007: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3008: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3009: /* }else */
1.268 brouard 3010: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3011: } /*End ii */
3012: } /* 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 */
3013:
3014: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
3015: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3016: /* end bmij */
1.266 brouard 3017: return ps; /*pointer is unchanged */
1.218 brouard 3018: }
1.217 brouard 3019: /*************** transition probabilities ***************/
3020:
1.218 brouard 3021: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3022: {
3023: /* According to parameters values stored in x and the covariate's values stored in cov,
3024: computes the probability to be observed in state j being in state i by appying the
3025: model to the ncovmodel covariates (including constant and age).
3026: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3027: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3028: ncth covariate in the global vector x is given by the formula:
3029: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3030: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3031: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3032: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3033: Outputs ps[i][j] the probability to be observed in j being in j according to
3034: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3035: */
3036: double s1, lnpijopii;
3037: /*double t34;*/
3038: int i,j, nc, ii, jj;
3039:
1.234 brouard 3040: for(i=1; i<= nlstate; i++){
3041: for(j=1; j<i;j++){
3042: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3043: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3044: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3045: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3046: }
3047: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3048: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3049: }
3050: for(j=i+1; j<=nlstate+ndeath;j++){
3051: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3052: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3053: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3054: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3055: }
3056: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3057: }
3058: }
3059:
3060: for(i=1; i<= nlstate; i++){
3061: s1=0;
3062: for(j=1; j<i; j++){
3063: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3064: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3065: }
3066: for(j=i+1; j<=nlstate+ndeath; j++){
3067: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3068: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3069: }
3070: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3071: ps[i][i]=1./(s1+1.);
3072: /* Computing other pijs */
3073: for(j=1; j<i; j++)
3074: ps[i][j]= exp(ps[i][j])*ps[i][i];
3075: for(j=i+1; j<=nlstate+ndeath; j++)
3076: ps[i][j]= exp(ps[i][j])*ps[i][i];
3077: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3078: } /* end i */
3079:
3080: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3081: for(jj=1; jj<= nlstate+ndeath; jj++){
3082: ps[ii][jj]=0;
3083: ps[ii][ii]=1;
3084: }
3085: }
3086: /* Added for backcast */ /* Transposed matrix too */
3087: for(jj=1; jj<= nlstate+ndeath; jj++){
3088: s1=0.;
3089: for(ii=1; ii<= nlstate+ndeath; ii++){
3090: s1+=ps[ii][jj];
3091: }
3092: for(ii=1; ii<= nlstate; ii++){
3093: ps[ii][jj]=ps[ii][jj]/s1;
3094: }
3095: }
3096: /* Transposition */
3097: for(jj=1; jj<= nlstate+ndeath; jj++){
3098: for(ii=jj; ii<= nlstate+ndeath; ii++){
3099: s1=ps[ii][jj];
3100: ps[ii][jj]=ps[jj][ii];
3101: ps[jj][ii]=s1;
3102: }
3103: }
3104: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3105: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3106: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3107: /* } */
3108: /* printf("\n "); */
3109: /* } */
3110: /* printf("\n ");printf("%lf ",cov[2]);*/
3111: /*
3112: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3113: goto end;*/
3114: return ps;
1.217 brouard 3115: }
3116:
3117:
1.126 brouard 3118: /**************** Product of 2 matrices ******************/
3119:
1.145 brouard 3120: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3121: {
3122: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3123: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3124: /* in, b, out are matrice of pointers which should have been initialized
3125: before: only the contents of out is modified. The function returns
3126: a pointer to pointers identical to out */
1.145 brouard 3127: int i, j, k;
1.126 brouard 3128: for(i=nrl; i<= nrh; i++)
1.145 brouard 3129: for(k=ncolol; k<=ncoloh; k++){
3130: out[i][k]=0.;
3131: for(j=ncl; j<=nch; j++)
3132: out[i][k] +=in[i][j]*b[j][k];
3133: }
1.126 brouard 3134: return out;
3135: }
3136:
3137:
3138: /************* Higher Matrix Product ***************/
3139:
1.235 brouard 3140: 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 3141: {
1.218 brouard 3142: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3143: 'nhstepm*hstepm*stepm' months (i.e. until
3144: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3145: nhstepm*hstepm matrices.
3146: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3147: (typically every 2 years instead of every month which is too big
3148: for the memory).
3149: Model is determined by parameters x and covariates have to be
3150: included manually here.
3151:
3152: */
3153:
3154: int i, j, d, h, k;
1.131 brouard 3155: double **out, cov[NCOVMAX+1];
1.126 brouard 3156: double **newm;
1.187 brouard 3157: double agexact;
1.214 brouard 3158: double agebegin, ageend;
1.126 brouard 3159:
3160: /* Hstepm could be zero and should return the unit matrix */
3161: for (i=1;i<=nlstate+ndeath;i++)
3162: for (j=1;j<=nlstate+ndeath;j++){
3163: oldm[i][j]=(i==j ? 1.0 : 0.0);
3164: po[i][j][0]=(i==j ? 1.0 : 0.0);
3165: }
3166: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3167: for(h=1; h <=nhstepm; h++){
3168: for(d=1; d <=hstepm; d++){
3169: newm=savm;
3170: /* Covariates have to be included here again */
3171: cov[1]=1.;
1.214 brouard 3172: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3173: cov[2]=agexact;
3174: if(nagesqr==1)
1.227 brouard 3175: cov[3]= agexact*agexact;
1.235 brouard 3176: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3177: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3178: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3179: /* 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)); */
3180: }
3181: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3182: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3183: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3184: /* 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]); */
3185: }
3186: for (k=1; k<=cptcovage;k++){
3187: if(Dummy[Tvar[Tage[k]]]){
3188: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3189: } else{
3190: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3191: }
3192: /* 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]); */
3193: }
3194: for (k=1; k<=cptcovprod;k++){ /* */
3195: /* 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]); */
3196: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3197: }
3198: /* for (k=1; k<=cptcovn;k++) */
3199: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3200: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3201: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3202: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3203: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3204:
3205:
1.126 brouard 3206: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3207: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3208: /* right multiplication of oldm by the current matrix */
1.126 brouard 3209: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3210: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3211: /* if((int)age == 70){ */
3212: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3213: /* for(i=1; i<=nlstate+ndeath; i++) { */
3214: /* printf("%d pmmij ",i); */
3215: /* for(j=1;j<=nlstate+ndeath;j++) { */
3216: /* printf("%f ",pmmij[i][j]); */
3217: /* } */
3218: /* printf(" oldm "); */
3219: /* for(j=1;j<=nlstate+ndeath;j++) { */
3220: /* printf("%f ",oldm[i][j]); */
3221: /* } */
3222: /* printf("\n"); */
3223: /* } */
3224: /* } */
1.126 brouard 3225: savm=oldm;
3226: oldm=newm;
3227: }
3228: for(i=1; i<=nlstate+ndeath; i++)
3229: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3230: po[i][j][h]=newm[i][j];
3231: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3232: }
1.128 brouard 3233: /*printf("h=%d ",h);*/
1.126 brouard 3234: } /* end h */
1.267 brouard 3235: /* printf("\n H=%d \n",h); */
1.126 brouard 3236: return po;
3237: }
3238:
1.217 brouard 3239: /************* Higher Back Matrix Product ***************/
1.218 brouard 3240: /* 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 3241: 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 3242: {
1.266 brouard 3243: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3244: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3245: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3246: nhstepm*hstepm matrices.
3247: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3248: (typically every 2 years instead of every month which is too big
1.217 brouard 3249: for the memory).
1.218 brouard 3250: Model is determined by parameters x and covariates have to be
1.266 brouard 3251: included manually here. Then we use a call to bmij(x and cov)
3252: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3253: */
1.217 brouard 3254:
3255: int i, j, d, h, k;
1.266 brouard 3256: double **out, cov[NCOVMAX+1], **bmij();
3257: double **newm, ***newmm;
1.217 brouard 3258: double agexact;
3259: double agebegin, ageend;
1.222 brouard 3260: double **oldm, **savm;
1.217 brouard 3261:
1.266 brouard 3262: newmm=po; /* To be saved */
3263: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3264: /* Hstepm could be zero and should return the unit matrix */
3265: for (i=1;i<=nlstate+ndeath;i++)
3266: for (j=1;j<=nlstate+ndeath;j++){
3267: oldm[i][j]=(i==j ? 1.0 : 0.0);
3268: po[i][j][0]=(i==j ? 1.0 : 0.0);
3269: }
3270: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3271: for(h=1; h <=nhstepm; h++){
3272: for(d=1; d <=hstepm; d++){
3273: newm=savm;
3274: /* Covariates have to be included here again */
3275: cov[1]=1.;
1.271 brouard 3276: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3277: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3278: cov[2]=agexact;
3279: if(nagesqr==1)
1.222 brouard 3280: cov[3]= agexact*agexact;
1.266 brouard 3281: for (k=1; k<=cptcovn;k++){
3282: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3283: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3284: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3285: /* 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)); */
3286: }
1.267 brouard 3287: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3288: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3289: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3290: /* 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]); */
3291: }
3292: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3293: if(Dummy[Tvar[Tage[k]]]){
3294: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3295: } else{
3296: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3297: }
3298: /* 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]); */
3299: }
3300: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3301: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3302: }
1.217 brouard 3303: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3304: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3305:
1.218 brouard 3306: /* Careful transposed matrix */
1.266 brouard 3307: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3308: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3309: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3310: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3311: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3312: /* if((int)age == 70){ */
3313: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3314: /* for(i=1; i<=nlstate+ndeath; i++) { */
3315: /* printf("%d pmmij ",i); */
3316: /* for(j=1;j<=nlstate+ndeath;j++) { */
3317: /* printf("%f ",pmmij[i][j]); */
3318: /* } */
3319: /* printf(" oldm "); */
3320: /* for(j=1;j<=nlstate+ndeath;j++) { */
3321: /* printf("%f ",oldm[i][j]); */
3322: /* } */
3323: /* printf("\n"); */
3324: /* } */
3325: /* } */
3326: savm=oldm;
3327: oldm=newm;
3328: }
3329: for(i=1; i<=nlstate+ndeath; i++)
3330: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3331: po[i][j][h]=newm[i][j];
1.268 brouard 3332: /* if(h==nhstepm) */
3333: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3334: }
1.268 brouard 3335: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3336: } /* end h */
1.268 brouard 3337: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3338: return po;
3339: }
3340:
3341:
1.162 brouard 3342: #ifdef NLOPT
3343: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3344: double fret;
3345: double *xt;
3346: int j;
3347: myfunc_data *d2 = (myfunc_data *) pd;
3348: /* xt = (p1-1); */
3349: xt=vector(1,n);
3350: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3351:
3352: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3353: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3354: printf("Function = %.12lf ",fret);
3355: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3356: printf("\n");
3357: free_vector(xt,1,n);
3358: return fret;
3359: }
3360: #endif
1.126 brouard 3361:
3362: /*************** log-likelihood *************/
3363: double func( double *x)
3364: {
1.226 brouard 3365: int i, ii, j, k, mi, d, kk;
3366: int ioffset=0;
3367: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3368: double **out;
3369: double lli; /* Individual log likelihood */
3370: int s1, s2;
1.228 brouard 3371: 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 3372: double bbh, survp;
3373: long ipmx;
3374: double agexact;
3375: /*extern weight */
3376: /* We are differentiating ll according to initial status */
3377: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3378: /*for(i=1;i<imx;i++)
3379: printf(" %d\n",s[4][i]);
3380: */
1.162 brouard 3381:
1.226 brouard 3382: ++countcallfunc;
1.162 brouard 3383:
1.226 brouard 3384: cov[1]=1.;
1.126 brouard 3385:
1.226 brouard 3386: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3387: ioffset=0;
1.226 brouard 3388: if(mle==1){
3389: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3390: /* Computes the values of the ncovmodel covariates of the model
3391: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3392: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3393: to be observed in j being in i according to the model.
3394: */
1.243 brouard 3395: ioffset=2+nagesqr ;
1.233 brouard 3396: /* Fixed */
1.234 brouard 3397: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3398: 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)*/
3399: }
1.226 brouard 3400: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3401: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3402: has been calculated etc */
3403: /* For an individual i, wav[i] gives the number of effective waves */
3404: /* We compute the contribution to Likelihood of each effective transition
3405: mw[mi][i] is real wave of the mi th effectve wave */
3406: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3407: s2=s[mw[mi+1][i]][i];
3408: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3409: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3410: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3411: */
3412: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3413: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3414: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3415: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3416: }
3417: for (ii=1;ii<=nlstate+ndeath;ii++)
3418: for (j=1;j<=nlstate+ndeath;j++){
3419: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3420: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3421: }
3422: for(d=0; d<dh[mi][i]; d++){
3423: newm=savm;
3424: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3425: cov[2]=agexact;
3426: if(nagesqr==1)
3427: cov[3]= agexact*agexact; /* Should be changed here */
3428: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3429: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3430: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3431: else
3432: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3433: }
3434: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3435: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3436: savm=oldm;
3437: oldm=newm;
3438: } /* end mult */
3439:
3440: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3441: /* But now since version 0.9 we anticipate for bias at large stepm.
3442: * If stepm is larger than one month (smallest stepm) and if the exact delay
3443: * (in months) between two waves is not a multiple of stepm, we rounded to
3444: * the nearest (and in case of equal distance, to the lowest) interval but now
3445: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3446: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3447: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3448: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3449: * -stepm/2 to stepm/2 .
3450: * For stepm=1 the results are the same as for previous versions of Imach.
3451: * For stepm > 1 the results are less biased than in previous versions.
3452: */
1.234 brouard 3453: s1=s[mw[mi][i]][i];
3454: s2=s[mw[mi+1][i]][i];
3455: bbh=(double)bh[mi][i]/(double)stepm;
3456: /* bias bh is positive if real duration
3457: * is higher than the multiple of stepm and negative otherwise.
3458: */
3459: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3460: if( s2 > nlstate){
3461: /* i.e. if s2 is a death state and if the date of death is known
3462: then the contribution to the likelihood is the probability to
3463: die between last step unit time and current step unit time,
3464: which is also equal to probability to die before dh
3465: minus probability to die before dh-stepm .
3466: In version up to 0.92 likelihood was computed
3467: as if date of death was unknown. Death was treated as any other
3468: health state: the date of the interview describes the actual state
3469: and not the date of a change in health state. The former idea was
3470: to consider that at each interview the state was recorded
3471: (healthy, disable or death) and IMaCh was corrected; but when we
3472: introduced the exact date of death then we should have modified
3473: the contribution of an exact death to the likelihood. This new
3474: contribution is smaller and very dependent of the step unit
3475: stepm. It is no more the probability to die between last interview
3476: and month of death but the probability to survive from last
3477: interview up to one month before death multiplied by the
3478: probability to die within a month. Thanks to Chris
3479: Jackson for correcting this bug. Former versions increased
3480: mortality artificially. The bad side is that we add another loop
3481: which slows down the processing. The difference can be up to 10%
3482: lower mortality.
3483: */
3484: /* If, at the beginning of the maximization mostly, the
3485: cumulative probability or probability to be dead is
3486: constant (ie = 1) over time d, the difference is equal to
3487: 0. out[s1][3] = savm[s1][3]: probability, being at state
3488: s1 at precedent wave, to be dead a month before current
3489: wave is equal to probability, being at state s1 at
3490: precedent wave, to be dead at mont of the current
3491: wave. Then the observed probability (that this person died)
3492: is null according to current estimated parameter. In fact,
3493: it should be very low but not zero otherwise the log go to
3494: infinity.
3495: */
1.183 brouard 3496: /* #ifdef INFINITYORIGINAL */
3497: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3498: /* #else */
3499: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3500: /* lli=log(mytinydouble); */
3501: /* else */
3502: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3503: /* #endif */
1.226 brouard 3504: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3505:
1.226 brouard 3506: } else if ( s2==-1 ) { /* alive */
3507: for (j=1,survp=0. ; j<=nlstate; j++)
3508: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3509: /*survp += out[s1][j]; */
3510: lli= log(survp);
3511: }
3512: else if (s2==-4) {
3513: for (j=3,survp=0. ; j<=nlstate; j++)
3514: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3515: lli= log(survp);
3516: }
3517: else if (s2==-5) {
3518: for (j=1,survp=0. ; j<=2; j++)
3519: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3520: lli= log(survp);
3521: }
3522: else{
3523: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3524: /* 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 */
3525: }
3526: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3527: /*if(lli ==000.0)*/
3528: /*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); */
3529: ipmx +=1;
3530: sw += weight[i];
3531: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3532: /* if (lli < log(mytinydouble)){ */
3533: /* 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); */
3534: /* 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]); */
3535: /* } */
3536: } /* end of wave */
3537: } /* end of individual */
3538: } else if(mle==2){
3539: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3540: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3541: for(mi=1; mi<= wav[i]-1; mi++){
3542: for (ii=1;ii<=nlstate+ndeath;ii++)
3543: for (j=1;j<=nlstate+ndeath;j++){
3544: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3545: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3546: }
3547: for(d=0; d<=dh[mi][i]; d++){
3548: newm=savm;
3549: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3550: cov[2]=agexact;
3551: if(nagesqr==1)
3552: cov[3]= agexact*agexact;
3553: for (kk=1; kk<=cptcovage;kk++) {
3554: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3555: }
3556: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3557: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3558: savm=oldm;
3559: oldm=newm;
3560: } /* end mult */
3561:
3562: s1=s[mw[mi][i]][i];
3563: s2=s[mw[mi+1][i]][i];
3564: bbh=(double)bh[mi][i]/(double)stepm;
3565: 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 */
3566: ipmx +=1;
3567: sw += weight[i];
3568: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3569: } /* end of wave */
3570: } /* end of individual */
3571: } else if(mle==3){ /* exponential inter-extrapolation */
3572: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3573: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3574: for(mi=1; mi<= wav[i]-1; mi++){
3575: for (ii=1;ii<=nlstate+ndeath;ii++)
3576: for (j=1;j<=nlstate+ndeath;j++){
3577: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3578: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3579: }
3580: for(d=0; d<dh[mi][i]; d++){
3581: newm=savm;
3582: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3583: cov[2]=agexact;
3584: if(nagesqr==1)
3585: cov[3]= agexact*agexact;
3586: for (kk=1; kk<=cptcovage;kk++) {
3587: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3588: }
3589: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3590: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3591: savm=oldm;
3592: oldm=newm;
3593: } /* end mult */
3594:
3595: s1=s[mw[mi][i]][i];
3596: s2=s[mw[mi+1][i]][i];
3597: bbh=(double)bh[mi][i]/(double)stepm;
3598: 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 */
3599: ipmx +=1;
3600: sw += weight[i];
3601: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3602: } /* end of wave */
3603: } /* end of individual */
3604: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3605: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3606: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3607: for(mi=1; mi<= wav[i]-1; mi++){
3608: for (ii=1;ii<=nlstate+ndeath;ii++)
3609: for (j=1;j<=nlstate+ndeath;j++){
3610: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3611: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3612: }
3613: for(d=0; d<dh[mi][i]; d++){
3614: newm=savm;
3615: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3616: cov[2]=agexact;
3617: if(nagesqr==1)
3618: cov[3]= agexact*agexact;
3619: for (kk=1; kk<=cptcovage;kk++) {
3620: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3621: }
1.126 brouard 3622:
1.226 brouard 3623: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3624: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3625: savm=oldm;
3626: oldm=newm;
3627: } /* end mult */
3628:
3629: s1=s[mw[mi][i]][i];
3630: s2=s[mw[mi+1][i]][i];
3631: if( s2 > nlstate){
3632: lli=log(out[s1][s2] - savm[s1][s2]);
3633: } else if ( s2==-1 ) { /* alive */
3634: for (j=1,survp=0. ; j<=nlstate; j++)
3635: survp += out[s1][j];
3636: lli= log(survp);
3637: }else{
3638: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3639: }
3640: ipmx +=1;
3641: sw += weight[i];
3642: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3643: /* 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 3644: } /* end of wave */
3645: } /* end of individual */
3646: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3647: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3648: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3649: for(mi=1; mi<= wav[i]-1; mi++){
3650: for (ii=1;ii<=nlstate+ndeath;ii++)
3651: for (j=1;j<=nlstate+ndeath;j++){
3652: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3653: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3654: }
3655: for(d=0; d<dh[mi][i]; d++){
3656: newm=savm;
3657: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3658: cov[2]=agexact;
3659: if(nagesqr==1)
3660: cov[3]= agexact*agexact;
3661: for (kk=1; kk<=cptcovage;kk++) {
3662: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3663: }
1.126 brouard 3664:
1.226 brouard 3665: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3666: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3667: savm=oldm;
3668: oldm=newm;
3669: } /* end mult */
3670:
3671: s1=s[mw[mi][i]][i];
3672: s2=s[mw[mi+1][i]][i];
3673: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3674: ipmx +=1;
3675: sw += weight[i];
3676: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3677: /*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]);*/
3678: } /* end of wave */
3679: } /* end of individual */
3680: } /* End of if */
3681: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3682: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3683: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3684: return -l;
1.126 brouard 3685: }
3686:
3687: /*************** log-likelihood *************/
3688: double funcone( double *x)
3689: {
1.228 brouard 3690: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3691: int i, ii, j, k, mi, d, kk;
1.228 brouard 3692: int ioffset=0;
1.131 brouard 3693: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3694: double **out;
3695: double lli; /* Individual log likelihood */
3696: double llt;
3697: int s1, s2;
1.228 brouard 3698: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3699:
1.126 brouard 3700: double bbh, survp;
1.187 brouard 3701: double agexact;
1.214 brouard 3702: double agebegin, ageend;
1.126 brouard 3703: /*extern weight */
3704: /* We are differentiating ll according to initial status */
3705: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3706: /*for(i=1;i<imx;i++)
3707: printf(" %d\n",s[4][i]);
3708: */
3709: cov[1]=1.;
3710:
3711: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3712: ioffset=0;
3713: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3714: /* ioffset=2+nagesqr+cptcovage; */
3715: ioffset=2+nagesqr;
1.232 brouard 3716: /* Fixed */
1.224 brouard 3717: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3718: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3719: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3720: 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)*/
3721: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3722: /* cov[2+6]=covar[Tvar[6]][i]; */
3723: /* cov[2+6]=covar[2][i]; V2 */
3724: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3725: /* cov[2+7]=covar[Tvar[7]][i]; */
3726: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3727: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3728: /* cov[2+9]=covar[Tvar[9]][i]; */
3729: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3730: }
1.232 brouard 3731: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3732: /* 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?)*\/ */
3733: /* } */
1.231 brouard 3734: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3735: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3736: /* } */
1.225 brouard 3737:
1.233 brouard 3738:
3739: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3740: /* Wave varying (but not age varying) */
3741: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3742: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3743: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3744: }
1.232 brouard 3745: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3746: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3747: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3748: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3749: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3750: /* 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 3751: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3752: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3753: /* /\* 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]); *\/ */
3754: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3755: /* } */
1.126 brouard 3756: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3757: for (j=1;j<=nlstate+ndeath;j++){
3758: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3759: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3760: }
1.214 brouard 3761:
3762: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3763: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3764: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3765: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3766: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3767: and mw[mi+1][i]. dh depends on stepm.*/
3768: newm=savm;
1.247 brouard 3769: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3770: cov[2]=agexact;
3771: if(nagesqr==1)
3772: cov[3]= agexact*agexact;
3773: for (kk=1; kk<=cptcovage;kk++) {
3774: if(!FixedV[Tvar[Tage[kk]]])
3775: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3776: else
3777: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3778: }
3779: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3780: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3781: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3782: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3783: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3784: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3785: savm=oldm;
3786: oldm=newm;
1.126 brouard 3787: } /* end mult */
3788:
3789: s1=s[mw[mi][i]][i];
3790: s2=s[mw[mi+1][i]][i];
1.217 brouard 3791: /* if(s2==-1){ */
1.268 brouard 3792: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3793: /* /\* exit(1); *\/ */
3794: /* } */
1.126 brouard 3795: bbh=(double)bh[mi][i]/(double)stepm;
3796: /* bias is positive if real duration
3797: * is higher than the multiple of stepm and negative otherwise.
3798: */
3799: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3800: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3801: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3802: for (j=1,survp=0. ; j<=nlstate; j++)
3803: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3804: lli= log(survp);
1.126 brouard 3805: }else if (mle==1){
1.242 brouard 3806: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3807: } else if(mle==2){
1.242 brouard 3808: 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 3809: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3810: 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 3811: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3812: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3813: } else{ /* mle=0 back to 1 */
1.242 brouard 3814: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3815: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3816: } /* End of if */
3817: ipmx +=1;
3818: sw += weight[i];
3819: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3820: /*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 3821: if(globpr){
1.246 brouard 3822: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3823: %11.6f %11.6f %11.6f ", \
1.242 brouard 3824: 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 3825: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3826: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3827: llt +=ll[k]*gipmx/gsw;
3828: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3829: }
3830: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3831: }
1.232 brouard 3832: } /* end of wave */
3833: } /* end of individual */
3834: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3835: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3836: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3837: if(globpr==0){ /* First time we count the contributions and weights */
3838: gipmx=ipmx;
3839: gsw=sw;
3840: }
3841: return -l;
1.126 brouard 3842: }
3843:
3844:
3845: /*************** function likelione ***********/
3846: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3847: {
3848: /* This routine should help understanding what is done with
3849: the selection of individuals/waves and
3850: to check the exact contribution to the likelihood.
3851: Plotting could be done.
3852: */
3853: int k;
3854:
3855: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3856: strcpy(fileresilk,"ILK_");
1.202 brouard 3857: strcat(fileresilk,fileresu);
1.126 brouard 3858: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3859: printf("Problem with resultfile: %s\n", fileresilk);
3860: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3861: }
1.214 brouard 3862: 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");
3863: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3864: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3865: for(k=1; k<=nlstate; k++)
3866: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3867: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3868: }
3869:
3870: *fretone=(*funcone)(p);
3871: if(*globpri !=0){
3872: fclose(ficresilk);
1.205 brouard 3873: if (mle ==0)
3874: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3875: else if(mle >=1)
3876: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3877: 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 3878: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3879:
3880: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3881: 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 3882: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3883: }
1.207 brouard 3884: 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 3885: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3886: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3887: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3888: fflush(fichtm);
1.205 brouard 3889: }
1.126 brouard 3890: return;
3891: }
3892:
3893:
3894: /*********** Maximum Likelihood Estimation ***************/
3895:
3896: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3897: {
1.165 brouard 3898: int i,j, iter=0;
1.126 brouard 3899: double **xi;
3900: double fret;
3901: double fretone; /* Only one call to likelihood */
3902: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3903:
3904: #ifdef NLOPT
3905: int creturn;
3906: nlopt_opt opt;
3907: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3908: double *lb;
3909: double minf; /* the minimum objective value, upon return */
3910: double * p1; /* Shifted parameters from 0 instead of 1 */
3911: myfunc_data dinst, *d = &dinst;
3912: #endif
3913:
3914:
1.126 brouard 3915: xi=matrix(1,npar,1,npar);
3916: for (i=1;i<=npar;i++)
3917: for (j=1;j<=npar;j++)
3918: xi[i][j]=(i==j ? 1.0 : 0.0);
3919: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3920: strcpy(filerespow,"POW_");
1.126 brouard 3921: strcat(filerespow,fileres);
3922: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3923: printf("Problem with resultfile: %s\n", filerespow);
3924: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3925: }
3926: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3927: for (i=1;i<=nlstate;i++)
3928: for(j=1;j<=nlstate+ndeath;j++)
3929: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3930: fprintf(ficrespow,"\n");
1.162 brouard 3931: #ifdef POWELL
1.126 brouard 3932: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3933: #endif
1.126 brouard 3934:
1.162 brouard 3935: #ifdef NLOPT
3936: #ifdef NEWUOA
3937: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3938: #else
3939: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3940: #endif
3941: lb=vector(0,npar-1);
3942: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3943: nlopt_set_lower_bounds(opt, lb);
3944: nlopt_set_initial_step1(opt, 0.1);
3945:
3946: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3947: d->function = func;
3948: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3949: nlopt_set_min_objective(opt, myfunc, d);
3950: nlopt_set_xtol_rel(opt, ftol);
3951: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3952: printf("nlopt failed! %d\n",creturn);
3953: }
3954: else {
3955: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3956: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3957: iter=1; /* not equal */
3958: }
3959: nlopt_destroy(opt);
3960: #endif
1.126 brouard 3961: free_matrix(xi,1,npar,1,npar);
3962: fclose(ficrespow);
1.203 brouard 3963: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3964: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3965: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3966:
3967: }
3968:
3969: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3970: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3971: {
3972: double **a,**y,*x,pd;
1.203 brouard 3973: /* double **hess; */
1.164 brouard 3974: int i, j;
1.126 brouard 3975: int *indx;
3976:
3977: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3978: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3979: void lubksb(double **a, int npar, int *indx, double b[]) ;
3980: void ludcmp(double **a, int npar, int *indx, double *d) ;
3981: double gompertz(double p[]);
1.203 brouard 3982: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3983:
3984: printf("\nCalculation of the hessian matrix. Wait...\n");
3985: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3986: for (i=1;i<=npar;i++){
1.203 brouard 3987: printf("%d-",i);fflush(stdout);
3988: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3989:
3990: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3991:
3992: /* printf(" %f ",p[i]);
3993: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3994: }
3995:
3996: for (i=1;i<=npar;i++) {
3997: for (j=1;j<=npar;j++) {
3998: if (j>i) {
1.203 brouard 3999: printf(".%d-%d",i,j);fflush(stdout);
4000: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4001: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4002:
4003: hess[j][i]=hess[i][j];
4004: /*printf(" %lf ",hess[i][j]);*/
4005: }
4006: }
4007: }
4008: printf("\n");
4009: fprintf(ficlog,"\n");
4010:
4011: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4012: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4013:
4014: a=matrix(1,npar,1,npar);
4015: y=matrix(1,npar,1,npar);
4016: x=vector(1,npar);
4017: indx=ivector(1,npar);
4018: for (i=1;i<=npar;i++)
4019: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4020: ludcmp(a,npar,indx,&pd);
4021:
4022: for (j=1;j<=npar;j++) {
4023: for (i=1;i<=npar;i++) x[i]=0;
4024: x[j]=1;
4025: lubksb(a,npar,indx,x);
4026: for (i=1;i<=npar;i++){
4027: matcov[i][j]=x[i];
4028: }
4029: }
4030:
4031: printf("\n#Hessian matrix#\n");
4032: fprintf(ficlog,"\n#Hessian matrix#\n");
4033: for (i=1;i<=npar;i++) {
4034: for (j=1;j<=npar;j++) {
1.203 brouard 4035: printf("%.6e ",hess[i][j]);
4036: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4037: }
4038: printf("\n");
4039: fprintf(ficlog,"\n");
4040: }
4041:
1.203 brouard 4042: /* printf("\n#Covariance matrix#\n"); */
4043: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4044: /* for (i=1;i<=npar;i++) { */
4045: /* for (j=1;j<=npar;j++) { */
4046: /* printf("%.6e ",matcov[i][j]); */
4047: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4048: /* } */
4049: /* printf("\n"); */
4050: /* fprintf(ficlog,"\n"); */
4051: /* } */
4052:
1.126 brouard 4053: /* Recompute Inverse */
1.203 brouard 4054: /* for (i=1;i<=npar;i++) */
4055: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4056: /* ludcmp(a,npar,indx,&pd); */
4057:
4058: /* printf("\n#Hessian matrix recomputed#\n"); */
4059:
4060: /* for (j=1;j<=npar;j++) { */
4061: /* for (i=1;i<=npar;i++) x[i]=0; */
4062: /* x[j]=1; */
4063: /* lubksb(a,npar,indx,x); */
4064: /* for (i=1;i<=npar;i++){ */
4065: /* y[i][j]=x[i]; */
4066: /* printf("%.3e ",y[i][j]); */
4067: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4068: /* } */
4069: /* printf("\n"); */
4070: /* fprintf(ficlog,"\n"); */
4071: /* } */
4072:
4073: /* Verifying the inverse matrix */
4074: #ifdef DEBUGHESS
4075: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4076:
1.203 brouard 4077: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4078: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4079:
4080: for (j=1;j<=npar;j++) {
4081: for (i=1;i<=npar;i++){
1.203 brouard 4082: printf("%.2f ",y[i][j]);
4083: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4084: }
4085: printf("\n");
4086: fprintf(ficlog,"\n");
4087: }
1.203 brouard 4088: #endif
1.126 brouard 4089:
4090: free_matrix(a,1,npar,1,npar);
4091: free_matrix(y,1,npar,1,npar);
4092: free_vector(x,1,npar);
4093: free_ivector(indx,1,npar);
1.203 brouard 4094: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4095:
4096:
4097: }
4098:
4099: /*************** hessian matrix ****************/
4100: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4101: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4102: int i;
4103: int l=1, lmax=20;
1.203 brouard 4104: double k1,k2, res, fx;
1.132 brouard 4105: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4106: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4107: int k=0,kmax=10;
4108: double l1;
4109:
4110: fx=func(x);
4111: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4112: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4113: l1=pow(10,l);
4114: delts=delt;
4115: for(k=1 ; k <kmax; k=k+1){
4116: delt = delta*(l1*k);
4117: p2[theta]=x[theta] +delt;
1.145 brouard 4118: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4119: p2[theta]=x[theta]-delt;
4120: k2=func(p2)-fx;
4121: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4122: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4123:
1.203 brouard 4124: #ifdef DEBUGHESSII
1.126 brouard 4125: 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);
4126: 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);
4127: #endif
4128: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4129: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4130: k=kmax;
4131: }
4132: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4133: k=kmax; l=lmax*10;
1.126 brouard 4134: }
4135: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4136: delts=delt;
4137: }
1.203 brouard 4138: } /* End loop k */
1.126 brouard 4139: }
4140: delti[theta]=delts;
4141: return res;
4142:
4143: }
4144:
1.203 brouard 4145: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4146: {
4147: int i;
1.164 brouard 4148: int l=1, lmax=20;
1.126 brouard 4149: double k1,k2,k3,k4,res,fx;
1.132 brouard 4150: double p2[MAXPARM+1];
1.203 brouard 4151: int k, kmax=1;
4152: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4153:
4154: int firstime=0;
1.203 brouard 4155:
1.126 brouard 4156: fx=func(x);
1.203 brouard 4157: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4158: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4159: p2[thetai]=x[thetai]+delti[thetai]*k;
4160: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4161: k1=func(p2)-fx;
4162:
1.203 brouard 4163: p2[thetai]=x[thetai]+delti[thetai]*k;
4164: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4165: k2=func(p2)-fx;
4166:
1.203 brouard 4167: p2[thetai]=x[thetai]-delti[thetai]*k;
4168: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4169: k3=func(p2)-fx;
4170:
1.203 brouard 4171: p2[thetai]=x[thetai]-delti[thetai]*k;
4172: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4173: k4=func(p2)-fx;
1.203 brouard 4174: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4175: if(k1*k2*k3*k4 <0.){
1.208 brouard 4176: firstime=1;
1.203 brouard 4177: kmax=kmax+10;
1.208 brouard 4178: }
4179: if(kmax >=10 || firstime ==1){
1.246 brouard 4180: 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);
4181: 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 4182: 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);
4183: 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);
4184: }
4185: #ifdef DEBUGHESSIJ
4186: v1=hess[thetai][thetai];
4187: v2=hess[thetaj][thetaj];
4188: cv12=res;
4189: /* Computing eigen value of Hessian matrix */
4190: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4191: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4192: if ((lc2 <0) || (lc1 <0) ){
4193: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4194: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4195: 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);
4196: 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);
4197: }
1.126 brouard 4198: #endif
4199: }
4200: return res;
4201: }
4202:
1.203 brouard 4203: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4204: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4205: /* { */
4206: /* int i; */
4207: /* int l=1, lmax=20; */
4208: /* double k1,k2,k3,k4,res,fx; */
4209: /* double p2[MAXPARM+1]; */
4210: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4211: /* int k=0,kmax=10; */
4212: /* double l1; */
4213:
4214: /* fx=func(x); */
4215: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4216: /* l1=pow(10,l); */
4217: /* delts=delt; */
4218: /* for(k=1 ; k <kmax; k=k+1){ */
4219: /* delt = delti*(l1*k); */
4220: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4221: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4222: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4223: /* k1=func(p2)-fx; */
4224:
4225: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4226: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4227: /* k2=func(p2)-fx; */
4228:
4229: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4230: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4231: /* k3=func(p2)-fx; */
4232:
4233: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4234: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4235: /* k4=func(p2)-fx; */
4236: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4237: /* #ifdef DEBUGHESSIJ */
4238: /* 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); */
4239: /* 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); */
4240: /* #endif */
4241: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4242: /* k=kmax; */
4243: /* } */
4244: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4245: /* k=kmax; l=lmax*10; */
4246: /* } */
4247: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4248: /* delts=delt; */
4249: /* } */
4250: /* } /\* End loop k *\/ */
4251: /* } */
4252: /* delti[theta]=delts; */
4253: /* return res; */
4254: /* } */
4255:
4256:
1.126 brouard 4257: /************** Inverse of matrix **************/
4258: void ludcmp(double **a, int n, int *indx, double *d)
4259: {
4260: int i,imax,j,k;
4261: double big,dum,sum,temp;
4262: double *vv;
4263:
4264: vv=vector(1,n);
4265: *d=1.0;
4266: for (i=1;i<=n;i++) {
4267: big=0.0;
4268: for (j=1;j<=n;j++)
4269: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4270: if (big == 0.0){
4271: printf(" Singular Hessian matrix at row %d:\n",i);
4272: for (j=1;j<=n;j++) {
4273: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4274: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4275: }
4276: fflush(ficlog);
4277: fclose(ficlog);
4278: nrerror("Singular matrix in routine ludcmp");
4279: }
1.126 brouard 4280: vv[i]=1.0/big;
4281: }
4282: for (j=1;j<=n;j++) {
4283: for (i=1;i<j;i++) {
4284: sum=a[i][j];
4285: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4286: a[i][j]=sum;
4287: }
4288: big=0.0;
4289: for (i=j;i<=n;i++) {
4290: sum=a[i][j];
4291: for (k=1;k<j;k++)
4292: sum -= a[i][k]*a[k][j];
4293: a[i][j]=sum;
4294: if ( (dum=vv[i]*fabs(sum)) >= big) {
4295: big=dum;
4296: imax=i;
4297: }
4298: }
4299: if (j != imax) {
4300: for (k=1;k<=n;k++) {
4301: dum=a[imax][k];
4302: a[imax][k]=a[j][k];
4303: a[j][k]=dum;
4304: }
4305: *d = -(*d);
4306: vv[imax]=vv[j];
4307: }
4308: indx[j]=imax;
4309: if (a[j][j] == 0.0) a[j][j]=TINY;
4310: if (j != n) {
4311: dum=1.0/(a[j][j]);
4312: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4313: }
4314: }
4315: free_vector(vv,1,n); /* Doesn't work */
4316: ;
4317: }
4318:
4319: void lubksb(double **a, int n, int *indx, double b[])
4320: {
4321: int i,ii=0,ip,j;
4322: double sum;
4323:
4324: for (i=1;i<=n;i++) {
4325: ip=indx[i];
4326: sum=b[ip];
4327: b[ip]=b[i];
4328: if (ii)
4329: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4330: else if (sum) ii=i;
4331: b[i]=sum;
4332: }
4333: for (i=n;i>=1;i--) {
4334: sum=b[i];
4335: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4336: b[i]=sum/a[i][i];
4337: }
4338: }
4339:
4340: void pstamp(FILE *fichier)
4341: {
1.196 brouard 4342: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4343: }
4344:
1.253 brouard 4345:
4346:
1.126 brouard 4347: /************ Frequencies ********************/
1.251 brouard 4348: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4349: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4350: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4351: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4352:
1.265 brouard 4353: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4354: int iind=0, iage=0;
4355: int mi; /* Effective wave */
4356: int first;
4357: double ***freq; /* Frequencies */
1.268 brouard 4358: 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 */
4359: 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 4360: double *meanq;
4361: double **meanqt;
4362: double *pp, **prop, *posprop, *pospropt;
4363: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4364: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4365: double agebegin, ageend;
4366:
4367: pp=vector(1,nlstate);
1.251 brouard 4368: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4369: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4370: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4371: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4372: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4373: meanqt=matrix(1,lastpass,1,nqtveff);
4374: strcpy(fileresp,"P_");
4375: strcat(fileresp,fileresu);
4376: /*strcat(fileresphtm,fileresu);*/
4377: if((ficresp=fopen(fileresp,"w"))==NULL) {
4378: printf("Problem with prevalence resultfile: %s\n", fileresp);
4379: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4380: exit(0);
4381: }
1.240 brouard 4382:
1.226 brouard 4383: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4384: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4385: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4386: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4387: fflush(ficlog);
4388: exit(70);
4389: }
4390: else{
4391: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4392: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4393: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4394: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4395: }
1.237 brouard 4396: 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 4397:
1.226 brouard 4398: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4399: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4400: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4401: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4402: fflush(ficlog);
4403: exit(70);
1.240 brouard 4404: } else{
1.226 brouard 4405: 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 4406: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4407: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4408: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4409: }
1.240 brouard 4410: 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);
4411:
1.253 brouard 4412: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4413: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4414: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4415: j1=0;
1.126 brouard 4416:
1.227 brouard 4417: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4418: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4419: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4420:
4421:
1.226 brouard 4422: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4423: reference=low_education V1=0,V2=0
4424: med_educ V1=1 V2=0,
4425: high_educ V1=0 V2=1
4426: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4427: */
1.249 brouard 4428: dateintsum=0;
4429: k2cpt=0;
4430:
1.253 brouard 4431: if(cptcoveff == 0 )
1.265 brouard 4432: nl=1; /* Constant and age model only */
1.253 brouard 4433: else
4434: nl=2;
1.265 brouard 4435:
4436: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4437: /* Loop on nj=1 or 2 if dummy covariates j!=0
4438: * Loop on j1(1 to 2**cptcoveff) covariate combination
4439: * freq[s1][s2][iage] =0.
4440: * Loop on iind
4441: * ++freq[s1][s2][iage] weighted
4442: * end iind
4443: * if covariate and j!0
4444: * headers Variable on one line
4445: * endif cov j!=0
4446: * header of frequency table by age
4447: * Loop on age
4448: * pp[s1]+=freq[s1][s2][iage] weighted
4449: * pos+=freq[s1][s2][iage] weighted
4450: * Loop on s1 initial state
4451: * fprintf(ficresp
4452: * end s1
4453: * end age
4454: * if j!=0 computes starting values
4455: * end compute starting values
4456: * end j1
4457: * end nl
4458: */
1.253 brouard 4459: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4460: if(nj==1)
4461: j=0; /* First pass for the constant */
1.265 brouard 4462: else{
1.253 brouard 4463: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4464: }
1.251 brouard 4465: first=1;
1.265 brouard 4466: 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 4467: posproptt=0.;
4468: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4469: scanf("%d", i);*/
4470: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4471: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4472: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4473: freq[i][s2][m]=0;
1.251 brouard 4474:
4475: for (i=1; i<=nlstate; i++) {
1.240 brouard 4476: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4477: prop[i][m]=0;
4478: posprop[i]=0;
4479: pospropt[i]=0;
4480: }
4481: /* for (z1=1; z1<= nqfveff; z1++) { */
4482: /* meanq[z1]+=0.; */
4483: /* for(m=1;m<=lastpass;m++){ */
4484: /* meanqt[m][z1]=0.; */
4485: /* } */
4486: /* } */
4487:
4488: /* dateintsum=0; */
4489: /* k2cpt=0; */
4490:
1.265 brouard 4491: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4492: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4493: bool=1;
4494: if(j !=0){
4495: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4496: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4497: /* for (z1=1; z1<= nqfveff; z1++) { */
4498: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4499: /* } */
4500: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4501: /* if(Tvaraff[z1] ==-20){ */
4502: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4503: /* }else if(Tvaraff[z1] ==-10){ */
4504: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4505: /* }else */
4506: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4507: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4508: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4509: /* 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",
4510: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4511: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4512: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4513: } /* Onlyf fixed */
4514: } /* end z1 */
4515: } /* cptcovn > 0 */
4516: } /* end any */
4517: }/* end j==0 */
1.265 brouard 4518: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4519: /* for(m=firstpass; m<=lastpass; m++){ */
4520: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4521: m=mw[mi][iind];
4522: if(j!=0){
4523: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4524: for (z1=1; z1<=cptcoveff; z1++) {
4525: if( Fixed[Tmodelind[z1]]==1){
4526: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4527: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4528: value is -1, we don't select. It differs from the
4529: constant and age model which counts them. */
4530: bool=0; /* not selected */
4531: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4532: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4533: bool=0;
4534: }
4535: }
4536: }
4537: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4538: } /* end j==0 */
4539: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4540: if(bool==1){
4541: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4542: and mw[mi+1][iind]. dh depends on stepm. */
4543: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4544: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4545: if(m >=firstpass && m <=lastpass){
4546: k2=anint[m][iind]+(mint[m][iind]/12.);
4547: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4548: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4549: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4550: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4551: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4552: if (m<lastpass) {
4553: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4554: /* 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]); */
4555: if(s[m][iind]==-1)
4556: 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.));
4557: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4558: /* if((int)agev[m][iind] == 55) */
4559: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4560: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4561: 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 4562: }
1.251 brouard 4563: } /* end if between passes */
4564: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4565: dateintsum=dateintsum+k2; /* on all covariates ?*/
4566: k2cpt++;
4567: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4568: }
1.251 brouard 4569: }else{
4570: bool=1;
4571: }/* end bool 2 */
4572: } /* end m */
4573: } /* end bool */
4574: } /* end iind = 1 to imx */
4575: /* prop[s][age] is feeded for any initial and valid live state as well as
4576: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4577:
4578:
4579: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4580: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4581: pstamp(ficresp);
1.251 brouard 4582: if (cptcoveff>0 && j!=0){
1.265 brouard 4583: pstamp(ficresp);
1.251 brouard 4584: printf( "\n#********** Variable ");
4585: fprintf(ficresp, "\n#********** Variable ");
4586: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4587: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4588: fprintf(ficlog, "\n#********** Variable ");
4589: for (z1=1; z1<=cptcoveff; z1++){
4590: if(!FixedV[Tvaraff[z1]]){
4591: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4592: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4593: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4594: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4595: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4596: }else{
1.251 brouard 4597: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4598: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4599: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4600: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4601: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4602: }
4603: }
4604: printf( "**********\n#");
4605: fprintf(ficresp, "**********\n#");
4606: fprintf(ficresphtm, "**********</h3>\n");
4607: fprintf(ficresphtmfr, "**********</h3>\n");
4608: fprintf(ficlog, "**********\n");
4609: }
4610: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4611: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4612: fprintf(ficresp, " Age");
4613: 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 4614: for(i=1; i<=nlstate;i++) {
1.265 brouard 4615: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4616: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4617: }
1.265 brouard 4618: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4619: fprintf(ficresphtm, "\n");
4620:
4621: /* Header of frequency table by age */
4622: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4623: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4624: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4625: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4626: if(s2!=0 && m!=0)
4627: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4628: }
1.226 brouard 4629: }
1.251 brouard 4630: fprintf(ficresphtmfr, "\n");
4631:
4632: /* For each age */
4633: for(iage=iagemin; iage <= iagemax+3; iage++){
4634: fprintf(ficresphtm,"<tr>");
4635: if(iage==iagemax+1){
4636: fprintf(ficlog,"1");
4637: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4638: }else if(iage==iagemax+2){
4639: fprintf(ficlog,"0");
4640: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4641: }else if(iage==iagemax+3){
4642: fprintf(ficlog,"Total");
4643: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4644: }else{
1.240 brouard 4645: if(first==1){
1.251 brouard 4646: first=0;
4647: printf("See log file for details...\n");
4648: }
4649: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4650: fprintf(ficlog,"Age %d", iage);
4651: }
1.265 brouard 4652: for(s1=1; s1 <=nlstate ; s1++){
4653: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4654: pp[s1] += freq[s1][m][iage];
1.251 brouard 4655: }
1.265 brouard 4656: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4657: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4658: pos += freq[s1][m][iage];
4659: if(pp[s1]>=1.e-10){
1.251 brouard 4660: if(first==1){
1.265 brouard 4661: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4662: }
1.265 brouard 4663: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4664: }else{
4665: if(first==1)
1.265 brouard 4666: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4667: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4668: }
4669: }
4670:
1.265 brouard 4671: for(s1=1; s1 <=nlstate ; s1++){
4672: /* posprop[s1]=0; */
4673: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4674: pp[s1] += freq[s1][m][iage];
4675: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4676:
4677: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4678: pos += pp[s1]; /* pos is the total number of transitions until this age */
4679: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4680: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4681: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4682: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4683: }
4684:
4685: /* Writing ficresp */
4686: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4687: if( iage <= iagemax){
4688: fprintf(ficresp," %d",iage);
4689: }
4690: }else if( nj==2){
4691: if( iage <= iagemax){
4692: fprintf(ficresp," %d",iage);
4693: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4694: }
1.240 brouard 4695: }
1.265 brouard 4696: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4697: if(pos>=1.e-5){
1.251 brouard 4698: if(first==1)
1.265 brouard 4699: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4700: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4701: }else{
4702: if(first==1)
1.265 brouard 4703: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4704: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4705: }
4706: if( iage <= iagemax){
4707: if(pos>=1.e-5){
1.265 brouard 4708: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4709: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4710: }else if( nj==2){
4711: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4712: }
4713: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4714: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4715: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4716: } else{
4717: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4718: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4719: }
1.240 brouard 4720: }
1.265 brouard 4721: pospropt[s1] +=posprop[s1];
4722: } /* end loop s1 */
1.251 brouard 4723: /* pospropt=0.; */
1.265 brouard 4724: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4725: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4726: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4727: if(first==1){
1.265 brouard 4728: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4729: }
1.265 brouard 4730: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4731: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4732: }
1.265 brouard 4733: if(s1!=0 && m!=0)
4734: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4735: }
1.265 brouard 4736: } /* end loop s1 */
1.251 brouard 4737: posproptt=0.;
1.265 brouard 4738: for(s1=1; s1 <=nlstate; s1++){
4739: posproptt += pospropt[s1];
1.251 brouard 4740: }
4741: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4742: fprintf(ficresphtm,"</tr>\n");
4743: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4744: if(iage <= iagemax)
4745: fprintf(ficresp,"\n");
1.240 brouard 4746: }
1.251 brouard 4747: if(first==1)
4748: printf("Others in log...\n");
4749: fprintf(ficlog,"\n");
4750: } /* end loop age iage */
1.265 brouard 4751:
1.251 brouard 4752: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4753: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4754: if(posproptt < 1.e-5){
1.265 brouard 4755: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4756: }else{
1.265 brouard 4757: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4758: }
1.226 brouard 4759: }
1.251 brouard 4760: fprintf(ficresphtm,"</tr>\n");
4761: fprintf(ficresphtm,"</table>\n");
4762: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4763: if(posproptt < 1.e-5){
1.251 brouard 4764: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4765: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4766: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4767: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4768: invalidvarcomb[j1]=1;
1.226 brouard 4769: }else{
1.251 brouard 4770: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4771: invalidvarcomb[j1]=0;
1.226 brouard 4772: }
1.251 brouard 4773: fprintf(ficresphtmfr,"</table>\n");
4774: fprintf(ficlog,"\n");
4775: if(j!=0){
4776: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4777: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4778: for(k=1; k <=(nlstate+ndeath); k++){
4779: if (k != i) {
1.265 brouard 4780: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4781: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4782: if(j1==1){ /* All dummy covariates to zero */
4783: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4784: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4785: printf("%d%d ",i,k);
4786: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4787: 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]));
4788: 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]));
4789: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4790: }
1.253 brouard 4791: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4792: for(iage=iagemin; iage <= iagemax+3; iage++){
4793: x[iage]= (double)iage;
4794: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4795: /* 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 4796: }
1.268 brouard 4797: /* Some are not finite, but linreg will ignore these ages */
4798: no=0;
1.253 brouard 4799: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4800: pstart[s1]=b;
4801: pstart[s1-1]=a;
1.252 brouard 4802: }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 */
4803: 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]);
4804: 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 4805: 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 4806: printf("%d%d ",i,k);
4807: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4808: 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 4809: }else{ /* Other cases, like quantitative fixed or varying covariates */
4810: ;
4811: }
4812: /* printf("%12.7f )", param[i][jj][k]); */
4813: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4814: s1++;
1.251 brouard 4815: } /* end jj */
4816: } /* end k!= i */
4817: } /* end k */
1.265 brouard 4818: } /* end i, s1 */
1.251 brouard 4819: } /* end j !=0 */
4820: } /* end selected combination of covariate j1 */
4821: if(j==0){ /* We can estimate starting values from the occurences in each case */
4822: printf("#Freqsummary: Starting values for the constants:\n");
4823: fprintf(ficlog,"\n");
1.265 brouard 4824: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4825: for(k=1; k <=(nlstate+ndeath); k++){
4826: if (k != i) {
4827: printf("%d%d ",i,k);
4828: fprintf(ficlog,"%d%d ",i,k);
4829: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4830: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4831: if(jj==1){ /* Age has to be done */
1.265 brouard 4832: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4833: 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]));
4834: 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 4835: }
4836: /* printf("%12.7f )", param[i][jj][k]); */
4837: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4838: s1++;
1.250 brouard 4839: }
1.251 brouard 4840: printf("\n");
4841: fprintf(ficlog,"\n");
1.250 brouard 4842: }
4843: }
4844: }
1.251 brouard 4845: printf("#Freqsummary\n");
4846: fprintf(ficlog,"\n");
1.265 brouard 4847: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4848: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4849: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
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]);
4852: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4853: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4854: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4855: /* } */
4856: }
1.265 brouard 4857: } /* end loop s1 */
1.251 brouard 4858:
4859: printf("\n");
4860: fprintf(ficlog,"\n");
4861: } /* end j=0 */
1.249 brouard 4862: } /* end j */
1.252 brouard 4863:
1.253 brouard 4864: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4865: for(i=1, jk=1; i <=nlstate; i++){
4866: for(j=1; j <=nlstate+ndeath; j++){
4867: if(j!=i){
4868: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4869: printf("%1d%1d",i,j);
4870: fprintf(ficparo,"%1d%1d",i,j);
4871: for(k=1; k<=ncovmodel;k++){
4872: /* printf(" %lf",param[i][j][k]); */
4873: /* fprintf(ficparo," %lf",param[i][j][k]); */
4874: p[jk]=pstart[jk];
4875: printf(" %f ",pstart[jk]);
4876: fprintf(ficparo," %f ",pstart[jk]);
4877: jk++;
4878: }
4879: printf("\n");
4880: fprintf(ficparo,"\n");
4881: }
4882: }
4883: }
4884: } /* end mle=-2 */
1.226 brouard 4885: dateintmean=dateintsum/k2cpt;
1.240 brouard 4886:
1.226 brouard 4887: fclose(ficresp);
4888: fclose(ficresphtm);
4889: fclose(ficresphtmfr);
4890: free_vector(meanq,1,nqfveff);
4891: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4892: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4893: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4894: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4895: free_vector(pospropt,1,nlstate);
4896: free_vector(posprop,1,nlstate);
1.251 brouard 4897: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4898: free_vector(pp,1,nlstate);
4899: /* End of freqsummary */
4900: }
1.126 brouard 4901:
1.268 brouard 4902: /* Simple linear regression */
4903: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4904:
4905: /* y=a+bx regression */
4906: double sumx = 0.0; /* sum of x */
4907: double sumx2 = 0.0; /* sum of x**2 */
4908: double sumxy = 0.0; /* sum of x * y */
4909: double sumy = 0.0; /* sum of y */
4910: double sumy2 = 0.0; /* sum of y**2 */
4911: double sume2 = 0.0; /* sum of square or residuals */
4912: double yhat;
4913:
4914: double denom=0;
4915: int i;
4916: int ne=*no;
4917:
4918: for ( i=ifi, ne=0;i<=ila;i++) {
4919: if(!isfinite(x[i]) || !isfinite(y[i])){
4920: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4921: continue;
4922: }
4923: ne=ne+1;
4924: sumx += x[i];
4925: sumx2 += x[i]*x[i];
4926: sumxy += x[i] * y[i];
4927: sumy += y[i];
4928: sumy2 += y[i]*y[i];
4929: denom = (ne * sumx2 - sumx*sumx);
4930: /* 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); */
4931: }
4932:
4933: denom = (ne * sumx2 - sumx*sumx);
4934: if (denom == 0) {
4935: // vertical, slope m is infinity
4936: *b = INFINITY;
4937: *a = 0;
4938: if (r) *r = 0;
4939: return 1;
4940: }
4941:
4942: *b = (ne * sumxy - sumx * sumy) / denom;
4943: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4944: if (r!=NULL) {
4945: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4946: sqrt((sumx2 - sumx*sumx/ne) *
4947: (sumy2 - sumy*sumy/ne));
4948: }
4949: *no=ne;
4950: for ( i=ifi, ne=0;i<=ila;i++) {
4951: if(!isfinite(x[i]) || !isfinite(y[i])){
4952: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4953: continue;
4954: }
4955: ne=ne+1;
4956: yhat = y[i] - *a -*b* x[i];
4957: sume2 += yhat * yhat ;
4958:
4959: denom = (ne * sumx2 - sumx*sumx);
4960: /* 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); */
4961: }
4962: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
4963: *sa= *sb * sqrt(sumx2/ne);
4964:
4965: return 0;
4966: }
4967:
1.126 brouard 4968: /************ Prevalence ********************/
1.227 brouard 4969: 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)
4970: {
4971: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4972: in each health status at the date of interview (if between dateprev1 and dateprev2).
4973: We still use firstpass and lastpass as another selection.
4974: */
1.126 brouard 4975:
1.227 brouard 4976: int i, m, jk, j1, bool, z1,j, iv;
4977: int mi; /* Effective wave */
4978: int iage;
4979: double agebegin, ageend;
4980:
4981: double **prop;
4982: double posprop;
4983: double y2; /* in fractional years */
4984: int iagemin, iagemax;
4985: int first; /** to stop verbosity which is redirected to log file */
4986:
4987: iagemin= (int) agemin;
4988: iagemax= (int) agemax;
4989: /*pp=vector(1,nlstate);*/
1.251 brouard 4990: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4991: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4992: j1=0;
1.222 brouard 4993:
1.227 brouard 4994: /*j=cptcoveff;*/
4995: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4996:
1.227 brouard 4997: first=1;
4998: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4999: for (i=1; i<=nlstate; i++)
1.251 brouard 5000: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5001: prop[i][iage]=0.0;
5002: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5003: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5004: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5005:
5006: for (i=1; i<=imx; i++) { /* Each individual */
5007: bool=1;
5008: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5009: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5010: m=mw[mi][i];
5011: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5012: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5013: for (z1=1; z1<=cptcoveff; z1++){
5014: if( Fixed[Tmodelind[z1]]==1){
5015: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5016: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5017: bool=0;
5018: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5019: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5020: bool=0;
5021: }
5022: }
5023: if(bool==1){ /* Otherwise we skip that wave/person */
5024: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5025: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5026: if(m >=firstpass && m <=lastpass){
5027: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5028: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5029: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5030: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5031: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5032: 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);
5033: exit(1);
5034: }
5035: if (s[m][i]>0 && s[m][i]<=nlstate) {
5036: /*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]]);*/
5037: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5038: prop[s[m][i]][iagemax+3] += weight[i];
5039: } /* end valid statuses */
5040: } /* end selection of dates */
5041: } /* end selection of waves */
5042: } /* end bool */
5043: } /* end wave */
5044: } /* end individual */
5045: for(i=iagemin; i <= iagemax+3; i++){
5046: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5047: posprop += prop[jk][i];
5048: }
5049:
5050: for(jk=1; jk <=nlstate ; jk++){
5051: if( i <= iagemax){
5052: if(posprop>=1.e-5){
5053: probs[i][jk][j1]= prop[jk][i]/posprop;
5054: } else{
5055: if(first==1){
5056: first=0;
1.266 brouard 5057: 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]);
5058: 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]);
5059: }else{
5060: 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 5061: }
5062: }
5063: }
5064: }/* end jk */
5065: }/* end i */
1.222 brouard 5066: /*} *//* end i1 */
1.227 brouard 5067: } /* end j1 */
1.222 brouard 5068:
1.227 brouard 5069: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5070: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5071: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5072: } /* End of prevalence */
1.126 brouard 5073:
5074: /************* Waves Concatenation ***************/
5075:
5076: 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)
5077: {
5078: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5079: Death is a valid wave (if date is known).
5080: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5081: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5082: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5083: */
1.126 brouard 5084:
1.224 brouard 5085: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5086: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5087: double sum=0., jmean=0.;*/
1.224 brouard 5088: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5089: int j, k=0,jk, ju, jl;
5090: double sum=0.;
5091: first=0;
1.214 brouard 5092: firstwo=0;
1.217 brouard 5093: firsthree=0;
1.218 brouard 5094: firstfour=0;
1.164 brouard 5095: jmin=100000;
1.126 brouard 5096: jmax=-1;
5097: jmean=0.;
1.224 brouard 5098:
5099: /* Treating live states */
1.214 brouard 5100: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5101: mi=0; /* First valid wave */
1.227 brouard 5102: mli=0; /* Last valid wave */
1.126 brouard 5103: m=firstpass;
1.214 brouard 5104: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5105: 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 */
5106: mli=m-1;/* mw[++mi][i]=m-1; */
5107: }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 */
5108: mw[++mi][i]=m;
5109: mli=m;
1.224 brouard 5110: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5111: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5112: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5113: }
1.227 brouard 5114: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5115: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5116: break;
1.224 brouard 5117: #else
1.227 brouard 5118: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5119: if(firsthree == 0){
1.262 brouard 5120: 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 5121: firsthree=1;
5122: }
1.262 brouard 5123: 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 5124: mw[++mi][i]=m;
5125: mli=m;
5126: }
5127: if(s[m][i]==-2){ /* Vital status is really unknown */
5128: nbwarn++;
5129: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5130: 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);
5131: 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);
5132: }
5133: break;
5134: }
5135: break;
1.224 brouard 5136: #endif
1.227 brouard 5137: }/* End m >= lastpass */
1.126 brouard 5138: }/* end while */
1.224 brouard 5139:
1.227 brouard 5140: /* 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 5141: /* After last pass */
1.224 brouard 5142: /* Treating death states */
1.214 brouard 5143: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5144: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5145: /* } */
1.126 brouard 5146: mi++; /* Death is another wave */
5147: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5148: /* Only death is a correct wave */
1.126 brouard 5149: mw[mi][i]=m;
1.257 brouard 5150: } /* else not in a death state */
1.224 brouard 5151: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5152: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5153: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5154: 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 */
5155: nbwarn++;
5156: if(firstfiv==0){
5157: 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 );
5158: firstfiv=1;
5159: }else{
5160: 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 );
5161: }
5162: }else{ /* Death occured afer last wave potential bias */
5163: nberr++;
5164: if(firstwo==0){
1.257 brouard 5165: 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 5166: firstwo=1;
5167: }
1.257 brouard 5168: 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 5169: }
1.257 brouard 5170: }else{ /* if date of interview is unknown */
1.227 brouard 5171: /* death is known but not confirmed by death status at any wave */
5172: if(firstfour==0){
5173: 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 );
5174: firstfour=1;
5175: }
5176: 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 5177: }
1.224 brouard 5178: } /* end if date of death is known */
5179: #endif
5180: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5181: /* wav[i]=mw[mi][i]; */
1.126 brouard 5182: if(mi==0){
5183: nbwarn++;
5184: if(first==0){
1.227 brouard 5185: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5186: first=1;
1.126 brouard 5187: }
5188: if(first==1){
1.227 brouard 5189: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5190: }
5191: } /* end mi==0 */
5192: } /* End individuals */
1.214 brouard 5193: /* wav and mw are no more changed */
1.223 brouard 5194:
1.214 brouard 5195:
1.126 brouard 5196: for(i=1; i<=imx; i++){
5197: for(mi=1; mi<wav[i];mi++){
5198: if (stepm <=0)
1.227 brouard 5199: dh[mi][i]=1;
1.126 brouard 5200: else{
1.260 brouard 5201: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5202: if (agedc[i] < 2*AGESUP) {
5203: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5204: if(j==0) j=1; /* Survives at least one month after exam */
5205: else if(j<0){
5206: nberr++;
5207: 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]);
5208: j=1; /* Temporary Dangerous patch */
5209: 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);
5210: 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]);
5211: 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);
5212: }
5213: k=k+1;
5214: if (j >= jmax){
5215: jmax=j;
5216: ijmax=i;
5217: }
5218: if (j <= jmin){
5219: jmin=j;
5220: ijmin=i;
5221: }
5222: sum=sum+j;
5223: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5224: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5225: }
5226: }
5227: else{
5228: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5229: /* 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 5230:
1.227 brouard 5231: k=k+1;
5232: if (j >= jmax) {
5233: jmax=j;
5234: ijmax=i;
5235: }
5236: else if (j <= jmin){
5237: jmin=j;
5238: ijmin=i;
5239: }
5240: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5241: /*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]);*/
5242: if(j<0){
5243: nberr++;
5244: 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]);
5245: 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]);
5246: }
5247: sum=sum+j;
5248: }
5249: jk= j/stepm;
5250: jl= j -jk*stepm;
5251: ju= j -(jk+1)*stepm;
5252: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5253: if(jl==0){
5254: dh[mi][i]=jk;
5255: bh[mi][i]=0;
5256: }else{ /* We want a negative bias in order to only have interpolation ie
5257: * to avoid the price of an extra matrix product in likelihood */
5258: dh[mi][i]=jk+1;
5259: bh[mi][i]=ju;
5260: }
5261: }else{
5262: if(jl <= -ju){
5263: dh[mi][i]=jk;
5264: bh[mi][i]=jl; /* bias is positive if real duration
5265: * is higher than the multiple of stepm and negative otherwise.
5266: */
5267: }
5268: else{
5269: dh[mi][i]=jk+1;
5270: bh[mi][i]=ju;
5271: }
5272: if(dh[mi][i]==0){
5273: dh[mi][i]=1; /* At least one step */
5274: bh[mi][i]=ju; /* At least one step */
5275: /* 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);*/
5276: }
5277: } /* end if mle */
1.126 brouard 5278: }
5279: } /* end wave */
5280: }
5281: jmean=sum/k;
5282: 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 5283: 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 5284: }
1.126 brouard 5285:
5286: /*********** Tricode ****************************/
1.220 brouard 5287: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5288: {
5289: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5290: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5291: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5292: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5293: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5294: */
1.130 brouard 5295:
1.242 brouard 5296: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5297: int modmaxcovj=0; /* Modality max of covariates j */
5298: int cptcode=0; /* Modality max of covariates j */
5299: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5300:
5301:
1.242 brouard 5302: /* cptcoveff=0; */
5303: /* *cptcov=0; */
1.126 brouard 5304:
1.242 brouard 5305: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5306:
1.242 brouard 5307: /* Loop on covariates without age and products and no quantitative variable */
5308: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5309: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5310: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5311: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5312: switch(Fixed[k]) {
5313: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5314: 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*/
5315: ij=(int)(covar[Tvar[k]][i]);
5316: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5317: * If product of Vn*Vm, still boolean *:
5318: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5319: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5320: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5321: modality of the nth covariate of individual i. */
5322: if (ij > modmaxcovj)
5323: modmaxcovj=ij;
5324: else if (ij < modmincovj)
5325: modmincovj=ij;
5326: if ((ij < -1) && (ij > NCOVMAX)){
5327: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5328: exit(1);
5329: }else
5330: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5331: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5332: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5333: /* getting the maximum value of the modality of the covariate
5334: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5335: female ies 1, then modmaxcovj=1.
5336: */
5337: } /* end for loop on individuals i */
5338: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5339: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5340: cptcode=modmaxcovj;
5341: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5342: /*for (i=0; i<=cptcode; i++) {*/
5343: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5344: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5345: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5346: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5347: if( j != -1){
5348: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5349: covariate for which somebody answered excluding
5350: undefined. Usually 2: 0 and 1. */
5351: }
5352: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5353: covariate for which somebody answered including
5354: undefined. Usually 3: -1, 0 and 1. */
5355: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5356: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5357: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5358:
1.242 brouard 5359: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5360: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5361: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5362: /* modmincovj=3; modmaxcovj = 7; */
5363: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5364: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5365: /* defining two dummy variables: variables V1_1 and V1_2.*/
5366: /* nbcode[Tvar[j]][ij]=k; */
5367: /* nbcode[Tvar[j]][1]=0; */
5368: /* nbcode[Tvar[j]][2]=1; */
5369: /* nbcode[Tvar[j]][3]=2; */
5370: /* To be continued (not working yet). */
5371: ij=0; /* ij is similar to i but can jump over null modalities */
5372: 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*/
5373: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5374: break;
5375: }
5376: ij++;
5377: 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*/
5378: cptcode = ij; /* New max modality for covar j */
5379: } /* end of loop on modality i=-1 to 1 or more */
5380: break;
5381: case 1: /* Testing on varying covariate, could be simple and
5382: * should look at waves or product of fixed *
5383: * varying. No time to test -1, assuming 0 and 1 only */
5384: ij=0;
5385: for(i=0; i<=1;i++){
5386: nbcode[Tvar[k]][++ij]=i;
5387: }
5388: break;
5389: default:
5390: break;
5391: } /* end switch */
5392: } /* end dummy test */
5393:
5394: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5395: /* /\*recode from 0 *\/ */
5396: /* k is a modality. If we have model=V1+V1*sex */
5397: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5398: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5399: /* } */
5400: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5401: /* if (ij > ncodemax[j]) { */
5402: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5403: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5404: /* break; */
5405: /* } */
5406: /* } /\* end of loop on modality k *\/ */
5407: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5408:
5409: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5410: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5411: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5412: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5413: 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 */
5414: 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 */
5415: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5416: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5417:
5418: ij=0;
5419: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5420: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5421: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5422: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5423: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5424: /* If product not in single variable we don't print results */
5425: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5426: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5427: 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*/
5428: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5429: 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 */
5430: if(Fixed[k]!=0)
5431: anyvaryingduminmodel=1;
5432: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5433: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5434: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5435: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5436: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5437: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5438: }
5439: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5440: /* ij--; */
5441: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5442: *cptcov=ij; /*Number of total real effective covariates: effective
5443: * because they can be excluded from the model and real
5444: * if in the model but excluded because missing values, but how to get k from ij?*/
5445: for(j=ij+1; j<= cptcovt; j++){
5446: Tvaraff[j]=0;
5447: Tmodelind[j]=0;
5448: }
5449: for(j=ntveff+1; j<= cptcovt; j++){
5450: TmodelInvind[j]=0;
5451: }
5452: /* To be sorted */
5453: ;
5454: }
1.126 brouard 5455:
1.145 brouard 5456:
1.126 brouard 5457: /*********** Health Expectancies ****************/
5458:
1.235 brouard 5459: 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 5460:
5461: {
5462: /* Health expectancies, no variances */
1.164 brouard 5463: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5464: int nhstepma, nstepma; /* Decreasing with age */
5465: double age, agelim, hf;
5466: double ***p3mat;
5467: double eip;
5468:
1.238 brouard 5469: /* pstamp(ficreseij); */
1.126 brouard 5470: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5471: fprintf(ficreseij,"# Age");
5472: for(i=1; i<=nlstate;i++){
5473: for(j=1; j<=nlstate;j++){
5474: fprintf(ficreseij," e%1d%1d ",i,j);
5475: }
5476: fprintf(ficreseij," e%1d. ",i);
5477: }
5478: fprintf(ficreseij,"\n");
5479:
5480:
5481: if(estepm < stepm){
5482: printf ("Problem %d lower than %d\n",estepm, stepm);
5483: }
5484: else hstepm=estepm;
5485: /* We compute the life expectancy from trapezoids spaced every estepm months
5486: * This is mainly to measure the difference between two models: for example
5487: * if stepm=24 months pijx are given only every 2 years and by summing them
5488: * we are calculating an estimate of the Life Expectancy assuming a linear
5489: * progression in between and thus overestimating or underestimating according
5490: * to the curvature of the survival function. If, for the same date, we
5491: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5492: * to compare the new estimate of Life expectancy with the same linear
5493: * hypothesis. A more precise result, taking into account a more precise
5494: * curvature will be obtained if estepm is as small as stepm. */
5495:
5496: /* For example we decided to compute the life expectancy with the smallest unit */
5497: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5498: nhstepm is the number of hstepm from age to agelim
5499: nstepm is the number of stepm from age to agelin.
1.270 brouard 5500: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5501: and note for a fixed period like estepm months */
5502: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5503: survival function given by stepm (the optimization length). Unfortunately it
5504: means that if the survival funtion is printed only each two years of age and if
5505: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5506: results. So we changed our mind and took the option of the best precision.
5507: */
5508: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5509:
5510: agelim=AGESUP;
5511: /* If stepm=6 months */
5512: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5513: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5514:
5515: /* nhstepm age range expressed in number of stepm */
5516: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5517: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5518: /* if (stepm >= YEARM) hstepm=1;*/
5519: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5520: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5521:
5522: for (age=bage; age<=fage; age ++){
5523: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5524: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5525: /* if (stepm >= YEARM) hstepm=1;*/
5526: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5527:
5528: /* If stepm=6 months */
5529: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5530: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5531:
1.235 brouard 5532: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5533:
5534: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5535:
5536: printf("%d|",(int)age);fflush(stdout);
5537: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5538:
5539: /* Computing expectancies */
5540: for(i=1; i<=nlstate;i++)
5541: for(j=1; j<=nlstate;j++)
5542: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5543: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5544:
5545: /* 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]);*/
5546:
5547: }
5548:
5549: fprintf(ficreseij,"%3.0f",age );
5550: for(i=1; i<=nlstate;i++){
5551: eip=0;
5552: for(j=1; j<=nlstate;j++){
5553: eip +=eij[i][j][(int)age];
5554: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5555: }
5556: fprintf(ficreseij,"%9.4f", eip );
5557: }
5558: fprintf(ficreseij,"\n");
5559:
5560: }
5561: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5562: printf("\n");
5563: fprintf(ficlog,"\n");
5564:
5565: }
5566:
1.235 brouard 5567: 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 5568:
5569: {
5570: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5571: to initial status i, ei. .
1.126 brouard 5572: */
5573: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5574: int nhstepma, nstepma; /* Decreasing with age */
5575: double age, agelim, hf;
5576: double ***p3matp, ***p3matm, ***varhe;
5577: double **dnewm,**doldm;
5578: double *xp, *xm;
5579: double **gp, **gm;
5580: double ***gradg, ***trgradg;
5581: int theta;
5582:
5583: double eip, vip;
5584:
5585: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5586: xp=vector(1,npar);
5587: xm=vector(1,npar);
5588: dnewm=matrix(1,nlstate*nlstate,1,npar);
5589: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5590:
5591: pstamp(ficresstdeij);
5592: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5593: fprintf(ficresstdeij,"# Age");
5594: for(i=1; i<=nlstate;i++){
5595: for(j=1; j<=nlstate;j++)
5596: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5597: fprintf(ficresstdeij," e%1d. ",i);
5598: }
5599: fprintf(ficresstdeij,"\n");
5600:
5601: pstamp(ficrescveij);
5602: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5603: fprintf(ficrescveij,"# Age");
5604: for(i=1; i<=nlstate;i++)
5605: for(j=1; j<=nlstate;j++){
5606: cptj= (j-1)*nlstate+i;
5607: for(i2=1; i2<=nlstate;i2++)
5608: for(j2=1; j2<=nlstate;j2++){
5609: cptj2= (j2-1)*nlstate+i2;
5610: if(cptj2 <= cptj)
5611: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5612: }
5613: }
5614: fprintf(ficrescveij,"\n");
5615:
5616: if(estepm < stepm){
5617: printf ("Problem %d lower than %d\n",estepm, stepm);
5618: }
5619: else hstepm=estepm;
5620: /* We compute the life expectancy from trapezoids spaced every estepm months
5621: * This is mainly to measure the difference between two models: for example
5622: * if stepm=24 months pijx are given only every 2 years and by summing them
5623: * we are calculating an estimate of the Life Expectancy assuming a linear
5624: * progression in between and thus overestimating or underestimating according
5625: * to the curvature of the survival function. If, for the same date, we
5626: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5627: * to compare the new estimate of Life expectancy with the same linear
5628: * hypothesis. A more precise result, taking into account a more precise
5629: * curvature will be obtained if estepm is as small as stepm. */
5630:
5631: /* For example we decided to compute the life expectancy with the smallest unit */
5632: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5633: nhstepm is the number of hstepm from age to agelim
5634: nstepm is the number of stepm from age to agelin.
5635: Look at hpijx to understand the reason of that which relies in memory size
5636: and note for a fixed period like estepm months */
5637: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5638: survival function given by stepm (the optimization length). Unfortunately it
5639: means that if the survival funtion is printed only each two years of age and if
5640: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5641: results. So we changed our mind and took the option of the best precision.
5642: */
5643: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5644:
5645: /* If stepm=6 months */
5646: /* nhstepm age range expressed in number of stepm */
5647: agelim=AGESUP;
5648: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5649: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5650: /* if (stepm >= YEARM) hstepm=1;*/
5651: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5652:
5653: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5654: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5655: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5656: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5657: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5658: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5659:
5660: for (age=bage; age<=fage; age ++){
5661: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5662: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5663: /* if (stepm >= YEARM) hstepm=1;*/
5664: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5665:
1.126 brouard 5666: /* If stepm=6 months */
5667: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5668: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5669:
5670: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5671:
1.126 brouard 5672: /* Computing Variances of health expectancies */
5673: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5674: decrease memory allocation */
5675: for(theta=1; theta <=npar; theta++){
5676: for(i=1; i<=npar; i++){
1.222 brouard 5677: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5678: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5679: }
1.235 brouard 5680: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5681: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5682:
1.126 brouard 5683: for(j=1; j<= nlstate; j++){
1.222 brouard 5684: for(i=1; i<=nlstate; i++){
5685: for(h=0; h<=nhstepm-1; h++){
5686: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5687: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5688: }
5689: }
1.126 brouard 5690: }
1.218 brouard 5691:
1.126 brouard 5692: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5693: for(h=0; h<=nhstepm-1; h++){
5694: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5695: }
1.126 brouard 5696: }/* End theta */
5697:
5698:
5699: for(h=0; h<=nhstepm-1; h++)
5700: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5701: for(theta=1; theta <=npar; theta++)
5702: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5703:
1.218 brouard 5704:
1.222 brouard 5705: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5706: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5707: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5708:
1.222 brouard 5709: printf("%d|",(int)age);fflush(stdout);
5710: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5711: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5712: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5713: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5714: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5715: for(ij=1;ij<=nlstate*nlstate;ij++)
5716: for(ji=1;ji<=nlstate*nlstate;ji++)
5717: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5718: }
5719: }
1.218 brouard 5720:
1.126 brouard 5721: /* Computing expectancies */
1.235 brouard 5722: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5723: for(i=1; i<=nlstate;i++)
5724: for(j=1; j<=nlstate;j++)
1.222 brouard 5725: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5726: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5727:
1.222 brouard 5728: /* 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 5729:
1.222 brouard 5730: }
1.269 brouard 5731:
5732: /* Standard deviation of expectancies ij */
1.126 brouard 5733: fprintf(ficresstdeij,"%3.0f",age );
5734: for(i=1; i<=nlstate;i++){
5735: eip=0.;
5736: vip=0.;
5737: for(j=1; j<=nlstate;j++){
1.222 brouard 5738: eip += eij[i][j][(int)age];
5739: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5740: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5741: 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 5742: }
5743: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5744: }
5745: fprintf(ficresstdeij,"\n");
1.218 brouard 5746:
1.269 brouard 5747: /* Variance of expectancies ij */
1.126 brouard 5748: fprintf(ficrescveij,"%3.0f",age );
5749: for(i=1; i<=nlstate;i++)
5750: for(j=1; j<=nlstate;j++){
1.222 brouard 5751: cptj= (j-1)*nlstate+i;
5752: for(i2=1; i2<=nlstate;i2++)
5753: for(j2=1; j2<=nlstate;j2++){
5754: cptj2= (j2-1)*nlstate+i2;
5755: if(cptj2 <= cptj)
5756: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5757: }
1.126 brouard 5758: }
5759: fprintf(ficrescveij,"\n");
1.218 brouard 5760:
1.126 brouard 5761: }
5762: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5763: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5764: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5765: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5766: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5767: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5768: printf("\n");
5769: fprintf(ficlog,"\n");
1.218 brouard 5770:
1.126 brouard 5771: free_vector(xm,1,npar);
5772: free_vector(xp,1,npar);
5773: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5774: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5775: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5776: }
1.218 brouard 5777:
1.126 brouard 5778: /************ Variance ******************/
1.235 brouard 5779: 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 5780: {
1.279 brouard 5781: /** Variance of health expectancies
5782: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5783: * double **newm;
5784: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5785: */
1.218 brouard 5786:
5787: /* int movingaverage(); */
5788: double **dnewm,**doldm;
5789: double **dnewmp,**doldmp;
5790: int i, j, nhstepm, hstepm, h, nstepm ;
5791: int k;
5792: double *xp;
1.279 brouard 5793: double **gp, **gm; /**< for var eij */
5794: double ***gradg, ***trgradg; /**< for var eij */
5795: double **gradgp, **trgradgp; /**< for var p point j */
5796: double *gpp, *gmp; /**< for var p point j */
5797: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5798: double ***p3mat;
5799: double age,agelim, hf;
5800: /* double ***mobaverage; */
5801: int theta;
5802: char digit[4];
5803: char digitp[25];
5804:
5805: char fileresprobmorprev[FILENAMELENGTH];
5806:
5807: if(popbased==1){
5808: if(mobilav!=0)
5809: strcpy(digitp,"-POPULBASED-MOBILAV_");
5810: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5811: }
5812: else
5813: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5814:
1.218 brouard 5815: /* if (mobilav!=0) { */
5816: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5817: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5818: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5819: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5820: /* } */
5821: /* } */
5822:
5823: strcpy(fileresprobmorprev,"PRMORPREV-");
5824: sprintf(digit,"%-d",ij);
5825: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5826: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5827: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5828: strcat(fileresprobmorprev,fileresu);
5829: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5830: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5831: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5832: }
5833: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5834: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5835: pstamp(ficresprobmorprev);
5836: 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 5837: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5838: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5839: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5840: }
5841: for(j=1;j<=cptcoveff;j++)
5842: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5843: fprintf(ficresprobmorprev,"\n");
5844:
1.218 brouard 5845: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5846: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5847: fprintf(ficresprobmorprev," p.%-d SE",j);
5848: for(i=1; i<=nlstate;i++)
5849: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5850: }
5851: fprintf(ficresprobmorprev,"\n");
5852:
5853: fprintf(ficgp,"\n# Routine varevsij");
5854: fprintf(ficgp,"\nunset title \n");
5855: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5856: 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");
5857: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 5858:
1.218 brouard 5859: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5860: pstamp(ficresvij);
5861: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5862: if(popbased==1)
5863: 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);
5864: else
5865: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5866: fprintf(ficresvij,"# Age");
5867: for(i=1; i<=nlstate;i++)
5868: for(j=1; j<=nlstate;j++)
5869: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5870: fprintf(ficresvij,"\n");
5871:
5872: xp=vector(1,npar);
5873: dnewm=matrix(1,nlstate,1,npar);
5874: doldm=matrix(1,nlstate,1,nlstate);
5875: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5876: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5877:
5878: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5879: gpp=vector(nlstate+1,nlstate+ndeath);
5880: gmp=vector(nlstate+1,nlstate+ndeath);
5881: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5882:
1.218 brouard 5883: if(estepm < stepm){
5884: printf ("Problem %d lower than %d\n",estepm, stepm);
5885: }
5886: else hstepm=estepm;
5887: /* For example we decided to compute the life expectancy with the smallest unit */
5888: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5889: nhstepm is the number of hstepm from age to agelim
5890: nstepm is the number of stepm from age to agelim.
5891: Look at function hpijx to understand why because of memory size limitations,
5892: we decided (b) to get a life expectancy respecting the most precise curvature of the
5893: survival function given by stepm (the optimization length). Unfortunately it
5894: means that if the survival funtion is printed every two years of age and if
5895: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5896: results. So we changed our mind and took the option of the best precision.
5897: */
5898: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5899: agelim = AGESUP;
5900: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5901: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5902: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5903: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5904: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5905: gp=matrix(0,nhstepm,1,nlstate);
5906: gm=matrix(0,nhstepm,1,nlstate);
5907:
5908:
5909: for(theta=1; theta <=npar; theta++){
5910: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5911: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5912: }
1.279 brouard 5913: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
5914: * returns into prlim .
5915: */
1.242 brouard 5916: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 5917:
5918: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 5919: if (popbased==1) {
5920: if(mobilav ==0){
5921: for(i=1; i<=nlstate;i++)
5922: prlim[i][i]=probs[(int)age][i][ij];
5923: }else{ /* mobilav */
5924: for(i=1; i<=nlstate;i++)
5925: prlim[i][i]=mobaverage[(int)age][i][ij];
5926: }
5927: }
1.279 brouard 5928: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}_x\f$ at horizon h.
5929: */
5930: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=0 to nhstepm */
5931: /**< And for each alive state j, sums over i \f$ w^i_x {}{h}_p^{ij}_x\f$, which are the probability
5932: * at horizon h in state j including mortality.
5933: */
1.218 brouard 5934: for(j=1; j<= nlstate; j++){
5935: for(h=0; h<=nhstepm; h++){
5936: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5937: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5938: }
5939: }
1.279 brouard 5940: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 5941: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 5942: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 5943: */
5944: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5945: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5946: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 5947: }
5948:
5949: /* Again with minus shift */
1.218 brouard 5950:
5951: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5952: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5953:
1.242 brouard 5954: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5955:
5956: if (popbased==1) {
5957: if(mobilav ==0){
5958: for(i=1; i<=nlstate;i++)
5959: prlim[i][i]=probs[(int)age][i][ij];
5960: }else{ /* mobilav */
5961: for(i=1; i<=nlstate;i++)
5962: prlim[i][i]=mobaverage[(int)age][i][ij];
5963: }
5964: }
5965:
1.235 brouard 5966: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5967:
5968: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5969: for(h=0; h<=nhstepm; h++){
5970: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5971: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5972: }
5973: }
5974: /* This for computing probability of death (h=1 means
5975: computed over hstepm matrices product = hstepm*stepm months)
5976: as a weighted average of prlim.
5977: */
5978: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5979: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5980: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5981: }
1.279 brouard 5982: /* end shifting computations */
5983:
5984: /**< Computing gradient matrix at horizon h
5985: */
1.218 brouard 5986: for(j=1; j<= nlstate; j++) /* vareij */
5987: for(h=0; h<=nhstepm; h++){
5988: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5989: }
1.279 brouard 5990: /**< Gradient of overall mortality p.3 (or p.j)
5991: */
5992: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 5993: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5994: }
5995:
5996: } /* End theta */
1.279 brouard 5997:
5998: /* We got the gradient matrix for each theta and state j */
1.218 brouard 5999: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6000:
6001: for(h=0; h<=nhstepm; h++) /* veij */
6002: for(j=1; j<=nlstate;j++)
6003: for(theta=1; theta <=npar; theta++)
6004: trgradg[h][j][theta]=gradg[h][theta][j];
6005:
6006: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6007: for(theta=1; theta <=npar; theta++)
6008: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6009: /**< as well as its transposed matrix
6010: */
1.218 brouard 6011:
6012: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6013: for(i=1;i<=nlstate;i++)
6014: for(j=1;j<=nlstate;j++)
6015: vareij[i][j][(int)age] =0.;
1.279 brouard 6016:
6017: /* Computing trgradg by matcov by gradg at age and summing over h
6018: * and k (nhstepm) formula 15 of article
6019: * Lievre-Brouard-Heathcote
6020: */
6021:
1.218 brouard 6022: for(h=0;h<=nhstepm;h++){
6023: for(k=0;k<=nhstepm;k++){
6024: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6025: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6026: for(i=1;i<=nlstate;i++)
6027: for(j=1;j<=nlstate;j++)
6028: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6029: }
6030: }
6031:
1.279 brouard 6032: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6033: * p.j overall mortality formula 49 but computed directly because
6034: * we compute the grad (wix pijx) instead of grad (pijx),even if
6035: * wix is independent of theta.
6036: */
1.218 brouard 6037: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6038: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6039: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6040: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6041: varppt[j][i]=doldmp[j][i];
6042: /* end ppptj */
6043: /* x centered again */
6044:
1.242 brouard 6045: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6046:
6047: if (popbased==1) {
6048: if(mobilav ==0){
6049: for(i=1; i<=nlstate;i++)
6050: prlim[i][i]=probs[(int)age][i][ij];
6051: }else{ /* mobilav */
6052: for(i=1; i<=nlstate;i++)
6053: prlim[i][i]=mobaverage[(int)age][i][ij];
6054: }
6055: }
6056:
6057: /* This for computing probability of death (h=1 means
6058: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6059: as a weighted average of prlim.
6060: */
1.235 brouard 6061: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6062: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6063: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6064: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6065: }
6066: /* end probability of death */
6067:
6068: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6069: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6070: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6071: for(i=1; i<=nlstate;i++){
6072: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6073: }
6074: }
6075: fprintf(ficresprobmorprev,"\n");
6076:
6077: fprintf(ficresvij,"%.0f ",age );
6078: for(i=1; i<=nlstate;i++)
6079: for(j=1; j<=nlstate;j++){
6080: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6081: }
6082: fprintf(ficresvij,"\n");
6083: free_matrix(gp,0,nhstepm,1,nlstate);
6084: free_matrix(gm,0,nhstepm,1,nlstate);
6085: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6086: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6087: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6088: } /* End age */
6089: free_vector(gpp,nlstate+1,nlstate+ndeath);
6090: free_vector(gmp,nlstate+1,nlstate+ndeath);
6091: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6092: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6093: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6094: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6095: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6096: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6097: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6098: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6099: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6100: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6101: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6102: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6103: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6104: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6105: 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);
6106: /* 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 6107: */
1.218 brouard 6108: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6109: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6110:
1.218 brouard 6111: free_vector(xp,1,npar);
6112: free_matrix(doldm,1,nlstate,1,nlstate);
6113: free_matrix(dnewm,1,nlstate,1,npar);
6114: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6115: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6116: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6117: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6118: fclose(ficresprobmorprev);
6119: fflush(ficgp);
6120: fflush(fichtm);
6121: } /* end varevsij */
1.126 brouard 6122:
6123: /************ Variance of prevlim ******************/
1.269 brouard 6124: 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 6125: {
1.205 brouard 6126: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6127: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6128:
1.268 brouard 6129: double **dnewmpar,**doldm;
1.126 brouard 6130: int i, j, nhstepm, hstepm;
6131: double *xp;
6132: double *gp, *gm;
6133: double **gradg, **trgradg;
1.208 brouard 6134: double **mgm, **mgp;
1.126 brouard 6135: double age,agelim;
6136: int theta;
6137:
6138: pstamp(ficresvpl);
6139: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6140: fprintf(ficresvpl,"# Age ");
6141: if(nresult >=1)
6142: fprintf(ficresvpl," Result# ");
1.126 brouard 6143: for(i=1; i<=nlstate;i++)
6144: fprintf(ficresvpl," %1d-%1d",i,i);
6145: fprintf(ficresvpl,"\n");
6146:
6147: xp=vector(1,npar);
1.268 brouard 6148: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6149: doldm=matrix(1,nlstate,1,nlstate);
6150:
6151: hstepm=1*YEARM; /* Every year of age */
6152: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6153: agelim = AGESUP;
6154: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6155: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6156: if (stepm >= YEARM) hstepm=1;
6157: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6158: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6159: mgp=matrix(1,npar,1,nlstate);
6160: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6161: gp=vector(1,nlstate);
6162: gm=vector(1,nlstate);
6163:
6164: for(theta=1; theta <=npar; theta++){
6165: for(i=1; i<=npar; i++){ /* Computes gradient */
6166: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6167: }
1.209 brouard 6168: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6169: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6170: else
1.235 brouard 6171: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6172: for(i=1;i<=nlstate;i++){
1.126 brouard 6173: gp[i] = prlim[i][i];
1.208 brouard 6174: mgp[theta][i] = prlim[i][i];
6175: }
1.126 brouard 6176: for(i=1; i<=npar; i++) /* Computes gradient */
6177: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6178: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6179: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6180: else
1.235 brouard 6181: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6182: for(i=1;i<=nlstate;i++){
1.126 brouard 6183: gm[i] = prlim[i][i];
1.208 brouard 6184: mgm[theta][i] = prlim[i][i];
6185: }
1.126 brouard 6186: for(i=1;i<=nlstate;i++)
6187: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6188: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6189: } /* End theta */
6190:
6191: trgradg =matrix(1,nlstate,1,npar);
6192:
6193: for(j=1; j<=nlstate;j++)
6194: for(theta=1; theta <=npar; theta++)
6195: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6196: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6197: /* printf("\nmgm mgp %d ",(int)age); */
6198: /* for(j=1; j<=nlstate;j++){ */
6199: /* printf(" %d ",j); */
6200: /* for(theta=1; theta <=npar; theta++) */
6201: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6202: /* printf("\n "); */
6203: /* } */
6204: /* } */
6205: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6206: /* printf("\n gradg %d ",(int)age); */
6207: /* for(j=1; j<=nlstate;j++){ */
6208: /* printf("%d ",j); */
6209: /* for(theta=1; theta <=npar; theta++) */
6210: /* printf("%d %lf ",theta,gradg[theta][j]); */
6211: /* printf("\n "); */
6212: /* } */
6213: /* } */
1.126 brouard 6214:
6215: for(i=1;i<=nlstate;i++)
6216: varpl[i][(int)age] =0.;
1.209 brouard 6217: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
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: }else{
1.268 brouard 6221: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6222: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6223: }
1.126 brouard 6224: for(i=1;i<=nlstate;i++)
6225: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6226:
6227: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6228: if(nresult >=1)
6229: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6230: for(i=1; i<=nlstate;i++)
6231: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6232: fprintf(ficresvpl,"\n");
6233: free_vector(gp,1,nlstate);
6234: free_vector(gm,1,nlstate);
1.208 brouard 6235: free_matrix(mgm,1,npar,1,nlstate);
6236: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6237: free_matrix(gradg,1,npar,1,nlstate);
6238: free_matrix(trgradg,1,nlstate,1,npar);
6239: } /* End age */
6240:
6241: free_vector(xp,1,npar);
6242: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6243: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6244:
6245: }
6246:
6247:
6248: /************ Variance of backprevalence limit ******************/
1.269 brouard 6249: 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 6250: {
6251: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6252: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6253:
6254: double **dnewmpar,**doldm;
6255: int i, j, nhstepm, hstepm;
6256: double *xp;
6257: double *gp, *gm;
6258: double **gradg, **trgradg;
6259: double **mgm, **mgp;
6260: double age,agelim;
6261: int theta;
6262:
6263: pstamp(ficresvbl);
6264: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6265: fprintf(ficresvbl,"# Age ");
6266: if(nresult >=1)
6267: fprintf(ficresvbl," Result# ");
6268: for(i=1; i<=nlstate;i++)
6269: fprintf(ficresvbl," %1d-%1d",i,i);
6270: fprintf(ficresvbl,"\n");
6271:
6272: xp=vector(1,npar);
6273: dnewmpar=matrix(1,nlstate,1,npar);
6274: doldm=matrix(1,nlstate,1,nlstate);
6275:
6276: hstepm=1*YEARM; /* Every year of age */
6277: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6278: agelim = AGEINF;
6279: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6280: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6281: if (stepm >= YEARM) hstepm=1;
6282: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6283: gradg=matrix(1,npar,1,nlstate);
6284: mgp=matrix(1,npar,1,nlstate);
6285: mgm=matrix(1,npar,1,nlstate);
6286: gp=vector(1,nlstate);
6287: gm=vector(1,nlstate);
6288:
6289: for(theta=1; theta <=npar; theta++){
6290: for(i=1; i<=npar; i++){ /* Computes gradient */
6291: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6292: }
6293: if(mobilavproj > 0 )
6294: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6295: else
6296: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6297: for(i=1;i<=nlstate;i++){
6298: gp[i] = bprlim[i][i];
6299: mgp[theta][i] = bprlim[i][i];
6300: }
6301: for(i=1; i<=npar; i++) /* Computes gradient */
6302: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6303: if(mobilavproj > 0 )
6304: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6305: else
6306: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6307: for(i=1;i<=nlstate;i++){
6308: gm[i] = bprlim[i][i];
6309: mgm[theta][i] = bprlim[i][i];
6310: }
6311: for(i=1;i<=nlstate;i++)
6312: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6313: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6314: } /* End theta */
6315:
6316: trgradg =matrix(1,nlstate,1,npar);
6317:
6318: for(j=1; j<=nlstate;j++)
6319: for(theta=1; theta <=npar; theta++)
6320: trgradg[j][theta]=gradg[theta][j];
6321: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6322: /* printf("\nmgm mgp %d ",(int)age); */
6323: /* for(j=1; j<=nlstate;j++){ */
6324: /* printf(" %d ",j); */
6325: /* for(theta=1; theta <=npar; theta++) */
6326: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6327: /* printf("\n "); */
6328: /* } */
6329: /* } */
6330: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6331: /* printf("\n gradg %d ",(int)age); */
6332: /* for(j=1; j<=nlstate;j++){ */
6333: /* printf("%d ",j); */
6334: /* for(theta=1; theta <=npar; theta++) */
6335: /* printf("%d %lf ",theta,gradg[theta][j]); */
6336: /* printf("\n "); */
6337: /* } */
6338: /* } */
6339:
6340: for(i=1;i<=nlstate;i++)
6341: varbpl[i][(int)age] =0.;
6342: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6343: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6344: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6345: }else{
6346: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6347: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6348: }
6349: for(i=1;i<=nlstate;i++)
6350: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6351:
6352: fprintf(ficresvbl,"%.0f ",age );
6353: if(nresult >=1)
6354: fprintf(ficresvbl,"%d ",nres );
6355: for(i=1; i<=nlstate;i++)
6356: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6357: fprintf(ficresvbl,"\n");
6358: free_vector(gp,1,nlstate);
6359: free_vector(gm,1,nlstate);
6360: free_matrix(mgm,1,npar,1,nlstate);
6361: free_matrix(mgp,1,npar,1,nlstate);
6362: free_matrix(gradg,1,npar,1,nlstate);
6363: free_matrix(trgradg,1,nlstate,1,npar);
6364: } /* End age */
6365:
6366: free_vector(xp,1,npar);
6367: free_matrix(doldm,1,nlstate,1,npar);
6368: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6369:
6370: }
6371:
6372: /************ Variance of one-step probabilities ******************/
6373: 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 6374: {
6375: int i, j=0, k1, l1, tj;
6376: int k2, l2, j1, z1;
6377: int k=0, l;
6378: int first=1, first1, first2;
6379: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6380: double **dnewm,**doldm;
6381: double *xp;
6382: double *gp, *gm;
6383: double **gradg, **trgradg;
6384: double **mu;
6385: double age, cov[NCOVMAX+1];
6386: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6387: int theta;
6388: char fileresprob[FILENAMELENGTH];
6389: char fileresprobcov[FILENAMELENGTH];
6390: char fileresprobcor[FILENAMELENGTH];
6391: double ***varpij;
6392:
6393: strcpy(fileresprob,"PROB_");
6394: strcat(fileresprob,fileres);
6395: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6396: printf("Problem with resultfile: %s\n", fileresprob);
6397: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6398: }
6399: strcpy(fileresprobcov,"PROBCOV_");
6400: strcat(fileresprobcov,fileresu);
6401: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6402: printf("Problem with resultfile: %s\n", fileresprobcov);
6403: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6404: }
6405: strcpy(fileresprobcor,"PROBCOR_");
6406: strcat(fileresprobcor,fileresu);
6407: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6408: printf("Problem with resultfile: %s\n", fileresprobcor);
6409: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6410: }
6411: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6412: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6413: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6414: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6415: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6416: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6417: pstamp(ficresprob);
6418: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6419: fprintf(ficresprob,"# Age");
6420: pstamp(ficresprobcov);
6421: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6422: fprintf(ficresprobcov,"# Age");
6423: pstamp(ficresprobcor);
6424: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6425: fprintf(ficresprobcor,"# Age");
1.126 brouard 6426:
6427:
1.222 brouard 6428: for(i=1; i<=nlstate;i++)
6429: for(j=1; j<=(nlstate+ndeath);j++){
6430: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6431: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6432: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6433: }
6434: /* fprintf(ficresprob,"\n");
6435: fprintf(ficresprobcov,"\n");
6436: fprintf(ficresprobcor,"\n");
6437: */
6438: xp=vector(1,npar);
6439: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6440: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6441: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6442: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6443: first=1;
6444: fprintf(ficgp,"\n# Routine varprob");
6445: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6446: fprintf(fichtm,"\n");
6447:
1.266 brouard 6448: 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 6449: 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);
6450: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6451: and drawn. It helps understanding how is the covariance between two incidences.\
6452: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6453: 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 6454: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6455: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6456: standard deviations wide on each axis. <br>\
6457: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6458: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6459: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6460:
1.222 brouard 6461: cov[1]=1;
6462: /* tj=cptcoveff; */
1.225 brouard 6463: tj = (int) pow(2,cptcoveff);
1.222 brouard 6464: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6465: j1=0;
1.224 brouard 6466: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6467: if (cptcovn>0) {
6468: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6469: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6470: fprintf(ficresprob, "**********\n#\n");
6471: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6472: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6473: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6474:
1.222 brouard 6475: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6476: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6477: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6478:
6479:
1.222 brouard 6480: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6481: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6482: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6483:
1.222 brouard 6484: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6485: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6486: fprintf(ficresprobcor, "**********\n#");
6487: if(invalidvarcomb[j1]){
6488: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6489: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6490: continue;
6491: }
6492: }
6493: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6494: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6495: gp=vector(1,(nlstate)*(nlstate+ndeath));
6496: gm=vector(1,(nlstate)*(nlstate+ndeath));
6497: for (age=bage; age<=fage; age ++){
6498: cov[2]=age;
6499: if(nagesqr==1)
6500: cov[3]= age*age;
6501: for (k=1; k<=cptcovn;k++) {
6502: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6503: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6504: * 1 1 1 1 1
6505: * 2 2 1 1 1
6506: * 3 1 2 1 1
6507: */
6508: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6509: }
6510: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6511: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6512: for (k=1; k<=cptcovprod;k++)
6513: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6514:
6515:
1.222 brouard 6516: for(theta=1; theta <=npar; theta++){
6517: for(i=1; i<=npar; i++)
6518: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6519:
1.222 brouard 6520: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6521:
1.222 brouard 6522: k=0;
6523: for(i=1; i<= (nlstate); i++){
6524: for(j=1; j<=(nlstate+ndeath);j++){
6525: k=k+1;
6526: gp[k]=pmmij[i][j];
6527: }
6528: }
1.220 brouard 6529:
1.222 brouard 6530: for(i=1; i<=npar; i++)
6531: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6532:
1.222 brouard 6533: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6534: k=0;
6535: for(i=1; i<=(nlstate); i++){
6536: for(j=1; j<=(nlstate+ndeath);j++){
6537: k=k+1;
6538: gm[k]=pmmij[i][j];
6539: }
6540: }
1.220 brouard 6541:
1.222 brouard 6542: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6543: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6544: }
1.126 brouard 6545:
1.222 brouard 6546: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6547: for(theta=1; theta <=npar; theta++)
6548: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6549:
1.222 brouard 6550: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6551: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6552:
1.222 brouard 6553: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6554:
1.222 brouard 6555: k=0;
6556: for(i=1; i<=(nlstate); i++){
6557: for(j=1; j<=(nlstate+ndeath);j++){
6558: k=k+1;
6559: mu[k][(int) age]=pmmij[i][j];
6560: }
6561: }
6562: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6563: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6564: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6565:
1.222 brouard 6566: /*printf("\n%d ",(int)age);
6567: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6568: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6569: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6570: }*/
1.220 brouard 6571:
1.222 brouard 6572: fprintf(ficresprob,"\n%d ",(int)age);
6573: fprintf(ficresprobcov,"\n%d ",(int)age);
6574: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6575:
1.222 brouard 6576: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6577: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6578: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6579: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6580: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6581: }
6582: i=0;
6583: for (k=1; k<=(nlstate);k++){
6584: for (l=1; l<=(nlstate+ndeath);l++){
6585: i++;
6586: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6587: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6588: for (j=1; j<=i;j++){
6589: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6590: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6591: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6592: }
6593: }
6594: }/* end of loop for state */
6595: } /* end of loop for age */
6596: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6597: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6598: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6599: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6600:
6601: /* Confidence intervalle of pij */
6602: /*
6603: fprintf(ficgp,"\nunset parametric;unset label");
6604: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6605: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6606: 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);
6607: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6608: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6609: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6610: */
6611:
6612: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6613: first1=1;first2=2;
6614: for (k2=1; k2<=(nlstate);k2++){
6615: for (l2=1; l2<=(nlstate+ndeath);l2++){
6616: if(l2==k2) continue;
6617: j=(k2-1)*(nlstate+ndeath)+l2;
6618: for (k1=1; k1<=(nlstate);k1++){
6619: for (l1=1; l1<=(nlstate+ndeath);l1++){
6620: if(l1==k1) continue;
6621: i=(k1-1)*(nlstate+ndeath)+l1;
6622: if(i<=j) continue;
6623: for (age=bage; age<=fage; age ++){
6624: if ((int)age %5==0){
6625: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6626: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6627: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6628: mu1=mu[i][(int) age]/stepm*YEARM ;
6629: mu2=mu[j][(int) age]/stepm*YEARM;
6630: c12=cv12/sqrt(v1*v2);
6631: /* Computing eigen value of matrix of covariance */
6632: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6633: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6634: if ((lc2 <0) || (lc1 <0) ){
6635: if(first2==1){
6636: first1=0;
6637: 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);
6638: }
6639: 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);
6640: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6641: /* lc2=fabs(lc2); */
6642: }
1.220 brouard 6643:
1.222 brouard 6644: /* Eigen vectors */
1.280 ! brouard 6645: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
! 6646: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
! 6647: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
! 6648: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
! 6649: }else
! 6650: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6651: /*v21=sqrt(1.-v11*v11); *//* error */
6652: v21=(lc1-v1)/cv12*v11;
6653: v12=-v21;
6654: v22=v11;
6655: tnalp=v21/v11;
6656: if(first1==1){
6657: first1=0;
6658: 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);
6659: }
6660: 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);
6661: /*printf(fignu*/
6662: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6663: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6664: if(first==1){
6665: first=0;
6666: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6667: fprintf(ficgp,"\nset parametric;unset label");
6668: 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);
6669: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6670: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6671: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6672: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6673: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6674: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6675: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6676: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6677: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6678: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6679: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6680: fprintf(ficgp,"\nplot [-pi:pi] %11.3e+ %.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)), %11.3e +%.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)) not", \
1.280 ! brouard 6681: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
! 6682: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6683: }else{
6684: first=0;
6685: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6686: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6687: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6688: 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 6689: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6690: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6691: }/* if first */
6692: } /* age mod 5 */
6693: } /* end loop age */
6694: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6695: first=1;
6696: } /*l12 */
6697: } /* k12 */
6698: } /*l1 */
6699: }/* k1 */
6700: } /* loop on combination of covariates j1 */
6701: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6702: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6703: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6704: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6705: free_vector(xp,1,npar);
6706: fclose(ficresprob);
6707: fclose(ficresprobcov);
6708: fclose(ficresprobcor);
6709: fflush(ficgp);
6710: fflush(fichtmcov);
6711: }
1.126 brouard 6712:
6713:
6714: /******************* Printing html file ***********/
1.201 brouard 6715: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6716: int lastpass, int stepm, int weightopt, char model[],\
6717: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6718: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6719: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6720: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6721: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6722:
6723: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6724: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6725: </ul>");
1.237 brouard 6726: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6727: </ul>", model);
1.214 brouard 6728: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6729: 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",
6730: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6731: 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 6732: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6733: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6734: fprintf(fichtm,"\
6735: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6736: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6737: fprintf(fichtm,"\
1.217 brouard 6738: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6739: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6740: fprintf(fichtm,"\
1.126 brouard 6741: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6742: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6743: fprintf(fichtm,"\
1.217 brouard 6744: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6745: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6746: fprintf(fichtm,"\
1.211 brouard 6747: - (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 6748: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6749: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6750: if(prevfcast==1){
6751: fprintf(fichtm,"\
6752: - Prevalence projections by age and states: \
1.201 brouard 6753: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6754: }
1.126 brouard 6755:
6756:
1.225 brouard 6757: m=pow(2,cptcoveff);
1.222 brouard 6758: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6759:
1.264 brouard 6760: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6761:
6762: jj1=0;
6763:
6764: fprintf(fichtm," \n<ul>");
6765: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6766: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6767: if(m != 1 && TKresult[nres]!= k1)
6768: continue;
6769: jj1++;
6770: if (cptcovn > 0) {
6771: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6772: for (cpt=1; cpt<=cptcoveff;cpt++){
6773: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6774: }
6775: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6776: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6777: }
6778: fprintf(fichtm,"\">");
6779:
6780: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6781: fprintf(fichtm,"************ Results for covariates");
6782: for (cpt=1; cpt<=cptcoveff;cpt++){
6783: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6784: }
6785: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6786: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6787: }
6788: if(invalidvarcomb[k1]){
6789: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6790: continue;
6791: }
6792: fprintf(fichtm,"</a></li>");
6793: } /* cptcovn >0 */
6794: }
6795: fprintf(fichtm," \n</ul>");
6796:
1.222 brouard 6797: jj1=0;
1.237 brouard 6798:
6799: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6800: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6801: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6802: continue;
1.220 brouard 6803:
1.222 brouard 6804: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6805: jj1++;
6806: if (cptcovn > 0) {
1.264 brouard 6807: fprintf(fichtm,"\n<p><a name=\"rescov");
6808: for (cpt=1; cpt<=cptcoveff;cpt++){
6809: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6810: }
6811: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6812: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6813: }
6814: fprintf(fichtm,"\"</a>");
6815:
1.222 brouard 6816: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6817: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6818: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6819: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6820: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6821: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6822: }
1.237 brouard 6823: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6824: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6825: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6826: }
6827:
1.230 brouard 6828: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6829: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6830: if(invalidvarcomb[k1]){
6831: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6832: printf("\nCombination (%d) ignored because no cases \n",k1);
6833: continue;
6834: }
6835: }
6836: /* aij, bij */
1.259 brouard 6837: 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 6838: <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 6839: /* Pij */
1.241 brouard 6840: 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> \
6841: <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 6842: /* Quasi-incidences */
6843: 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 6844: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6845: 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 6846: 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> \
6847: <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 6848: /* Survival functions (period) in state j */
6849: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6850: 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> \
6851: <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 6852: }
6853: /* State specific survival functions (period) */
6854: for(cpt=1; cpt<=nlstate;cpt++){
6855: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6856: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6857: <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 6858: }
6859: /* Period (stable) prevalence in each health state */
6860: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6861: 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> \
6862: <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 6863: }
6864: if(backcast==1){
6865: /* Period (stable) back prevalence in each health state */
6866: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6867: 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 6868: <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 6869: }
1.217 brouard 6870: }
1.222 brouard 6871: if(prevfcast==1){
6872: /* Projection of prevalence up to period (stable) prevalence in each health state */
6873: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6874: 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> \
6875: <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 6876: }
6877: }
1.268 brouard 6878: if(backcast==1){
6879: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6880: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6881: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6882: 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 \
6883: 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) \
6884: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6885: <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 6886: }
6887: }
1.220 brouard 6888:
1.222 brouard 6889: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6890: 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> \
6891: <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 6892: }
6893: /* } /\* end i1 *\/ */
6894: }/* End k1 */
6895: fprintf(fichtm,"</ul>");
1.126 brouard 6896:
1.222 brouard 6897: fprintf(fichtm,"\
1.126 brouard 6898: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6899: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6900: - 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 6901: But because parameters are usually highly correlated (a higher incidence of disability \
6902: and a higher incidence of recovery can give very close observed transition) it might \
6903: be very useful to look not only at linear confidence intervals estimated from the \
6904: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6905: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6906: covariance matrix of the one-step probabilities. \
6907: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6908:
1.222 brouard 6909: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6910: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6911: fprintf(fichtm,"\
1.126 brouard 6912: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6913: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6914:
1.222 brouard 6915: fprintf(fichtm,"\
1.126 brouard 6916: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6917: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6918: fprintf(fichtm,"\
1.126 brouard 6919: - 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): \
6920: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6921: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6922: fprintf(fichtm,"\
1.126 brouard 6923: - (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): \
6924: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6925: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6926: fprintf(fichtm,"\
1.128 brouard 6927: - 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 6928: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6929: fprintf(fichtm,"\
1.128 brouard 6930: - 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 6931: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6932: fprintf(fichtm,"\
1.126 brouard 6933: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6934: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6935:
6936: /* if(popforecast==1) fprintf(fichtm,"\n */
6937: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6938: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6939: /* <br>",fileres,fileres,fileres,fileres); */
6940: /* else */
6941: /* 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 6942: fflush(fichtm);
6943: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6944:
1.225 brouard 6945: m=pow(2,cptcoveff);
1.222 brouard 6946: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6947:
1.222 brouard 6948: jj1=0;
1.237 brouard 6949:
1.241 brouard 6950: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6951: for(k1=1; k1<=m;k1++){
1.253 brouard 6952: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6953: continue;
1.222 brouard 6954: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6955: jj1++;
1.126 brouard 6956: if (cptcovn > 0) {
6957: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6958: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6959: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6960: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6961: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6962: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6963: }
6964:
1.126 brouard 6965: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6966:
1.222 brouard 6967: if(invalidvarcomb[k1]){
6968: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6969: continue;
6970: }
1.126 brouard 6971: }
6972: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6973: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6974: 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 6975: <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 6976: }
6977: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6978: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6979: true period expectancies (those weighted with period prevalences are also\
6980: drawn in addition to the population based expectancies computed using\
1.241 brouard 6981: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6982: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6983: /* } /\* end i1 *\/ */
6984: }/* End k1 */
1.241 brouard 6985: }/* End nres */
1.222 brouard 6986: fprintf(fichtm,"</ul>");
6987: fflush(fichtm);
1.126 brouard 6988: }
6989:
6990: /******************* Gnuplot file **************/
1.270 brouard 6991: 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 6992:
6993: char dirfileres[132],optfileres[132];
1.264 brouard 6994: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6995: 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 6996: int lv=0, vlv=0, kl=0;
1.130 brouard 6997: int ng=0;
1.201 brouard 6998: int vpopbased;
1.223 brouard 6999: int ioffset; /* variable offset for columns */
1.270 brouard 7000: int iyearc=1; /* variable column for year of projection */
7001: int iagec=1; /* variable column for age of projection */
1.235 brouard 7002: int nres=0; /* Index of resultline */
1.266 brouard 7003: int istart=1; /* For starting graphs in projections */
1.219 brouard 7004:
1.126 brouard 7005: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7006: /* printf("Problem with file %s",optionfilegnuplot); */
7007: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7008: /* } */
7009:
7010: /*#ifdef windows */
7011: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7012: /*#endif */
1.225 brouard 7013: m=pow(2,cptcoveff);
1.126 brouard 7014:
1.274 brouard 7015: /* diagram of the model */
7016: fprintf(ficgp,"\n#Diagram of the model \n");
7017: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7018: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7019: 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);
7020:
7021: 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);
7022: fprintf(ficgp,"\n#show arrow\nunset label\n");
7023: 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);
7024: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7025: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7026: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7027: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7028:
1.202 brouard 7029: /* Contribution to likelihood */
7030: /* Plot the probability implied in the likelihood */
1.223 brouard 7031: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7032: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7033: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7034: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7035: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7036: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7037: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7038: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7039: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7040: 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));
7041: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7042: 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));
7043: for (i=1; i<= nlstate ; i ++) {
7044: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7045: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7046: 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);
7047: for (j=2; j<= nlstate+ndeath ; j ++) {
7048: 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);
7049: }
7050: fprintf(ficgp,";\nset out; unset ylabel;\n");
7051: }
7052: /* 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 */
7053: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7054: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7055: fprintf(ficgp,"\nset out;unset log\n");
7056: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7057:
1.126 brouard 7058: strcpy(dirfileres,optionfilefiname);
7059: strcpy(optfileres,"vpl");
1.223 brouard 7060: /* 1eme*/
1.238 brouard 7061: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7062: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7063: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7064: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7065: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7066: continue;
7067: /* We are interested in selected combination by the resultline */
1.246 brouard 7068: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7069: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7070: strcpy(gplotlabel,"(");
1.238 brouard 7071: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7072: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7073: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7074: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7075: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7076: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7077: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7078: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7079: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7080: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7081: }
7082: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7083: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7084: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7085: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7086: }
7087: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7088: /* printf("\n#\n"); */
1.238 brouard 7089: fprintf(ficgp,"\n#\n");
7090: if(invalidvarcomb[k1]){
1.260 brouard 7091: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7092: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7093: continue;
7094: }
1.235 brouard 7095:
1.241 brouard 7096: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7097: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7098: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7099: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7100: 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);
7101: /* 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); */
7102: /* k1-1 error should be nres-1*/
1.238 brouard 7103: for (i=1; i<= nlstate ; i ++) {
7104: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7105: else fprintf(ficgp," %%*lf (%%*lf)");
7106: }
1.260 brouard 7107: 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 7108: for (i=1; i<= nlstate ; i ++) {
7109: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7110: else fprintf(ficgp," %%*lf (%%*lf)");
7111: }
1.260 brouard 7112: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7113: for (i=1; i<= nlstate ; i ++) {
7114: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7115: else fprintf(ficgp," %%*lf (%%*lf)");
7116: }
1.265 brouard 7117: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7118:
7119: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7120: if(cptcoveff ==0){
1.271 brouard 7121: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7122: }else{
7123: kl=0;
7124: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7125: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7126: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7127: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7128: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7129: vlv= nbcode[Tvaraff[k]][lv];
7130: kl++;
7131: /* 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 *\/ */
7132: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7133: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7134: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7135: if(k==cptcoveff){
7136: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7137: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7138: }else{
7139: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7140: kl++;
7141: }
7142: } /* end covariate */
7143: } /* end if no covariate */
7144:
1.238 brouard 7145: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7146: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7147: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7148: if(cptcoveff ==0){
1.245 brouard 7149: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7150: }else{
7151: kl=0;
7152: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7153: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7154: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7155: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7156: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7157: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7158: kl++;
1.238 brouard 7159: /* 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 *\/ */
7160: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7161: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7162: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7163: if(k==cptcoveff){
1.245 brouard 7164: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7165: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7166: }else{
7167: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7168: kl++;
7169: }
7170: } /* end covariate */
7171: } /* end if no covariate */
1.268 brouard 7172: if(backcast == 1){
7173: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7174: /* k1-1 error should be nres-1*/
7175: for (i=1; i<= nlstate ; i ++) {
7176: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7177: else fprintf(ficgp," %%*lf (%%*lf)");
7178: }
1.271 brouard 7179: 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 7180: for (i=1; i<= nlstate ; i ++) {
7181: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7182: else fprintf(ficgp," %%*lf (%%*lf)");
7183: }
1.276 brouard 7184: 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 7185: for (i=1; i<= nlstate ; i ++) {
7186: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7187: else fprintf(ficgp," %%*lf (%%*lf)");
7188: }
1.274 brouard 7189: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7190: } /* end if backprojcast */
1.238 brouard 7191: } /* end if backcast */
1.276 brouard 7192: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7193: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7194: } /* nres */
1.201 brouard 7195: } /* k1 */
7196: } /* cpt */
1.235 brouard 7197:
7198:
1.126 brouard 7199: /*2 eme*/
1.238 brouard 7200: for (k1=1; k1<= m ; k1 ++){
7201: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7202: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7203: continue;
7204: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7205: strcpy(gplotlabel,"(");
1.238 brouard 7206: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7207: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7208: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7209: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7210: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7211: vlv= nbcode[Tvaraff[k]][lv];
7212: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7213: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7214: }
1.237 brouard 7215: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7216: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7217: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7218: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7219: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7220: }
1.264 brouard 7221: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7222: fprintf(ficgp,"\n#\n");
1.223 brouard 7223: if(invalidvarcomb[k1]){
7224: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7225: continue;
7226: }
1.219 brouard 7227:
1.241 brouard 7228: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7229: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7230: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7231: if(vpopbased==0){
1.238 brouard 7232: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7233: }else
1.238 brouard 7234: fprintf(ficgp,"\nreplot ");
7235: for (i=1; i<= nlstate+1 ; i ++) {
7236: k=2*i;
1.261 brouard 7237: 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 7238: for (j=1; j<= nlstate+1 ; j ++) {
7239: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7240: else fprintf(ficgp," %%*lf (%%*lf)");
7241: }
7242: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7243: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7244: 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 7245: for (j=1; j<= nlstate+1 ; j ++) {
7246: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7247: else fprintf(ficgp," %%*lf (%%*lf)");
7248: }
7249: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7250: 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 7251: for (j=1; j<= nlstate+1 ; j ++) {
7252: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7253: else fprintf(ficgp," %%*lf (%%*lf)");
7254: }
7255: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7256: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7257: } /* state */
7258: } /* vpopbased */
1.264 brouard 7259: 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 7260: } /* end nres */
7261: } /* k1 end 2 eme*/
7262:
7263:
7264: /*3eme*/
7265: for (k1=1; k1<= m ; k1 ++){
7266: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7267: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7268: continue;
7269:
7270: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7271: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7272: strcpy(gplotlabel,"(");
1.238 brouard 7273: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7274: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7275: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7276: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7277: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7278: vlv= nbcode[Tvaraff[k]][lv];
7279: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7280: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7281: }
7282: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7283: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7284: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7285: }
1.264 brouard 7286: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7287: fprintf(ficgp,"\n#\n");
7288: if(invalidvarcomb[k1]){
7289: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7290: continue;
7291: }
7292:
7293: /* k=2+nlstate*(2*cpt-2); */
7294: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7295: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7296: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7297: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7298: 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 7299: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7300: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7301: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7302: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7303: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7304: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7305:
1.238 brouard 7306: */
7307: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7308: 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 7309: /* 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 7310:
1.238 brouard 7311: }
1.261 brouard 7312: 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 7313: }
1.264 brouard 7314: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7315: } /* end nres */
7316: } /* end kl 3eme */
1.126 brouard 7317:
1.223 brouard 7318: /* 4eme */
1.201 brouard 7319: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7320: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7321: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7322: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7323: continue;
1.238 brouard 7324: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7325: strcpy(gplotlabel,"(");
1.238 brouard 7326: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7327: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7328: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7329: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7330: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7331: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7332: vlv= nbcode[Tvaraff[k]][lv];
7333: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7334: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7335: }
7336: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7337: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7338: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7339: }
1.264 brouard 7340: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7341: fprintf(ficgp,"\n#\n");
7342: if(invalidvarcomb[k1]){
7343: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7344: continue;
1.223 brouard 7345: }
1.238 brouard 7346:
1.241 brouard 7347: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7348: 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 7349: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7350: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7351: k=3;
7352: for (i=1; i<= nlstate ; i ++){
7353: if(i==1){
7354: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7355: }else{
7356: fprintf(ficgp,", '' ");
7357: }
7358: l=(nlstate+ndeath)*(i-1)+1;
7359: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7360: for (j=2; j<= nlstate+ndeath ; j ++)
7361: fprintf(ficgp,"+$%d",k+l+j-1);
7362: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7363: } /* nlstate */
1.264 brouard 7364: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7365: } /* end cpt state*/
7366: } /* end nres */
7367: } /* end covariate k1 */
7368:
1.220 brouard 7369: /* 5eme */
1.201 brouard 7370: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7371: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7372: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7373: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7374: continue;
1.238 brouard 7375: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7376: strcpy(gplotlabel,"(");
1.238 brouard 7377: 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);
7378: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7379: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7380: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7381: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7382: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7383: vlv= nbcode[Tvaraff[k]][lv];
7384: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7385: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7386: }
7387: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7388: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7389: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7390: }
1.264 brouard 7391: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7392: fprintf(ficgp,"\n#\n");
7393: if(invalidvarcomb[k1]){
7394: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7395: continue;
7396: }
1.227 brouard 7397:
1.241 brouard 7398: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7399: 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 7400: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7401: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7402: k=3;
7403: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7404: if(j==1)
7405: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7406: else
7407: fprintf(ficgp,", '' ");
7408: l=(nlstate+ndeath)*(cpt-1) +j;
7409: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7410: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7411: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7412: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7413: } /* nlstate */
7414: fprintf(ficgp,", '' ");
7415: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7416: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7417: l=(nlstate+ndeath)*(cpt-1) +j;
7418: if(j < nlstate)
7419: fprintf(ficgp,"$%d +",k+l);
7420: else
7421: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7422: }
1.264 brouard 7423: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7424: } /* end cpt state*/
7425: } /* end covariate */
7426: } /* end nres */
1.227 brouard 7427:
1.220 brouard 7428: /* 6eme */
1.202 brouard 7429: /* CV preval stable (period) for each covariate */
1.237 brouard 7430: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7431: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7432: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7433: continue;
1.255 brouard 7434: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7435: strcpy(gplotlabel,"(");
1.211 brouard 7436: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7437: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7438: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7439: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7440: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7441: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7442: vlv= nbcode[Tvaraff[k]][lv];
7443: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7444: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7445: }
1.237 brouard 7446: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7447: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7448: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7449: }
1.264 brouard 7450: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7451: fprintf(ficgp,"\n#\n");
1.223 brouard 7452: if(invalidvarcomb[k1]){
1.227 brouard 7453: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7454: continue;
1.223 brouard 7455: }
1.227 brouard 7456:
1.241 brouard 7457: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7458: 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 7459: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7460: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7461: k=3; /* Offset */
1.255 brouard 7462: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7463: if(i==1)
7464: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7465: else
7466: fprintf(ficgp,", '' ");
1.255 brouard 7467: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7468: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7469: for (j=2; j<= nlstate ; j ++)
7470: fprintf(ficgp,"+$%d",k+l+j-1);
7471: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7472: } /* nlstate */
1.264 brouard 7473: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7474: } /* end cpt state*/
7475: } /* end covariate */
1.227 brouard 7476:
7477:
1.220 brouard 7478: /* 7eme */
1.218 brouard 7479: if(backcast == 1){
1.217 brouard 7480: /* CV back preval stable (period) for each covariate */
1.237 brouard 7481: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7482: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7483: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7484: continue;
1.268 brouard 7485: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7486: strcpy(gplotlabel,"(");
7487: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7488: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7489: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7490: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7491: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7492: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7493: vlv= nbcode[Tvaraff[k]][lv];
7494: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7495: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7496: }
1.237 brouard 7497: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7498: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7499: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7500: }
1.264 brouard 7501: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7502: fprintf(ficgp,"\n#\n");
7503: if(invalidvarcomb[k1]){
7504: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7505: continue;
7506: }
7507:
1.241 brouard 7508: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7509: 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 7510: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7511: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7512: k=3; /* Offset */
1.268 brouard 7513: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7514: if(i==1)
7515: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7516: else
7517: fprintf(ficgp,", '' ");
7518: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7519: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7520: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7521: /* 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 7522: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7523: /* for (j=2; j<= nlstate ; j ++) */
7524: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7525: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7526: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7527: } /* nlstate */
1.264 brouard 7528: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7529: } /* end cpt state*/
7530: } /* end covariate */
7531: } /* End if backcast */
7532:
1.223 brouard 7533: /* 8eme */
1.218 brouard 7534: if(prevfcast==1){
7535: /* Projection from cross-sectional to stable (period) for each covariate */
7536:
1.237 brouard 7537: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7538: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7539: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7540: continue;
1.211 brouard 7541: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7542: strcpy(gplotlabel,"(");
1.227 brouard 7543: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7544: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7545: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7546: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7547: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7548: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7549: vlv= nbcode[Tvaraff[k]][lv];
7550: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7551: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7552: }
1.237 brouard 7553: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7554: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7555: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7556: }
1.264 brouard 7557: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7558: fprintf(ficgp,"\n#\n");
7559: if(invalidvarcomb[k1]){
7560: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7561: continue;
7562: }
7563:
7564: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7565: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7566: 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 7567: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7568: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7569:
7570: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7571: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7572: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7573: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7574: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7575: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7576: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7577: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7578: if(i==istart){
1.227 brouard 7579: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7580: }else{
7581: fprintf(ficgp,",\\\n '' ");
7582: }
7583: if(cptcoveff ==0){ /* No covariate */
7584: ioffset=2; /* Age is in 2 */
7585: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7586: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7587: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7588: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7589: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7590: if(i==nlstate+1){
1.270 brouard 7591: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7592: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7593: fprintf(ficgp,",\\\n '' ");
7594: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7595: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7596: offyear, \
1.268 brouard 7597: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7598: }else
1.227 brouard 7599: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7600: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7601: }else{ /* more than 2 covariates */
1.270 brouard 7602: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7603: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7604: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7605: iyearc=ioffset-1;
7606: iagec=ioffset;
1.227 brouard 7607: fprintf(ficgp," u %d:(",ioffset);
7608: kl=0;
7609: strcpy(gplotcondition,"(");
7610: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7611: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7612: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7613: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7614: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7615: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7616: kl++;
7617: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7618: kl++;
7619: if(k <cptcoveff && cptcoveff>1)
7620: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7621: }
7622: strcpy(gplotcondition+strlen(gplotcondition),")");
7623: /* 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 *\/ */
7624: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7625: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7626: /* '' 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*/
7627: if(i==nlstate+1){
1.270 brouard 7628: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7629: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7630: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7631: fprintf(ficgp," u %d:(",iagec);
7632: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7633: iyearc, iagec, offyear, \
7634: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7635: /* '' 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 7636: }else{
7637: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7638: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7639: }
7640: } /* end if covariate */
7641: } /* nlstate */
1.264 brouard 7642: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7643: } /* end cpt state*/
7644: } /* end covariate */
7645: } /* End if prevfcast */
1.227 brouard 7646:
1.268 brouard 7647: if(backcast==1){
7648: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7649:
7650: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7651: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7652: if(m != 1 && TKresult[nres]!= k1)
7653: continue;
7654: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7655: strcpy(gplotlabel,"(");
7656: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7657: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7658: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7659: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7660: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7661: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7662: vlv= nbcode[Tvaraff[k]][lv];
7663: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7664: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7665: }
7666: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7667: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7668: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7669: }
7670: strcpy(gplotlabel+strlen(gplotlabel),")");
7671: fprintf(ficgp,"\n#\n");
7672: if(invalidvarcomb[k1]){
7673: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7674: continue;
7675: }
7676:
7677: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7678: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7679: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7680: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7681: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7682:
7683: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7684: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7685: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7686: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7687: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7688: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7689: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7690: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7691: if(i==istart){
7692: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7693: }else{
7694: fprintf(ficgp,",\\\n '' ");
7695: }
7696: if(cptcoveff ==0){ /* No covariate */
7697: ioffset=2; /* Age is in 2 */
7698: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7699: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7700: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7701: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7702: fprintf(ficgp," u %d:(", ioffset);
7703: if(i==nlstate+1){
1.270 brouard 7704: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7705: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7706: fprintf(ficgp,",\\\n '' ");
7707: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7708: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7709: offbyear, \
7710: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7711: }else
7712: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7713: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7714: }else{ /* more than 2 covariates */
1.270 brouard 7715: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7716: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7717: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7718: iyearc=ioffset-1;
7719: iagec=ioffset;
1.268 brouard 7720: fprintf(ficgp," u %d:(",ioffset);
7721: kl=0;
7722: strcpy(gplotcondition,"(");
7723: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7724: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7725: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7726: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7727: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7728: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7729: kl++;
7730: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7731: kl++;
7732: if(k <cptcoveff && cptcoveff>1)
7733: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7734: }
7735: strcpy(gplotcondition+strlen(gplotcondition),")");
7736: /* 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 *\/ */
7737: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7738: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7739: /* '' 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*/
7740: if(i==nlstate+1){
1.270 brouard 7741: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7742: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7743: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7744: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7745: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7746: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7747: iyearc,iagec,offbyear, \
7748: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7749: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7750: }else{
7751: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7752: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7753: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7754: }
7755: } /* end if covariate */
7756: } /* nlstate */
7757: fprintf(ficgp,"\nset out; unset label;\n");
7758: } /* end cpt state*/
7759: } /* end covariate */
7760: } /* End if backcast */
7761:
1.227 brouard 7762:
1.238 brouard 7763: /* 9eme writing MLE parameters */
7764: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7765: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7766: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7767: for(k=1; k <=(nlstate+ndeath); k++){
7768: if (k != i) {
1.227 brouard 7769: fprintf(ficgp,"# current state %d\n",k);
7770: for(j=1; j <=ncovmodel; j++){
7771: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7772: jk++;
7773: }
7774: fprintf(ficgp,"\n");
1.126 brouard 7775: }
7776: }
1.223 brouard 7777: }
1.187 brouard 7778: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7779:
1.145 brouard 7780: /*goto avoid;*/
1.238 brouard 7781: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7782: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7783: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7784: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7785: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7786: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7787: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7788: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7789: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7790: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7791: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7792: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7793: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7794: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7795: fprintf(ficgp,"#\n");
1.223 brouard 7796: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7797: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7798: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7799: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7800: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7801: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7802: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7803: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7804: continue;
1.264 brouard 7805: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7806: strcpy(gplotlabel,"(");
1.276 brouard 7807: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7808: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7809: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7810: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7811: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7812: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7813: vlv= nbcode[Tvaraff[k]][lv];
7814: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7815: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7816: }
1.237 brouard 7817: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7818: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7819: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7820: }
1.264 brouard 7821: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7822: fprintf(ficgp,"\n#\n");
1.264 brouard 7823: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7824: fprintf(ficgp,"\nset key outside ");
7825: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7826: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7827: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7828: if (ng==1){
7829: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7830: fprintf(ficgp,"\nunset log y");
7831: }else if (ng==2){
7832: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7833: fprintf(ficgp,"\nset log y");
7834: }else if (ng==3){
7835: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7836: fprintf(ficgp,"\nset log y");
7837: }else
7838: fprintf(ficgp,"\nunset title ");
7839: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7840: i=1;
7841: for(k2=1; k2<=nlstate; k2++) {
7842: k3=i;
7843: for(k=1; k<=(nlstate+ndeath); k++) {
7844: if (k != k2){
7845: switch( ng) {
7846: case 1:
7847: if(nagesqr==0)
7848: fprintf(ficgp," p%d+p%d*x",i,i+1);
7849: else /* nagesqr =1 */
7850: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7851: break;
7852: case 2: /* ng=2 */
7853: if(nagesqr==0)
7854: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7855: else /* nagesqr =1 */
7856: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7857: break;
7858: case 3:
7859: if(nagesqr==0)
7860: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7861: else /* nagesqr =1 */
7862: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7863: break;
7864: }
7865: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7866: ijp=1; /* product no age */
7867: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7868: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7869: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7870: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7871: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7872: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7873: if(DummyV[j]==0){
7874: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7875: }else{ /* quantitative */
7876: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7877: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7878: }
7879: ij++;
1.237 brouard 7880: }
1.268 brouard 7881: }
7882: }else if(cptcovprod >0){
7883: if(j==Tprod[ijp]) { /* */
7884: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7885: if(ijp <=cptcovprod) { /* Product */
7886: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7887: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7888: /* 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)]); */
7889: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7890: }else{ /* Vn is dummy and Vm is quanti */
7891: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7892: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7893: }
7894: }else{ /* Vn*Vm Vn is quanti */
7895: if(DummyV[Tvard[ijp][2]]==0){
7896: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7897: }else{ /* Both quanti */
7898: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7899: }
1.237 brouard 7900: }
1.268 brouard 7901: ijp++;
1.237 brouard 7902: }
1.268 brouard 7903: } /* end Tprod */
1.237 brouard 7904: } else{ /* simple covariate */
1.264 brouard 7905: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7906: if(Dummy[j]==0){
7907: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7908: }else{ /* quantitative */
7909: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7910: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7911: }
1.237 brouard 7912: } /* end simple */
7913: } /* end j */
1.223 brouard 7914: }else{
7915: i=i-ncovmodel;
7916: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7917: fprintf(ficgp," (1.");
7918: }
1.227 brouard 7919:
1.223 brouard 7920: if(ng != 1){
7921: fprintf(ficgp,")/(1");
1.227 brouard 7922:
1.264 brouard 7923: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7924: if(nagesqr==0)
1.264 brouard 7925: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7926: else /* nagesqr =1 */
1.264 brouard 7927: 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 7928:
1.223 brouard 7929: ij=1;
7930: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7931: if(cptcovage >0){
7932: if((j-2)==Tage[ij]) { /* Bug valgrind */
7933: if(ij <=cptcovage) { /* Bug valgrind */
7934: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7935: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7936: ij++;
7937: }
7938: }
7939: }else
7940: 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 7941: }
7942: fprintf(ficgp,")");
7943: }
7944: fprintf(ficgp,")");
7945: if(ng ==2)
1.276 brouard 7946: 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 7947: else /* ng= 3 */
1.276 brouard 7948: 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 7949: }else{ /* end ng <> 1 */
7950: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 7951: 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 7952: }
7953: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7954: fprintf(ficgp,",");
7955: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7956: fprintf(ficgp,",");
7957: i=i+ncovmodel;
7958: } /* end k */
7959: } /* end k2 */
1.276 brouard 7960: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
7961: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 7962: } /* end k1 */
1.223 brouard 7963: } /* end ng */
7964: /* avoid: */
7965: fflush(ficgp);
1.126 brouard 7966: } /* end gnuplot */
7967:
7968:
7969: /*************** Moving average **************/
1.219 brouard 7970: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7971: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7972:
1.222 brouard 7973: int i, cpt, cptcod;
7974: int modcovmax =1;
7975: int mobilavrange, mob;
7976: int iage=0;
7977:
1.266 brouard 7978: double sum=0., sumr=0.;
1.222 brouard 7979: double age;
1.266 brouard 7980: double *sumnewp, *sumnewm, *sumnewmr;
7981: double *agemingood, *agemaxgood;
7982: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7983:
7984:
1.278 brouard 7985: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
7986: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 7987:
7988: sumnewp = vector(1,ncovcombmax);
7989: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7990: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7991: agemingood = vector(1,ncovcombmax);
1.266 brouard 7992: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7993: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7994: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7995:
7996: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7997: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7998: sumnewp[cptcod]=0.;
1.266 brouard 7999: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8000: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8001: }
8002: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8003:
1.266 brouard 8004: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8005: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8006: else mobilavrange=mobilav;
8007: for (age=bage; age<=fage; age++)
8008: for (i=1; i<=nlstate;i++)
8009: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8010: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8011: /* We keep the original values on the extreme ages bage, fage and for
8012: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8013: we use a 5 terms etc. until the borders are no more concerned.
8014: */
8015: for (mob=3;mob <=mobilavrange;mob=mob+2){
8016: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8017: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8018: sumnewm[cptcod]=0.;
8019: for (i=1; i<=nlstate;i++){
1.222 brouard 8020: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8021: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8022: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8023: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8024: }
8025: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8026: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8027: } /* end i */
8028: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8029: } /* end cptcod */
1.222 brouard 8030: }/* end age */
8031: }/* end mob */
1.266 brouard 8032: }else{
8033: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8034: return -1;
1.266 brouard 8035: }
8036:
8037: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8038: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8039: if(invalidvarcomb[cptcod]){
8040: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8041: continue;
8042: }
1.219 brouard 8043:
1.266 brouard 8044: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8045: sumnewm[cptcod]=0.;
8046: sumnewmr[cptcod]=0.;
8047: for (i=1; i<=nlstate;i++){
8048: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8049: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8050: }
8051: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8052: agemingoodr[cptcod]=age;
8053: }
8054: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8055: agemingood[cptcod]=age;
8056: }
8057: } /* age */
8058: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8059: sumnewm[cptcod]=0.;
1.266 brouard 8060: sumnewmr[cptcod]=0.;
1.222 brouard 8061: for (i=1; i<=nlstate;i++){
8062: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8063: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8064: }
8065: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8066: agemaxgoodr[cptcod]=age;
1.222 brouard 8067: }
8068: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8069: agemaxgood[cptcod]=age;
8070: }
8071: } /* age */
8072: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8073: /* but they will change */
8074: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8075: sumnewm[cptcod]=0.;
8076: sumnewmr[cptcod]=0.;
8077: for (i=1; i<=nlstate;i++){
8078: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8079: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8080: }
8081: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8082: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8083: agemaxgoodr[cptcod]=age; /* age min */
8084: for (i=1; i<=nlstate;i++)
8085: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
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)agemaxgoodr[cptcod]][i][cptcod];
8089: } /* i */
8090: } /* end bad */
8091: }else{
8092: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8093: agemaxgood[cptcod]=age;
8094: }else{ /* bad we change the value with the values of good ages */
8095: for (i=1; i<=nlstate;i++){
8096: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8097: } /* i */
8098: } /* end bad */
8099: }/* end else */
8100: sum=0.;sumr=0.;
8101: for (i=1; i<=nlstate;i++){
8102: sum+=mobaverage[(int)age][i][cptcod];
8103: sumr+=probs[(int)age][i][cptcod];
8104: }
8105: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8106: 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 8107: } /* end bad */
8108: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8109: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8110: 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 8111: } /* end bad */
8112: }/* age */
1.266 brouard 8113:
8114: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8115: sumnewm[cptcod]=0.;
1.266 brouard 8116: sumnewmr[cptcod]=0.;
1.222 brouard 8117: for (i=1; i<=nlstate;i++){
8118: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8119: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8120: }
8121: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8122: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8123: agemingoodr[cptcod]=age;
8124: for (i=1; i<=nlstate;i++)
8125: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8126: }else{ /* bad we change the value with the values of good ages */
8127: for (i=1; i<=nlstate;i++){
8128: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8129: } /* i */
8130: } /* end bad */
8131: }else{
8132: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8133: agemingood[cptcod]=age;
8134: }else{ /* bad */
8135: for (i=1; i<=nlstate;i++){
8136: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8137: } /* i */
8138: } /* end bad */
8139: }/* end else */
8140: sum=0.;sumr=0.;
8141: for (i=1; i<=nlstate;i++){
8142: sum+=mobaverage[(int)age][i][cptcod];
8143: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8144: }
1.266 brouard 8145: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8146: 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 8147: } /* end bad */
8148: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8149: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8150: 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 8151: } /* end bad */
8152: }/* age */
1.266 brouard 8153:
1.222 brouard 8154:
8155: for (age=bage; age<=fage; age++){
1.235 brouard 8156: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8157: sumnewp[cptcod]=0.;
8158: sumnewm[cptcod]=0.;
8159: for (i=1; i<=nlstate;i++){
8160: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8161: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8162: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8163: }
8164: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8165: }
8166: /* printf("\n"); */
8167: /* } */
1.266 brouard 8168:
1.222 brouard 8169: /* brutal averaging */
1.266 brouard 8170: /* for (i=1; i<=nlstate;i++){ */
8171: /* for (age=1; age<=bage; age++){ */
8172: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8173: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8174: /* } */
8175: /* for (age=fage; age<=AGESUP; age++){ */
8176: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8177: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8178: /* } */
8179: /* } /\* end i status *\/ */
8180: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8181: /* for (age=1; age<=AGESUP; age++){ */
8182: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8183: /* mobaverage[(int)age][i][cptcod]=0.; */
8184: /* } */
8185: /* } */
1.222 brouard 8186: }/* end cptcod */
1.266 brouard 8187: free_vector(agemaxgoodr,1, ncovcombmax);
8188: free_vector(agemaxgood,1, ncovcombmax);
8189: free_vector(agemingood,1, ncovcombmax);
8190: free_vector(agemingoodr,1, ncovcombmax);
8191: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8192: free_vector(sumnewm,1, ncovcombmax);
8193: free_vector(sumnewp,1, ncovcombmax);
8194: return 0;
8195: }/* End movingaverage */
1.218 brouard 8196:
1.126 brouard 8197:
8198: /************** Forecasting ******************/
1.269 brouard 8199: 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 8200: /* proj1, year, month, day of starting projection
8201: agemin, agemax range of age
8202: dateprev1 dateprev2 range of dates during which prevalence is computed
8203: anproj2 year of en of projection (same day and month as proj1).
8204: */
1.267 brouard 8205: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8206: double agec; /* generic age */
8207: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8208: double *popeffectif,*popcount;
8209: double ***p3mat;
1.218 brouard 8210: /* double ***mobaverage; */
1.126 brouard 8211: char fileresf[FILENAMELENGTH];
8212:
8213: agelim=AGESUP;
1.211 brouard 8214: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8215: in each health status at the date of interview (if between dateprev1 and dateprev2).
8216: We still use firstpass and lastpass as another selection.
8217: */
1.214 brouard 8218: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8219: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8220:
1.201 brouard 8221: strcpy(fileresf,"F_");
8222: strcat(fileresf,fileresu);
1.126 brouard 8223: if((ficresf=fopen(fileresf,"w"))==NULL) {
8224: printf("Problem with forecast resultfile: %s\n", fileresf);
8225: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8226: }
1.235 brouard 8227: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8228: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8229:
1.225 brouard 8230: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8231:
8232:
8233: stepsize=(int) (stepm+YEARM-1)/YEARM;
8234: if (stepm<=12) stepsize=1;
8235: if(estepm < stepm){
8236: printf ("Problem %d lower than %d\n",estepm, stepm);
8237: }
1.270 brouard 8238: else{
8239: hstepm=estepm;
8240: }
8241: if(estepm > stepm){ /* Yes every two year */
8242: stepsize=2;
8243: }
1.126 brouard 8244:
8245: hstepm=hstepm/stepm;
8246: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8247: fractional in yp1 */
8248: anprojmean=yp;
8249: yp2=modf((yp1*12),&yp);
8250: mprojmean=yp;
8251: yp1=modf((yp2*30.5),&yp);
8252: jprojmean=yp;
8253: if(jprojmean==0) jprojmean=1;
8254: if(mprojmean==0) jprojmean=1;
8255:
1.227 brouard 8256: i1=pow(2,cptcoveff);
1.126 brouard 8257: if (cptcovn < 1){i1=1;}
8258:
8259: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8260:
8261: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8262:
1.126 brouard 8263: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8264: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8265: for(k=1; k<=i1;k++){
1.253 brouard 8266: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8267: continue;
1.227 brouard 8268: if(invalidvarcomb[k]){
8269: printf("\nCombination (%d) projection ignored because no cases \n",k);
8270: continue;
8271: }
8272: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8273: for(j=1;j<=cptcoveff;j++) {
8274: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8275: }
1.235 brouard 8276: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8277: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8278: }
1.227 brouard 8279: fprintf(ficresf," yearproj age");
8280: for(j=1; j<=nlstate+ndeath;j++){
8281: for(i=1; i<=nlstate;i++)
8282: fprintf(ficresf," p%d%d",i,j);
8283: fprintf(ficresf," wp.%d",j);
8284: }
8285: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8286: fprintf(ficresf,"\n");
8287: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8288: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8289: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8290: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8291: nhstepm = nhstepm/hstepm;
8292: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8293: oldm=oldms;savm=savms;
1.268 brouard 8294: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8295: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8296: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8297: for (h=0; h<=nhstepm; h++){
8298: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8299: break;
8300: }
8301: }
8302: fprintf(ficresf,"\n");
8303: for(j=1;j<=cptcoveff;j++)
8304: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8305: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8306:
8307: for(j=1; j<=nlstate+ndeath;j++) {
8308: ppij=0.;
8309: for(i=1; i<=nlstate;i++) {
1.278 brouard 8310: if (mobilav>=1)
8311: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8312: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8313: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8314: }
1.268 brouard 8315: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8316: } /* end i */
8317: fprintf(ficresf," %.3f", ppij);
8318: }/* end j */
1.227 brouard 8319: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8320: } /* end agec */
1.266 brouard 8321: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8322: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8323: } /* end yearp */
8324: } /* end k */
1.219 brouard 8325:
1.126 brouard 8326: fclose(ficresf);
1.215 brouard 8327: printf("End of Computing forecasting \n");
8328: fprintf(ficlog,"End of Computing forecasting\n");
8329:
1.126 brouard 8330: }
8331:
1.269 brouard 8332: /************** Back Forecasting ******************/
8333: 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 8334: /* back1, year, month, day of starting backection
8335: agemin, agemax range of age
8336: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8337: anback2 year of end of backprojection (same day and month as back1).
8338: prevacurrent and prev are prevalences.
1.267 brouard 8339: */
8340: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8341: double agec; /* generic age */
1.268 brouard 8342: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8343: double *popeffectif,*popcount;
8344: double ***p3mat;
8345: /* double ***mobaverage; */
8346: char fileresfb[FILENAMELENGTH];
8347:
1.268 brouard 8348: agelim=AGEINF;
1.267 brouard 8349: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8350: in each health status at the date of interview (if between dateprev1 and dateprev2).
8351: We still use firstpass and lastpass as another selection.
8352: */
8353: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8354: /* firstpass, lastpass, stepm, weightopt, model); */
8355:
8356: /*Do we need to compute prevalence again?*/
8357:
8358: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8359:
8360: strcpy(fileresfb,"FB_");
8361: strcat(fileresfb,fileresu);
8362: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8363: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8364: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8365: }
8366: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8367: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8368:
8369: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8370:
8371:
8372: stepsize=(int) (stepm+YEARM-1)/YEARM;
8373: if (stepm<=12) stepsize=1;
8374: if(estepm < stepm){
8375: printf ("Problem %d lower than %d\n",estepm, stepm);
8376: }
1.270 brouard 8377: else{
8378: hstepm=estepm;
8379: }
8380: if(estepm >= stepm){ /* Yes every two year */
8381: stepsize=2;
8382: }
1.267 brouard 8383:
8384: hstepm=hstepm/stepm;
8385: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8386: fractional in yp1 */
8387: anprojmean=yp;
8388: yp2=modf((yp1*12),&yp);
8389: mprojmean=yp;
8390: yp1=modf((yp2*30.5),&yp);
8391: jprojmean=yp;
8392: if(jprojmean==0) jprojmean=1;
8393: if(mprojmean==0) jprojmean=1;
8394:
8395: i1=pow(2,cptcoveff);
8396: if (cptcovn < 1){i1=1;}
8397:
8398: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8399: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8400:
8401: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8402:
8403: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8404: for(k=1; k<=i1;k++){
8405: if(i1 != 1 && TKresult[nres]!= k)
8406: continue;
8407: if(invalidvarcomb[k]){
8408: printf("\nCombination (%d) projection ignored because no cases \n",k);
8409: continue;
8410: }
1.268 brouard 8411: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8412: for(j=1;j<=cptcoveff;j++) {
8413: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8414: }
8415: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8416: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8417: }
8418: fprintf(ficresfb," yearbproj age");
8419: for(j=1; j<=nlstate+ndeath;j++){
8420: for(i=1; i<=nlstate;i++)
1.268 brouard 8421: fprintf(ficresfb," b%d%d",i,j);
8422: fprintf(ficresfb," b.%d",j);
1.267 brouard 8423: }
8424: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8425: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8426: fprintf(ficresfb,"\n");
8427: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8428: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8429: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8430: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8431: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8432: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8433: nhstepm = nhstepm/hstepm;
8434: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8435: oldm=oldms;savm=savms;
1.268 brouard 8436: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8437: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8438: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8439: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8440: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8441: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8442: for (h=0; h<=nhstepm; h++){
1.268 brouard 8443: if (h*hstepm/YEARM*stepm ==-yearp) {
8444: break;
8445: }
8446: }
8447: fprintf(ficresfb,"\n");
8448: for(j=1;j<=cptcoveff;j++)
8449: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8450: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8451: for(i=1; i<=nlstate+ndeath;i++) {
8452: ppij=0.;ppi=0.;
8453: for(j=1; j<=nlstate;j++) {
8454: /* if (mobilav==1) */
1.269 brouard 8455: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8456: ppi=ppi+prevacurrent[(int)agec][j][k];
8457: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8458: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8459: /* else { */
8460: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8461: /* } */
1.268 brouard 8462: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8463: } /* end j */
8464: if(ppi <0.99){
8465: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8466: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8467: }
8468: fprintf(ficresfb," %.3f", ppij);
8469: }/* end j */
1.267 brouard 8470: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8471: } /* end agec */
8472: } /* end yearp */
8473: } /* end k */
1.217 brouard 8474:
1.267 brouard 8475: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8476:
1.267 brouard 8477: fclose(ficresfb);
8478: printf("End of Computing Back forecasting \n");
8479: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8480:
1.267 brouard 8481: }
1.217 brouard 8482:
1.269 brouard 8483: /* Variance of prevalence limit: varprlim */
8484: 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){
8485: /*------- Variance of period (stable) prevalence------*/
8486:
8487: char fileresvpl[FILENAMELENGTH];
8488: FILE *ficresvpl;
8489: double **oldm, **savm;
8490: double **varpl; /* Variances of prevalence limits by age */
8491: int i1, k, nres, j ;
8492:
8493: strcpy(fileresvpl,"VPL_");
8494: strcat(fileresvpl,fileresu);
8495: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8496: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8497: exit(0);
8498: }
8499: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8500: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8501:
8502: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8503: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8504:
8505: i1=pow(2,cptcoveff);
8506: if (cptcovn < 1){i1=1;}
8507:
8508: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8509: for(k=1; k<=i1;k++){
8510: if(i1 != 1 && TKresult[nres]!= k)
8511: continue;
8512: fprintf(ficresvpl,"\n#****** ");
8513: printf("\n#****** ");
8514: fprintf(ficlog,"\n#****** ");
8515: for(j=1;j<=cptcoveff;j++) {
8516: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8517: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8518: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8519: }
8520: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8521: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8522: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8523: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8524: }
8525: fprintf(ficresvpl,"******\n");
8526: printf("******\n");
8527: fprintf(ficlog,"******\n");
8528:
8529: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8530: oldm=oldms;savm=savms;
8531: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8532: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8533: /*}*/
8534: }
8535:
8536: fclose(ficresvpl);
8537: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8538: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8539:
8540: }
8541: /* Variance of back prevalence: varbprlim */
8542: 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){
8543: /*------- Variance of back (stable) prevalence------*/
8544:
8545: char fileresvbl[FILENAMELENGTH];
8546: FILE *ficresvbl;
8547:
8548: double **oldm, **savm;
8549: double **varbpl; /* Variances of back prevalence limits by age */
8550: int i1, k, nres, j ;
8551:
8552: strcpy(fileresvbl,"VBL_");
8553: strcat(fileresvbl,fileresu);
8554: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8555: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8556: exit(0);
8557: }
8558: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8559: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8560:
8561:
8562: i1=pow(2,cptcoveff);
8563: if (cptcovn < 1){i1=1;}
8564:
8565: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8566: for(k=1; k<=i1;k++){
8567: if(i1 != 1 && TKresult[nres]!= k)
8568: continue;
8569: fprintf(ficresvbl,"\n#****** ");
8570: printf("\n#****** ");
8571: fprintf(ficlog,"\n#****** ");
8572: for(j=1;j<=cptcoveff;j++) {
8573: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8574: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8575: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8576: }
8577: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8578: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8579: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8580: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8581: }
8582: fprintf(ficresvbl,"******\n");
8583: printf("******\n");
8584: fprintf(ficlog,"******\n");
8585:
8586: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8587: oldm=oldms;savm=savms;
8588:
8589: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8590: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8591: /*}*/
8592: }
8593:
8594: fclose(ficresvbl);
8595: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8596: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8597:
8598: } /* End of varbprlim */
8599:
1.126 brouard 8600: /************** Forecasting *****not tested NB*************/
1.227 brouard 8601: /* 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 8602:
1.227 brouard 8603: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8604: /* int *popage; */
8605: /* double calagedatem, agelim, kk1, kk2; */
8606: /* double *popeffectif,*popcount; */
8607: /* double ***p3mat,***tabpop,***tabpopprev; */
8608: /* /\* double ***mobaverage; *\/ */
8609: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8610:
1.227 brouard 8611: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8612: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8613: /* agelim=AGESUP; */
8614: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8615:
1.227 brouard 8616: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8617:
8618:
1.227 brouard 8619: /* strcpy(filerespop,"POP_"); */
8620: /* strcat(filerespop,fileresu); */
8621: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8622: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8623: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8624: /* } */
8625: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8626: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8627:
1.227 brouard 8628: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8629:
1.227 brouard 8630: /* /\* if (mobilav!=0) { *\/ */
8631: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8632: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8633: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8634: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8635: /* /\* } *\/ */
8636: /* /\* } *\/ */
1.126 brouard 8637:
1.227 brouard 8638: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8639: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8640:
1.227 brouard 8641: /* agelim=AGESUP; */
1.126 brouard 8642:
1.227 brouard 8643: /* hstepm=1; */
8644: /* hstepm=hstepm/stepm; */
1.218 brouard 8645:
1.227 brouard 8646: /* if (popforecast==1) { */
8647: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8648: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8649: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8650: /* } */
8651: /* popage=ivector(0,AGESUP); */
8652: /* popeffectif=vector(0,AGESUP); */
8653: /* popcount=vector(0,AGESUP); */
1.126 brouard 8654:
1.227 brouard 8655: /* i=1; */
8656: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8657:
1.227 brouard 8658: /* imx=i; */
8659: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8660: /* } */
1.218 brouard 8661:
1.227 brouard 8662: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8663: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8664: /* k=k+1; */
8665: /* fprintf(ficrespop,"\n#******"); */
8666: /* for(j=1;j<=cptcoveff;j++) { */
8667: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8668: /* } */
8669: /* fprintf(ficrespop,"******\n"); */
8670: /* fprintf(ficrespop,"# Age"); */
8671: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8672: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8673:
1.227 brouard 8674: /* for (cpt=0; cpt<=0;cpt++) { */
8675: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8676:
1.227 brouard 8677: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8678: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8679: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8680:
1.227 brouard 8681: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8682: /* oldm=oldms;savm=savms; */
8683: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8684:
1.227 brouard 8685: /* for (h=0; h<=nhstepm; h++){ */
8686: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8687: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8688: /* } */
8689: /* for(j=1; j<=nlstate+ndeath;j++) { */
8690: /* kk1=0.;kk2=0; */
8691: /* for(i=1; i<=nlstate;i++) { */
8692: /* if (mobilav==1) */
8693: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8694: /* else { */
8695: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8696: /* } */
8697: /* } */
8698: /* if (h==(int)(calagedatem+12*cpt)){ */
8699: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8700: /* /\*fprintf(ficrespop," %.3f", kk1); */
8701: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8702: /* } */
8703: /* } */
8704: /* for(i=1; i<=nlstate;i++){ */
8705: /* kk1=0.; */
8706: /* for(j=1; j<=nlstate;j++){ */
8707: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8708: /* } */
8709: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8710: /* } */
1.218 brouard 8711:
1.227 brouard 8712: /* if (h==(int)(calagedatem+12*cpt)) */
8713: /* for(j=1; j<=nlstate;j++) */
8714: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8715: /* } */
8716: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8717: /* } */
8718: /* } */
1.218 brouard 8719:
1.227 brouard 8720: /* /\******\/ */
1.218 brouard 8721:
1.227 brouard 8722: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8723: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8724: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8725: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8726: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8727:
1.227 brouard 8728: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8729: /* oldm=oldms;savm=savms; */
8730: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8731: /* for (h=0; h<=nhstepm; h++){ */
8732: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8733: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8734: /* } */
8735: /* for(j=1; j<=nlstate+ndeath;j++) { */
8736: /* kk1=0.;kk2=0; */
8737: /* for(i=1; i<=nlstate;i++) { */
8738: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8739: /* } */
8740: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8741: /* } */
8742: /* } */
8743: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8744: /* } */
8745: /* } */
8746: /* } */
8747: /* } */
1.218 brouard 8748:
1.227 brouard 8749: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8750:
1.227 brouard 8751: /* if (popforecast==1) { */
8752: /* free_ivector(popage,0,AGESUP); */
8753: /* free_vector(popeffectif,0,AGESUP); */
8754: /* free_vector(popcount,0,AGESUP); */
8755: /* } */
8756: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8757: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8758: /* fclose(ficrespop); */
8759: /* } /\* End of popforecast *\/ */
1.218 brouard 8760:
1.126 brouard 8761: int fileappend(FILE *fichier, char *optionfich)
8762: {
8763: if((fichier=fopen(optionfich,"a"))==NULL) {
8764: printf("Problem with file: %s\n", optionfich);
8765: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8766: return (0);
8767: }
8768: fflush(fichier);
8769: return (1);
8770: }
8771:
8772:
8773: /**************** function prwizard **********************/
8774: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8775: {
8776:
8777: /* Wizard to print covariance matrix template */
8778:
1.164 brouard 8779: char ca[32], cb[32];
8780: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8781: int numlinepar;
8782:
8783: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8784: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8785: for(i=1; i <=nlstate; i++){
8786: jj=0;
8787: for(j=1; j <=nlstate+ndeath; j++){
8788: if(j==i) continue;
8789: jj++;
8790: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8791: printf("%1d%1d",i,j);
8792: fprintf(ficparo,"%1d%1d",i,j);
8793: for(k=1; k<=ncovmodel;k++){
8794: /* printf(" %lf",param[i][j][k]); */
8795: /* fprintf(ficparo," %lf",param[i][j][k]); */
8796: printf(" 0.");
8797: fprintf(ficparo," 0.");
8798: }
8799: printf("\n");
8800: fprintf(ficparo,"\n");
8801: }
8802: }
8803: printf("# Scales (for hessian or gradient estimation)\n");
8804: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8805: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8806: for(i=1; i <=nlstate; i++){
8807: jj=0;
8808: for(j=1; j <=nlstate+ndeath; j++){
8809: if(j==i) continue;
8810: jj++;
8811: fprintf(ficparo,"%1d%1d",i,j);
8812: printf("%1d%1d",i,j);
8813: fflush(stdout);
8814: for(k=1; k<=ncovmodel;k++){
8815: /* printf(" %le",delti3[i][j][k]); */
8816: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8817: printf(" 0.");
8818: fprintf(ficparo," 0.");
8819: }
8820: numlinepar++;
8821: printf("\n");
8822: fprintf(ficparo,"\n");
8823: }
8824: }
8825: printf("# Covariance matrix\n");
8826: /* # 121 Var(a12)\n\ */
8827: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8828: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8829: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8830: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8831: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8832: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8833: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8834: fflush(stdout);
8835: fprintf(ficparo,"# Covariance matrix\n");
8836: /* # 121 Var(a12)\n\ */
8837: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8838: /* # ...\n\ */
8839: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8840:
8841: for(itimes=1;itimes<=2;itimes++){
8842: jj=0;
8843: for(i=1; i <=nlstate; i++){
8844: for(j=1; j <=nlstate+ndeath; j++){
8845: if(j==i) continue;
8846: for(k=1; k<=ncovmodel;k++){
8847: jj++;
8848: ca[0]= k+'a'-1;ca[1]='\0';
8849: if(itimes==1){
8850: printf("#%1d%1d%d",i,j,k);
8851: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8852: }else{
8853: printf("%1d%1d%d",i,j,k);
8854: fprintf(ficparo,"%1d%1d%d",i,j,k);
8855: /* printf(" %.5le",matcov[i][j]); */
8856: }
8857: ll=0;
8858: for(li=1;li <=nlstate; li++){
8859: for(lj=1;lj <=nlstate+ndeath; lj++){
8860: if(lj==li) continue;
8861: for(lk=1;lk<=ncovmodel;lk++){
8862: ll++;
8863: if(ll<=jj){
8864: cb[0]= lk +'a'-1;cb[1]='\0';
8865: if(ll<jj){
8866: if(itimes==1){
8867: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8868: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8869: }else{
8870: printf(" 0.");
8871: fprintf(ficparo," 0.");
8872: }
8873: }else{
8874: if(itimes==1){
8875: printf(" Var(%s%1d%1d)",ca,i,j);
8876: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8877: }else{
8878: printf(" 0.");
8879: fprintf(ficparo," 0.");
8880: }
8881: }
8882: }
8883: } /* end lk */
8884: } /* end lj */
8885: } /* end li */
8886: printf("\n");
8887: fprintf(ficparo,"\n");
8888: numlinepar++;
8889: } /* end k*/
8890: } /*end j */
8891: } /* end i */
8892: } /* end itimes */
8893:
8894: } /* end of prwizard */
8895: /******************* Gompertz Likelihood ******************************/
8896: double gompertz(double x[])
8897: {
8898: double A,B,L=0.0,sump=0.,num=0.;
8899: int i,n=0; /* n is the size of the sample */
8900:
1.220 brouard 8901: for (i=1;i<=imx ; i++) {
1.126 brouard 8902: sump=sump+weight[i];
8903: /* sump=sump+1;*/
8904: num=num+1;
8905: }
8906:
8907:
8908: /* for (i=0; i<=imx; i++)
8909: 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]);*/
8910:
8911: for (i=1;i<=imx ; i++)
8912: {
8913: if (cens[i] == 1 && wav[i]>1)
8914: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8915:
8916: if (cens[i] == 0 && wav[i]>1)
8917: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8918: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8919:
8920: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8921: if (wav[i] > 1 ) { /* ??? */
8922: L=L+A*weight[i];
8923: /* 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]);*/
8924: }
8925: }
8926:
8927: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8928:
8929: return -2*L*num/sump;
8930: }
8931:
1.136 brouard 8932: #ifdef GSL
8933: /******************* Gompertz_f Likelihood ******************************/
8934: double gompertz_f(const gsl_vector *v, void *params)
8935: {
8936: double A,B,LL=0.0,sump=0.,num=0.;
8937: double *x= (double *) v->data;
8938: int i,n=0; /* n is the size of the sample */
8939:
8940: for (i=0;i<=imx-1 ; i++) {
8941: sump=sump+weight[i];
8942: /* sump=sump+1;*/
8943: num=num+1;
8944: }
8945:
8946:
8947: /* for (i=0; i<=imx; i++)
8948: 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]);*/
8949: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8950: for (i=1;i<=imx ; i++)
8951: {
8952: if (cens[i] == 1 && wav[i]>1)
8953: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8954:
8955: if (cens[i] == 0 && wav[i]>1)
8956: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8957: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8958:
8959: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8960: if (wav[i] > 1 ) { /* ??? */
8961: LL=LL+A*weight[i];
8962: /* 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]);*/
8963: }
8964: }
8965:
8966: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8967: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8968:
8969: return -2*LL*num/sump;
8970: }
8971: #endif
8972:
1.126 brouard 8973: /******************* Printing html file ***********/
1.201 brouard 8974: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8975: int lastpass, int stepm, int weightopt, char model[],\
8976: int imx, double p[],double **matcov,double agemortsup){
8977: int i,k;
8978:
8979: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8980: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8981: for (i=1;i<=2;i++)
8982: 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 8983: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8984: fprintf(fichtm,"</ul>");
8985:
8986: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8987:
8988: 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>");
8989:
8990: for (k=agegomp;k<(agemortsup-2);k++)
8991: 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]);
8992:
8993:
8994: fflush(fichtm);
8995: }
8996:
8997: /******************* Gnuplot file **************/
1.201 brouard 8998: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8999:
9000: char dirfileres[132],optfileres[132];
1.164 brouard 9001:
1.126 brouard 9002: int ng;
9003:
9004:
9005: /*#ifdef windows */
9006: fprintf(ficgp,"cd \"%s\" \n",pathc);
9007: /*#endif */
9008:
9009:
9010: strcpy(dirfileres,optionfilefiname);
9011: strcpy(optfileres,"vpl");
1.199 brouard 9012: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9013: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9014: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9015: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9016: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9017:
9018: }
9019:
1.136 brouard 9020: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9021: {
1.126 brouard 9022:
1.136 brouard 9023: /*-------- data file ----------*/
9024: FILE *fic;
9025: char dummy[]=" ";
1.240 brouard 9026: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9027: int lstra;
1.136 brouard 9028: int linei, month, year,iout;
9029: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9030: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9031: char *stratrunc;
1.223 brouard 9032:
1.240 brouard 9033: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9034: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9035:
1.240 brouard 9036: for(v=1; v <=ncovcol;v++){
9037: DummyV[v]=0;
9038: FixedV[v]=0;
9039: }
9040: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9041: DummyV[v]=1;
9042: FixedV[v]=0;
9043: }
9044: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9045: DummyV[v]=0;
9046: FixedV[v]=1;
9047: }
9048: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9049: DummyV[v]=1;
9050: FixedV[v]=1;
9051: }
9052: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9053: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9054: 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]);
9055: }
1.126 brouard 9056:
1.136 brouard 9057: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9058: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9059: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9060: }
1.126 brouard 9061:
1.136 brouard 9062: i=1;
9063: linei=0;
9064: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9065: linei=linei+1;
9066: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9067: if(line[j] == '\t')
9068: line[j] = ' ';
9069: }
9070: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9071: ;
9072: };
9073: line[j+1]=0; /* Trims blanks at end of line */
9074: if(line[0]=='#'){
9075: fprintf(ficlog,"Comment line\n%s\n",line);
9076: printf("Comment line\n%s\n",line);
9077: continue;
9078: }
9079: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9080: strcpy(line, linetmp);
1.223 brouard 9081:
9082: /* Loops on waves */
9083: for (j=maxwav;j>=1;j--){
9084: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9085: cutv(stra, strb, line, ' ');
9086: if(strb[0]=='.') { /* Missing value */
9087: lval=-1;
9088: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9089: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9090: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9091: 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);
9092: 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);
9093: return 1;
9094: }
9095: }else{
9096: errno=0;
9097: /* what_kind_of_number(strb); */
9098: dval=strtod(strb,&endptr);
9099: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9100: /* if(strb != endptr && *endptr == '\0') */
9101: /* dval=dlval; */
9102: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9103: if( strb[0]=='\0' || (*endptr != '\0')){
9104: 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);
9105: 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);
9106: return 1;
9107: }
9108: cotqvar[j][iv][i]=dval;
9109: cotvar[j][ntv+iv][i]=dval;
9110: }
9111: strcpy(line,stra);
1.223 brouard 9112: }/* end loop ntqv */
1.225 brouard 9113:
1.223 brouard 9114: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9115: cutv(stra, strb, line, ' ');
9116: if(strb[0]=='.') { /* Missing value */
9117: lval=-1;
9118: }else{
9119: errno=0;
9120: lval=strtol(strb,&endptr,10);
9121: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9122: if( strb[0]=='\0' || (*endptr != '\0')){
9123: 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);
9124: 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);
9125: return 1;
9126: }
9127: }
9128: if(lval <-1 || lval >1){
9129: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9130: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9131: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9132: For example, for multinomial values like 1, 2 and 3,\n \
9133: build V1=0 V2=0 for the reference value (1),\n \
9134: V1=1 V2=0 for (2) \n \
1.223 brouard 9135: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9136: output of IMaCh is often meaningless.\n \
1.223 brouard 9137: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9138: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9139: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9140: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9141: For example, for multinomial values like 1, 2 and 3,\n \
9142: build V1=0 V2=0 for the reference value (1),\n \
9143: V1=1 V2=0 for (2) \n \
1.223 brouard 9144: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9145: output of IMaCh is often meaningless.\n \
1.223 brouard 9146: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9147: return 1;
9148: }
9149: cotvar[j][iv][i]=(double)(lval);
9150: strcpy(line,stra);
1.223 brouard 9151: }/* end loop ntv */
1.225 brouard 9152:
1.223 brouard 9153: /* Statuses at wave */
1.137 brouard 9154: cutv(stra, strb, line, ' ');
1.223 brouard 9155: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9156: lval=-1;
1.136 brouard 9157: }else{
1.238 brouard 9158: errno=0;
9159: lval=strtol(strb,&endptr,10);
9160: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9161: if( strb[0]=='\0' || (*endptr != '\0')){
9162: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a status of wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,j,maxwav);
9163: 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);
9164: return 1;
9165: }
1.136 brouard 9166: }
1.225 brouard 9167:
1.136 brouard 9168: s[j][i]=lval;
1.225 brouard 9169:
1.223 brouard 9170: /* Date of Interview */
1.136 brouard 9171: strcpy(line,stra);
9172: cutv(stra, strb,line,' ');
1.169 brouard 9173: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9174: }
1.169 brouard 9175: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9176: month=99;
9177: year=9999;
1.136 brouard 9178: }else{
1.225 brouard 9179: 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);
9180: 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);
9181: return 1;
1.136 brouard 9182: }
9183: anint[j][i]= (double) year;
9184: mint[j][i]= (double)month;
9185: strcpy(line,stra);
1.223 brouard 9186: } /* End loop on waves */
1.225 brouard 9187:
1.223 brouard 9188: /* Date of death */
1.136 brouard 9189: cutv(stra, strb,line,' ');
1.169 brouard 9190: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9191: }
1.169 brouard 9192: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9193: month=99;
9194: year=9999;
9195: }else{
1.141 brouard 9196: 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 9197: 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);
9198: return 1;
1.136 brouard 9199: }
9200: andc[i]=(double) year;
9201: moisdc[i]=(double) month;
9202: strcpy(line,stra);
9203:
1.223 brouard 9204: /* Date of birth */
1.136 brouard 9205: cutv(stra, strb,line,' ');
1.169 brouard 9206: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9207: }
1.169 brouard 9208: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9209: month=99;
9210: year=9999;
9211: }else{
1.141 brouard 9212: 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);
9213: 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 9214: return 1;
1.136 brouard 9215: }
9216: if (year==9999) {
1.141 brouard 9217: 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);
9218: 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 9219: return 1;
9220:
1.136 brouard 9221: }
9222: annais[i]=(double)(year);
9223: moisnais[i]=(double)(month);
9224: strcpy(line,stra);
1.225 brouard 9225:
1.223 brouard 9226: /* Sample weight */
1.136 brouard 9227: cutv(stra, strb,line,' ');
9228: errno=0;
9229: dval=strtod(strb,&endptr);
9230: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9231: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9232: 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 9233: fflush(ficlog);
9234: return 1;
9235: }
9236: weight[i]=dval;
9237: strcpy(line,stra);
1.225 brouard 9238:
1.223 brouard 9239: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9240: cutv(stra, strb, line, ' ');
9241: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9242: lval=-1;
1.223 brouard 9243: }else{
1.225 brouard 9244: errno=0;
9245: /* what_kind_of_number(strb); */
9246: dval=strtod(strb,&endptr);
9247: /* if(strb != endptr && *endptr == '\0') */
9248: /* dval=dlval; */
9249: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9250: if( strb[0]=='\0' || (*endptr != '\0')){
9251: 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);
9252: 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);
9253: return 1;
9254: }
9255: coqvar[iv][i]=dval;
1.226 brouard 9256: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9257: }
9258: strcpy(line,stra);
9259: }/* end loop nqv */
1.136 brouard 9260:
1.223 brouard 9261: /* Covariate values */
1.136 brouard 9262: for (j=ncovcol;j>=1;j--){
9263: cutv(stra, strb,line,' ');
1.223 brouard 9264: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9265: lval=-1;
1.136 brouard 9266: }else{
1.225 brouard 9267: errno=0;
9268: lval=strtol(strb,&endptr,10);
9269: if( strb[0]=='\0' || (*endptr != '\0')){
9270: 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);
9271: 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);
9272: return 1;
9273: }
1.136 brouard 9274: }
9275: if(lval <-1 || lval >1){
1.225 brouard 9276: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9277: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9278: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9279: For example, for multinomial values like 1, 2 and 3,\n \
9280: build V1=0 V2=0 for the reference value (1),\n \
9281: V1=1 V2=0 for (2) \n \
1.136 brouard 9282: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9283: output of IMaCh is often meaningless.\n \
1.136 brouard 9284: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9285: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9286: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9287: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9288: For example, for multinomial values like 1, 2 and 3,\n \
9289: build V1=0 V2=0 for the reference value (1),\n \
9290: V1=1 V2=0 for (2) \n \
1.136 brouard 9291: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9292: output of IMaCh is often meaningless.\n \
1.136 brouard 9293: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9294: return 1;
1.136 brouard 9295: }
9296: covar[j][i]=(double)(lval);
9297: strcpy(line,stra);
9298: }
9299: lstra=strlen(stra);
1.225 brouard 9300:
1.136 brouard 9301: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9302: stratrunc = &(stra[lstra-9]);
9303: num[i]=atol(stratrunc);
9304: }
9305: else
9306: num[i]=atol(stra);
9307: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9308: 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;}*/
9309:
9310: i=i+1;
9311: } /* End loop reading data */
1.225 brouard 9312:
1.136 brouard 9313: *imax=i-1; /* Number of individuals */
9314: fclose(fic);
1.225 brouard 9315:
1.136 brouard 9316: return (0);
1.164 brouard 9317: /* endread: */
1.225 brouard 9318: printf("Exiting readdata: ");
9319: fclose(fic);
9320: return (1);
1.223 brouard 9321: }
1.126 brouard 9322:
1.234 brouard 9323: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9324: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9325: while (*p2 == ' ')
1.234 brouard 9326: p2++;
9327: /* while ((*p1++ = *p2++) !=0) */
9328: /* ; */
9329: /* do */
9330: /* while (*p2 == ' ') */
9331: /* p2++; */
9332: /* while (*p1++ == *p2++); */
9333: *stri=p2;
1.145 brouard 9334: }
9335:
1.235 brouard 9336: int decoderesult ( char resultline[], int nres)
1.230 brouard 9337: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9338: {
1.235 brouard 9339: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9340: char resultsav[MAXLINE];
1.234 brouard 9341: int resultmodel[MAXLINE];
9342: int modelresult[MAXLINE];
1.230 brouard 9343: char stra[80], strb[80], strc[80], strd[80],stre[80];
9344:
1.234 brouard 9345: removefirstspace(&resultline);
1.233 brouard 9346: printf("decoderesult:%s\n",resultline);
1.230 brouard 9347:
9348: if (strstr(resultline,"v") !=0){
9349: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9350: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9351: return 1;
9352: }
9353: trimbb(resultsav, resultline);
9354: if (strlen(resultsav) >1){
9355: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9356: }
1.253 brouard 9357: if(j == 0){ /* Resultline but no = */
9358: TKresult[nres]=0; /* Combination for the nresult and the model */
9359: return (0);
9360: }
9361:
1.234 brouard 9362: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9363: 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);
9364: 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);
9365: }
9366: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9367: if(nbocc(resultsav,'=') >1){
9368: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9369: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9370: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9371: }else
9372: cutl(strc,strd,resultsav,'=');
1.230 brouard 9373: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9374:
1.230 brouard 9375: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9376: Tvarsel[k]=atoi(strc);
9377: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9378: /* cptcovsel++; */
9379: if (nbocc(stra,'=') >0)
9380: strcpy(resultsav,stra); /* and analyzes it */
9381: }
1.235 brouard 9382: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9383: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9384: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9385: match=0;
1.236 brouard 9386: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9387: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9388: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9389: match=1;
9390: break;
9391: }
9392: }
9393: if(match == 0){
9394: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9395: }
9396: }
9397: }
1.235 brouard 9398: /* Checking for missing or useless values in comparison of current model needs */
9399: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9400: match=0;
1.235 brouard 9401: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9402: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9403: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9404: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9405: ++match;
9406: }
9407: }
9408: }
9409: if(match == 0){
9410: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9411: }else if(match > 1){
9412: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9413: }
9414: }
1.235 brouard 9415:
1.234 brouard 9416: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9417: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9418: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9419: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9420: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9421: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9422: /* 1 0 0 0 */
9423: /* 2 1 0 0 */
9424: /* 3 0 1 0 */
9425: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9426: /* 5 0 0 1 */
9427: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9428: /* 7 0 1 1 */
9429: /* 8 1 1 1 */
1.237 brouard 9430: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9431: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9432: /* V5*age V5 known which value for nres? */
9433: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9434: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9435: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9436: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9437: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9438: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9439: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9440: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9441: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9442: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9443: k4++;;
9444: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9445: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9446: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9447: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9448: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9449: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9450: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9451: k4q++;;
9452: }
9453: }
1.234 brouard 9454:
1.235 brouard 9455: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9456: return (0);
9457: }
1.235 brouard 9458:
1.230 brouard 9459: int decodemodel( char model[], int lastobs)
9460: /**< This routine decodes the model and returns:
1.224 brouard 9461: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9462: * - nagesqr = 1 if age*age in the model, otherwise 0.
9463: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9464: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9465: * - cptcovage number of covariates with age*products =2
9466: * - cptcovs number of simple covariates
9467: * - 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
9468: * which is a new column after the 9 (ncovcol) variables.
9469: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9470: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9471: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9472: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9473: */
1.136 brouard 9474: {
1.238 brouard 9475: int i, j, k, ks, v;
1.227 brouard 9476: int j1, k1, k2, k3, k4;
1.136 brouard 9477: char modelsav[80];
1.145 brouard 9478: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9479: char *strpt;
1.136 brouard 9480:
1.145 brouard 9481: /*removespace(model);*/
1.136 brouard 9482: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9483: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9484: if (strstr(model,"AGE") !=0){
1.192 brouard 9485: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9486: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9487: return 1;
9488: }
1.141 brouard 9489: if (strstr(model,"v") !=0){
9490: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9491: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9492: return 1;
9493: }
1.187 brouard 9494: strcpy(modelsav,model);
9495: if ((strpt=strstr(model,"age*age")) !=0){
9496: printf(" strpt=%s, model=%s\n",strpt, model);
9497: if(strpt != model){
1.234 brouard 9498: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9499: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9500: corresponding column of parameters.\n",model);
1.234 brouard 9501: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9502: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9503: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9504: return 1;
1.225 brouard 9505: }
1.187 brouard 9506: nagesqr=1;
9507: if (strstr(model,"+age*age") !=0)
1.234 brouard 9508: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9509: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9510: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9511: else
1.234 brouard 9512: substrchaine(modelsav, model, "age*age");
1.187 brouard 9513: }else
9514: nagesqr=0;
9515: if (strlen(modelsav) >1){
9516: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9517: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9518: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9519: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9520: * cst, age and age*age
9521: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9522: /* including age products which are counted in cptcovage.
9523: * but the covariates which are products must be treated
9524: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9525: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9526: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9527:
9528:
1.187 brouard 9529: /* Design
9530: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9531: * < ncovcol=8 >
9532: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9533: * k= 1 2 3 4 5 6 7 8
9534: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9535: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9536: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9537: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9538: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9539: * Tage[++cptcovage]=k
9540: * if products, new covar are created after ncovcol with k1
9541: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9542: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9543: * 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
9544: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9545: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9546: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9547: * < ncovcol=8 >
9548: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9549: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9550: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9551: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9552: * p Tprod[1]@2={ 6, 5}
9553: *p Tvard[1][1]@4= {7, 8, 5, 6}
9554: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9555: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9556: *How to reorganize?
9557: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9558: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9559: * {2, 1, 4, 8, 5, 6, 3, 7}
9560: * Struct []
9561: */
1.225 brouard 9562:
1.187 brouard 9563: /* This loop fills the array Tvar from the string 'model'.*/
9564: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9565: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9566: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9567: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9568: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9569: /* k=1 Tvar[1]=2 (from V2) */
9570: /* k=5 Tvar[5] */
9571: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9572: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9573: /* } */
1.198 brouard 9574: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9575: /*
9576: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9577: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9578: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9579: }
1.187 brouard 9580: cptcovage=0;
9581: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9582: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9583: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9584: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9585: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9586: /*scanf("%d",i);*/
9587: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9588: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9589: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9590: /* covar is not filled and then is empty */
9591: cptcovprod--;
9592: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9593: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9594: Typevar[k]=1; /* 1 for age product */
9595: cptcovage++; /* Sums the number of covariates which include age as a product */
9596: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9597: /*printf("stre=%s ", stre);*/
9598: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9599: cptcovprod--;
9600: cutl(stre,strb,strc,'V');
9601: Tvar[k]=atoi(stre);
9602: Typevar[k]=1; /* 1 for age product */
9603: cptcovage++;
9604: Tage[cptcovage]=k;
9605: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9606: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9607: cptcovn++;
9608: cptcovprodnoage++;k1++;
9609: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9610: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9611: because this model-covariate is a construction we invent a new column
9612: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9613: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9614: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9615: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9616: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9617: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9618: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9619: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9620: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9621: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9622: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9623: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9624: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9625: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9626: for (i=1; i<=lastobs;i++){
9627: /* Computes the new covariate which is a product of
9628: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9629: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9630: }
9631: } /* End age is not in the model */
9632: } /* End if model includes a product */
9633: else { /* no more sum */
9634: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9635: /* scanf("%d",i);*/
9636: cutl(strd,strc,strb,'V');
9637: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9638: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9639: Tvar[k]=atoi(strd);
9640: Typevar[k]=0; /* 0 for simple covariates */
9641: }
9642: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9643: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9644: scanf("%d",i);*/
1.187 brouard 9645: } /* end of loop + on total covariates */
9646: } /* end if strlen(modelsave == 0) age*age might exist */
9647: } /* end if strlen(model == 0) */
1.136 brouard 9648:
9649: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9650: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9651:
1.136 brouard 9652: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9653: printf("cptcovprod=%d ", cptcovprod);
9654: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9655: scanf("%d ",i);*/
9656:
9657:
1.230 brouard 9658: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9659: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9660: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9661: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9662: k = 1 2 3 4 5 6 7 8 9
9663: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9664: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9665: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9666: Dummy[k] 1 0 0 0 3 1 1 2 3
9667: Tmodelind[combination of covar]=k;
1.225 brouard 9668: */
9669: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9670: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9671: /* 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 9672: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9673: printf("Model=%s\n\
9674: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9675: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9676: 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);
9677: fprintf(ficlog,"Model=%s\n\
9678: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9679: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9680: 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 9681: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9682: 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 */
9683: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9684: Fixed[k]= 0;
9685: Dummy[k]= 0;
1.225 brouard 9686: ncoveff++;
1.232 brouard 9687: ncovf++;
1.234 brouard 9688: nsd++;
9689: modell[k].maintype= FTYPE;
9690: TvarsD[nsd]=Tvar[k];
9691: TvarsDind[nsd]=k;
9692: TvarF[ncovf]=Tvar[k];
9693: TvarFind[ncovf]=k;
9694: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9695: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9696: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9697: Fixed[k]= 0;
9698: Dummy[k]= 0;
9699: ncoveff++;
9700: ncovf++;
9701: modell[k].maintype= FTYPE;
9702: TvarF[ncovf]=Tvar[k];
9703: TvarFind[ncovf]=k;
1.230 brouard 9704: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9705: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9706: }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 9707: Fixed[k]= 0;
9708: Dummy[k]= 1;
1.230 brouard 9709: nqfveff++;
1.234 brouard 9710: modell[k].maintype= FTYPE;
9711: modell[k].subtype= FQ;
9712: nsq++;
9713: TvarsQ[nsq]=Tvar[k];
9714: TvarsQind[nsq]=k;
1.232 brouard 9715: ncovf++;
1.234 brouard 9716: TvarF[ncovf]=Tvar[k];
9717: TvarFind[ncovf]=k;
1.231 brouard 9718: 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 9719: 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 9720: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9721: Fixed[k]= 1;
9722: Dummy[k]= 0;
1.225 brouard 9723: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9724: modell[k].maintype= VTYPE;
9725: modell[k].subtype= VD;
9726: nsd++;
9727: TvarsD[nsd]=Tvar[k];
9728: TvarsDind[nsd]=k;
9729: ncovv++; /* Only simple time varying variables */
9730: TvarV[ncovv]=Tvar[k];
1.242 brouard 9731: 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 9732: 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 */
9733: 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 9734: 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);
9735: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9736: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9737: Fixed[k]= 1;
9738: Dummy[k]= 1;
9739: nqtveff++;
9740: modell[k].maintype= VTYPE;
9741: modell[k].subtype= VQ;
9742: ncovv++; /* Only simple time varying variables */
9743: nsq++;
9744: TvarsQ[nsq]=Tvar[k];
9745: TvarsQind[nsq]=k;
9746: TvarV[ncovv]=Tvar[k];
1.242 brouard 9747: 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 9748: 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 */
9749: 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 9750: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9751: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9752: 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 9753: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9754: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9755: ncova++;
9756: TvarA[ncova]=Tvar[k];
9757: TvarAind[ncova]=k;
1.231 brouard 9758: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9759: Fixed[k]= 2;
9760: Dummy[k]= 2;
9761: modell[k].maintype= ATYPE;
9762: modell[k].subtype= APFD;
9763: /* ncoveff++; */
1.227 brouard 9764: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9765: Fixed[k]= 2;
9766: Dummy[k]= 3;
9767: modell[k].maintype= ATYPE;
9768: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9769: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9770: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9771: Fixed[k]= 3;
9772: Dummy[k]= 2;
9773: modell[k].maintype= ATYPE;
9774: modell[k].subtype= APVD; /* Product age * varying dummy */
9775: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9776: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9777: Fixed[k]= 3;
9778: Dummy[k]= 3;
9779: modell[k].maintype= ATYPE;
9780: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9781: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9782: }
9783: }else if (Typevar[k] == 2) { /* product without age */
9784: k1=Tposprod[k];
9785: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9786: if(Tvard[k1][2] <=ncovcol){
9787: Fixed[k]= 1;
9788: Dummy[k]= 0;
9789: modell[k].maintype= FTYPE;
9790: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9791: ncovf++; /* Fixed variables without age */
9792: TvarF[ncovf]=Tvar[k];
9793: TvarFind[ncovf]=k;
9794: }else if(Tvard[k1][2] <=ncovcol+nqv){
9795: Fixed[k]= 0; /* or 2 ?*/
9796: Dummy[k]= 1;
9797: modell[k].maintype= FTYPE;
9798: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9799: ncovf++; /* Varying variables without age */
9800: TvarF[ncovf]=Tvar[k];
9801: TvarFind[ncovf]=k;
9802: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9803: Fixed[k]= 1;
9804: Dummy[k]= 0;
9805: modell[k].maintype= VTYPE;
9806: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9807: ncovv++; /* Varying variables without age */
9808: TvarV[ncovv]=Tvar[k];
9809: TvarVind[ncovv]=k;
9810: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9811: Fixed[k]= 1;
9812: Dummy[k]= 1;
9813: modell[k].maintype= VTYPE;
9814: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9815: ncovv++; /* Varying variables without age */
9816: TvarV[ncovv]=Tvar[k];
9817: TvarVind[ncovv]=k;
9818: }
1.227 brouard 9819: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9820: if(Tvard[k1][2] <=ncovcol){
9821: Fixed[k]= 0; /* or 2 ?*/
9822: Dummy[k]= 1;
9823: modell[k].maintype= FTYPE;
9824: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9825: ncovf++; /* Fixed variables without age */
9826: TvarF[ncovf]=Tvar[k];
9827: TvarFind[ncovf]=k;
9828: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9829: Fixed[k]= 1;
9830: Dummy[k]= 1;
9831: modell[k].maintype= VTYPE;
9832: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9833: ncovv++; /* Varying variables without age */
9834: TvarV[ncovv]=Tvar[k];
9835: TvarVind[ncovv]=k;
9836: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9837: Fixed[k]= 1;
9838: Dummy[k]= 1;
9839: modell[k].maintype= VTYPE;
9840: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9841: ncovv++; /* Varying variables without age */
9842: TvarV[ncovv]=Tvar[k];
9843: TvarVind[ncovv]=k;
9844: ncovv++; /* Varying variables without age */
9845: TvarV[ncovv]=Tvar[k];
9846: TvarVind[ncovv]=k;
9847: }
1.227 brouard 9848: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9849: if(Tvard[k1][2] <=ncovcol){
9850: Fixed[k]= 1;
9851: Dummy[k]= 1;
9852: modell[k].maintype= VTYPE;
9853: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9854: ncovv++; /* Varying variables without age */
9855: TvarV[ncovv]=Tvar[k];
9856: TvarVind[ncovv]=k;
9857: }else if(Tvard[k1][2] <=ncovcol+nqv){
9858: Fixed[k]= 1;
9859: Dummy[k]= 1;
9860: modell[k].maintype= VTYPE;
9861: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9862: ncovv++; /* Varying variables without age */
9863: TvarV[ncovv]=Tvar[k];
9864: TvarVind[ncovv]=k;
9865: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9866: Fixed[k]= 1;
9867: Dummy[k]= 0;
9868: modell[k].maintype= VTYPE;
9869: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9870: ncovv++; /* Varying variables without age */
9871: TvarV[ncovv]=Tvar[k];
9872: TvarVind[ncovv]=k;
9873: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9874: Fixed[k]= 1;
9875: Dummy[k]= 1;
9876: modell[k].maintype= VTYPE;
9877: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9878: ncovv++; /* Varying variables without age */
9879: TvarV[ncovv]=Tvar[k];
9880: TvarVind[ncovv]=k;
9881: }
1.227 brouard 9882: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9883: if(Tvard[k1][2] <=ncovcol){
9884: Fixed[k]= 1;
9885: Dummy[k]= 1;
9886: modell[k].maintype= VTYPE;
9887: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9888: ncovv++; /* Varying variables without age */
9889: TvarV[ncovv]=Tvar[k];
9890: TvarVind[ncovv]=k;
9891: }else if(Tvard[k1][2] <=ncovcol+nqv){
9892: Fixed[k]= 1;
9893: Dummy[k]= 1;
9894: modell[k].maintype= VTYPE;
9895: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9896: ncovv++; /* Varying variables without age */
9897: TvarV[ncovv]=Tvar[k];
9898: TvarVind[ncovv]=k;
9899: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9900: Fixed[k]= 1;
9901: Dummy[k]= 1;
9902: modell[k].maintype= VTYPE;
9903: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9904: ncovv++; /* Varying variables without age */
9905: TvarV[ncovv]=Tvar[k];
9906: TvarVind[ncovv]=k;
9907: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9908: Fixed[k]= 1;
9909: Dummy[k]= 1;
9910: modell[k].maintype= VTYPE;
9911: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9912: ncovv++; /* Varying variables without age */
9913: TvarV[ncovv]=Tvar[k];
9914: TvarVind[ncovv]=k;
9915: }
1.227 brouard 9916: }else{
1.240 brouard 9917: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9918: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9919: } /*end k1*/
1.225 brouard 9920: }else{
1.226 brouard 9921: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9922: 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 9923: }
1.227 brouard 9924: 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 9925: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9926: 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]);
9927: }
9928: /* Searching for doublons in the model */
9929: for(k1=1; k1<= cptcovt;k1++){
9930: for(k2=1; k2 <k1;k2++){
9931: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9932: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9933: if(Tvar[k1]==Tvar[k2]){
9934: 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]]);
9935: 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);
9936: return(1);
9937: }
9938: }else if (Typevar[k1] ==2){
9939: k3=Tposprod[k1];
9940: k4=Tposprod[k2];
9941: 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])) ){
9942: 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]]);
9943: 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);
9944: return(1);
9945: }
9946: }
1.227 brouard 9947: }
9948: }
1.225 brouard 9949: }
9950: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9951: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9952: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9953: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9954: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9955: /*endread:*/
1.225 brouard 9956: printf("Exiting decodemodel: ");
9957: return (1);
1.136 brouard 9958: }
9959:
1.169 brouard 9960: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9961: {/* Check ages at death */
1.136 brouard 9962: int i, m;
1.218 brouard 9963: int firstone=0;
9964:
1.136 brouard 9965: for (i=1; i<=imx; i++) {
9966: for(m=2; (m<= maxwav); m++) {
9967: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9968: anint[m][i]=9999;
1.216 brouard 9969: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9970: s[m][i]=-1;
1.136 brouard 9971: }
9972: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9973: *nberr = *nberr + 1;
1.218 brouard 9974: if(firstone == 0){
9975: firstone=1;
1.260 brouard 9976: 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 9977: }
1.262 brouard 9978: 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 9979: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9980: }
9981: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9982: (*nberr)++;
1.259 brouard 9983: 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 9984: 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 9985: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9986: }
9987: }
9988: }
9989:
9990: for (i=1; i<=imx; i++) {
9991: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9992: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9993: 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 9994: if (s[m][i] >= nlstate+1) {
1.169 brouard 9995: if(agedc[i]>0){
9996: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9997: agev[m][i]=agedc[i];
1.214 brouard 9998: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9999: }else {
1.136 brouard 10000: if ((int)andc[i]!=9999){
10001: nbwarn++;
10002: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10003: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10004: agev[m][i]=-1;
10005: }
10006: }
1.169 brouard 10007: } /* agedc > 0 */
1.214 brouard 10008: } /* end if */
1.136 brouard 10009: else if(s[m][i] !=9){ /* Standard case, age in fractional
10010: years but with the precision of a month */
10011: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10012: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10013: agev[m][i]=1;
10014: else if(agev[m][i] < *agemin){
10015: *agemin=agev[m][i];
10016: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10017: }
10018: else if(agev[m][i] >*agemax){
10019: *agemax=agev[m][i];
1.156 brouard 10020: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10021: }
10022: /*agev[m][i]=anint[m][i]-annais[i];*/
10023: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10024: } /* en if 9*/
1.136 brouard 10025: else { /* =9 */
1.214 brouard 10026: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10027: agev[m][i]=1;
10028: s[m][i]=-1;
10029: }
10030: }
1.214 brouard 10031: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10032: agev[m][i]=1;
1.214 brouard 10033: else{
10034: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10035: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10036: agev[m][i]=0;
10037: }
10038: } /* End for lastpass */
10039: }
1.136 brouard 10040:
10041: for (i=1; i<=imx; i++) {
10042: for(m=firstpass; (m<=lastpass); m++){
10043: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10044: (*nberr)++;
1.136 brouard 10045: 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);
10046: 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);
10047: return 1;
10048: }
10049: }
10050: }
10051:
10052: /*for (i=1; i<=imx; i++){
10053: for (m=firstpass; (m<lastpass); m++){
10054: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10055: }
10056:
10057: }*/
10058:
10059:
1.139 brouard 10060: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10061: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10062:
10063: return (0);
1.164 brouard 10064: /* endread:*/
1.136 brouard 10065: printf("Exiting calandcheckages: ");
10066: return (1);
10067: }
10068:
1.172 brouard 10069: #if defined(_MSC_VER)
10070: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10071: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10072: //#include "stdafx.h"
10073: //#include <stdio.h>
10074: //#include <tchar.h>
10075: //#include <windows.h>
10076: //#include <iostream>
10077: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10078:
10079: LPFN_ISWOW64PROCESS fnIsWow64Process;
10080:
10081: BOOL IsWow64()
10082: {
10083: BOOL bIsWow64 = FALSE;
10084:
10085: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10086: // (HANDLE, PBOOL);
10087:
10088: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10089:
10090: HMODULE module = GetModuleHandle(_T("kernel32"));
10091: const char funcName[] = "IsWow64Process";
10092: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10093: GetProcAddress(module, funcName);
10094:
10095: if (NULL != fnIsWow64Process)
10096: {
10097: if (!fnIsWow64Process(GetCurrentProcess(),
10098: &bIsWow64))
10099: //throw std::exception("Unknown error");
10100: printf("Unknown error\n");
10101: }
10102: return bIsWow64 != FALSE;
10103: }
10104: #endif
1.177 brouard 10105:
1.191 brouard 10106: void syscompilerinfo(int logged)
1.167 brouard 10107: {
10108: /* #include "syscompilerinfo.h"*/
1.185 brouard 10109: /* command line Intel compiler 32bit windows, XP compatible:*/
10110: /* /GS /W3 /Gy
10111: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10112: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10113: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10114: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10115: */
10116: /* 64 bits */
1.185 brouard 10117: /*
10118: /GS /W3 /Gy
10119: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10120: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10121: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10122: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10123: /* Optimization are useless and O3 is slower than O2 */
10124: /*
10125: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10126: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10127: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10128: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10129: */
1.186 brouard 10130: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10131: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10132: /PDB:"visual studio
10133: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10134: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10135: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10136: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10137: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10138: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10139: uiAccess='false'"
10140: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10141: /NOLOGO /TLBID:1
10142: */
1.177 brouard 10143: #if defined __INTEL_COMPILER
1.178 brouard 10144: #if defined(__GNUC__)
10145: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10146: #endif
1.177 brouard 10147: #elif defined(__GNUC__)
1.179 brouard 10148: #ifndef __APPLE__
1.174 brouard 10149: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10150: #endif
1.177 brouard 10151: struct utsname sysInfo;
1.178 brouard 10152: int cross = CROSS;
10153: if (cross){
10154: printf("Cross-");
1.191 brouard 10155: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10156: }
1.174 brouard 10157: #endif
10158:
1.171 brouard 10159: #include <stdint.h>
1.178 brouard 10160:
1.191 brouard 10161: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10162: #if defined(__clang__)
1.191 brouard 10163: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10164: #endif
10165: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10166: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10167: #endif
10168: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10169: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10170: #endif
10171: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10172: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10173: #endif
10174: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10175: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10176: #endif
10177: #if defined(_MSC_VER)
1.191 brouard 10178: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10179: #endif
10180: #if defined(__PGI)
1.191 brouard 10181: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10182: #endif
10183: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10184: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10185: #endif
1.191 brouard 10186: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10187:
1.167 brouard 10188: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10189: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10190: // Windows (x64 and x86)
1.191 brouard 10191: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10192: #elif __unix__ // all unices, not all compilers
10193: // Unix
1.191 brouard 10194: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10195: #elif __linux__
10196: // linux
1.191 brouard 10197: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10198: #elif __APPLE__
1.174 brouard 10199: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10200: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10201: #endif
10202:
10203: /* __MINGW32__ */
10204: /* __CYGWIN__ */
10205: /* __MINGW64__ */
10206: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10207: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10208: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10209: /* _WIN64 // Defined for applications for Win64. */
10210: /* _M_X64 // Defined for compilations that target x64 processors. */
10211: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10212:
1.167 brouard 10213: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10214: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10215: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10216: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10217: #else
1.191 brouard 10218: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10219: #endif
10220:
1.169 brouard 10221: #if defined(__GNUC__)
10222: # if defined(__GNUC_PATCHLEVEL__)
10223: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10224: + __GNUC_MINOR__ * 100 \
10225: + __GNUC_PATCHLEVEL__)
10226: # else
10227: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10228: + __GNUC_MINOR__ * 100)
10229: # endif
1.174 brouard 10230: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10231: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10232:
10233: if (uname(&sysInfo) != -1) {
10234: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10235: 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 10236: }
10237: else
10238: perror("uname() error");
1.179 brouard 10239: //#ifndef __INTEL_COMPILER
10240: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10241: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10242: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10243: #endif
1.169 brouard 10244: #endif
1.172 brouard 10245:
10246: // void main()
10247: // {
1.169 brouard 10248: #if defined(_MSC_VER)
1.174 brouard 10249: if (IsWow64()){
1.191 brouard 10250: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10251: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10252: }
10253: else{
1.191 brouard 10254: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10255: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10256: }
1.172 brouard 10257: // printf("\nPress Enter to continue...");
10258: // getchar();
10259: // }
10260:
1.169 brouard 10261: #endif
10262:
1.167 brouard 10263:
1.219 brouard 10264: }
1.136 brouard 10265:
1.219 brouard 10266: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10267: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10268: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10269: /* double ftolpl = 1.e-10; */
1.180 brouard 10270: double age, agebase, agelim;
1.203 brouard 10271: double tot;
1.180 brouard 10272:
1.202 brouard 10273: strcpy(filerespl,"PL_");
10274: strcat(filerespl,fileresu);
10275: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10276: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10277: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10278: }
1.227 brouard 10279: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10280: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10281: pstamp(ficrespl);
1.203 brouard 10282: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10283: fprintf(ficrespl,"#Age ");
10284: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10285: fprintf(ficrespl,"\n");
1.180 brouard 10286:
1.219 brouard 10287: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10288:
1.219 brouard 10289: agebase=ageminpar;
10290: agelim=agemaxpar;
1.180 brouard 10291:
1.227 brouard 10292: /* i1=pow(2,ncoveff); */
1.234 brouard 10293: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10294: if (cptcovn < 1){i1=1;}
1.180 brouard 10295:
1.238 brouard 10296: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10297: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10298: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10299: continue;
1.235 brouard 10300:
1.238 brouard 10301: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10302: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10303: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10304: /* k=k+1; */
10305: /* to clean */
10306: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10307: fprintf(ficrespl,"#******");
10308: printf("#******");
10309: fprintf(ficlog,"#******");
10310: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10311: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10312: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10313: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10314: }
10315: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10316: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10317: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10318: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10319: }
10320: fprintf(ficrespl,"******\n");
10321: printf("******\n");
10322: fprintf(ficlog,"******\n");
10323: if(invalidvarcomb[k]){
10324: printf("\nCombination (%d) ignored because no case \n",k);
10325: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10326: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10327: continue;
10328: }
1.219 brouard 10329:
1.238 brouard 10330: fprintf(ficrespl,"#Age ");
10331: for(j=1;j<=cptcoveff;j++) {
10332: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10333: }
10334: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10335: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10336:
1.238 brouard 10337: for (age=agebase; age<=agelim; age++){
10338: /* for (age=agebase; age<=agebase; age++){ */
10339: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10340: fprintf(ficrespl,"%.0f ",age );
10341: for(j=1;j<=cptcoveff;j++)
10342: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10343: tot=0.;
10344: for(i=1; i<=nlstate;i++){
10345: tot += prlim[i][i];
10346: fprintf(ficrespl," %.5f", prlim[i][i]);
10347: }
10348: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10349: } /* Age */
10350: /* was end of cptcod */
10351: } /* cptcov */
10352: } /* nres */
1.219 brouard 10353: return 0;
1.180 brouard 10354: }
10355:
1.218 brouard 10356: 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){
10357: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10358:
10359: /* Computes the back prevalence limit for any combination of covariate values
10360: * at any age between ageminpar and agemaxpar
10361: */
1.235 brouard 10362: int i, j, k, i1, nres=0 ;
1.217 brouard 10363: /* double ftolpl = 1.e-10; */
10364: double age, agebase, agelim;
10365: double tot;
1.218 brouard 10366: /* double ***mobaverage; */
10367: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10368:
10369: strcpy(fileresplb,"PLB_");
10370: strcat(fileresplb,fileresu);
10371: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10372: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10373: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10374: }
10375: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10376: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10377: pstamp(ficresplb);
10378: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10379: fprintf(ficresplb,"#Age ");
10380: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10381: fprintf(ficresplb,"\n");
10382:
1.218 brouard 10383:
10384: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10385:
10386: agebase=ageminpar;
10387: agelim=agemaxpar;
10388:
10389:
1.227 brouard 10390: i1=pow(2,cptcoveff);
1.218 brouard 10391: if (cptcovn < 1){i1=1;}
1.227 brouard 10392:
1.238 brouard 10393: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10394: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10395: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10396: continue;
10397: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10398: fprintf(ficresplb,"#******");
10399: printf("#******");
10400: fprintf(ficlog,"#******");
10401: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10402: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10403: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10404: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10405: }
10406: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10407: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10408: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10409: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10410: }
10411: fprintf(ficresplb,"******\n");
10412: printf("******\n");
10413: fprintf(ficlog,"******\n");
10414: if(invalidvarcomb[k]){
10415: printf("\nCombination (%d) ignored because no cases \n",k);
10416: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10417: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10418: continue;
10419: }
1.218 brouard 10420:
1.238 brouard 10421: fprintf(ficresplb,"#Age ");
10422: for(j=1;j<=cptcoveff;j++) {
10423: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10424: }
10425: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10426: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10427:
10428:
1.238 brouard 10429: for (age=agebase; age<=agelim; age++){
10430: /* for (age=agebase; age<=agebase; age++){ */
10431: if(mobilavproj > 0){
10432: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10433: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10434: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10435: }else if (mobilavproj == 0){
10436: 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);
10437: 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);
10438: exit(1);
10439: }else{
10440: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10441: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10442: /* printf("TOTOT\n"); */
10443: /* exit(1); */
1.238 brouard 10444: }
10445: fprintf(ficresplb,"%.0f ",age );
10446: for(j=1;j<=cptcoveff;j++)
10447: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10448: tot=0.;
10449: for(i=1; i<=nlstate;i++){
10450: tot += bprlim[i][i];
10451: fprintf(ficresplb," %.5f", bprlim[i][i]);
10452: }
10453: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10454: } /* Age */
10455: /* was end of cptcod */
1.255 brouard 10456: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10457: } /* end of any combination */
10458: } /* end of nres */
1.218 brouard 10459: /* hBijx(p, bage, fage); */
10460: /* fclose(ficrespijb); */
10461:
10462: return 0;
1.217 brouard 10463: }
1.218 brouard 10464:
1.180 brouard 10465: int hPijx(double *p, int bage, int fage){
10466: /*------------- h Pij x at various ages ------------*/
10467:
10468: int stepsize;
10469: int agelim;
10470: int hstepm;
10471: int nhstepm;
1.235 brouard 10472: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10473:
10474: double agedeb;
10475: double ***p3mat;
10476:
1.201 brouard 10477: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10478: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10479: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10480: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10481: }
10482: printf("Computing pij: result on file '%s' \n", filerespij);
10483: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10484:
10485: stepsize=(int) (stepm+YEARM-1)/YEARM;
10486: /*if (stepm<=24) stepsize=2;*/
10487:
10488: agelim=AGESUP;
10489: hstepm=stepsize*YEARM; /* Every year of age */
10490: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10491:
1.180 brouard 10492: /* hstepm=1; aff par mois*/
10493: pstamp(ficrespij);
10494: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10495: i1= pow(2,cptcoveff);
1.218 brouard 10496: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10497: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10498: /* k=k+1; */
1.235 brouard 10499: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10500: for(k=1; k<=i1;k++){
1.253 brouard 10501: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10502: continue;
1.183 brouard 10503: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10504: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10505: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10506: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10507: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10508: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10509: }
1.183 brouard 10510: fprintf(ficrespij,"******\n");
10511:
10512: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10513: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10514: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10515:
10516: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10517:
1.183 brouard 10518: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10519: oldm=oldms;savm=savms;
1.235 brouard 10520: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10521: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10522: for(i=1; i<=nlstate;i++)
10523: for(j=1; j<=nlstate+ndeath;j++)
10524: fprintf(ficrespij," %1d-%1d",i,j);
10525: fprintf(ficrespij,"\n");
10526: for (h=0; h<=nhstepm; h++){
10527: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10528: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10529: for(i=1; i<=nlstate;i++)
10530: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10531: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10532: fprintf(ficrespij,"\n");
10533: }
1.183 brouard 10534: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10535: fprintf(ficrespij,"\n");
10536: }
1.180 brouard 10537: /*}*/
10538: }
1.218 brouard 10539: return 0;
1.180 brouard 10540: }
1.218 brouard 10541:
10542: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10543: /*------------- h Bij x at various ages ------------*/
10544:
10545: int stepsize;
1.218 brouard 10546: /* int agelim; */
10547: int ageminl;
1.217 brouard 10548: int hstepm;
10549: int nhstepm;
1.238 brouard 10550: int h, i, i1, j, k, nres;
1.218 brouard 10551:
1.217 brouard 10552: double agedeb;
10553: double ***p3mat;
1.218 brouard 10554:
10555: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10556: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10557: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10558: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10559: }
10560: printf("Computing pij back: result on file '%s' \n", filerespijb);
10561: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10562:
10563: stepsize=(int) (stepm+YEARM-1)/YEARM;
10564: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10565:
1.218 brouard 10566: /* agelim=AGESUP; */
10567: ageminl=30;
10568: hstepm=stepsize*YEARM; /* Every year of age */
10569: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10570:
10571: /* hstepm=1; aff par mois*/
10572: pstamp(ficrespijb);
1.255 brouard 10573: 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 10574: i1= pow(2,cptcoveff);
1.218 brouard 10575: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10576: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10577: /* k=k+1; */
1.238 brouard 10578: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10579: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10580: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10581: continue;
10582: fprintf(ficrespijb,"\n#****** ");
10583: for(j=1;j<=cptcoveff;j++)
10584: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10585: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10586: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10587: }
10588: fprintf(ficrespijb,"******\n");
1.264 brouard 10589: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10590: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10591: continue;
10592: }
10593:
10594: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10595: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10596: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10597: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10598: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10599:
10600: /* nhstepm=nhstepm*YEARM; aff par mois*/
10601:
1.266 brouard 10602: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10603: /* and memory limitations if stepm is small */
10604:
1.238 brouard 10605: /* oldm=oldms;savm=savms; */
10606: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10607: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10608: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10609: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10610: for(i=1; i<=nlstate;i++)
10611: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10612: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10613: fprintf(ficrespijb,"\n");
1.238 brouard 10614: for (h=0; h<=nhstepm; h++){
10615: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10616: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10617: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10618: for(i=1; i<=nlstate;i++)
10619: for(j=1; j<=nlstate+ndeath;j++)
10620: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10621: fprintf(ficrespijb,"\n");
10622: }
10623: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10624: fprintf(ficrespijb,"\n");
10625: } /* end age deb */
10626: } /* end combination */
10627: } /* end nres */
1.218 brouard 10628: return 0;
10629: } /* hBijx */
1.217 brouard 10630:
1.180 brouard 10631:
1.136 brouard 10632: /***********************************************/
10633: /**************** Main Program *****************/
10634: /***********************************************/
10635:
10636: int main(int argc, char *argv[])
10637: {
10638: #ifdef GSL
10639: const gsl_multimin_fminimizer_type *T;
10640: size_t iteri = 0, it;
10641: int rval = GSL_CONTINUE;
10642: int status = GSL_SUCCESS;
10643: double ssval;
10644: #endif
10645: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10646: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10647: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10648: int jj, ll, li, lj, lk;
1.136 brouard 10649: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10650: int num_filled;
1.136 brouard 10651: int itimes;
10652: int NDIM=2;
10653: int vpopbased=0;
1.235 brouard 10654: int nres=0;
1.258 brouard 10655: int endishere=0;
1.277 brouard 10656: int noffset=0;
1.274 brouard 10657: int ncurrv=0; /* Temporary variable */
10658:
1.164 brouard 10659: char ca[32], cb[32];
1.136 brouard 10660: /* FILE *fichtm; *//* Html File */
10661: /* FILE *ficgp;*/ /*Gnuplot File */
10662: struct stat info;
1.191 brouard 10663: double agedeb=0.;
1.194 brouard 10664:
10665: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10666: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10667:
1.165 brouard 10668: double fret;
1.191 brouard 10669: double dum=0.; /* Dummy variable */
1.136 brouard 10670: double ***p3mat;
1.218 brouard 10671: /* double ***mobaverage; */
1.164 brouard 10672:
10673: char line[MAXLINE];
1.197 brouard 10674: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10675:
1.234 brouard 10676: char modeltemp[MAXLINE];
1.230 brouard 10677: char resultline[MAXLINE];
10678:
1.136 brouard 10679: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10680: char *tok, *val; /* pathtot */
1.136 brouard 10681: int firstobs=1, lastobs=10;
1.195 brouard 10682: int c, h , cpt, c2;
1.191 brouard 10683: int jl=0;
10684: int i1, j1, jk, stepsize=0;
1.194 brouard 10685: int count=0;
10686:
1.164 brouard 10687: int *tab;
1.136 brouard 10688: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10689: int backcast=0;
1.136 brouard 10690: int mobilav=0,popforecast=0;
1.191 brouard 10691: int hstepm=0, nhstepm=0;
1.136 brouard 10692: int agemortsup;
10693: float sumlpop=0.;
10694: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10695: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10696:
1.191 brouard 10697: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10698: double ftolpl=FTOL;
10699: double **prlim;
1.217 brouard 10700: double **bprlim;
1.136 brouard 10701: double ***param; /* Matrix of parameters */
1.251 brouard 10702: double ***paramstart; /* Matrix of starting parameter values */
10703: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10704: double **matcov; /* Matrix of covariance */
1.203 brouard 10705: double **hess; /* Hessian matrix */
1.136 brouard 10706: double ***delti3; /* Scale */
10707: double *delti; /* Scale */
10708: double ***eij, ***vareij;
10709: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10710:
1.136 brouard 10711: double *epj, vepp;
1.164 brouard 10712:
1.273 brouard 10713: double dateprev1, dateprev2;
10714: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10715: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10716:
1.136 brouard 10717: double **ximort;
1.145 brouard 10718: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10719: int *dcwave;
10720:
1.164 brouard 10721: char z[1]="c";
1.136 brouard 10722:
10723: /*char *strt;*/
10724: char strtend[80];
1.126 brouard 10725:
1.164 brouard 10726:
1.126 brouard 10727: /* setlocale (LC_ALL, ""); */
10728: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10729: /* textdomain (PACKAGE); */
10730: /* setlocale (LC_CTYPE, ""); */
10731: /* setlocale (LC_MESSAGES, ""); */
10732:
10733: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10734: rstart_time = time(NULL);
10735: /* (void) gettimeofday(&start_time,&tzp);*/
10736: start_time = *localtime(&rstart_time);
1.126 brouard 10737: curr_time=start_time;
1.157 brouard 10738: /*tml = *localtime(&start_time.tm_sec);*/
10739: /* strcpy(strstart,asctime(&tml)); */
10740: strcpy(strstart,asctime(&start_time));
1.126 brouard 10741:
10742: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10743: /* tp.tm_sec = tp.tm_sec +86400; */
10744: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10745: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10746: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10747: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10748: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10749: /* strt=asctime(&tmg); */
10750: /* printf("Time(after) =%s",strstart); */
10751: /* (void) time (&time_value);
10752: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10753: * tm = *localtime(&time_value);
10754: * strstart=asctime(&tm);
10755: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10756: */
10757:
10758: nberr=0; /* Number of errors and warnings */
10759: nbwarn=0;
1.184 brouard 10760: #ifdef WIN32
10761: _getcwd(pathcd, size);
10762: #else
1.126 brouard 10763: getcwd(pathcd, size);
1.184 brouard 10764: #endif
1.191 brouard 10765: syscompilerinfo(0);
1.196 brouard 10766: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10767: if(argc <=1){
10768: printf("\nEnter the parameter file name: ");
1.205 brouard 10769: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10770: printf("ERROR Empty parameter file name\n");
10771: goto end;
10772: }
1.126 brouard 10773: i=strlen(pathr);
10774: if(pathr[i-1]=='\n')
10775: pathr[i-1]='\0';
1.156 brouard 10776: i=strlen(pathr);
1.205 brouard 10777: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10778: pathr[i-1]='\0';
1.205 brouard 10779: }
10780: i=strlen(pathr);
10781: if( i==0 ){
10782: printf("ERROR Empty parameter file name\n");
10783: goto end;
10784: }
10785: for (tok = pathr; tok != NULL; ){
1.126 brouard 10786: printf("Pathr |%s|\n",pathr);
10787: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10788: printf("val= |%s| pathr=%s\n",val,pathr);
10789: strcpy (pathtot, val);
10790: if(pathr[0] == '\0') break; /* Dirty */
10791: }
10792: }
10793: else{
10794: strcpy(pathtot,argv[1]);
10795: }
10796: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10797: /*cygwin_split_path(pathtot,path,optionfile);
10798: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10799: /* cutv(path,optionfile,pathtot,'\\');*/
10800:
10801: /* Split argv[0], imach program to get pathimach */
10802: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10803: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10804: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10805: /* strcpy(pathimach,argv[0]); */
10806: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10807: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10808: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10809: #ifdef WIN32
10810: _chdir(path); /* Can be a relative path */
10811: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10812: #else
1.126 brouard 10813: chdir(path); /* Can be a relative path */
1.184 brouard 10814: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10815: #endif
10816: printf("Current directory %s!\n",pathcd);
1.126 brouard 10817: strcpy(command,"mkdir ");
10818: strcat(command,optionfilefiname);
10819: if((outcmd=system(command)) != 0){
1.169 brouard 10820: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10821: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10822: /* fclose(ficlog); */
10823: /* exit(1); */
10824: }
10825: /* if((imk=mkdir(optionfilefiname))<0){ */
10826: /* perror("mkdir"); */
10827: /* } */
10828:
10829: /*-------- arguments in the command line --------*/
10830:
1.186 brouard 10831: /* Main Log file */
1.126 brouard 10832: strcat(filelog, optionfilefiname);
10833: strcat(filelog,".log"); /* */
10834: if((ficlog=fopen(filelog,"w"))==NULL) {
10835: printf("Problem with logfile %s\n",filelog);
10836: goto end;
10837: }
10838: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10839: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10840: fprintf(ficlog,"\nEnter the parameter file name: \n");
10841: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10842: path=%s \n\
10843: optionfile=%s\n\
10844: optionfilext=%s\n\
1.156 brouard 10845: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10846:
1.197 brouard 10847: syscompilerinfo(1);
1.167 brouard 10848:
1.126 brouard 10849: printf("Local time (at start):%s",strstart);
10850: fprintf(ficlog,"Local time (at start): %s",strstart);
10851: fflush(ficlog);
10852: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10853: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10854:
10855: /* */
10856: strcpy(fileres,"r");
10857: strcat(fileres, optionfilefiname);
1.201 brouard 10858: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10859: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10860: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10861:
1.186 brouard 10862: /* Main ---------arguments file --------*/
1.126 brouard 10863:
10864: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10865: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10866: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10867: fflush(ficlog);
1.149 brouard 10868: /* goto end; */
10869: exit(70);
1.126 brouard 10870: }
10871:
10872:
10873:
10874: strcpy(filereso,"o");
1.201 brouard 10875: strcat(filereso,fileresu);
1.126 brouard 10876: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10877: printf("Problem with Output resultfile: %s\n", filereso);
10878: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10879: fflush(ficlog);
10880: goto end;
10881: }
1.278 brouard 10882: /*-------- Rewriting parameter file ----------*/
10883: strcpy(rfileres,"r"); /* "Rparameterfile */
10884: strcat(rfileres,optionfilefiname); /* Parameter file first name */
10885: strcat(rfileres,"."); /* */
10886: strcat(rfileres,optionfilext); /* Other files have txt extension */
10887: if((ficres =fopen(rfileres,"w"))==NULL) {
10888: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10889: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
10890: fflush(ficlog);
10891: goto end;
10892: }
10893: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 10894:
1.278 brouard 10895:
1.126 brouard 10896: /* Reads comments: lines beginning with '#' */
10897: numlinepar=0;
1.277 brouard 10898: /* Is it a BOM UTF-8 Windows file? */
10899: /* First parameter line */
1.197 brouard 10900: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 10901: noffset=0;
10902: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10903: {
10904: noffset=noffset+3;
10905: printf("# File is an UTF8 Bom.\n"); // 0xBF
10906: }
10907: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
10908: {
10909: noffset=noffset+2;
10910: printf("# File is an UTF16BE BOM file\n");
10911: }
10912: else if( line[0] == 0 && line[1] == 0)
10913: {
10914: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10915: noffset=noffset+4;
10916: printf("# File is an UTF16BE BOM file\n");
10917: }
10918: } else{
10919: ;/*printf(" Not a BOM file\n");*/
10920: }
10921:
1.197 brouard 10922: /* If line starts with a # it is a comment */
1.277 brouard 10923: if (line[noffset] == '#') {
1.197 brouard 10924: numlinepar++;
10925: fputs(line,stdout);
10926: fputs(line,ficparo);
1.278 brouard 10927: fputs(line,ficres);
1.197 brouard 10928: fputs(line,ficlog);
10929: continue;
10930: }else
10931: break;
10932: }
10933: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10934: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10935: if (num_filled != 5) {
10936: printf("Should be 5 parameters\n");
10937: }
1.126 brouard 10938: numlinepar++;
1.197 brouard 10939: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10940: }
10941: /* Second parameter line */
10942: while(fgets(line, MAXLINE, ficpar)) {
10943: /* If line starts with a # it is a comment */
10944: if (line[0] == '#') {
10945: numlinepar++;
10946: fputs(line,stdout);
10947: fputs(line,ficparo);
1.278 brouard 10948: fputs(line,ficres);
1.197 brouard 10949: fputs(line,ficlog);
10950: continue;
10951: }else
10952: break;
10953: }
1.223 brouard 10954: 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", \
10955: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10956: if (num_filled != 11) {
10957: 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 10958: printf("but line=%s\n",line);
1.197 brouard 10959: }
1.223 brouard 10960: 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 10961: }
1.203 brouard 10962: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10963: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10964: /* Third parameter line */
10965: while(fgets(line, MAXLINE, ficpar)) {
10966: /* If line starts with a # it is a comment */
10967: if (line[0] == '#') {
10968: numlinepar++;
10969: fputs(line,stdout);
10970: fputs(line,ficparo);
1.278 brouard 10971: fputs(line,ficres);
1.197 brouard 10972: fputs(line,ficlog);
10973: continue;
10974: }else
10975: break;
10976: }
1.201 brouard 10977: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 10978: if (num_filled != 1){
10979: printf("ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
10980: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age' %s\n",num_filled, line);
1.197 brouard 10981: model[0]='\0';
10982: goto end;
10983: }
10984: else{
10985: if (model[0]=='+'){
10986: for(i=1; i<=strlen(model);i++)
10987: modeltemp[i-1]=model[i];
1.201 brouard 10988: strcpy(model,modeltemp);
1.197 brouard 10989: }
10990: }
1.199 brouard 10991: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10992: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10993: }
10994: /* 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); */
10995: /* numlinepar=numlinepar+3; /\* In general *\/ */
10996: /* 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 10997: 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);
10998: 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 10999: fflush(ficlog);
1.190 brouard 11000: /* if(model[0]=='#'|| model[0]== '\0'){ */
11001: if(model[0]=='#'){
1.279 brouard 11002: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11003: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11004: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11005: if(mle != -1){
1.279 brouard 11006: 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 11007: exit(1);
11008: }
11009: }
1.126 brouard 11010: while((c=getc(ficpar))=='#' && c!= EOF){
11011: ungetc(c,ficpar);
11012: fgets(line, MAXLINE, ficpar);
11013: numlinepar++;
1.195 brouard 11014: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11015: z[0]=line[1];
11016: }
11017: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11018: fputs(line, stdout);
11019: //puts(line);
1.126 brouard 11020: fputs(line,ficparo);
11021: fputs(line,ficlog);
11022: }
11023: ungetc(c,ficpar);
11024:
11025:
1.145 brouard 11026: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 11027: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
11028: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
11029: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11030: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11031: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11032: v1+v2*age+v2*v3 makes cptcovn = 3
11033: */
11034: if (strlen(model)>1)
1.187 brouard 11035: 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 11036: else
1.187 brouard 11037: ncovmodel=2; /* Constant and age */
1.133 brouard 11038: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11039: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11040: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11041: 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);
11042: 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);
11043: fflush(stdout);
11044: fclose (ficlog);
11045: goto end;
11046: }
1.126 brouard 11047: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11048: delti=delti3[1][1];
11049: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11050: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11051: /* We could also provide initial parameters values giving by simple logistic regression
11052: * only one way, that is without matrix product. We will have nlstate maximizations */
11053: /* for(i=1;i<nlstate;i++){ */
11054: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11055: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11056: /* } */
1.126 brouard 11057: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11058: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11059: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11060: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11061: fclose (ficparo);
11062: fclose (ficlog);
11063: goto end;
11064: exit(0);
1.220 brouard 11065: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11066: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11067: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11068: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11069: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11070: matcov=matrix(1,npar,1,npar);
1.203 brouard 11071: hess=matrix(1,npar,1,npar);
1.220 brouard 11072: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11073: /* Read guessed parameters */
1.126 brouard 11074: /* Reads comments: lines beginning with '#' */
11075: while((c=getc(ficpar))=='#' && c!= EOF){
11076: ungetc(c,ficpar);
11077: fgets(line, MAXLINE, ficpar);
11078: numlinepar++;
1.141 brouard 11079: fputs(line,stdout);
1.126 brouard 11080: fputs(line,ficparo);
11081: fputs(line,ficlog);
11082: }
11083: ungetc(c,ficpar);
11084:
11085: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11086: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11087: for(i=1; i <=nlstate; i++){
1.234 brouard 11088: j=0;
1.126 brouard 11089: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11090: if(jj==i) continue;
11091: j++;
11092: fscanf(ficpar,"%1d%1d",&i1,&j1);
11093: if ((i1 != i) || (j1 != jj)){
11094: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11095: It might be a problem of design; if ncovcol and the model are correct\n \
11096: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11097: exit(1);
11098: }
11099: fprintf(ficparo,"%1d%1d",i1,j1);
11100: if(mle==1)
11101: printf("%1d%1d",i,jj);
11102: fprintf(ficlog,"%1d%1d",i,jj);
11103: for(k=1; k<=ncovmodel;k++){
11104: fscanf(ficpar," %lf",¶m[i][j][k]);
11105: if(mle==1){
11106: printf(" %lf",param[i][j][k]);
11107: fprintf(ficlog," %lf",param[i][j][k]);
11108: }
11109: else
11110: fprintf(ficlog," %lf",param[i][j][k]);
11111: fprintf(ficparo," %lf",param[i][j][k]);
11112: }
11113: fscanf(ficpar,"\n");
11114: numlinepar++;
11115: if(mle==1)
11116: printf("\n");
11117: fprintf(ficlog,"\n");
11118: fprintf(ficparo,"\n");
1.126 brouard 11119: }
11120: }
11121: fflush(ficlog);
1.234 brouard 11122:
1.251 brouard 11123: /* Reads parameters values */
1.126 brouard 11124: p=param[1][1];
1.251 brouard 11125: pstart=paramstart[1][1];
1.126 brouard 11126:
11127: /* Reads comments: lines beginning with '#' */
11128: while((c=getc(ficpar))=='#' && c!= EOF){
11129: ungetc(c,ficpar);
11130: fgets(line, MAXLINE, ficpar);
11131: numlinepar++;
1.141 brouard 11132: fputs(line,stdout);
1.126 brouard 11133: fputs(line,ficparo);
11134: fputs(line,ficlog);
11135: }
11136: ungetc(c,ficpar);
11137:
11138: for(i=1; i <=nlstate; i++){
11139: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11140: fscanf(ficpar,"%1d%1d",&i1,&j1);
11141: if ( (i1-i) * (j1-j) != 0){
11142: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11143: exit(1);
11144: }
11145: printf("%1d%1d",i,j);
11146: fprintf(ficparo,"%1d%1d",i1,j1);
11147: fprintf(ficlog,"%1d%1d",i1,j1);
11148: for(k=1; k<=ncovmodel;k++){
11149: fscanf(ficpar,"%le",&delti3[i][j][k]);
11150: printf(" %le",delti3[i][j][k]);
11151: fprintf(ficparo," %le",delti3[i][j][k]);
11152: fprintf(ficlog," %le",delti3[i][j][k]);
11153: }
11154: fscanf(ficpar,"\n");
11155: numlinepar++;
11156: printf("\n");
11157: fprintf(ficparo,"\n");
11158: fprintf(ficlog,"\n");
1.126 brouard 11159: }
11160: }
11161: fflush(ficlog);
1.234 brouard 11162:
1.145 brouard 11163: /* Reads covariance matrix */
1.126 brouard 11164: delti=delti3[1][1];
1.220 brouard 11165:
11166:
1.126 brouard 11167: /* 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 11168:
1.126 brouard 11169: /* Reads comments: lines beginning with '#' */
11170: while((c=getc(ficpar))=='#' && c!= EOF){
11171: ungetc(c,ficpar);
11172: fgets(line, MAXLINE, ficpar);
11173: numlinepar++;
1.141 brouard 11174: fputs(line,stdout);
1.126 brouard 11175: fputs(line,ficparo);
11176: fputs(line,ficlog);
11177: }
11178: ungetc(c,ficpar);
1.220 brouard 11179:
1.126 brouard 11180: matcov=matrix(1,npar,1,npar);
1.203 brouard 11181: hess=matrix(1,npar,1,npar);
1.131 brouard 11182: for(i=1; i <=npar; i++)
11183: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11184:
1.194 brouard 11185: /* Scans npar lines */
1.126 brouard 11186: for(i=1; i <=npar; i++){
1.226 brouard 11187: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11188: if(count != 3){
1.226 brouard 11189: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11190: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11191: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11192: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11193: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11194: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11195: exit(1);
1.220 brouard 11196: }else{
1.226 brouard 11197: if(mle==1)
11198: printf("%1d%1d%d",i1,j1,jk);
11199: }
11200: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11201: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11202: for(j=1; j <=i; j++){
1.226 brouard 11203: fscanf(ficpar," %le",&matcov[i][j]);
11204: if(mle==1){
11205: printf(" %.5le",matcov[i][j]);
11206: }
11207: fprintf(ficlog," %.5le",matcov[i][j]);
11208: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11209: }
11210: fscanf(ficpar,"\n");
11211: numlinepar++;
11212: if(mle==1)
1.220 brouard 11213: printf("\n");
1.126 brouard 11214: fprintf(ficlog,"\n");
11215: fprintf(ficparo,"\n");
11216: }
1.194 brouard 11217: /* End of read covariance matrix npar lines */
1.126 brouard 11218: for(i=1; i <=npar; i++)
11219: for(j=i+1;j<=npar;j++)
1.226 brouard 11220: matcov[i][j]=matcov[j][i];
1.126 brouard 11221:
11222: if(mle==1)
11223: printf("\n");
11224: fprintf(ficlog,"\n");
11225:
11226: fflush(ficlog);
11227:
11228: } /* End of mle != -3 */
1.218 brouard 11229:
1.186 brouard 11230: /* Main data
11231: */
1.126 brouard 11232: n= lastobs;
11233: num=lvector(1,n);
11234: moisnais=vector(1,n);
11235: annais=vector(1,n);
11236: moisdc=vector(1,n);
11237: andc=vector(1,n);
1.220 brouard 11238: weight=vector(1,n);
1.126 brouard 11239: agedc=vector(1,n);
11240: cod=ivector(1,n);
1.220 brouard 11241: for(i=1;i<=n;i++){
1.234 brouard 11242: num[i]=0;
11243: moisnais[i]=0;
11244: annais[i]=0;
11245: moisdc[i]=0;
11246: andc[i]=0;
11247: agedc[i]=0;
11248: cod[i]=0;
11249: weight[i]=1.0; /* Equal weights, 1 by default */
11250: }
1.126 brouard 11251: mint=matrix(1,maxwav,1,n);
11252: anint=matrix(1,maxwav,1,n);
1.131 brouard 11253: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11254: tab=ivector(1,NCOVMAX);
1.144 brouard 11255: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11256: 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 11257:
1.136 brouard 11258: /* Reads data from file datafile */
11259: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11260: goto end;
11261:
11262: /* Calculation of the number of parameters from char model */
1.234 brouard 11263: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11264: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11265: k=3 V4 Tvar[k=3]= 4 (from V4)
11266: k=2 V1 Tvar[k=2]= 1 (from V1)
11267: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11268: */
11269:
11270: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11271: TvarsDind=ivector(1,NCOVMAX); /* */
11272: TvarsD=ivector(1,NCOVMAX); /* */
11273: TvarsQind=ivector(1,NCOVMAX); /* */
11274: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11275: TvarF=ivector(1,NCOVMAX); /* */
11276: TvarFind=ivector(1,NCOVMAX); /* */
11277: TvarV=ivector(1,NCOVMAX); /* */
11278: TvarVind=ivector(1,NCOVMAX); /* */
11279: TvarA=ivector(1,NCOVMAX); /* */
11280: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11281: TvarFD=ivector(1,NCOVMAX); /* */
11282: TvarFDind=ivector(1,NCOVMAX); /* */
11283: TvarFQ=ivector(1,NCOVMAX); /* */
11284: TvarFQind=ivector(1,NCOVMAX); /* */
11285: TvarVD=ivector(1,NCOVMAX); /* */
11286: TvarVDind=ivector(1,NCOVMAX); /* */
11287: TvarVQ=ivector(1,NCOVMAX); /* */
11288: TvarVQind=ivector(1,NCOVMAX); /* */
11289:
1.230 brouard 11290: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11291: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11292: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11293: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11294: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11295: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11296: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11297: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11298: */
11299: /* For model-covariate k tells which data-covariate to use but
11300: because this model-covariate is a construction we invent a new column
11301: ncovcol + k1
11302: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11303: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11304: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11305: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11306: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11307: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11308: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11309: */
1.145 brouard 11310: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11311: 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 11312: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11313: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11314: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11315: 4 covariates (3 plus signs)
11316: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11317: */
1.230 brouard 11318: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11319: * individual dummy, fixed or varying:
11320: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11321: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11322: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11323: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11324: * Tmodelind[1]@9={9,0,3,2,}*/
11325: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11326: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11327: * individual quantitative, fixed or varying:
11328: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11329: * 3, 1, 0, 0, 0, 0, 0, 0},
11330: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11331: /* Main decodemodel */
11332:
1.187 brouard 11333:
1.223 brouard 11334: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11335: goto end;
11336:
1.137 brouard 11337: if((double)(lastobs-imx)/(double)imx > 1.10){
11338: nbwarn++;
11339: 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);
11340: 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);
11341: }
1.136 brouard 11342: /* if(mle==1){*/
1.137 brouard 11343: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11344: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11345: }
11346:
11347: /*-calculation of age at interview from date of interview and age at death -*/
11348: agev=matrix(1,maxwav,1,imx);
11349:
11350: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11351: goto end;
11352:
1.126 brouard 11353:
1.136 brouard 11354: agegomp=(int)agemin;
11355: free_vector(moisnais,1,n);
11356: free_vector(annais,1,n);
1.126 brouard 11357: /* free_matrix(mint,1,maxwav,1,n);
11358: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11359: /* free_vector(moisdc,1,n); */
11360: /* free_vector(andc,1,n); */
1.145 brouard 11361: /* */
11362:
1.126 brouard 11363: wav=ivector(1,imx);
1.214 brouard 11364: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11365: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11366: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11367: 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.*/
11368: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11369: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11370:
11371: /* Concatenates waves */
1.214 brouard 11372: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11373: Death is a valid wave (if date is known).
11374: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11375: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11376: and mw[mi+1][i]. dh depends on stepm.
11377: */
11378:
1.126 brouard 11379: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11380: /* Concatenates waves */
1.145 brouard 11381:
1.215 brouard 11382: free_vector(moisdc,1,n);
11383: free_vector(andc,1,n);
11384:
1.126 brouard 11385: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11386: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11387: ncodemax[1]=1;
1.145 brouard 11388: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11389: cptcoveff=0;
1.220 brouard 11390: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11391: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11392: }
11393:
11394: ncovcombmax=pow(2,cptcoveff);
11395: invalidvarcomb=ivector(1, ncovcombmax);
11396: for(i=1;i<ncovcombmax;i++)
11397: invalidvarcomb[i]=0;
11398:
1.211 brouard 11399: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11400: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11401: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11402:
1.200 brouard 11403: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11404: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11405: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11406: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11407: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11408: * (currently 0 or 1) in the data.
11409: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11410: * corresponding modality (h,j).
11411: */
11412:
1.145 brouard 11413: h=0;
11414: /*if (cptcovn > 0) */
1.126 brouard 11415: m=pow(2,cptcoveff);
11416:
1.144 brouard 11417: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11418: * For k=4 covariates, h goes from 1 to m=2**k
11419: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11420: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11421: * h\k 1 2 3 4
1.143 brouard 11422: *______________________________
11423: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11424: * 2 2 1 1 1
11425: * 3 i=2 1 2 1 1
11426: * 4 2 2 1 1
11427: * 5 i=3 1 i=2 1 2 1
11428: * 6 2 1 2 1
11429: * 7 i=4 1 2 2 1
11430: * 8 2 2 2 1
1.197 brouard 11431: * 9 i=5 1 i=3 1 i=2 1 2
11432: * 10 2 1 1 2
11433: * 11 i=6 1 2 1 2
11434: * 12 2 2 1 2
11435: * 13 i=7 1 i=4 1 2 2
11436: * 14 2 1 2 2
11437: * 15 i=8 1 2 2 2
11438: * 16 2 2 2 2
1.143 brouard 11439: */
1.212 brouard 11440: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11441: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11442: * and the value of each covariate?
11443: * V1=1, V2=1, V3=2, V4=1 ?
11444: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11445: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11446: * In order to get the real value in the data, we use nbcode
11447: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11448: * We are keeping this crazy system in order to be able (in the future?)
11449: * to have more than 2 values (0 or 1) for a covariate.
11450: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11451: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11452: * bbbbbbbb
11453: * 76543210
11454: * h-1 00000101 (6-1=5)
1.219 brouard 11455: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11456: * &
11457: * 1 00000001 (1)
1.219 brouard 11458: * 00000000 = 1 & ((h-1) >> (k-1))
11459: * +1= 00000001 =1
1.211 brouard 11460: *
11461: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11462: * h' 1101 =2^3+2^2+0x2^1+2^0
11463: * >>k' 11
11464: * & 00000001
11465: * = 00000001
11466: * +1 = 00000010=2 = codtabm(14,3)
11467: * Reverse h=6 and m=16?
11468: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11469: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11470: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11471: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11472: * V3=decodtabm(14,3,2**4)=2
11473: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11474: *(h-1) >> (j-1) 0011 =13 >> 2
11475: * &1 000000001
11476: * = 000000001
11477: * +1= 000000010 =2
11478: * 2211
11479: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11480: * V3=2
1.220 brouard 11481: * codtabm and decodtabm are identical
1.211 brouard 11482: */
11483:
1.145 brouard 11484:
11485: free_ivector(Ndum,-1,NCOVMAX);
11486:
11487:
1.126 brouard 11488:
1.186 brouard 11489: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11490: strcpy(optionfilegnuplot,optionfilefiname);
11491: if(mle==-3)
1.201 brouard 11492: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11493: strcat(optionfilegnuplot,".gp");
11494:
11495: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11496: printf("Problem with file %s",optionfilegnuplot);
11497: }
11498: else{
1.204 brouard 11499: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11500: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11501: //fprintf(ficgp,"set missing 'NaNq'\n");
11502: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11503: }
11504: /* fclose(ficgp);*/
1.186 brouard 11505:
11506:
11507: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11508:
11509: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11510: if(mle==-3)
1.201 brouard 11511: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11512: strcat(optionfilehtm,".htm");
11513: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11514: printf("Problem with %s \n",optionfilehtm);
11515: exit(0);
1.126 brouard 11516: }
11517:
11518: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11519: strcat(optionfilehtmcov,"-cov.htm");
11520: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11521: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11522: }
11523: else{
11524: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
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: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11528: }
11529:
1.213 brouard 11530: 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 11531: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11532: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11533: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11534: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11535: \n\
11536: <hr size=\"2\" color=\"#EC5E5E\">\
11537: <ul><li><h4>Parameter files</h4>\n\
11538: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11539: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11540: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11541: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11542: - Date and time at start: %s</ul>\n",\
11543: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11544: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11545: fileres,fileres,\
11546: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11547: fflush(fichtm);
11548:
11549: strcpy(pathr,path);
11550: strcat(pathr,optionfilefiname);
1.184 brouard 11551: #ifdef WIN32
11552: _chdir(optionfilefiname); /* Move to directory named optionfile */
11553: #else
1.126 brouard 11554: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11555: #endif
11556:
1.126 brouard 11557:
1.220 brouard 11558: /* Calculates basic frequencies. Computes observed prevalence at single age
11559: and for any valid combination of covariates
1.126 brouard 11560: and prints on file fileres'p'. */
1.251 brouard 11561: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11562: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11563:
11564: fprintf(fichtm,"\n");
1.274 brouard 11565: 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",\
11566: ftol, stepm);
11567: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11568: ncurrv=1;
11569: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11570: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11571: ncurrv=i;
11572: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
11573: fprintf(fichtm,"\n<li> Number of time varying (wave varying) covariates: ntv=%d ", ntv);
11574: ncurrv=i;
11575: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
11576: fprintf(fichtm,"\n<li>Number of quantitative time varying covariates: nqtv=%d ", nqtv);
11577: ncurrv=i;
11578: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11579: 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", \
11580: nlstate, ndeath, maxwav, mle, weightopt);
11581:
11582: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11583: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11584:
11585:
11586: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11587: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11588: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11589: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11590: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11591: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11592: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11593: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11594: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11595:
1.126 brouard 11596: /* For Powell, parameters are in a vector p[] starting at p[1]
11597: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11598: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11599:
11600: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11601: /* For mortality only */
1.126 brouard 11602: if (mle==-3){
1.136 brouard 11603: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11604: for(i=1;i<=NDIM;i++)
11605: for(j=1;j<=NDIM;j++)
11606: ximort[i][j]=0.;
1.186 brouard 11607: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11608: cens=ivector(1,n);
11609: ageexmed=vector(1,n);
11610: agecens=vector(1,n);
11611: dcwave=ivector(1,n);
1.223 brouard 11612:
1.126 brouard 11613: for (i=1; i<=imx; i++){
11614: dcwave[i]=-1;
11615: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11616: if (s[m][i]>nlstate) {
11617: dcwave[i]=m;
11618: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11619: break;
11620: }
1.126 brouard 11621: }
1.226 brouard 11622:
1.126 brouard 11623: for (i=1; i<=imx; i++) {
11624: if (wav[i]>0){
1.226 brouard 11625: ageexmed[i]=agev[mw[1][i]][i];
11626: j=wav[i];
11627: agecens[i]=1.;
11628:
11629: if (ageexmed[i]> 1 && wav[i] > 0){
11630: agecens[i]=agev[mw[j][i]][i];
11631: cens[i]= 1;
11632: }else if (ageexmed[i]< 1)
11633: cens[i]= -1;
11634: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11635: cens[i]=0 ;
1.126 brouard 11636: }
11637: else cens[i]=-1;
11638: }
11639:
11640: for (i=1;i<=NDIM;i++) {
11641: for (j=1;j<=NDIM;j++)
1.226 brouard 11642: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11643: }
11644:
1.145 brouard 11645: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11646: /*printf("%lf %lf", p[1], p[2]);*/
11647:
11648:
1.136 brouard 11649: #ifdef GSL
11650: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11651: #else
1.126 brouard 11652: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11653: #endif
1.201 brouard 11654: strcpy(filerespow,"POW-MORT_");
11655: strcat(filerespow,fileresu);
1.126 brouard 11656: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11657: printf("Problem with resultfile: %s\n", filerespow);
11658: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11659: }
1.136 brouard 11660: #ifdef GSL
11661: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11662: #else
1.126 brouard 11663: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11664: #endif
1.126 brouard 11665: /* for (i=1;i<=nlstate;i++)
11666: for(j=1;j<=nlstate+ndeath;j++)
11667: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11668: */
11669: fprintf(ficrespow,"\n");
1.136 brouard 11670: #ifdef GSL
11671: /* gsl starts here */
11672: T = gsl_multimin_fminimizer_nmsimplex;
11673: gsl_multimin_fminimizer *sfm = NULL;
11674: gsl_vector *ss, *x;
11675: gsl_multimin_function minex_func;
11676:
11677: /* Initial vertex size vector */
11678: ss = gsl_vector_alloc (NDIM);
11679:
11680: if (ss == NULL){
11681: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11682: }
11683: /* Set all step sizes to 1 */
11684: gsl_vector_set_all (ss, 0.001);
11685:
11686: /* Starting point */
1.126 brouard 11687:
1.136 brouard 11688: x = gsl_vector_alloc (NDIM);
11689:
11690: if (x == NULL){
11691: gsl_vector_free(ss);
11692: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11693: }
11694:
11695: /* Initialize method and iterate */
11696: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11697: /* gsl_vector_set(x, 0, 0.0268); */
11698: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11699: gsl_vector_set(x, 0, p[1]);
11700: gsl_vector_set(x, 1, p[2]);
11701:
11702: minex_func.f = &gompertz_f;
11703: minex_func.n = NDIM;
11704: minex_func.params = (void *)&p; /* ??? */
11705:
11706: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11707: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11708:
11709: printf("Iterations beginning .....\n\n");
11710: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11711:
11712: iteri=0;
11713: while (rval == GSL_CONTINUE){
11714: iteri++;
11715: status = gsl_multimin_fminimizer_iterate(sfm);
11716:
11717: if (status) printf("error: %s\n", gsl_strerror (status));
11718: fflush(0);
11719:
11720: if (status)
11721: break;
11722:
11723: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11724: ssval = gsl_multimin_fminimizer_size (sfm);
11725:
11726: if (rval == GSL_SUCCESS)
11727: printf ("converged to a local maximum at\n");
11728:
11729: printf("%5d ", iteri);
11730: for (it = 0; it < NDIM; it++){
11731: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11732: }
11733: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11734: }
11735:
11736: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11737:
11738: gsl_vector_free(x); /* initial values */
11739: gsl_vector_free(ss); /* inital step size */
11740: for (it=0; it<NDIM; it++){
11741: p[it+1]=gsl_vector_get(sfm->x,it);
11742: fprintf(ficrespow," %.12lf", p[it]);
11743: }
11744: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11745: #endif
11746: #ifdef POWELL
11747: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11748: #endif
1.126 brouard 11749: fclose(ficrespow);
11750:
1.203 brouard 11751: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11752:
11753: for(i=1; i <=NDIM; i++)
11754: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11755: matcov[i][j]=matcov[j][i];
1.126 brouard 11756:
11757: printf("\nCovariance matrix\n ");
1.203 brouard 11758: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11759: for(i=1; i <=NDIM; i++) {
11760: for(j=1;j<=NDIM;j++){
1.220 brouard 11761: printf("%f ",matcov[i][j]);
11762: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11763: }
1.203 brouard 11764: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11765: }
11766:
11767: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11768: for (i=1;i<=NDIM;i++) {
1.126 brouard 11769: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11770: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11771: }
1.126 brouard 11772: lsurv=vector(1,AGESUP);
11773: lpop=vector(1,AGESUP);
11774: tpop=vector(1,AGESUP);
11775: lsurv[agegomp]=100000;
11776:
11777: for (k=agegomp;k<=AGESUP;k++) {
11778: agemortsup=k;
11779: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11780: }
11781:
11782: for (k=agegomp;k<agemortsup;k++)
11783: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11784:
11785: for (k=agegomp;k<agemortsup;k++){
11786: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11787: sumlpop=sumlpop+lpop[k];
11788: }
11789:
11790: tpop[agegomp]=sumlpop;
11791: for (k=agegomp;k<(agemortsup-3);k++){
11792: /* tpop[k+1]=2;*/
11793: tpop[k+1]=tpop[k]-lpop[k];
11794: }
11795:
11796:
11797: printf("\nAge lx qx dx Lx Tx e(x)\n");
11798: for (k=agegomp;k<(agemortsup-2);k++)
11799: 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]);
11800:
11801:
11802: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11803: ageminpar=50;
11804: agemaxpar=100;
1.194 brouard 11805: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11806: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11807: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11808: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11809: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11810: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11811: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11812: }else{
11813: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11814: 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 11815: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11816: }
1.201 brouard 11817: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11818: stepm, weightopt,\
11819: model,imx,p,matcov,agemortsup);
11820:
11821: free_vector(lsurv,1,AGESUP);
11822: free_vector(lpop,1,AGESUP);
11823: free_vector(tpop,1,AGESUP);
1.220 brouard 11824: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11825: free_ivector(cens,1,n);
11826: free_vector(agecens,1,n);
11827: free_ivector(dcwave,1,n);
1.220 brouard 11828: #ifdef GSL
1.136 brouard 11829: #endif
1.186 brouard 11830: } /* Endof if mle==-3 mortality only */
1.205 brouard 11831: /* Standard */
11832: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11833: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11834: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11835: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11836: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11837: for (k=1; k<=npar;k++)
11838: printf(" %d %8.5f",k,p[k]);
11839: printf("\n");
1.205 brouard 11840: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11841: /* mlikeli uses func not funcone */
1.247 brouard 11842: /* for(i=1;i<nlstate;i++){ */
11843: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11844: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11845: /* } */
1.205 brouard 11846: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11847: }
11848: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11849: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11850: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11851: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11852: }
11853: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11854: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11855: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11856: for (k=1; k<=npar;k++)
11857: printf(" %d %8.5f",k,p[k]);
11858: printf("\n");
11859:
11860: /*--------- results files --------------*/
1.224 brouard 11861: 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 11862:
11863:
11864: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11865: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11866: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11867: for(i=1,jk=1; i <=nlstate; i++){
11868: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11869: if (k != i) {
11870: printf("%d%d ",i,k);
11871: fprintf(ficlog,"%d%d ",i,k);
11872: fprintf(ficres,"%1d%1d ",i,k);
11873: for(j=1; j <=ncovmodel; j++){
11874: printf("%12.7f ",p[jk]);
11875: fprintf(ficlog,"%12.7f ",p[jk]);
11876: fprintf(ficres,"%12.7f ",p[jk]);
11877: jk++;
11878: }
11879: printf("\n");
11880: fprintf(ficlog,"\n");
11881: fprintf(ficres,"\n");
11882: }
1.126 brouard 11883: }
11884: }
1.203 brouard 11885: if(mle != 0){
11886: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11887: ftolhess=ftol; /* Usually correct */
1.203 brouard 11888: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11889: 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");
11890: 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");
11891: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11892: for(k=1; k <=(nlstate+ndeath); k++){
11893: if (k != i) {
11894: printf("%d%d ",i,k);
11895: fprintf(ficlog,"%d%d ",i,k);
11896: for(j=1; j <=ncovmodel; j++){
11897: 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]));
11898: 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]));
11899: jk++;
11900: }
11901: printf("\n");
11902: fprintf(ficlog,"\n");
11903: }
11904: }
1.193 brouard 11905: }
1.203 brouard 11906: } /* end of hesscov and Wald tests */
1.225 brouard 11907:
1.203 brouard 11908: /* */
1.126 brouard 11909: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11910: printf("# Scales (for hessian or gradient estimation)\n");
11911: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11912: for(i=1,jk=1; i <=nlstate; i++){
11913: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11914: if (j!=i) {
11915: fprintf(ficres,"%1d%1d",i,j);
11916: printf("%1d%1d",i,j);
11917: fprintf(ficlog,"%1d%1d",i,j);
11918: for(k=1; k<=ncovmodel;k++){
11919: printf(" %.5e",delti[jk]);
11920: fprintf(ficlog," %.5e",delti[jk]);
11921: fprintf(ficres," %.5e",delti[jk]);
11922: jk++;
11923: }
11924: printf("\n");
11925: fprintf(ficlog,"\n");
11926: fprintf(ficres,"\n");
11927: }
1.126 brouard 11928: }
11929: }
11930:
11931: 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 11932: if(mle >= 1) /* To big for the screen */
1.126 brouard 11933: 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");
11934: 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");
11935: /* # 121 Var(a12)\n\ */
11936: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11937: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11938: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11939: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11940: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11941: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11942: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11943:
11944:
11945: /* Just to have a covariance matrix which will be more understandable
11946: even is we still don't want to manage dictionary of variables
11947: */
11948: for(itimes=1;itimes<=2;itimes++){
11949: jj=0;
11950: for(i=1; i <=nlstate; i++){
1.225 brouard 11951: for(j=1; j <=nlstate+ndeath; j++){
11952: if(j==i) continue;
11953: for(k=1; k<=ncovmodel;k++){
11954: jj++;
11955: ca[0]= k+'a'-1;ca[1]='\0';
11956: if(itimes==1){
11957: if(mle>=1)
11958: printf("#%1d%1d%d",i,j,k);
11959: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11960: fprintf(ficres,"#%1d%1d%d",i,j,k);
11961: }else{
11962: if(mle>=1)
11963: printf("%1d%1d%d",i,j,k);
11964: fprintf(ficlog,"%1d%1d%d",i,j,k);
11965: fprintf(ficres,"%1d%1d%d",i,j,k);
11966: }
11967: ll=0;
11968: for(li=1;li <=nlstate; li++){
11969: for(lj=1;lj <=nlstate+ndeath; lj++){
11970: if(lj==li) continue;
11971: for(lk=1;lk<=ncovmodel;lk++){
11972: ll++;
11973: if(ll<=jj){
11974: cb[0]= lk +'a'-1;cb[1]='\0';
11975: if(ll<jj){
11976: if(itimes==1){
11977: if(mle>=1)
11978: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11979: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11980: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11981: }else{
11982: if(mle>=1)
11983: printf(" %.5e",matcov[jj][ll]);
11984: fprintf(ficlog," %.5e",matcov[jj][ll]);
11985: fprintf(ficres," %.5e",matcov[jj][ll]);
11986: }
11987: }else{
11988: if(itimes==1){
11989: if(mle>=1)
11990: printf(" Var(%s%1d%1d)",ca,i,j);
11991: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11992: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11993: }else{
11994: if(mle>=1)
11995: printf(" %.7e",matcov[jj][ll]);
11996: fprintf(ficlog," %.7e",matcov[jj][ll]);
11997: fprintf(ficres," %.7e",matcov[jj][ll]);
11998: }
11999: }
12000: }
12001: } /* end lk */
12002: } /* end lj */
12003: } /* end li */
12004: if(mle>=1)
12005: printf("\n");
12006: fprintf(ficlog,"\n");
12007: fprintf(ficres,"\n");
12008: numlinepar++;
12009: } /* end k*/
12010: } /*end j */
1.126 brouard 12011: } /* end i */
12012: } /* end itimes */
12013:
12014: fflush(ficlog);
12015: fflush(ficres);
1.225 brouard 12016: while(fgets(line, MAXLINE, ficpar)) {
12017: /* If line starts with a # it is a comment */
12018: if (line[0] == '#') {
12019: numlinepar++;
12020: fputs(line,stdout);
12021: fputs(line,ficparo);
12022: fputs(line,ficlog);
12023: continue;
12024: }else
12025: break;
12026: }
12027:
1.209 brouard 12028: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12029: /* ungetc(c,ficpar); */
12030: /* fgets(line, MAXLINE, ficpar); */
12031: /* fputs(line,stdout); */
12032: /* fputs(line,ficparo); */
12033: /* } */
12034: /* ungetc(c,ficpar); */
1.126 brouard 12035:
12036: estepm=0;
1.209 brouard 12037: 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 12038:
12039: if (num_filled != 6) {
12040: 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);
12041: 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);
12042: goto end;
12043: }
12044: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12045: }
12046: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12047: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12048:
1.209 brouard 12049: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12050: if (estepm==0 || estepm < stepm) estepm=stepm;
12051: if (fage <= 2) {
12052: bage = ageminpar;
12053: fage = agemaxpar;
12054: }
12055:
12056: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12057: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12058: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12059:
1.186 brouard 12060: /* Other stuffs, more or less useful */
1.254 brouard 12061: while(fgets(line, MAXLINE, ficpar)) {
12062: /* If line starts with a # it is a comment */
12063: if (line[0] == '#') {
12064: numlinepar++;
12065: fputs(line,stdout);
12066: fputs(line,ficparo);
12067: fputs(line,ficlog);
12068: continue;
12069: }else
12070: break;
12071: }
12072:
12073: 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){
12074:
12075: if (num_filled != 7) {
12076: 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);
12077: 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);
12078: goto end;
12079: }
12080: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12081: 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);
12082: 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);
12083: 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 12084: }
1.254 brouard 12085:
12086: while(fgets(line, MAXLINE, ficpar)) {
12087: /* If line starts with a # it is a comment */
12088: if (line[0] == '#') {
12089: numlinepar++;
12090: fputs(line,stdout);
12091: fputs(line,ficparo);
12092: fputs(line,ficlog);
12093: continue;
12094: }else
12095: break;
1.126 brouard 12096: }
12097:
12098:
12099: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12100: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12101:
1.254 brouard 12102: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12103: if (num_filled != 1) {
12104: 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);
12105: 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);
12106: goto end;
12107: }
12108: printf("pop_based=%d\n",popbased);
12109: fprintf(ficlog,"pop_based=%d\n",popbased);
12110: fprintf(ficparo,"pop_based=%d\n",popbased);
12111: fprintf(ficres,"pop_based=%d\n",popbased);
12112: }
12113:
1.258 brouard 12114: /* Results */
12115: nresult=0;
12116: do{
12117: if(!fgets(line, MAXLINE, ficpar)){
12118: endishere=1;
12119: parameterline=14;
12120: }else if (line[0] == '#') {
12121: /* If line starts with a # it is a comment */
1.254 brouard 12122: numlinepar++;
12123: fputs(line,stdout);
12124: fputs(line,ficparo);
12125: fputs(line,ficlog);
12126: continue;
1.258 brouard 12127: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12128: parameterline=11;
12129: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12130: parameterline=12;
12131: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12132: parameterline=13;
12133: else{
12134: parameterline=14;
1.254 brouard 12135: }
1.258 brouard 12136: switch (parameterline){
12137: case 11:
12138: 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){
12139: if (num_filled != 8) {
12140: 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);
12141: 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);
12142: goto end;
12143: }
12144: 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);
12145: 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);
12146: 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);
12147: 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);
12148: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12149: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12150: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12151:
1.258 brouard 12152: }
1.254 brouard 12153: break;
1.258 brouard 12154: case 12:
12155: /*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);*/
12156: 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){
12157: if (num_filled != 8) {
1.262 brouard 12158: 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);
12159: 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 12160: goto end;
12161: }
12162: 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);
12163: 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);
12164: 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);
12165: 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);
12166: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12167: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12168: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12169: }
1.230 brouard 12170: break;
1.258 brouard 12171: case 13:
12172: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12173: if (num_filled == 0){
12174: resultline[0]='\0';
12175: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12176: 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);
12177: break;
12178: } else if (num_filled != 1){
12179: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12180: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12181: }
12182: nresult++; /* Sum of resultlines */
12183: printf("Result %d: result=%s\n",nresult, resultline);
12184: if(nresult > MAXRESULTLINES){
12185: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12186: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12187: goto end;
12188: }
12189: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12190: fprintf(ficparo,"result: %s\n",resultline);
12191: fprintf(ficres,"result: %s\n",resultline);
12192: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12193: break;
1.258 brouard 12194: case 14:
1.259 brouard 12195: if(ncovmodel >2 && nresult==0 ){
12196: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12197: goto end;
12198: }
1.259 brouard 12199: break;
1.258 brouard 12200: default:
12201: nresult=1;
12202: decoderesult(".",nresult ); /* No covariate */
12203: }
12204: } /* End switch parameterline */
12205: }while(endishere==0); /* End do */
1.126 brouard 12206:
1.230 brouard 12207: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12208: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12209:
12210: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12211: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12212: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12213: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12214: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12215: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12216: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12217: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12218: }else{
1.270 brouard 12219: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12220: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12221: }
12222: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12223: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12224: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12225:
1.225 brouard 12226: /*------------ free_vector -------------*/
12227: /* chdir(path); */
1.220 brouard 12228:
1.215 brouard 12229: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12230: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12231: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12232: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12233: free_lvector(num,1,n);
12234: free_vector(agedc,1,n);
12235: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12236: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12237: fclose(ficparo);
12238: fclose(ficres);
1.220 brouard 12239:
12240:
1.186 brouard 12241: /* Other results (useful)*/
1.220 brouard 12242:
12243:
1.126 brouard 12244: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12245: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12246: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12247: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12248: fclose(ficrespl);
12249:
12250: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12251: /*#include "hpijx.h"*/
12252: hPijx(p, bage, fage);
1.145 brouard 12253: fclose(ficrespij);
1.227 brouard 12254:
1.220 brouard 12255: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12256: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12257: k=1;
1.126 brouard 12258: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12259:
1.269 brouard 12260: /* Prevalence for each covariate combination in probs[age][status][cov] */
12261: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12262: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12263: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12264: for(k=1;k<=ncovcombmax;k++)
12265: probs[i][j][k]=0.;
1.269 brouard 12266: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12267: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12268: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12269: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12270: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12271: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12272: for(k=1;k<=ncovcombmax;k++)
12273: mobaverages[i][j][k]=0.;
1.219 brouard 12274: mobaverage=mobaverages;
12275: if (mobilav!=0) {
1.235 brouard 12276: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12277: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12278: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12279: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12280: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12281: }
1.269 brouard 12282: } else if (mobilavproj !=0) {
1.235 brouard 12283: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12284: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12285: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12286: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12287: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12288: }
1.269 brouard 12289: }else{
12290: printf("Internal error moving average\n");
12291: fflush(stdout);
12292: exit(1);
1.219 brouard 12293: }
12294: }/* end if moving average */
1.227 brouard 12295:
1.126 brouard 12296: /*---------- Forecasting ------------------*/
12297: if(prevfcast==1){
12298: /* if(stepm ==1){*/
1.269 brouard 12299: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12300: }
1.269 brouard 12301:
12302: /* Backcasting */
1.217 brouard 12303: if(backcast==1){
1.219 brouard 12304: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12305: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12306: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12307:
12308: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12309:
12310: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12311:
1.219 brouard 12312: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12313: fclose(ficresplb);
12314:
1.222 brouard 12315: hBijx(p, bage, fage, mobaverage);
12316: fclose(ficrespijb);
1.219 brouard 12317:
1.269 brouard 12318: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12319: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12320: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12321:
12322:
1.269 brouard 12323: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12324: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12325: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12326: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12327: } /* end Backcasting */
1.268 brouard 12328:
1.186 brouard 12329:
12330: /* ------ Other prevalence ratios------------ */
1.126 brouard 12331:
1.215 brouard 12332: free_ivector(wav,1,imx);
12333: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12334: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12335: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12336:
12337:
1.127 brouard 12338: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12339:
1.201 brouard 12340: strcpy(filerese,"E_");
12341: strcat(filerese,fileresu);
1.126 brouard 12342: if((ficreseij=fopen(filerese,"w"))==NULL) {
12343: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12344: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12345: }
1.208 brouard 12346: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12347: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12348:
12349: pstamp(ficreseij);
1.219 brouard 12350:
1.235 brouard 12351: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12352: if (cptcovn < 1){i1=1;}
12353:
12354: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12355: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12356: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12357: continue;
1.219 brouard 12358: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12359: printf("\n#****** ");
1.225 brouard 12360: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12361: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12362: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12363: }
12364: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12365: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12366: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12367: }
12368: fprintf(ficreseij,"******\n");
1.235 brouard 12369: printf("******\n");
1.219 brouard 12370:
12371: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12372: oldm=oldms;savm=savms;
1.235 brouard 12373: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12374:
1.219 brouard 12375: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12376: }
12377: fclose(ficreseij);
1.208 brouard 12378: printf("done evsij\n");fflush(stdout);
12379: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12380:
1.218 brouard 12381:
1.227 brouard 12382: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12383:
1.201 brouard 12384: strcpy(filerest,"T_");
12385: strcat(filerest,fileresu);
1.127 brouard 12386: if((ficrest=fopen(filerest,"w"))==NULL) {
12387: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12388: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12389: }
1.208 brouard 12390: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12391: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12392: strcpy(fileresstde,"STDE_");
12393: strcat(fileresstde,fileresu);
1.126 brouard 12394: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12395: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12396: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12397: }
1.227 brouard 12398: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12399: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12400:
1.201 brouard 12401: strcpy(filerescve,"CVE_");
12402: strcat(filerescve,fileresu);
1.126 brouard 12403: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12404: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12405: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12406: }
1.227 brouard 12407: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12408: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12409:
1.201 brouard 12410: strcpy(fileresv,"V_");
12411: strcat(fileresv,fileresu);
1.126 brouard 12412: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12413: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12414: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12415: }
1.227 brouard 12416: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12417: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12418:
1.235 brouard 12419: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12420: if (cptcovn < 1){i1=1;}
12421:
12422: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12423: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12424: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12425: continue;
1.242 brouard 12426: printf("\n#****** Result for:");
12427: fprintf(ficrest,"\n#****** Result for:");
12428: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12429: for(j=1;j<=cptcoveff;j++){
12430: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12431: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12432: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12433: }
1.235 brouard 12434: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12435: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12436: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12437: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12438: }
1.208 brouard 12439: fprintf(ficrest,"******\n");
1.227 brouard 12440: fprintf(ficlog,"******\n");
12441: printf("******\n");
1.208 brouard 12442:
12443: fprintf(ficresstdeij,"\n#****** ");
12444: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12445: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12446: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12447: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12448: }
1.235 brouard 12449: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12450: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12451: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12452: }
1.208 brouard 12453: fprintf(ficresstdeij,"******\n");
12454: fprintf(ficrescveij,"******\n");
12455:
12456: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12457: /* pstamp(ficresvij); */
1.225 brouard 12458: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12459: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12460: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12461: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12462: }
1.208 brouard 12463: fprintf(ficresvij,"******\n");
12464:
12465: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12466: oldm=oldms;savm=savms;
1.235 brouard 12467: printf(" cvevsij ");
12468: fprintf(ficlog, " cvevsij ");
12469: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12470: printf(" end cvevsij \n ");
12471: fprintf(ficlog, " end cvevsij \n ");
12472:
12473: /*
12474: */
12475: /* goto endfree; */
12476:
12477: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12478: pstamp(ficrest);
12479:
1.269 brouard 12480: epj=vector(1,nlstate+1);
1.208 brouard 12481: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12482: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12483: cptcod= 0; /* To be deleted */
12484: printf("varevsij vpopbased=%d \n",vpopbased);
12485: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12486: 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 12487: 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 ");
12488: if(vpopbased==1)
12489: 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);
12490: else
12491: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12492: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12493: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12494: fprintf(ficrest,"\n");
12495: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12496: printf("Computing age specific period (stable) prevalences in each health state \n");
12497: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12498: for(age=bage; age <=fage ;age++){
1.235 brouard 12499: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12500: if (vpopbased==1) {
12501: if(mobilav ==0){
12502: for(i=1; i<=nlstate;i++)
12503: prlim[i][i]=probs[(int)age][i][k];
12504: }else{ /* mobilav */
12505: for(i=1; i<=nlstate;i++)
12506: prlim[i][i]=mobaverage[(int)age][i][k];
12507: }
12508: }
1.219 brouard 12509:
1.227 brouard 12510: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12511: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12512: /* printf(" age %4.0f ",age); */
12513: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12514: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12515: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12516: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12517: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12518: }
12519: epj[nlstate+1] +=epj[j];
12520: }
12521: /* printf(" age %4.0f \n",age); */
1.219 brouard 12522:
1.227 brouard 12523: for(i=1, vepp=0.;i <=nlstate;i++)
12524: for(j=1;j <=nlstate;j++)
12525: vepp += vareij[i][j][(int)age];
12526: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12527: for(j=1;j <=nlstate;j++){
12528: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12529: }
12530: fprintf(ficrest,"\n");
12531: }
1.208 brouard 12532: } /* End vpopbased */
1.269 brouard 12533: free_vector(epj,1,nlstate+1);
1.208 brouard 12534: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12535: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12536: printf("done selection\n");fflush(stdout);
12537: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12538:
1.235 brouard 12539: } /* End k selection */
1.227 brouard 12540:
12541: printf("done State-specific expectancies\n");fflush(stdout);
12542: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12543:
1.269 brouard 12544: /* variance-covariance of period prevalence*/
12545: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12546:
1.227 brouard 12547:
12548: free_vector(weight,1,n);
12549: free_imatrix(Tvard,1,NCOVMAX,1,2);
12550: free_imatrix(s,1,maxwav+1,1,n);
12551: free_matrix(anint,1,maxwav,1,n);
12552: free_matrix(mint,1,maxwav,1,n);
12553: free_ivector(cod,1,n);
12554: free_ivector(tab,1,NCOVMAX);
12555: fclose(ficresstdeij);
12556: fclose(ficrescveij);
12557: fclose(ficresvij);
12558: fclose(ficrest);
12559: fclose(ficpar);
12560:
12561:
1.126 brouard 12562: /*---------- End : free ----------------*/
1.219 brouard 12563: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12564: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12565: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12566: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12567: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12568: } /* mle==-3 arrives here for freeing */
1.227 brouard 12569: /* endfree:*/
12570: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12571: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12572: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12573: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12574: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12575: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12576: free_matrix(covar,0,NCOVMAX,1,n);
12577: free_matrix(matcov,1,npar,1,npar);
12578: free_matrix(hess,1,npar,1,npar);
12579: /*free_vector(delti,1,npar);*/
12580: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12581: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12582: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12583: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12584:
12585: free_ivector(ncodemax,1,NCOVMAX);
12586: free_ivector(ncodemaxwundef,1,NCOVMAX);
12587: free_ivector(Dummy,-1,NCOVMAX);
12588: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12589: free_ivector(DummyV,1,NCOVMAX);
12590: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12591: free_ivector(Typevar,-1,NCOVMAX);
12592: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12593: free_ivector(TvarsQ,1,NCOVMAX);
12594: free_ivector(TvarsQind,1,NCOVMAX);
12595: free_ivector(TvarsD,1,NCOVMAX);
12596: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12597: free_ivector(TvarFD,1,NCOVMAX);
12598: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12599: free_ivector(TvarF,1,NCOVMAX);
12600: free_ivector(TvarFind,1,NCOVMAX);
12601: free_ivector(TvarV,1,NCOVMAX);
12602: free_ivector(TvarVind,1,NCOVMAX);
12603: free_ivector(TvarA,1,NCOVMAX);
12604: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12605: free_ivector(TvarFQ,1,NCOVMAX);
12606: free_ivector(TvarFQind,1,NCOVMAX);
12607: free_ivector(TvarVD,1,NCOVMAX);
12608: free_ivector(TvarVDind,1,NCOVMAX);
12609: free_ivector(TvarVQ,1,NCOVMAX);
12610: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12611: free_ivector(Tvarsel,1,NCOVMAX);
12612: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12613: free_ivector(Tposprod,1,NCOVMAX);
12614: free_ivector(Tprod,1,NCOVMAX);
12615: free_ivector(Tvaraff,1,NCOVMAX);
12616: free_ivector(invalidvarcomb,1,ncovcombmax);
12617: free_ivector(Tage,1,NCOVMAX);
12618: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12619: free_ivector(TmodelInvind,1,NCOVMAX);
12620: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12621:
12622: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12623: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12624: fflush(fichtm);
12625: fflush(ficgp);
12626:
1.227 brouard 12627:
1.126 brouard 12628: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12629: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12630: 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 12631: }else{
12632: printf("End of Imach\n");
12633: fprintf(ficlog,"End of Imach\n");
12634: }
12635: printf("See log file on %s\n",filelog);
12636: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12637: /*(void) gettimeofday(&end_time,&tzp);*/
12638: rend_time = time(NULL);
12639: end_time = *localtime(&rend_time);
12640: /* tml = *localtime(&end_time.tm_sec); */
12641: strcpy(strtend,asctime(&end_time));
1.126 brouard 12642: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12643: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12644: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12645:
1.157 brouard 12646: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12647: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12648: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12649: /* printf("Total time was %d uSec.\n", total_usecs);*/
12650: /* if(fileappend(fichtm,optionfilehtm)){ */
12651: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12652: fclose(fichtm);
12653: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12654: fclose(fichtmcov);
12655: fclose(ficgp);
12656: fclose(ficlog);
12657: /*------ End -----------*/
1.227 brouard 12658:
12659:
12660: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12661: #ifdef WIN32
1.227 brouard 12662: if (_chdir(pathcd) != 0)
12663: printf("Can't move to directory %s!\n",path);
12664: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12665: #else
1.227 brouard 12666: if(chdir(pathcd) != 0)
12667: printf("Can't move to directory %s!\n", path);
12668: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12669: #endif
1.126 brouard 12670: printf("Current directory %s!\n",pathcd);
12671: /*strcat(plotcmd,CHARSEPARATOR);*/
12672: sprintf(plotcmd,"gnuplot");
1.157 brouard 12673: #ifdef _WIN32
1.126 brouard 12674: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12675: #endif
12676: if(!stat(plotcmd,&info)){
1.158 brouard 12677: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12678: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12679: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12680: }else
12681: strcpy(pplotcmd,plotcmd);
1.157 brouard 12682: #ifdef __unix
1.126 brouard 12683: strcpy(plotcmd,GNUPLOTPROGRAM);
12684: if(!stat(plotcmd,&info)){
1.158 brouard 12685: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12686: }else
12687: strcpy(pplotcmd,plotcmd);
12688: #endif
12689: }else
12690: strcpy(pplotcmd,plotcmd);
12691:
12692: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12693: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12694:
1.126 brouard 12695: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12696: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12697: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12698: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12699: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12700: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12701: }
1.158 brouard 12702: printf(" Successful, please wait...");
1.126 brouard 12703: while (z[0] != 'q') {
12704: /* chdir(path); */
1.154 brouard 12705: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12706: scanf("%s",z);
12707: /* if (z[0] == 'c') system("./imach"); */
12708: if (z[0] == 'e') {
1.158 brouard 12709: #ifdef __APPLE__
1.152 brouard 12710: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12711: #elif __linux
12712: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12713: #else
1.152 brouard 12714: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12715: #endif
12716: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12717: system(pplotcmd);
1.126 brouard 12718: }
12719: else if (z[0] == 'g') system(plotcmd);
12720: else if (z[0] == 'q') exit(0);
12721: }
1.227 brouard 12722: end:
1.126 brouard 12723: while (z[0] != 'q') {
1.195 brouard 12724: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12725: scanf("%s",z);
12726: }
12727: }
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