Annotation of imach/src/imach.c, revision 1.275
1.275 ! brouard 1: /* $Id: imach.c,v 1.274 2017/06/29 09:47:08 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.275 ! brouard 4: Revision 1.274 2017/06/29 09:47:08 brouard
! 5: Summary: Version 0.99r14
! 6:
1.274 brouard 7: Revision 1.273 2017/06/27 11:06:02 brouard
8: Summary: More documentation on projections
9:
1.273 brouard 10: Revision 1.272 2017/06/27 10:22:40 brouard
11: Summary: Color of backprojection changed from 6 to 5(yellow)
12:
1.272 brouard 13: Revision 1.271 2017/06/27 10:17:50 brouard
14: Summary: Some bug with rint
15:
1.271 brouard 16: Revision 1.270 2017/05/24 05:45:29 brouard
17: *** empty log message ***
18:
1.270 brouard 19: Revision 1.269 2017/05/23 08:39:25 brouard
20: Summary: Code into subroutine, cleanings
21:
1.269 brouard 22: Revision 1.268 2017/05/18 20:09:32 brouard
23: Summary: backprojection and confidence intervals of backprevalence
24:
1.268 brouard 25: Revision 1.267 2017/05/13 10:25:05 brouard
26: Summary: temporary save for backprojection
27:
1.267 brouard 28: Revision 1.266 2017/05/13 07:26:12 brouard
29: Summary: Version 0.99r13 (improvements and bugs fixed)
30:
1.266 brouard 31: Revision 1.265 2017/04/26 16:22:11 brouard
32: Summary: imach 0.99r13 Some bugs fixed
33:
1.265 brouard 34: Revision 1.264 2017/04/26 06:01:29 brouard
35: Summary: Labels in graphs
36:
1.264 brouard 37: Revision 1.263 2017/04/24 15:23:15 brouard
38: Summary: to save
39:
1.263 brouard 40: Revision 1.262 2017/04/18 16:48:12 brouard
41: *** empty log message ***
42:
1.262 brouard 43: Revision 1.261 2017/04/05 10:14:09 brouard
44: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
45:
1.261 brouard 46: Revision 1.260 2017/04/04 17:46:59 brouard
47: Summary: Gnuplot indexations fixed (humm)
48:
1.260 brouard 49: Revision 1.259 2017/04/04 13:01:16 brouard
50: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
51:
1.259 brouard 52: Revision 1.258 2017/04/03 10:17:47 brouard
53: Summary: Version 0.99r12
54:
55: Some cleanings, conformed with updated documentation.
56:
1.258 brouard 57: Revision 1.257 2017/03/29 16:53:30 brouard
58: Summary: Temp
59:
1.257 brouard 60: Revision 1.256 2017/03/27 05:50:23 brouard
61: Summary: Temporary
62:
1.256 brouard 63: Revision 1.255 2017/03/08 16:02:28 brouard
64: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
65:
1.255 brouard 66: Revision 1.254 2017/03/08 07:13:00 brouard
67: Summary: Fixing data parameter line
68:
1.254 brouard 69: Revision 1.253 2016/12/15 11:59:41 brouard
70: Summary: 0.99 in progress
71:
1.253 brouard 72: Revision 1.252 2016/09/15 21:15:37 brouard
73: *** empty log message ***
74:
1.252 brouard 75: Revision 1.251 2016/09/15 15:01:13 brouard
76: Summary: not working
77:
1.251 brouard 78: Revision 1.250 2016/09/08 16:07:27 brouard
79: Summary: continue
80:
1.250 brouard 81: Revision 1.249 2016/09/07 17:14:18 brouard
82: Summary: Starting values from frequencies
83:
1.249 brouard 84: Revision 1.248 2016/09/07 14:10:18 brouard
85: *** empty log message ***
86:
1.248 brouard 87: Revision 1.247 2016/09/02 11:11:21 brouard
88: *** empty log message ***
89:
1.247 brouard 90: Revision 1.246 2016/09/02 08:49:22 brouard
91: *** empty log message ***
92:
1.246 brouard 93: Revision 1.245 2016/09/02 07:25:01 brouard
94: *** empty log message ***
95:
1.245 brouard 96: Revision 1.244 2016/09/02 07:17:34 brouard
97: *** empty log message ***
98:
1.244 brouard 99: Revision 1.243 2016/09/02 06:45:35 brouard
100: *** empty log message ***
101:
1.243 brouard 102: Revision 1.242 2016/08/30 15:01:20 brouard
103: Summary: Fixing a lots
104:
1.242 brouard 105: Revision 1.241 2016/08/29 17:17:25 brouard
106: Summary: gnuplot problem in Back projection to fix
107:
1.241 brouard 108: Revision 1.240 2016/08/29 07:53:18 brouard
109: Summary: Better
110:
1.240 brouard 111: Revision 1.239 2016/08/26 15:51:03 brouard
112: Summary: Improvement in Powell output in order to copy and paste
113:
114: Author:
115:
1.239 brouard 116: Revision 1.238 2016/08/26 14:23:35 brouard
117: Summary: Starting tests of 0.99
118:
1.238 brouard 119: Revision 1.237 2016/08/26 09:20:19 brouard
120: Summary: to valgrind
121:
1.237 brouard 122: Revision 1.236 2016/08/25 10:50:18 brouard
123: *** empty log message ***
124:
1.236 brouard 125: Revision 1.235 2016/08/25 06:59:23 brouard
126: *** empty log message ***
127:
1.235 brouard 128: Revision 1.234 2016/08/23 16:51:20 brouard
129: *** empty log message ***
130:
1.234 brouard 131: Revision 1.233 2016/08/23 07:40:50 brouard
132: Summary: not working
133:
1.233 brouard 134: Revision 1.232 2016/08/22 14:20:21 brouard
135: Summary: not working
136:
1.232 brouard 137: Revision 1.231 2016/08/22 07:17:15 brouard
138: Summary: not working
139:
1.231 brouard 140: Revision 1.230 2016/08/22 06:55:53 brouard
141: Summary: Not working
142:
1.230 brouard 143: Revision 1.229 2016/07/23 09:45:53 brouard
144: Summary: Completing for func too
145:
1.229 brouard 146: Revision 1.228 2016/07/22 17:45:30 brouard
147: Summary: Fixing some arrays, still debugging
148:
1.227 brouard 149: Revision 1.226 2016/07/12 18:42:34 brouard
150: Summary: temp
151:
1.226 brouard 152: Revision 1.225 2016/07/12 08:40:03 brouard
153: Summary: saving but not running
154:
1.225 brouard 155: Revision 1.224 2016/07/01 13:16:01 brouard
156: Summary: Fixes
157:
1.224 brouard 158: Revision 1.223 2016/02/19 09:23:35 brouard
159: Summary: temporary
160:
1.223 brouard 161: Revision 1.222 2016/02/17 08:14:50 brouard
162: Summary: Probably last 0.98 stable version 0.98r6
163:
1.222 brouard 164: Revision 1.221 2016/02/15 23:35:36 brouard
165: Summary: minor bug
166:
1.220 brouard 167: Revision 1.219 2016/02/15 00:48:12 brouard
168: *** empty log message ***
169:
1.219 brouard 170: Revision 1.218 2016/02/12 11:29:23 brouard
171: Summary: 0.99 Back projections
172:
1.218 brouard 173: Revision 1.217 2015/12/23 17:18:31 brouard
174: Summary: Experimental backcast
175:
1.217 brouard 176: Revision 1.216 2015/12/18 17:32:11 brouard
177: Summary: 0.98r4 Warning and status=-2
178:
179: Version 0.98r4 is now:
180: - displaying an error when status is -1, date of interview unknown and date of death known;
181: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
182: Older changes concerning s=-2, dating from 2005 have been supersed.
183:
1.216 brouard 184: Revision 1.215 2015/12/16 08:52:24 brouard
185: Summary: 0.98r4 working
186:
1.215 brouard 187: Revision 1.214 2015/12/16 06:57:54 brouard
188: Summary: temporary not working
189:
1.214 brouard 190: Revision 1.213 2015/12/11 18:22:17 brouard
191: Summary: 0.98r4
192:
1.213 brouard 193: Revision 1.212 2015/11/21 12:47:24 brouard
194: Summary: minor typo
195:
1.212 brouard 196: Revision 1.211 2015/11/21 12:41:11 brouard
197: Summary: 0.98r3 with some graph of projected cross-sectional
198:
199: Author: Nicolas Brouard
200:
1.211 brouard 201: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 202: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 203: Summary: Adding ftolpl parameter
204: Author: N Brouard
205:
206: We had difficulties to get smoothed confidence intervals. It was due
207: to the period prevalence which wasn't computed accurately. The inner
208: parameter ftolpl is now an outer parameter of the .imach parameter
209: file after estepm. If ftolpl is small 1.e-4 and estepm too,
210: computation are long.
211:
1.209 brouard 212: Revision 1.208 2015/11/17 14:31:57 brouard
213: Summary: temporary
214:
1.208 brouard 215: Revision 1.207 2015/10/27 17:36:57 brouard
216: *** empty log message ***
217:
1.207 brouard 218: Revision 1.206 2015/10/24 07:14:11 brouard
219: *** empty log message ***
220:
1.206 brouard 221: Revision 1.205 2015/10/23 15:50:53 brouard
222: Summary: 0.98r3 some clarification for graphs on likelihood contributions
223:
1.205 brouard 224: Revision 1.204 2015/10/01 16:20:26 brouard
225: Summary: Some new graphs of contribution to likelihood
226:
1.204 brouard 227: Revision 1.203 2015/09/30 17:45:14 brouard
228: Summary: looking at better estimation of the hessian
229:
230: Also a better criteria for convergence to the period prevalence And
231: therefore adding the number of years needed to converge. (The
232: prevalence in any alive state shold sum to one
233:
1.203 brouard 234: Revision 1.202 2015/09/22 19:45:16 brouard
235: Summary: Adding some overall graph on contribution to likelihood. Might change
236:
1.202 brouard 237: Revision 1.201 2015/09/15 17:34:58 brouard
238: Summary: 0.98r0
239:
240: - Some new graphs like suvival functions
241: - Some bugs fixed like model=1+age+V2.
242:
1.201 brouard 243: Revision 1.200 2015/09/09 16:53:55 brouard
244: Summary: Big bug thanks to Flavia
245:
246: Even model=1+age+V2. did not work anymore
247:
1.200 brouard 248: Revision 1.199 2015/09/07 14:09:23 brouard
249: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
250:
1.199 brouard 251: Revision 1.198 2015/09/03 07:14:39 brouard
252: Summary: 0.98q5 Flavia
253:
1.198 brouard 254: Revision 1.197 2015/09/01 18:24:39 brouard
255: *** empty log message ***
256:
1.197 brouard 257: Revision 1.196 2015/08/18 23:17:52 brouard
258: Summary: 0.98q5
259:
1.196 brouard 260: Revision 1.195 2015/08/18 16:28:39 brouard
261: Summary: Adding a hack for testing purpose
262:
263: After reading the title, ftol and model lines, if the comment line has
264: a q, starting with #q, the answer at the end of the run is quit. It
265: permits to run test files in batch with ctest. The former workaround was
266: $ echo q | imach foo.imach
267:
1.195 brouard 268: Revision 1.194 2015/08/18 13:32:00 brouard
269: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
270:
1.194 brouard 271: Revision 1.193 2015/08/04 07:17:42 brouard
272: Summary: 0.98q4
273:
1.193 brouard 274: Revision 1.192 2015/07/16 16:49:02 brouard
275: Summary: Fixing some outputs
276:
1.192 brouard 277: Revision 1.191 2015/07/14 10:00:33 brouard
278: Summary: Some fixes
279:
1.191 brouard 280: Revision 1.190 2015/05/05 08:51:13 brouard
281: Summary: Adding digits in output parameters (7 digits instead of 6)
282:
283: Fix 1+age+.
284:
1.190 brouard 285: Revision 1.189 2015/04/30 14:45:16 brouard
286: Summary: 0.98q2
287:
1.189 brouard 288: Revision 1.188 2015/04/30 08:27:53 brouard
289: *** empty log message ***
290:
1.188 brouard 291: Revision 1.187 2015/04/29 09:11:15 brouard
292: *** empty log message ***
293:
1.187 brouard 294: Revision 1.186 2015/04/23 12:01:52 brouard
295: Summary: V1*age is working now, version 0.98q1
296:
297: Some codes had been disabled in order to simplify and Vn*age was
298: working in the optimization phase, ie, giving correct MLE parameters,
299: but, as usual, outputs were not correct and program core dumped.
300:
1.186 brouard 301: Revision 1.185 2015/03/11 13:26:42 brouard
302: Summary: Inclusion of compile and links command line for Intel Compiler
303:
1.185 brouard 304: Revision 1.184 2015/03/11 11:52:39 brouard
305: Summary: Back from Windows 8. Intel Compiler
306:
1.184 brouard 307: Revision 1.183 2015/03/10 20:34:32 brouard
308: Summary: 0.98q0, trying with directest, mnbrak fixed
309:
310: We use directest instead of original Powell test; probably no
311: incidence on the results, but better justifications;
312: We fixed Numerical Recipes mnbrak routine which was wrong and gave
313: wrong results.
314:
1.183 brouard 315: Revision 1.182 2015/02/12 08:19:57 brouard
316: Summary: Trying to keep directest which seems simpler and more general
317: Author: Nicolas Brouard
318:
1.182 brouard 319: Revision 1.181 2015/02/11 23:22:24 brouard
320: Summary: Comments on Powell added
321:
322: Author:
323:
1.181 brouard 324: Revision 1.180 2015/02/11 17:33:45 brouard
325: Summary: Finishing move from main to function (hpijx and prevalence_limit)
326:
1.180 brouard 327: Revision 1.179 2015/01/04 09:57:06 brouard
328: Summary: back to OS/X
329:
1.179 brouard 330: Revision 1.178 2015/01/04 09:35:48 brouard
331: *** empty log message ***
332:
1.178 brouard 333: Revision 1.177 2015/01/03 18:40:56 brouard
334: Summary: Still testing ilc32 on OSX
335:
1.177 brouard 336: Revision 1.176 2015/01/03 16:45:04 brouard
337: *** empty log message ***
338:
1.176 brouard 339: Revision 1.175 2015/01/03 16:33:42 brouard
340: *** empty log message ***
341:
1.175 brouard 342: Revision 1.174 2015/01/03 16:15:49 brouard
343: Summary: Still in cross-compilation
344:
1.174 brouard 345: Revision 1.173 2015/01/03 12:06:26 brouard
346: Summary: trying to detect cross-compilation
347:
1.173 brouard 348: Revision 1.172 2014/12/27 12:07:47 brouard
349: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
350:
1.172 brouard 351: Revision 1.171 2014/12/23 13:26:59 brouard
352: Summary: Back from Visual C
353:
354: Still problem with utsname.h on Windows
355:
1.171 brouard 356: Revision 1.170 2014/12/23 11:17:12 brouard
357: Summary: Cleaning some \%% back to %%
358:
359: The escape was mandatory for a specific compiler (which one?), but too many warnings.
360:
1.170 brouard 361: Revision 1.169 2014/12/22 23:08:31 brouard
362: Summary: 0.98p
363:
364: Outputs some informations on compiler used, OS etc. Testing on different platforms.
365:
1.169 brouard 366: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 367: Summary: update
1.169 brouard 368:
1.168 brouard 369: Revision 1.167 2014/12/22 13:50:56 brouard
370: Summary: Testing uname and compiler version and if compiled 32 or 64
371:
372: Testing on Linux 64
373:
1.167 brouard 374: Revision 1.166 2014/12/22 11:40:47 brouard
375: *** empty log message ***
376:
1.166 brouard 377: Revision 1.165 2014/12/16 11:20:36 brouard
378: Summary: After compiling on Visual C
379:
380: * imach.c (Module): Merging 1.61 to 1.162
381:
1.165 brouard 382: Revision 1.164 2014/12/16 10:52:11 brouard
383: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
384:
385: * imach.c (Module): Merging 1.61 to 1.162
386:
1.164 brouard 387: Revision 1.163 2014/12/16 10:30:11 brouard
388: * imach.c (Module): Merging 1.61 to 1.162
389:
1.163 brouard 390: Revision 1.162 2014/09/25 11:43:39 brouard
391: Summary: temporary backup 0.99!
392:
1.162 brouard 393: Revision 1.1 2014/09/16 11:06:58 brouard
394: Summary: With some code (wrong) for nlopt
395:
396: Author:
397:
398: Revision 1.161 2014/09/15 20:41:41 brouard
399: Summary: Problem with macro SQR on Intel compiler
400:
1.161 brouard 401: Revision 1.160 2014/09/02 09:24:05 brouard
402: *** empty log message ***
403:
1.160 brouard 404: Revision 1.159 2014/09/01 10:34:10 brouard
405: Summary: WIN32
406: Author: Brouard
407:
1.159 brouard 408: Revision 1.158 2014/08/27 17:11:51 brouard
409: *** empty log message ***
410:
1.158 brouard 411: Revision 1.157 2014/08/27 16:26:55 brouard
412: Summary: Preparing windows Visual studio version
413: Author: Brouard
414:
415: In order to compile on Visual studio, time.h is now correct and time_t
416: and tm struct should be used. difftime should be used but sometimes I
417: just make the differences in raw time format (time(&now).
418: Trying to suppress #ifdef LINUX
419: Add xdg-open for __linux in order to open default browser.
420:
1.157 brouard 421: Revision 1.156 2014/08/25 20:10:10 brouard
422: *** empty log message ***
423:
1.156 brouard 424: Revision 1.155 2014/08/25 18:32:34 brouard
425: Summary: New compile, minor changes
426: Author: Brouard
427:
1.155 brouard 428: Revision 1.154 2014/06/20 17:32:08 brouard
429: Summary: Outputs now all graphs of convergence to period prevalence
430:
1.154 brouard 431: Revision 1.153 2014/06/20 16:45:46 brouard
432: Summary: If 3 live state, convergence to period prevalence on same graph
433: Author: Brouard
434:
1.153 brouard 435: Revision 1.152 2014/06/18 17:54:09 brouard
436: Summary: open browser, use gnuplot on same dir than imach if not found in the path
437:
1.152 brouard 438: Revision 1.151 2014/06/18 16:43:30 brouard
439: *** empty log message ***
440:
1.151 brouard 441: Revision 1.150 2014/06/18 16:42:35 brouard
442: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
443: Author: brouard
444:
1.150 brouard 445: Revision 1.149 2014/06/18 15:51:14 brouard
446: Summary: Some fixes in parameter files errors
447: Author: Nicolas Brouard
448:
1.149 brouard 449: Revision 1.148 2014/06/17 17:38:48 brouard
450: Summary: Nothing new
451: Author: Brouard
452:
453: Just a new packaging for OS/X version 0.98nS
454:
1.148 brouard 455: Revision 1.147 2014/06/16 10:33:11 brouard
456: *** empty log message ***
457:
1.147 brouard 458: Revision 1.146 2014/06/16 10:20:28 brouard
459: Summary: Merge
460: Author: Brouard
461:
462: Merge, before building revised version.
463:
1.146 brouard 464: Revision 1.145 2014/06/10 21:23:15 brouard
465: Summary: Debugging with valgrind
466: Author: Nicolas Brouard
467:
468: Lot of changes in order to output the results with some covariates
469: After the Edimburgh REVES conference 2014, it seems mandatory to
470: improve the code.
471: No more memory valgrind error but a lot has to be done in order to
472: continue the work of splitting the code into subroutines.
473: Also, decodemodel has been improved. Tricode is still not
474: optimal. nbcode should be improved. Documentation has been added in
475: the source code.
476:
1.144 brouard 477: Revision 1.143 2014/01/26 09:45:38 brouard
478: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
479:
480: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
481: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
482:
1.143 brouard 483: Revision 1.142 2014/01/26 03:57:36 brouard
484: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
485:
486: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
487:
1.142 brouard 488: Revision 1.141 2014/01/26 02:42:01 brouard
489: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
490:
1.141 brouard 491: Revision 1.140 2011/09/02 10:37:54 brouard
492: Summary: times.h is ok with mingw32 now.
493:
1.140 brouard 494: Revision 1.139 2010/06/14 07:50:17 brouard
495: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
496: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
497:
1.139 brouard 498: Revision 1.138 2010/04/30 18:19:40 brouard
499: *** empty log message ***
500:
1.138 brouard 501: Revision 1.137 2010/04/29 18:11:38 brouard
502: (Module): Checking covariates for more complex models
503: than V1+V2. A lot of change to be done. Unstable.
504:
1.137 brouard 505: Revision 1.136 2010/04/26 20:30:53 brouard
506: (Module): merging some libgsl code. Fixing computation
507: of likelione (using inter/intrapolation if mle = 0) in order to
508: get same likelihood as if mle=1.
509: Some cleaning of code and comments added.
510:
1.136 brouard 511: Revision 1.135 2009/10/29 15:33:14 brouard
512: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
513:
1.135 brouard 514: Revision 1.134 2009/10/29 13:18:53 brouard
515: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
516:
1.134 brouard 517: Revision 1.133 2009/07/06 10:21:25 brouard
518: just nforces
519:
1.133 brouard 520: Revision 1.132 2009/07/06 08:22:05 brouard
521: Many tings
522:
1.132 brouard 523: Revision 1.131 2009/06/20 16:22:47 brouard
524: Some dimensions resccaled
525:
1.131 brouard 526: Revision 1.130 2009/05/26 06:44:34 brouard
527: (Module): Max Covariate is now set to 20 instead of 8. A
528: lot of cleaning with variables initialized to 0. Trying to make
529: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
530:
1.130 brouard 531: Revision 1.129 2007/08/31 13:49:27 lievre
532: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
533:
1.129 lievre 534: Revision 1.128 2006/06/30 13:02:05 brouard
535: (Module): Clarifications on computing e.j
536:
1.128 brouard 537: Revision 1.127 2006/04/28 18:11:50 brouard
538: (Module): Yes the sum of survivors was wrong since
539: imach-114 because nhstepm was no more computed in the age
540: loop. Now we define nhstepma in the age loop.
541: (Module): In order to speed up (in case of numerous covariates) we
542: compute health expectancies (without variances) in a first step
543: and then all the health expectancies with variances or standard
544: deviation (needs data from the Hessian matrices) which slows the
545: computation.
546: In the future we should be able to stop the program is only health
547: expectancies and graph are needed without standard deviations.
548:
1.127 brouard 549: Revision 1.126 2006/04/28 17:23:28 brouard
550: (Module): Yes the sum of survivors was wrong since
551: imach-114 because nhstepm was no more computed in the age
552: loop. Now we define nhstepma in the age loop.
553: Version 0.98h
554:
1.126 brouard 555: Revision 1.125 2006/04/04 15:20:31 lievre
556: Errors in calculation of health expectancies. Age was not initialized.
557: Forecasting file added.
558:
559: Revision 1.124 2006/03/22 17:13:53 lievre
560: Parameters are printed with %lf instead of %f (more numbers after the comma).
561: The log-likelihood is printed in the log file
562:
563: Revision 1.123 2006/03/20 10:52:43 brouard
564: * imach.c (Module): <title> changed, corresponds to .htm file
565: name. <head> headers where missing.
566:
567: * imach.c (Module): Weights can have a decimal point as for
568: English (a comma might work with a correct LC_NUMERIC environment,
569: otherwise the weight is truncated).
570: Modification of warning when the covariates values are not 0 or
571: 1.
572: Version 0.98g
573:
574: Revision 1.122 2006/03/20 09:45:41 brouard
575: (Module): Weights can have a decimal point as for
576: English (a comma might work with a correct LC_NUMERIC environment,
577: otherwise the weight is truncated).
578: Modification of warning when the covariates values are not 0 or
579: 1.
580: Version 0.98g
581:
582: Revision 1.121 2006/03/16 17:45:01 lievre
583: * imach.c (Module): Comments concerning covariates added
584:
585: * imach.c (Module): refinements in the computation of lli if
586: status=-2 in order to have more reliable computation if stepm is
587: not 1 month. Version 0.98f
588:
589: Revision 1.120 2006/03/16 15:10:38 lievre
590: (Module): refinements in the computation of lli if
591: status=-2 in order to have more reliable computation if stepm is
592: not 1 month. Version 0.98f
593:
594: Revision 1.119 2006/03/15 17:42:26 brouard
595: (Module): Bug if status = -2, the loglikelihood was
596: computed as likelihood omitting the logarithm. Version O.98e
597:
598: Revision 1.118 2006/03/14 18:20:07 brouard
599: (Module): varevsij Comments added explaining the second
600: table of variances if popbased=1 .
601: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
602: (Module): Function pstamp added
603: (Module): Version 0.98d
604:
605: Revision 1.117 2006/03/14 17:16:22 brouard
606: (Module): varevsij Comments added explaining the second
607: table of variances if popbased=1 .
608: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
609: (Module): Function pstamp added
610: (Module): Version 0.98d
611:
612: Revision 1.116 2006/03/06 10:29:27 brouard
613: (Module): Variance-covariance wrong links and
614: varian-covariance of ej. is needed (Saito).
615:
616: Revision 1.115 2006/02/27 12:17:45 brouard
617: (Module): One freematrix added in mlikeli! 0.98c
618:
619: Revision 1.114 2006/02/26 12:57:58 brouard
620: (Module): Some improvements in processing parameter
621: filename with strsep.
622:
623: Revision 1.113 2006/02/24 14:20:24 brouard
624: (Module): Memory leaks checks with valgrind and:
625: datafile was not closed, some imatrix were not freed and on matrix
626: allocation too.
627:
628: Revision 1.112 2006/01/30 09:55:26 brouard
629: (Module): Back to gnuplot.exe instead of wgnuplot.exe
630:
631: Revision 1.111 2006/01/25 20:38:18 brouard
632: (Module): Lots of cleaning and bugs added (Gompertz)
633: (Module): Comments can be added in data file. Missing date values
634: can be a simple dot '.'.
635:
636: Revision 1.110 2006/01/25 00:51:50 brouard
637: (Module): Lots of cleaning and bugs added (Gompertz)
638:
639: Revision 1.109 2006/01/24 19:37:15 brouard
640: (Module): Comments (lines starting with a #) are allowed in data.
641:
642: Revision 1.108 2006/01/19 18:05:42 lievre
643: Gnuplot problem appeared...
644: To be fixed
645:
646: Revision 1.107 2006/01/19 16:20:37 brouard
647: Test existence of gnuplot in imach path
648:
649: Revision 1.106 2006/01/19 13:24:36 brouard
650: Some cleaning and links added in html output
651:
652: Revision 1.105 2006/01/05 20:23:19 lievre
653: *** empty log message ***
654:
655: Revision 1.104 2005/09/30 16:11:43 lievre
656: (Module): sump fixed, loop imx fixed, and simplifications.
657: (Module): If the status is missing at the last wave but we know
658: that the person is alive, then we can code his/her status as -2
659: (instead of missing=-1 in earlier versions) and his/her
660: contributions to the likelihood is 1 - Prob of dying from last
661: health status (= 1-p13= p11+p12 in the easiest case of somebody in
662: the healthy state at last known wave). Version is 0.98
663:
664: Revision 1.103 2005/09/30 15:54:49 lievre
665: (Module): sump fixed, loop imx fixed, and simplifications.
666:
667: Revision 1.102 2004/09/15 17:31:30 brouard
668: Add the possibility to read data file including tab characters.
669:
670: Revision 1.101 2004/09/15 10:38:38 brouard
671: Fix on curr_time
672:
673: Revision 1.100 2004/07/12 18:29:06 brouard
674: Add version for Mac OS X. Just define UNIX in Makefile
675:
676: Revision 1.99 2004/06/05 08:57:40 brouard
677: *** empty log message ***
678:
679: Revision 1.98 2004/05/16 15:05:56 brouard
680: New version 0.97 . First attempt to estimate force of mortality
681: directly from the data i.e. without the need of knowing the health
682: state at each age, but using a Gompertz model: log u =a + b*age .
683: This is the basic analysis of mortality and should be done before any
684: other analysis, in order to test if the mortality estimated from the
685: cross-longitudinal survey is different from the mortality estimated
686: from other sources like vital statistic data.
687:
688: The same imach parameter file can be used but the option for mle should be -3.
689:
1.133 brouard 690: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 691: former routines in order to include the new code within the former code.
692:
693: The output is very simple: only an estimate of the intercept and of
694: the slope with 95% confident intervals.
695:
696: Current limitations:
697: A) Even if you enter covariates, i.e. with the
698: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
699: B) There is no computation of Life Expectancy nor Life Table.
700:
701: Revision 1.97 2004/02/20 13:25:42 lievre
702: Version 0.96d. Population forecasting command line is (temporarily)
703: suppressed.
704:
705: Revision 1.96 2003/07/15 15:38:55 brouard
706: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
707: rewritten within the same printf. Workaround: many printfs.
708:
709: Revision 1.95 2003/07/08 07:54:34 brouard
710: * imach.c (Repository):
711: (Repository): Using imachwizard code to output a more meaningful covariance
712: matrix (cov(a12,c31) instead of numbers.
713:
714: Revision 1.94 2003/06/27 13:00:02 brouard
715: Just cleaning
716:
717: Revision 1.93 2003/06/25 16:33:55 brouard
718: (Module): On windows (cygwin) function asctime_r doesn't
719: exist so I changed back to asctime which exists.
720: (Module): Version 0.96b
721:
722: Revision 1.92 2003/06/25 16:30:45 brouard
723: (Module): On windows (cygwin) function asctime_r doesn't
724: exist so I changed back to asctime which exists.
725:
726: Revision 1.91 2003/06/25 15:30:29 brouard
727: * imach.c (Repository): Duplicated warning errors corrected.
728: (Repository): Elapsed time after each iteration is now output. It
729: helps to forecast when convergence will be reached. Elapsed time
730: is stamped in powell. We created a new html file for the graphs
731: concerning matrix of covariance. It has extension -cov.htm.
732:
733: Revision 1.90 2003/06/24 12:34:15 brouard
734: (Module): Some bugs corrected for windows. Also, when
735: mle=-1 a template is output in file "or"mypar.txt with the design
736: of the covariance matrix to be input.
737:
738: Revision 1.89 2003/06/24 12:30:52 brouard
739: (Module): Some bugs corrected for windows. Also, when
740: mle=-1 a template is output in file "or"mypar.txt with the design
741: of the covariance matrix to be input.
742:
743: Revision 1.88 2003/06/23 17:54:56 brouard
744: * 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.
745:
746: Revision 1.87 2003/06/18 12:26:01 brouard
747: Version 0.96
748:
749: Revision 1.86 2003/06/17 20:04:08 brouard
750: (Module): Change position of html and gnuplot routines and added
751: routine fileappend.
752:
753: Revision 1.85 2003/06/17 13:12:43 brouard
754: * imach.c (Repository): Check when date of death was earlier that
755: current date of interview. It may happen when the death was just
756: prior to the death. In this case, dh was negative and likelihood
757: was wrong (infinity). We still send an "Error" but patch by
758: assuming that the date of death was just one stepm after the
759: interview.
760: (Repository): Because some people have very long ID (first column)
761: we changed int to long in num[] and we added a new lvector for
762: memory allocation. But we also truncated to 8 characters (left
763: truncation)
764: (Repository): No more line truncation errors.
765:
766: Revision 1.84 2003/06/13 21:44:43 brouard
767: * imach.c (Repository): Replace "freqsummary" at a correct
768: place. It differs from routine "prevalence" which may be called
769: many times. Probs is memory consuming and must be used with
770: parcimony.
771: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
772:
773: Revision 1.83 2003/06/10 13:39:11 lievre
774: *** empty log message ***
775:
776: Revision 1.82 2003/06/05 15:57:20 brouard
777: Add log in imach.c and fullversion number is now printed.
778:
779: */
780: /*
781: Interpolated Markov Chain
782:
783: Short summary of the programme:
784:
1.227 brouard 785: This program computes Healthy Life Expectancies or State-specific
786: (if states aren't health statuses) Expectancies from
787: cross-longitudinal data. Cross-longitudinal data consist in:
788:
789: -1- a first survey ("cross") where individuals from different ages
790: are interviewed on their health status or degree of disability (in
791: the case of a health survey which is our main interest)
792:
793: -2- at least a second wave of interviews ("longitudinal") which
794: measure each change (if any) in individual health status. Health
795: expectancies are computed from the time spent in each health state
796: according to a model. More health states you consider, more time is
797: necessary to reach the Maximum Likelihood of the parameters involved
798: in the model. The simplest model is the multinomial logistic model
799: where pij is the probability to be observed in state j at the second
800: wave conditional to be observed in state i at the first
801: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
802: etc , where 'age' is age and 'sex' is a covariate. If you want to
803: have a more complex model than "constant and age", you should modify
804: the program where the markup *Covariates have to be included here
805: again* invites you to do it. More covariates you add, slower the
1.126 brouard 806: convergence.
807:
808: The advantage of this computer programme, compared to a simple
809: multinomial logistic model, is clear when the delay between waves is not
810: identical for each individual. Also, if a individual missed an
811: intermediate interview, the information is lost, but taken into
812: account using an interpolation or extrapolation.
813:
814: hPijx is the probability to be observed in state i at age x+h
815: conditional to the observed state i at age x. The delay 'h' can be
816: split into an exact number (nh*stepm) of unobserved intermediate
817: states. This elementary transition (by month, quarter,
818: semester or year) is modelled as a multinomial logistic. The hPx
819: matrix is simply the matrix product of nh*stepm elementary matrices
820: and the contribution of each individual to the likelihood is simply
821: hPijx.
822:
823: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 824: of the life expectancies. It also computes the period (stable) prevalence.
825:
826: Back prevalence and projections:
1.227 brouard 827:
828: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
829: double agemaxpar, double ftolpl, int *ncvyearp, double
830: dateprev1,double dateprev2, int firstpass, int lastpass, int
831: mobilavproj)
832:
833: Computes the back prevalence limit for any combination of
834: covariate values k at any age between ageminpar and agemaxpar and
835: returns it in **bprlim. In the loops,
836:
837: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
838: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
839:
840: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 841: Computes for any combination of covariates k and any age between bage and fage
842: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
843: oldm=oldms;savm=savms;
1.227 brouard 844:
1.267 brouard 845: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 846: Computes the transition matrix starting at age 'age' over
847: 'nhstepm*hstepm*stepm' months (i.e. until
848: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 849: nhstepm*hstepm matrices.
850:
851: Returns p3mat[i][j][h] after calling
852: p3mat[i][j][h]=matprod2(newm,
853: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
854: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
855: oldm);
1.226 brouard 856:
857: Important routines
858:
859: - func (or funcone), computes logit (pij) distinguishing
860: o fixed variables (single or product dummies or quantitative);
861: o varying variables by:
862: (1) wave (single, product dummies, quantitative),
863: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
864: % fixed dummy (treated) or quantitative (not done because time-consuming);
865: % varying dummy (not done) or quantitative (not done);
866: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
867: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
868: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
869: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
870: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 871:
1.226 brouard 872:
873:
1.133 brouard 874: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
875: Institut national d'études démographiques, Paris.
1.126 brouard 876: This software have been partly granted by Euro-REVES, a concerted action
877: from the European Union.
878: It is copyrighted identically to a GNU software product, ie programme and
879: software can be distributed freely for non commercial use. Latest version
880: can be accessed at http://euroreves.ined.fr/imach .
881:
882: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
883: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
884:
885: **********************************************************************/
886: /*
887: main
888: read parameterfile
889: read datafile
890: concatwav
891: freqsummary
892: if (mle >= 1)
893: mlikeli
894: print results files
895: if mle==1
896: computes hessian
897: read end of parameter file: agemin, agemax, bage, fage, estepm
898: begin-prev-date,...
899: open gnuplot file
900: open html file
1.145 brouard 901: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
902: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
903: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
904: freexexit2 possible for memory heap.
905:
906: h Pij x | pij_nom ficrestpij
907: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
908: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
909: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
910:
911: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
912: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
913: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
914: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
915: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
916:
1.126 brouard 917: forecasting if prevfcast==1 prevforecast call prevalence()
918: health expectancies
919: Variance-covariance of DFLE
920: prevalence()
921: movingaverage()
922: varevsij()
923: if popbased==1 varevsij(,popbased)
924: total life expectancies
925: Variance of period (stable) prevalence
926: end
927: */
928:
1.187 brouard 929: /* #define DEBUG */
930: /* #define DEBUGBRENT */
1.203 brouard 931: /* #define DEBUGLINMIN */
932: /* #define DEBUGHESS */
933: #define DEBUGHESSIJ
1.224 brouard 934: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 935: #define POWELL /* Instead of NLOPT */
1.224 brouard 936: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 937: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
938: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 939:
940: #include <math.h>
941: #include <stdio.h>
942: #include <stdlib.h>
943: #include <string.h>
1.226 brouard 944: #include <ctype.h>
1.159 brouard 945:
946: #ifdef _WIN32
947: #include <io.h>
1.172 brouard 948: #include <windows.h>
949: #include <tchar.h>
1.159 brouard 950: #else
1.126 brouard 951: #include <unistd.h>
1.159 brouard 952: #endif
1.126 brouard 953:
954: #include <limits.h>
955: #include <sys/types.h>
1.171 brouard 956:
957: #if defined(__GNUC__)
958: #include <sys/utsname.h> /* Doesn't work on Windows */
959: #endif
960:
1.126 brouard 961: #include <sys/stat.h>
962: #include <errno.h>
1.159 brouard 963: /* extern int errno; */
1.126 brouard 964:
1.157 brouard 965: /* #ifdef LINUX */
966: /* #include <time.h> */
967: /* #include "timeval.h" */
968: /* #else */
969: /* #include <sys/time.h> */
970: /* #endif */
971:
1.126 brouard 972: #include <time.h>
973:
1.136 brouard 974: #ifdef GSL
975: #include <gsl/gsl_errno.h>
976: #include <gsl/gsl_multimin.h>
977: #endif
978:
1.167 brouard 979:
1.162 brouard 980: #ifdef NLOPT
981: #include <nlopt.h>
982: typedef struct {
983: double (* function)(double [] );
984: } myfunc_data ;
985: #endif
986:
1.126 brouard 987: /* #include <libintl.h> */
988: /* #define _(String) gettext (String) */
989:
1.251 brouard 990: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 991:
992: #define GNUPLOTPROGRAM "gnuplot"
993: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
994: #define FILENAMELENGTH 132
995:
996: #define GLOCK_ERROR_NOPATH -1 /* empty path */
997: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
998:
1.144 brouard 999: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1000: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1001:
1002: #define NINTERVMAX 8
1.144 brouard 1003: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1004: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1005: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1006: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1007: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1008: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1009: #define MAXN 20000
1.144 brouard 1010: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1011: /* #define AGESUP 130 */
1012: #define AGESUP 150
1.268 brouard 1013: #define AGEINF 0
1.218 brouard 1014: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1015: #define AGEBASE 40
1.194 brouard 1016: #define AGEOVERFLOW 1.e20
1.164 brouard 1017: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1018: #ifdef _WIN32
1019: #define DIRSEPARATOR '\\'
1020: #define CHARSEPARATOR "\\"
1021: #define ODIRSEPARATOR '/'
1022: #else
1.126 brouard 1023: #define DIRSEPARATOR '/'
1024: #define CHARSEPARATOR "/"
1025: #define ODIRSEPARATOR '\\'
1026: #endif
1027:
1.275 ! brouard 1028: /* $Id: imach.c,v 1.274 2017/06/29 09:47:08 brouard Exp $ */
1.126 brouard 1029: /* $State: Exp $ */
1.196 brouard 1030: #include "version.h"
1031: char version[]=__IMACH_VERSION__;
1.224 brouard 1032: 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";
1.275 ! brouard 1033: char fullversion[]="$Revision: 1.274 $ $Date: 2017/06/29 09:47:08 $";
1.126 brouard 1034: char strstart[80];
1035: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1036: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1037: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1038: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1039: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1040: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1041: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1042: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1043: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1044: int cptcovprodnoage=0; /**< Number of covariate products without age */
1045: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1046: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1047: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1048: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1049: int nsd=0; /**< Total number of single dummy variables (output) */
1050: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1051: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1052: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1053: int ntveff=0; /**< ntveff number of effective time varying variables */
1054: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1055: int cptcov=0; /* Working variable */
1.218 brouard 1056: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1057: int npar=NPARMAX;
1058: int nlstate=2; /* Number of live states */
1059: int ndeath=1; /* Number of dead states */
1.130 brouard 1060: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1061: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1062: int popbased=0;
1063:
1064: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1065: int maxwav=0; /* Maxim number of waves */
1066: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1067: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1068: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1069: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1070: int mle=1, weightopt=0;
1.126 brouard 1071: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1072: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1073: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1074: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1075: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1076: int selected(int kvar); /* Is covariate kvar selected for printing results */
1077:
1.130 brouard 1078: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1079: double **matprod2(); /* test */
1.126 brouard 1080: double **oldm, **newm, **savm; /* Working pointers to matrices */
1081: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1082: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1083:
1.136 brouard 1084: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1085: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1086: FILE *ficlog, *ficrespow;
1.130 brouard 1087: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1088: double fretone; /* Only one call to likelihood */
1.130 brouard 1089: long ipmx=0; /* Number of contributions */
1.126 brouard 1090: double sw; /* Sum of weights */
1091: char filerespow[FILENAMELENGTH];
1092: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1093: FILE *ficresilk;
1094: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1095: FILE *ficresprobmorprev;
1096: FILE *fichtm, *fichtmcov; /* Html File */
1097: FILE *ficreseij;
1098: char filerese[FILENAMELENGTH];
1099: FILE *ficresstdeij;
1100: char fileresstde[FILENAMELENGTH];
1101: FILE *ficrescveij;
1102: char filerescve[FILENAMELENGTH];
1103: FILE *ficresvij;
1104: char fileresv[FILENAMELENGTH];
1.269 brouard 1105:
1.126 brouard 1106: char title[MAXLINE];
1.234 brouard 1107: char model[MAXLINE]; /**< The model line */
1.217 brouard 1108: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1109: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1110: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1111: char command[FILENAMELENGTH];
1112: int outcmd=0;
1113:
1.217 brouard 1114: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1115: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1116: char filelog[FILENAMELENGTH]; /* Log file */
1117: char filerest[FILENAMELENGTH];
1118: char fileregp[FILENAMELENGTH];
1119: char popfile[FILENAMELENGTH];
1120:
1121: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1122:
1.157 brouard 1123: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1124: /* struct timezone tzp; */
1125: /* extern int gettimeofday(); */
1126: struct tm tml, *gmtime(), *localtime();
1127:
1128: extern time_t time();
1129:
1130: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1131: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1132: struct tm tm;
1133:
1.126 brouard 1134: char strcurr[80], strfor[80];
1135:
1136: char *endptr;
1137: long lval;
1138: double dval;
1139:
1140: #define NR_END 1
1141: #define FREE_ARG char*
1142: #define FTOL 1.0e-10
1143:
1144: #define NRANSI
1.240 brouard 1145: #define ITMAX 200
1146: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1147:
1148: #define TOL 2.0e-4
1149:
1150: #define CGOLD 0.3819660
1151: #define ZEPS 1.0e-10
1152: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1153:
1154: #define GOLD 1.618034
1155: #define GLIMIT 100.0
1156: #define TINY 1.0e-20
1157:
1158: static double maxarg1,maxarg2;
1159: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1160: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1161:
1162: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1163: #define rint(a) floor(a+0.5)
1.166 brouard 1164: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1165: #define mytinydouble 1.0e-16
1.166 brouard 1166: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1167: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1168: /* static double dsqrarg; */
1169: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1170: static double sqrarg;
1171: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1172: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1173: int agegomp= AGEGOMP;
1174:
1175: int imx;
1176: int stepm=1;
1177: /* Stepm, step in month: minimum step interpolation*/
1178:
1179: int estepm;
1180: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1181:
1182: int m,nb;
1183: long *num;
1.197 brouard 1184: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1185: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1186: covariate for which somebody answered excluding
1187: undefined. Usually 2: 0 and 1. */
1188: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1189: covariate for which somebody answered including
1190: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1191: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1192: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1193: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1194: double *ageexmed,*agecens;
1195: double dateintmean=0;
1196:
1197: double *weight;
1198: int **s; /* Status */
1.141 brouard 1199: double *agedc;
1.145 brouard 1200: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1201: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1202: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1203: double **coqvar; /* Fixed quantitative covariate nqv */
1204: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1205: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1206: double idx;
1207: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1208: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1209: /*k 1 2 3 4 5 6 7 8 9 */
1210: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1211: /* Tndvar[k] 1 2 3 4 5 */
1212: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1213: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1214: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1215: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1216: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1217: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1218: /* Tprod[i]=k 4 7 */
1219: /* Tage[i]=k 5 8 */
1220: /* */
1221: /* Type */
1222: /* V 1 2 3 4 5 */
1223: /* F F V V V */
1224: /* D Q D D Q */
1225: /* */
1226: int *TvarsD;
1227: int *TvarsDind;
1228: int *TvarsQ;
1229: int *TvarsQind;
1230:
1.235 brouard 1231: #define MAXRESULTLINES 10
1232: int nresult=0;
1.258 brouard 1233: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1234: int TKresult[MAXRESULTLINES];
1.237 brouard 1235: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1236: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1237: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1238: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1239: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1240: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1241:
1.234 brouard 1242: /* 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 1243: 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 */
1244: 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 */
1245: 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 */
1246: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1247: 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 */
1248: 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 1249: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1250: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1251: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1252: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1253: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1254: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1255: 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 */
1256: 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 */
1257:
1.230 brouard 1258: int *Tvarsel; /**< Selected covariates for output */
1259: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1260: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1261: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1262: 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 1263: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1264: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1265: int *Tage;
1.227 brouard 1266: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1267: 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 1268: 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*/
1269: 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 1270: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1271: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1272: int **Tvard;
1273: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1274: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1275: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1276: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1277: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1278: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1279: double *lsurv, *lpop, *tpop;
1280:
1.231 brouard 1281: #define FD 1; /* Fixed dummy covariate */
1282: #define FQ 2; /* Fixed quantitative covariate */
1283: #define FP 3; /* Fixed product covariate */
1284: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1285: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1286: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1287: #define VD 10; /* Varying dummy covariate */
1288: #define VQ 11; /* Varying quantitative covariate */
1289: #define VP 12; /* Varying product covariate */
1290: #define VPDD 13; /* Varying product dummy*dummy covariate */
1291: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1292: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1293: #define APFD 16; /* Age product * fixed dummy covariate */
1294: #define APFQ 17; /* Age product * fixed quantitative covariate */
1295: #define APVD 18; /* Age product * varying dummy covariate */
1296: #define APVQ 19; /* Age product * varying quantitative covariate */
1297:
1298: #define FTYPE 1; /* Fixed covariate */
1299: #define VTYPE 2; /* Varying covariate (loop in wave) */
1300: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1301:
1302: struct kmodel{
1303: int maintype; /* main type */
1304: int subtype; /* subtype */
1305: };
1306: struct kmodel modell[NCOVMAX];
1307:
1.143 brouard 1308: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1309: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1310:
1311: /**************** split *************************/
1312: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1313: {
1314: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1315: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1316: */
1317: char *ss; /* pointer */
1.186 brouard 1318: int l1=0, l2=0; /* length counters */
1.126 brouard 1319:
1320: l1 = strlen(path ); /* length of path */
1321: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1322: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1323: if ( ss == NULL ) { /* no directory, so determine current directory */
1324: strcpy( name, path ); /* we got the fullname name because no directory */
1325: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1326: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1327: /* get current working directory */
1328: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1329: #ifdef WIN32
1330: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1331: #else
1332: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1333: #endif
1.126 brouard 1334: return( GLOCK_ERROR_GETCWD );
1335: }
1336: /* got dirc from getcwd*/
1337: printf(" DIRC = %s \n",dirc);
1.205 brouard 1338: } else { /* strip directory from path */
1.126 brouard 1339: ss++; /* after this, the filename */
1340: l2 = strlen( ss ); /* length of filename */
1341: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1342: strcpy( name, ss ); /* save file name */
1343: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1344: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1345: printf(" DIRC2 = %s \n",dirc);
1346: }
1347: /* We add a separator at the end of dirc if not exists */
1348: l1 = strlen( dirc ); /* length of directory */
1349: if( dirc[l1-1] != DIRSEPARATOR ){
1350: dirc[l1] = DIRSEPARATOR;
1351: dirc[l1+1] = 0;
1352: printf(" DIRC3 = %s \n",dirc);
1353: }
1354: ss = strrchr( name, '.' ); /* find last / */
1355: if (ss >0){
1356: ss++;
1357: strcpy(ext,ss); /* save extension */
1358: l1= strlen( name);
1359: l2= strlen(ss)+1;
1360: strncpy( finame, name, l1-l2);
1361: finame[l1-l2]= 0;
1362: }
1363:
1364: return( 0 ); /* we're done */
1365: }
1366:
1367:
1368: /******************************************/
1369:
1370: void replace_back_to_slash(char *s, char*t)
1371: {
1372: int i;
1373: int lg=0;
1374: i=0;
1375: lg=strlen(t);
1376: for(i=0; i<= lg; i++) {
1377: (s[i] = t[i]);
1378: if (t[i]== '\\') s[i]='/';
1379: }
1380: }
1381:
1.132 brouard 1382: char *trimbb(char *out, char *in)
1.137 brouard 1383: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1384: char *s;
1385: s=out;
1386: while (*in != '\0'){
1.137 brouard 1387: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1388: in++;
1389: }
1390: *out++ = *in++;
1391: }
1392: *out='\0';
1393: return s;
1394: }
1395:
1.187 brouard 1396: /* char *substrchaine(char *out, char *in, char *chain) */
1397: /* { */
1398: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1399: /* char *s, *t; */
1400: /* t=in;s=out; */
1401: /* while ((*in != *chain) && (*in != '\0')){ */
1402: /* *out++ = *in++; */
1403: /* } */
1404:
1405: /* /\* *in matches *chain *\/ */
1406: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1407: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1408: /* } */
1409: /* in--; chain--; */
1410: /* while ( (*in != '\0')){ */
1411: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1412: /* *out++ = *in++; */
1413: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1414: /* } */
1415: /* *out='\0'; */
1416: /* out=s; */
1417: /* return out; */
1418: /* } */
1419: char *substrchaine(char *out, char *in, char *chain)
1420: {
1421: /* Substract chain 'chain' from 'in', return and output 'out' */
1422: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1423:
1424: char *strloc;
1425:
1426: strcpy (out, in);
1427: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1428: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1429: if(strloc != NULL){
1430: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1431: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1432: /* strcpy (strloc, strloc +strlen(chain));*/
1433: }
1434: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1435: return out;
1436: }
1437:
1438:
1.145 brouard 1439: char *cutl(char *blocc, char *alocc, char *in, char occ)
1440: {
1.187 brouard 1441: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1442: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1443: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1444: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1445: */
1.160 brouard 1446: char *s, *t;
1.145 brouard 1447: t=in;s=in;
1448: while ((*in != occ) && (*in != '\0')){
1449: *alocc++ = *in++;
1450: }
1451: if( *in == occ){
1452: *(alocc)='\0';
1453: s=++in;
1454: }
1455:
1456: if (s == t) {/* occ not found */
1457: *(alocc-(in-s))='\0';
1458: in=s;
1459: }
1460: while ( *in != '\0'){
1461: *blocc++ = *in++;
1462: }
1463:
1464: *blocc='\0';
1465: return t;
1466: }
1.137 brouard 1467: char *cutv(char *blocc, char *alocc, char *in, char occ)
1468: {
1.187 brouard 1469: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1470: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1471: gives blocc="abcdef2ghi" and alocc="j".
1472: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1473: */
1474: char *s, *t;
1475: t=in;s=in;
1476: while (*in != '\0'){
1477: while( *in == occ){
1478: *blocc++ = *in++;
1479: s=in;
1480: }
1481: *blocc++ = *in++;
1482: }
1483: if (s == t) /* occ not found */
1484: *(blocc-(in-s))='\0';
1485: else
1486: *(blocc-(in-s)-1)='\0';
1487: in=s;
1488: while ( *in != '\0'){
1489: *alocc++ = *in++;
1490: }
1491:
1492: *alocc='\0';
1493: return s;
1494: }
1495:
1.126 brouard 1496: int nbocc(char *s, char occ)
1497: {
1498: int i,j=0;
1499: int lg=20;
1500: i=0;
1501: lg=strlen(s);
1502: for(i=0; i<= lg; i++) {
1.234 brouard 1503: if (s[i] == occ ) j++;
1.126 brouard 1504: }
1505: return j;
1506: }
1507:
1.137 brouard 1508: /* void cutv(char *u,char *v, char*t, char occ) */
1509: /* { */
1510: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1511: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1512: /* gives u="abcdef2ghi" and v="j" *\/ */
1513: /* int i,lg,j,p=0; */
1514: /* i=0; */
1515: /* lg=strlen(t); */
1516: /* for(j=0; j<=lg-1; j++) { */
1517: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1518: /* } */
1.126 brouard 1519:
1.137 brouard 1520: /* for(j=0; j<p; j++) { */
1521: /* (u[j] = t[j]); */
1522: /* } */
1523: /* u[p]='\0'; */
1.126 brouard 1524:
1.137 brouard 1525: /* for(j=0; j<= lg; j++) { */
1526: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1527: /* } */
1528: /* } */
1.126 brouard 1529:
1.160 brouard 1530: #ifdef _WIN32
1531: char * strsep(char **pp, const char *delim)
1532: {
1533: char *p, *q;
1534:
1535: if ((p = *pp) == NULL)
1536: return 0;
1537: if ((q = strpbrk (p, delim)) != NULL)
1538: {
1539: *pp = q + 1;
1540: *q = '\0';
1541: }
1542: else
1543: *pp = 0;
1544: return p;
1545: }
1546: #endif
1547:
1.126 brouard 1548: /********************** nrerror ********************/
1549:
1550: void nrerror(char error_text[])
1551: {
1552: fprintf(stderr,"ERREUR ...\n");
1553: fprintf(stderr,"%s\n",error_text);
1554: exit(EXIT_FAILURE);
1555: }
1556: /*********************** vector *******************/
1557: double *vector(int nl, int nh)
1558: {
1559: double *v;
1560: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1561: if (!v) nrerror("allocation failure in vector");
1562: return v-nl+NR_END;
1563: }
1564:
1565: /************************ free vector ******************/
1566: void free_vector(double*v, int nl, int nh)
1567: {
1568: free((FREE_ARG)(v+nl-NR_END));
1569: }
1570:
1571: /************************ivector *******************************/
1572: int *ivector(long nl,long nh)
1573: {
1574: int *v;
1575: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1576: if (!v) nrerror("allocation failure in ivector");
1577: return v-nl+NR_END;
1578: }
1579:
1580: /******************free ivector **************************/
1581: void free_ivector(int *v, long nl, long nh)
1582: {
1583: free((FREE_ARG)(v+nl-NR_END));
1584: }
1585:
1586: /************************lvector *******************************/
1587: long *lvector(long nl,long nh)
1588: {
1589: long *v;
1590: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1591: if (!v) nrerror("allocation failure in ivector");
1592: return v-nl+NR_END;
1593: }
1594:
1595: /******************free lvector **************************/
1596: void free_lvector(long *v, long nl, long nh)
1597: {
1598: free((FREE_ARG)(v+nl-NR_END));
1599: }
1600:
1601: /******************* imatrix *******************************/
1602: int **imatrix(long nrl, long nrh, long ncl, long nch)
1603: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1604: {
1605: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1606: int **m;
1607:
1608: /* allocate pointers to rows */
1609: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1610: if (!m) nrerror("allocation failure 1 in matrix()");
1611: m += NR_END;
1612: m -= nrl;
1613:
1614:
1615: /* allocate rows and set pointers to them */
1616: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1617: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1618: m[nrl] += NR_END;
1619: m[nrl] -= ncl;
1620:
1621: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1622:
1623: /* return pointer to array of pointers to rows */
1624: return m;
1625: }
1626:
1627: /****************** free_imatrix *************************/
1628: void free_imatrix(m,nrl,nrh,ncl,nch)
1629: int **m;
1630: long nch,ncl,nrh,nrl;
1631: /* free an int matrix allocated by imatrix() */
1632: {
1633: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1634: free((FREE_ARG) (m+nrl-NR_END));
1635: }
1636:
1637: /******************* matrix *******************************/
1638: double **matrix(long nrl, long nrh, long ncl, long nch)
1639: {
1640: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1641: double **m;
1642:
1643: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1644: if (!m) nrerror("allocation failure 1 in matrix()");
1645: m += NR_END;
1646: m -= nrl;
1647:
1648: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1649: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1650: m[nrl] += NR_END;
1651: m[nrl] -= ncl;
1652:
1653: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1654: return m;
1.145 brouard 1655: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1656: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1657: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1658: */
1659: }
1660:
1661: /*************************free matrix ************************/
1662: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1663: {
1664: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1665: free((FREE_ARG)(m+nrl-NR_END));
1666: }
1667:
1668: /******************* ma3x *******************************/
1669: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1670: {
1671: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1672: double ***m;
1673:
1674: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1675: if (!m) nrerror("allocation failure 1 in matrix()");
1676: m += NR_END;
1677: m -= nrl;
1678:
1679: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1680: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1681: m[nrl] += NR_END;
1682: m[nrl] -= ncl;
1683:
1684: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1685:
1686: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1687: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1688: m[nrl][ncl] += NR_END;
1689: m[nrl][ncl] -= nll;
1690: for (j=ncl+1; j<=nch; j++)
1691: m[nrl][j]=m[nrl][j-1]+nlay;
1692:
1693: for (i=nrl+1; i<=nrh; i++) {
1694: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1695: for (j=ncl+1; j<=nch; j++)
1696: m[i][j]=m[i][j-1]+nlay;
1697: }
1698: return m;
1699: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1700: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1701: */
1702: }
1703:
1704: /*************************free ma3x ************************/
1705: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1706: {
1707: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1708: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1709: free((FREE_ARG)(m+nrl-NR_END));
1710: }
1711:
1712: /*************** function subdirf ***********/
1713: char *subdirf(char fileres[])
1714: {
1715: /* Caution optionfilefiname is hidden */
1716: strcpy(tmpout,optionfilefiname);
1717: strcat(tmpout,"/"); /* Add to the right */
1718: strcat(tmpout,fileres);
1719: return tmpout;
1720: }
1721:
1722: /*************** function subdirf2 ***********/
1723: char *subdirf2(char fileres[], char *preop)
1724: {
1725:
1726: /* Caution optionfilefiname is hidden */
1727: strcpy(tmpout,optionfilefiname);
1728: strcat(tmpout,"/");
1729: strcat(tmpout,preop);
1730: strcat(tmpout,fileres);
1731: return tmpout;
1732: }
1733:
1734: /*************** function subdirf3 ***********/
1735: char *subdirf3(char fileres[], char *preop, char *preop2)
1736: {
1737:
1738: /* Caution optionfilefiname is hidden */
1739: strcpy(tmpout,optionfilefiname);
1740: strcat(tmpout,"/");
1741: strcat(tmpout,preop);
1742: strcat(tmpout,preop2);
1743: strcat(tmpout,fileres);
1744: return tmpout;
1745: }
1.213 brouard 1746:
1747: /*************** function subdirfext ***********/
1748: char *subdirfext(char fileres[], char *preop, char *postop)
1749: {
1750:
1751: strcpy(tmpout,preop);
1752: strcat(tmpout,fileres);
1753: strcat(tmpout,postop);
1754: return tmpout;
1755: }
1.126 brouard 1756:
1.213 brouard 1757: /*************** function subdirfext3 ***********/
1758: char *subdirfext3(char fileres[], char *preop, char *postop)
1759: {
1760:
1761: /* Caution optionfilefiname is hidden */
1762: strcpy(tmpout,optionfilefiname);
1763: strcat(tmpout,"/");
1764: strcat(tmpout,preop);
1765: strcat(tmpout,fileres);
1766: strcat(tmpout,postop);
1767: return tmpout;
1768: }
1769:
1.162 brouard 1770: char *asc_diff_time(long time_sec, char ascdiff[])
1771: {
1772: long sec_left, days, hours, minutes;
1773: days = (time_sec) / (60*60*24);
1774: sec_left = (time_sec) % (60*60*24);
1775: hours = (sec_left) / (60*60) ;
1776: sec_left = (sec_left) %(60*60);
1777: minutes = (sec_left) /60;
1778: sec_left = (sec_left) % (60);
1779: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1780: return ascdiff;
1781: }
1782:
1.126 brouard 1783: /***************** f1dim *************************/
1784: extern int ncom;
1785: extern double *pcom,*xicom;
1786: extern double (*nrfunc)(double []);
1787:
1788: double f1dim(double x)
1789: {
1790: int j;
1791: double f;
1792: double *xt;
1793:
1794: xt=vector(1,ncom);
1795: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1796: f=(*nrfunc)(xt);
1797: free_vector(xt,1,ncom);
1798: return f;
1799: }
1800:
1801: /*****************brent *************************/
1802: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1803: {
1804: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1805: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1806: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1807: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1808: * returned function value.
1809: */
1.126 brouard 1810: int iter;
1811: double a,b,d,etemp;
1.159 brouard 1812: double fu=0,fv,fw,fx;
1.164 brouard 1813: double ftemp=0.;
1.126 brouard 1814: double p,q,r,tol1,tol2,u,v,w,x,xm;
1815: double e=0.0;
1816:
1817: a=(ax < cx ? ax : cx);
1818: b=(ax > cx ? ax : cx);
1819: x=w=v=bx;
1820: fw=fv=fx=(*f)(x);
1821: for (iter=1;iter<=ITMAX;iter++) {
1822: xm=0.5*(a+b);
1823: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1824: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1825: printf(".");fflush(stdout);
1826: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1827: #ifdef DEBUGBRENT
1.126 brouard 1828: 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);
1829: 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);
1830: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1831: #endif
1832: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1833: *xmin=x;
1834: return fx;
1835: }
1836: ftemp=fu;
1837: if (fabs(e) > tol1) {
1838: r=(x-w)*(fx-fv);
1839: q=(x-v)*(fx-fw);
1840: p=(x-v)*q-(x-w)*r;
1841: q=2.0*(q-r);
1842: if (q > 0.0) p = -p;
1843: q=fabs(q);
1844: etemp=e;
1845: e=d;
1846: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1847: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1848: else {
1.224 brouard 1849: d=p/q;
1850: u=x+d;
1851: if (u-a < tol2 || b-u < tol2)
1852: d=SIGN(tol1,xm-x);
1.126 brouard 1853: }
1854: } else {
1855: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1856: }
1857: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1858: fu=(*f)(u);
1859: if (fu <= fx) {
1860: if (u >= x) a=x; else b=x;
1861: SHFT(v,w,x,u)
1.183 brouard 1862: SHFT(fv,fw,fx,fu)
1863: } else {
1864: if (u < x) a=u; else b=u;
1865: if (fu <= fw || w == x) {
1.224 brouard 1866: v=w;
1867: w=u;
1868: fv=fw;
1869: fw=fu;
1.183 brouard 1870: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1871: v=u;
1872: fv=fu;
1.183 brouard 1873: }
1874: }
1.126 brouard 1875: }
1876: nrerror("Too many iterations in brent");
1877: *xmin=x;
1878: return fx;
1879: }
1880:
1881: /****************** mnbrak ***********************/
1882:
1883: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1884: double (*func)(double))
1.183 brouard 1885: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1886: the downhill direction (defined by the function as evaluated at the initial points) and returns
1887: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1888: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1889: */
1.126 brouard 1890: double ulim,u,r,q, dum;
1891: double fu;
1.187 brouard 1892:
1893: double scale=10.;
1894: int iterscale=0;
1895:
1896: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1897: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1898:
1899:
1900: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1901: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1902: /* *bx = *ax - (*ax - *bx)/scale; */
1903: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1904: /* } */
1905:
1.126 brouard 1906: if (*fb > *fa) {
1907: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1908: SHFT(dum,*fb,*fa,dum)
1909: }
1.126 brouard 1910: *cx=(*bx)+GOLD*(*bx-*ax);
1911: *fc=(*func)(*cx);
1.183 brouard 1912: #ifdef DEBUG
1.224 brouard 1913: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1914: 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 1915: #endif
1.224 brouard 1916: 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 1917: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1918: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1919: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1920: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1921: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1922: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1923: fu=(*func)(u);
1.163 brouard 1924: #ifdef DEBUG
1925: /* f(x)=A(x-u)**2+f(u) */
1926: double A, fparabu;
1927: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1928: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1929: 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);
1930: 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 1931: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1932: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1933: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1934: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1935: #endif
1.184 brouard 1936: #ifdef MNBRAKORIGINAL
1.183 brouard 1937: #else
1.191 brouard 1938: /* if (fu > *fc) { */
1939: /* #ifdef DEBUG */
1940: /* printf("mnbrak4 fu > fc \n"); */
1941: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1942: /* #endif */
1943: /* /\* 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 *\\/ *\/ */
1944: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1945: /* dum=u; /\* Shifting c and u *\/ */
1946: /* u = *cx; */
1947: /* *cx = dum; */
1948: /* dum = fu; */
1949: /* fu = *fc; */
1950: /* *fc =dum; */
1951: /* } else { /\* end *\/ */
1952: /* #ifdef DEBUG */
1953: /* printf("mnbrak3 fu < fc \n"); */
1954: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1955: /* #endif */
1956: /* dum=u; /\* Shifting c and u *\/ */
1957: /* u = *cx; */
1958: /* *cx = dum; */
1959: /* dum = fu; */
1960: /* fu = *fc; */
1961: /* *fc =dum; */
1962: /* } */
1.224 brouard 1963: #ifdef DEBUGMNBRAK
1964: double A, fparabu;
1965: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1966: fparabu= *fa - A*(*ax-u)*(*ax-u);
1967: 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);
1968: 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 1969: #endif
1.191 brouard 1970: dum=u; /* Shifting c and u */
1971: u = *cx;
1972: *cx = dum;
1973: dum = fu;
1974: fu = *fc;
1975: *fc =dum;
1.183 brouard 1976: #endif
1.162 brouard 1977: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1978: #ifdef DEBUG
1.224 brouard 1979: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1980: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1981: #endif
1.126 brouard 1982: fu=(*func)(u);
1983: if (fu < *fc) {
1.183 brouard 1984: #ifdef DEBUG
1.224 brouard 1985: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1986: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1987: #endif
1988: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1989: SHFT(*fb,*fc,fu,(*func)(u))
1990: #ifdef DEBUG
1991: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1992: #endif
1993: }
1.162 brouard 1994: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1995: #ifdef DEBUG
1.224 brouard 1996: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1997: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1998: #endif
1.126 brouard 1999: u=ulim;
2000: fu=(*func)(u);
1.183 brouard 2001: } else { /* u could be left to b (if r > q parabola has a maximum) */
2002: #ifdef DEBUG
1.224 brouard 2003: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2004: 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 2005: #endif
1.126 brouard 2006: u=(*cx)+GOLD*(*cx-*bx);
2007: fu=(*func)(u);
1.224 brouard 2008: #ifdef DEBUG
2009: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2010: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2011: #endif
1.183 brouard 2012: } /* end tests */
1.126 brouard 2013: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2014: SHFT(*fa,*fb,*fc,fu)
2015: #ifdef DEBUG
1.224 brouard 2016: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2017: 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 2018: #endif
2019: } /* 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 2020: }
2021:
2022: /*************** linmin ************************/
1.162 brouard 2023: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2024: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2025: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2026: the value of func at the returned location p . This is actually all accomplished by calling the
2027: routines mnbrak and brent .*/
1.126 brouard 2028: int ncom;
2029: double *pcom,*xicom;
2030: double (*nrfunc)(double []);
2031:
1.224 brouard 2032: #ifdef LINMINORIGINAL
1.126 brouard 2033: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2034: #else
2035: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2036: #endif
1.126 brouard 2037: {
2038: double brent(double ax, double bx, double cx,
2039: double (*f)(double), double tol, double *xmin);
2040: double f1dim(double x);
2041: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2042: double *fc, double (*func)(double));
2043: int j;
2044: double xx,xmin,bx,ax;
2045: double fx,fb,fa;
1.187 brouard 2046:
1.203 brouard 2047: #ifdef LINMINORIGINAL
2048: #else
2049: double scale=10., axs, xxs; /* Scale added for infinity */
2050: #endif
2051:
1.126 brouard 2052: ncom=n;
2053: pcom=vector(1,n);
2054: xicom=vector(1,n);
2055: nrfunc=func;
2056: for (j=1;j<=n;j++) {
2057: pcom[j]=p[j];
1.202 brouard 2058: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2059: }
1.187 brouard 2060:
1.203 brouard 2061: #ifdef LINMINORIGINAL
2062: xx=1.;
2063: #else
2064: axs=0.0;
2065: xxs=1.;
2066: do{
2067: xx= xxs;
2068: #endif
1.187 brouard 2069: ax=0.;
2070: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2071: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2072: /* 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)) */
2073: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2074: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2075: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2076: /* 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 2077: #ifdef LINMINORIGINAL
2078: #else
2079: if (fx != fx){
1.224 brouard 2080: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2081: printf("|");
2082: fprintf(ficlog,"|");
1.203 brouard 2083: #ifdef DEBUGLINMIN
1.224 brouard 2084: 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 2085: #endif
2086: }
1.224 brouard 2087: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2088: #endif
2089:
1.191 brouard 2090: #ifdef DEBUGLINMIN
2091: 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 2092: 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 2093: #endif
1.224 brouard 2094: #ifdef LINMINORIGINAL
2095: #else
2096: if(fb == fx){ /* Flat function in the direction */
2097: xmin=xx;
2098: *flat=1;
2099: }else{
2100: *flat=0;
2101: #endif
2102: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2103: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2104: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2105: /* fmin = f(p[j] + xmin * xi[j]) */
2106: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2107: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2108: #ifdef DEBUG
1.224 brouard 2109: 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);
2110: 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);
2111: #endif
2112: #ifdef LINMINORIGINAL
2113: #else
2114: }
1.126 brouard 2115: #endif
1.191 brouard 2116: #ifdef DEBUGLINMIN
2117: printf("linmin end ");
1.202 brouard 2118: fprintf(ficlog,"linmin end ");
1.191 brouard 2119: #endif
1.126 brouard 2120: for (j=1;j<=n;j++) {
1.203 brouard 2121: #ifdef LINMINORIGINAL
2122: xi[j] *= xmin;
2123: #else
2124: #ifdef DEBUGLINMIN
2125: if(xxs <1.0)
2126: printf(" before xi[%d]=%12.8f", j,xi[j]);
2127: #endif
2128: 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) */
2129: #ifdef DEBUGLINMIN
2130: if(xxs <1.0)
2131: 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 );
2132: #endif
2133: #endif
1.187 brouard 2134: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2135: }
1.191 brouard 2136: #ifdef DEBUGLINMIN
1.203 brouard 2137: printf("\n");
1.191 brouard 2138: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2139: 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 2140: for (j=1;j<=n;j++) {
1.202 brouard 2141: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2142: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2143: if(j % ncovmodel == 0){
1.191 brouard 2144: printf("\n");
1.202 brouard 2145: fprintf(ficlog,"\n");
2146: }
1.191 brouard 2147: }
1.203 brouard 2148: #else
1.191 brouard 2149: #endif
1.126 brouard 2150: free_vector(xicom,1,n);
2151: free_vector(pcom,1,n);
2152: }
2153:
2154:
2155: /*************** powell ************************/
1.162 brouard 2156: /*
2157: Minimization of a function func of n variables. Input consists of an initial starting point
2158: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2159: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2160: such that failure to decrease by more than this amount on one iteration signals doneness. On
2161: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2162: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2163: */
1.224 brouard 2164: #ifdef LINMINORIGINAL
2165: #else
2166: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2167: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2168: #endif
1.126 brouard 2169: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2170: double (*func)(double []))
2171: {
1.224 brouard 2172: #ifdef LINMINORIGINAL
2173: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2174: double (*func)(double []));
1.224 brouard 2175: #else
1.241 brouard 2176: void linmin(double p[], double xi[], int n, double *fret,
2177: double (*func)(double []),int *flat);
1.224 brouard 2178: #endif
1.239 brouard 2179: int i,ibig,j,jk,k;
1.126 brouard 2180: double del,t,*pt,*ptt,*xit;
1.181 brouard 2181: double directest;
1.126 brouard 2182: double fp,fptt;
2183: double *xits;
2184: int niterf, itmp;
1.224 brouard 2185: #ifdef LINMINORIGINAL
2186: #else
2187:
2188: flatdir=ivector(1,n);
2189: for (j=1;j<=n;j++) flatdir[j]=0;
2190: #endif
1.126 brouard 2191:
2192: pt=vector(1,n);
2193: ptt=vector(1,n);
2194: xit=vector(1,n);
2195: xits=vector(1,n);
2196: *fret=(*func)(p);
2197: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2198: rcurr_time = time(NULL);
1.126 brouard 2199: for (*iter=1;;++(*iter)) {
1.187 brouard 2200: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2201: ibig=0;
2202: del=0.0;
1.157 brouard 2203: rlast_time=rcurr_time;
2204: /* (void) gettimeofday(&curr_time,&tzp); */
2205: rcurr_time = time(NULL);
2206: curr_time = *localtime(&rcurr_time);
2207: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2208: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2209: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2210: for (i=1;i<=n;i++) {
1.126 brouard 2211: fprintf(ficrespow," %.12lf", p[i]);
2212: }
1.239 brouard 2213: fprintf(ficrespow,"\n");fflush(ficrespow);
2214: printf("\n#model= 1 + age ");
2215: fprintf(ficlog,"\n#model= 1 + age ");
2216: if(nagesqr==1){
1.241 brouard 2217: printf(" + age*age ");
2218: fprintf(ficlog," + age*age ");
1.239 brouard 2219: }
2220: for(j=1;j <=ncovmodel-2;j++){
2221: if(Typevar[j]==0) {
2222: printf(" + V%d ",Tvar[j]);
2223: fprintf(ficlog," + V%d ",Tvar[j]);
2224: }else if(Typevar[j]==1) {
2225: printf(" + V%d*age ",Tvar[j]);
2226: fprintf(ficlog," + V%d*age ",Tvar[j]);
2227: }else if(Typevar[j]==2) {
2228: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2229: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2230: }
2231: }
1.126 brouard 2232: printf("\n");
1.239 brouard 2233: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2234: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2235: fprintf(ficlog,"\n");
1.239 brouard 2236: for(i=1,jk=1; i <=nlstate; i++){
2237: for(k=1; k <=(nlstate+ndeath); k++){
2238: if (k != i) {
2239: printf("%d%d ",i,k);
2240: fprintf(ficlog,"%d%d ",i,k);
2241: for(j=1; j <=ncovmodel; j++){
2242: printf("%12.7f ",p[jk]);
2243: fprintf(ficlog,"%12.7f ",p[jk]);
2244: jk++;
2245: }
2246: printf("\n");
2247: fprintf(ficlog,"\n");
2248: }
2249: }
2250: }
1.241 brouard 2251: if(*iter <=3 && *iter >1){
1.157 brouard 2252: tml = *localtime(&rcurr_time);
2253: strcpy(strcurr,asctime(&tml));
2254: rforecast_time=rcurr_time;
1.126 brouard 2255: itmp = strlen(strcurr);
2256: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2257: strcurr[itmp-1]='\0';
1.162 brouard 2258: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2259: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2260: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2261: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2262: forecast_time = *localtime(&rforecast_time);
2263: strcpy(strfor,asctime(&forecast_time));
2264: itmp = strlen(strfor);
2265: if(strfor[itmp-1]=='\n')
2266: strfor[itmp-1]='\0';
2267: 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);
2268: 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 2269: }
2270: }
1.187 brouard 2271: for (i=1;i<=n;i++) { /* For each direction i */
2272: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2273: fptt=(*fret);
2274: #ifdef DEBUG
1.203 brouard 2275: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2276: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2277: #endif
1.203 brouard 2278: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2279: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2280: #ifdef LINMINORIGINAL
1.188 brouard 2281: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2282: #else
2283: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2284: flatdir[i]=flat; /* Function is vanishing in that direction i */
2285: #endif
2286: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2287: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2288: /* because that direction will be replaced unless the gain del is small */
2289: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2290: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2291: /* with the new direction. */
2292: del=fabs(fptt-(*fret));
2293: ibig=i;
1.126 brouard 2294: }
2295: #ifdef DEBUG
2296: printf("%d %.12e",i,(*fret));
2297: fprintf(ficlog,"%d %.12e",i,(*fret));
2298: for (j=1;j<=n;j++) {
1.224 brouard 2299: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2300: printf(" x(%d)=%.12e",j,xit[j]);
2301: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2302: }
2303: for(j=1;j<=n;j++) {
1.225 brouard 2304: printf(" p(%d)=%.12e",j,p[j]);
2305: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2306: }
2307: printf("\n");
2308: fprintf(ficlog,"\n");
2309: #endif
1.187 brouard 2310: } /* end loop on each direction i */
2311: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2312: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2313: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2314: for(j=1;j<=n;j++) {
1.225 brouard 2315: if(flatdir[j] >0){
2316: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2317: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2318: }
2319: /* printf("\n"); */
2320: /* fprintf(ficlog,"\n"); */
2321: }
1.243 brouard 2322: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2323: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2324: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2325: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2326: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2327: /* decreased of more than 3.84 */
2328: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2329: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2330: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2331:
1.188 brouard 2332: /* Starting the program with initial values given by a former maximization will simply change */
2333: /* the scales of the directions and the directions, because the are reset to canonical directions */
2334: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2335: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2336: #ifdef DEBUG
2337: int k[2],l;
2338: k[0]=1;
2339: k[1]=-1;
2340: printf("Max: %.12e",(*func)(p));
2341: fprintf(ficlog,"Max: %.12e",(*func)(p));
2342: for (j=1;j<=n;j++) {
2343: printf(" %.12e",p[j]);
2344: fprintf(ficlog," %.12e",p[j]);
2345: }
2346: printf("\n");
2347: fprintf(ficlog,"\n");
2348: for(l=0;l<=1;l++) {
2349: for (j=1;j<=n;j++) {
2350: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2351: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2352: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2353: }
2354: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2355: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2356: }
2357: #endif
2358:
1.224 brouard 2359: #ifdef LINMINORIGINAL
2360: #else
2361: free_ivector(flatdir,1,n);
2362: #endif
1.126 brouard 2363: free_vector(xit,1,n);
2364: free_vector(xits,1,n);
2365: free_vector(ptt,1,n);
2366: free_vector(pt,1,n);
2367: return;
1.192 brouard 2368: } /* enough precision */
1.240 brouard 2369: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2370: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2371: ptt[j]=2.0*p[j]-pt[j];
2372: xit[j]=p[j]-pt[j];
2373: pt[j]=p[j];
2374: }
1.181 brouard 2375: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2376: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2377: if (*iter <=4) {
1.225 brouard 2378: #else
2379: #endif
1.224 brouard 2380: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2381: #else
1.161 brouard 2382: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2383: #endif
1.162 brouard 2384: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2385: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2386: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2387: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2388: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2389: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2390: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2391: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2392: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2393: /* Even if f3 <f1, directest can be negative and t >0 */
2394: /* mu² and del² are equal when f3=f1 */
2395: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2396: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2397: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2398: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2399: #ifdef NRCORIGINAL
2400: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2401: #else
2402: 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 2403: t= t- del*SQR(fp-fptt);
1.183 brouard 2404: #endif
1.202 brouard 2405: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2406: #ifdef DEBUG
1.181 brouard 2407: 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);
2408: 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 2409: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2410: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2411: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2412: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2413: 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);
2414: 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);
2415: #endif
1.183 brouard 2416: #ifdef POWELLORIGINAL
2417: if (t < 0.0) { /* Then we use it for new direction */
2418: #else
1.182 brouard 2419: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2420: 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 2421: 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 2422: 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 2423: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2424: }
1.181 brouard 2425: if (directest < 0.0) { /* Then we use it for new direction */
2426: #endif
1.191 brouard 2427: #ifdef DEBUGLINMIN
1.234 brouard 2428: printf("Before linmin in direction P%d-P0\n",n);
2429: for (j=1;j<=n;j++) {
2430: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2431: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2432: if(j % ncovmodel == 0){
2433: printf("\n");
2434: fprintf(ficlog,"\n");
2435: }
2436: }
1.224 brouard 2437: #endif
2438: #ifdef LINMINORIGINAL
1.234 brouard 2439: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2440: #else
1.234 brouard 2441: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2442: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2443: #endif
1.234 brouard 2444:
1.191 brouard 2445: #ifdef DEBUGLINMIN
1.234 brouard 2446: for (j=1;j<=n;j++) {
2447: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2448: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2449: if(j % ncovmodel == 0){
2450: printf("\n");
2451: fprintf(ficlog,"\n");
2452: }
2453: }
1.224 brouard 2454: #endif
1.234 brouard 2455: for (j=1;j<=n;j++) {
2456: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2457: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2458: }
1.224 brouard 2459: #ifdef LINMINORIGINAL
2460: #else
1.234 brouard 2461: for (j=1, flatd=0;j<=n;j++) {
2462: if(flatdir[j]>0)
2463: flatd++;
2464: }
2465: if(flatd >0){
1.255 brouard 2466: printf("%d flat directions: ",flatd);
2467: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2468: for (j=1;j<=n;j++) {
2469: if(flatdir[j]>0){
2470: printf("%d ",j);
2471: fprintf(ficlog,"%d ",j);
2472: }
2473: }
2474: printf("\n");
2475: fprintf(ficlog,"\n");
2476: }
1.191 brouard 2477: #endif
1.234 brouard 2478: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2479: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2480:
1.126 brouard 2481: #ifdef DEBUG
1.234 brouard 2482: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2483: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2484: for(j=1;j<=n;j++){
2485: printf(" %lf",xit[j]);
2486: fprintf(ficlog," %lf",xit[j]);
2487: }
2488: printf("\n");
2489: fprintf(ficlog,"\n");
1.126 brouard 2490: #endif
1.192 brouard 2491: } /* end of t or directest negative */
1.224 brouard 2492: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2493: #else
1.234 brouard 2494: } /* end if (fptt < fp) */
1.192 brouard 2495: #endif
1.225 brouard 2496: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2497: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2498: #else
1.224 brouard 2499: #endif
1.234 brouard 2500: } /* loop iteration */
1.126 brouard 2501: }
1.234 brouard 2502:
1.126 brouard 2503: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2504:
1.235 brouard 2505: 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 2506: {
1.235 brouard 2507: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2508: (and selected quantitative values in nres)
2509: by left multiplying the unit
1.234 brouard 2510: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2511: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2512: /* Wx is row vector: population in state 1, population in state 2, population dead */
2513: /* or prevalence in state 1, prevalence in state 2, 0 */
2514: /* newm is the matrix after multiplications, its rows are identical at a factor */
2515: /* Initial matrix pimij */
2516: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2517: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2518: /* 0, 0 , 1} */
2519: /*
2520: * and after some iteration: */
2521: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2522: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2523: /* 0, 0 , 1} */
2524: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2525: /* {0.51571254859325999, 0.4842874514067399, */
2526: /* 0.51326036147820708, 0.48673963852179264} */
2527: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2528:
1.126 brouard 2529: int i, ii,j,k;
1.209 brouard 2530: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2531: /* double **matprod2(); */ /* test */
1.218 brouard 2532: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2533: double **newm;
1.209 brouard 2534: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2535: int ncvloop=0;
1.169 brouard 2536:
1.209 brouard 2537: min=vector(1,nlstate);
2538: max=vector(1,nlstate);
2539: meandiff=vector(1,nlstate);
2540:
1.218 brouard 2541: /* Starting with matrix unity */
1.126 brouard 2542: for (ii=1;ii<=nlstate+ndeath;ii++)
2543: for (j=1;j<=nlstate+ndeath;j++){
2544: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2545: }
1.169 brouard 2546:
2547: cov[1]=1.;
2548:
2549: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2550: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2551: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2552: ncvloop++;
1.126 brouard 2553: newm=savm;
2554: /* Covariates have to be included here again */
1.138 brouard 2555: cov[2]=agefin;
1.187 brouard 2556: if(nagesqr==1)
2557: cov[3]= agefin*agefin;;
1.234 brouard 2558: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2559: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2560: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2561: /* 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 2562: }
2563: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2564: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2565: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2566: /* 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 2567: }
1.237 brouard 2568: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2569: if(Dummy[Tvar[Tage[k]]]){
2570: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2571: } else{
1.235 brouard 2572: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2573: }
1.235 brouard 2574: /* 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 2575: }
1.237 brouard 2576: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2577: /* 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 2578: if(Dummy[Tvard[k][1]==0]){
2579: if(Dummy[Tvard[k][2]==0]){
2580: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2581: }else{
2582: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2583: }
2584: }else{
2585: if(Dummy[Tvard[k][2]==0]){
2586: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2587: }else{
2588: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2589: }
2590: }
1.234 brouard 2591: }
1.138 brouard 2592: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2593: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2594: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2595: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2596: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2597: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2598: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2599:
1.126 brouard 2600: savm=oldm;
2601: oldm=newm;
1.209 brouard 2602:
2603: for(j=1; j<=nlstate; j++){
2604: max[j]=0.;
2605: min[j]=1.;
2606: }
2607: for(i=1;i<=nlstate;i++){
2608: sumnew=0;
2609: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2610: for(j=1; j<=nlstate; j++){
2611: prlim[i][j]= newm[i][j]/(1-sumnew);
2612: max[j]=FMAX(max[j],prlim[i][j]);
2613: min[j]=FMIN(min[j],prlim[i][j]);
2614: }
2615: }
2616:
1.126 brouard 2617: maxmax=0.;
1.209 brouard 2618: for(j=1; j<=nlstate; j++){
2619: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2620: maxmax=FMAX(maxmax,meandiff[j]);
2621: /* 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 2622: } /* j loop */
1.203 brouard 2623: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2624: /* 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 2625: if(maxmax < ftolpl){
1.209 brouard 2626: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2627: free_vector(min,1,nlstate);
2628: free_vector(max,1,nlstate);
2629: free_vector(meandiff,1,nlstate);
1.126 brouard 2630: return prlim;
2631: }
1.169 brouard 2632: } /* age loop */
1.208 brouard 2633: /* After some age loop it doesn't converge */
1.209 brouard 2634: 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 2635: 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 2636: /* 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); */
2637: free_vector(min,1,nlstate);
2638: free_vector(max,1,nlstate);
2639: free_vector(meandiff,1,nlstate);
1.208 brouard 2640:
1.169 brouard 2641: return prlim; /* should not reach here */
1.126 brouard 2642: }
2643:
1.217 brouard 2644:
2645: /**** Back Prevalence limit (stable or period prevalence) ****************/
2646:
1.218 brouard 2647: /* 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) */
2648: /* 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 2649: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2650: {
1.264 brouard 2651: /* 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 2652: matrix by transitions matrix until convergence is reached with precision ftolpl */
2653: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2654: /* Wx is row vector: population in state 1, population in state 2, population dead */
2655: /* or prevalence in state 1, prevalence in state 2, 0 */
2656: /* newm is the matrix after multiplications, its rows are identical at a factor */
2657: /* Initial matrix pimij */
2658: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2659: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2660: /* 0, 0 , 1} */
2661: /*
2662: * and after some iteration: */
2663: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2664: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2665: /* 0, 0 , 1} */
2666: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2667: /* {0.51571254859325999, 0.4842874514067399, */
2668: /* 0.51326036147820708, 0.48673963852179264} */
2669: /* If we start from prlim again, prlim tends to a constant matrix */
2670:
2671: int i, ii,j,k;
1.247 brouard 2672: int first=0;
1.217 brouard 2673: double *min, *max, *meandiff, maxmax,sumnew=0.;
2674: /* double **matprod2(); */ /* test */
2675: double **out, cov[NCOVMAX+1], **bmij();
2676: double **newm;
1.218 brouard 2677: double **dnewm, **doldm, **dsavm; /* for use */
2678: double **oldm, **savm; /* for use */
2679:
1.217 brouard 2680: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2681: int ncvloop=0;
2682:
2683: min=vector(1,nlstate);
2684: max=vector(1,nlstate);
2685: meandiff=vector(1,nlstate);
2686:
1.266 brouard 2687: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2688: oldm=oldms; savm=savms;
2689:
2690: /* Starting with matrix unity */
2691: for (ii=1;ii<=nlstate+ndeath;ii++)
2692: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2693: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2694: }
2695:
2696: cov[1]=1.;
2697:
2698: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2699: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2700: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2701: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2702: ncvloop++;
1.218 brouard 2703: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2704: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2705: /* Covariates have to be included here again */
2706: cov[2]=agefin;
2707: if(nagesqr==1)
2708: cov[3]= agefin*agefin;;
1.242 brouard 2709: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2710: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2711: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2712: /* 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 2713: }
2714: /* for (k=1; k<=cptcovn;k++) { */
2715: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2716: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2717: /* /\* 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])]); *\/ */
2718: /* } */
2719: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2720: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2721: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2722: /* 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]); */
2723: }
2724: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2725: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2726: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2727: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2728: for (k=1; k<=cptcovage;k++){ /* For product with age */
2729: if(Dummy[Tvar[Tage[k]]]){
2730: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2731: } else{
2732: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2733: }
2734: /* 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]); */
2735: }
2736: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2737: /* 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]); */
2738: if(Dummy[Tvard[k][1]==0]){
2739: if(Dummy[Tvard[k][2]==0]){
2740: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2741: }else{
2742: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2743: }
2744: }else{
2745: if(Dummy[Tvard[k][2]==0]){
2746: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2747: }else{
2748: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2749: }
2750: }
1.217 brouard 2751: }
2752:
2753: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2754: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2755: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2756: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2757: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2758: /* ij should be linked to the correct index of cov */
2759: /* age and covariate values ij are in 'cov', but we need to pass
2760: * ij for the observed prevalence at age and status and covariate
2761: * number: prevacurrent[(int)agefin][ii][ij]
2762: */
2763: /* 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 *\/ */
2764: /* 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 *\/ */
2765: 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 2766: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2767: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2768: /* for(i=1; i<=nlstate+ndeath; i++) { */
2769: /* printf("%d newm= ",i); */
2770: /* for(j=1;j<=nlstate+ndeath;j++) { */
2771: /* printf("%f ",newm[i][j]); */
2772: /* } */
2773: /* printf("oldm * "); */
2774: /* for(j=1;j<=nlstate+ndeath;j++) { */
2775: /* printf("%f ",oldm[i][j]); */
2776: /* } */
1.268 brouard 2777: /* printf(" bmmij "); */
1.266 brouard 2778: /* for(j=1;j<=nlstate+ndeath;j++) { */
2779: /* printf("%f ",pmmij[i][j]); */
2780: /* } */
2781: /* printf("\n"); */
2782: /* } */
2783: /* } */
1.217 brouard 2784: savm=oldm;
2785: oldm=newm;
1.266 brouard 2786:
1.217 brouard 2787: for(j=1; j<=nlstate; j++){
2788: max[j]=0.;
2789: min[j]=1.;
2790: }
2791: for(j=1; j<=nlstate; j++){
2792: for(i=1;i<=nlstate;i++){
1.234 brouard 2793: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2794: bprlim[i][j]= newm[i][j];
2795: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2796: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2797: }
2798: }
1.218 brouard 2799:
1.217 brouard 2800: maxmax=0.;
2801: for(i=1; i<=nlstate; i++){
2802: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2803: maxmax=FMAX(maxmax,meandiff[i]);
2804: /* 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 2805: } /* i loop */
1.217 brouard 2806: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2807: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2808: if(maxmax < ftolpl){
1.220 brouard 2809: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2810: free_vector(min,1,nlstate);
2811: free_vector(max,1,nlstate);
2812: free_vector(meandiff,1,nlstate);
2813: return bprlim;
2814: }
2815: } /* age loop */
2816: /* After some age loop it doesn't converge */
1.247 brouard 2817: if(first){
2818: first=1;
2819: 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\
2820: 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);
2821: }
2822: 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 2823: 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);
2824: /* 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); */
2825: free_vector(min,1,nlstate);
2826: free_vector(max,1,nlstate);
2827: free_vector(meandiff,1,nlstate);
2828:
2829: return bprlim; /* should not reach here */
2830: }
2831:
1.126 brouard 2832: /*************** transition probabilities ***************/
2833:
2834: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2835: {
1.138 brouard 2836: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2837: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2838: model to the ncovmodel covariates (including constant and age).
2839: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2840: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2841: ncth covariate in the global vector x is given by the formula:
2842: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2843: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2844: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2845: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2846: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2847: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2848: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2849: */
2850: double s1, lnpijopii;
1.126 brouard 2851: /*double t34;*/
1.164 brouard 2852: int i,j, nc, ii, jj;
1.126 brouard 2853:
1.223 brouard 2854: for(i=1; i<= nlstate; i++){
2855: for(j=1; j<i;j++){
2856: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2857: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2858: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2859: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2860: }
2861: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2862: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2863: }
2864: for(j=i+1; j<=nlstate+ndeath;j++){
2865: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2866: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2867: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2868: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2869: }
2870: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2871: }
2872: }
1.218 brouard 2873:
1.223 brouard 2874: for(i=1; i<= nlstate; i++){
2875: s1=0;
2876: for(j=1; j<i; j++){
2877: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2878: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2879: }
2880: for(j=i+1; j<=nlstate+ndeath; j++){
2881: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2882: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2883: }
2884: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2885: ps[i][i]=1./(s1+1.);
2886: /* Computing other pijs */
2887: for(j=1; j<i; j++)
2888: ps[i][j]= exp(ps[i][j])*ps[i][i];
2889: for(j=i+1; j<=nlstate+ndeath; j++)
2890: ps[i][j]= exp(ps[i][j])*ps[i][i];
2891: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2892: } /* end i */
1.218 brouard 2893:
1.223 brouard 2894: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2895: for(jj=1; jj<= nlstate+ndeath; jj++){
2896: ps[ii][jj]=0;
2897: ps[ii][ii]=1;
2898: }
2899: }
1.218 brouard 2900:
2901:
1.223 brouard 2902: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2903: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2904: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2905: /* } */
2906: /* printf("\n "); */
2907: /* } */
2908: /* printf("\n ");printf("%lf ",cov[2]);*/
2909: /*
2910: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2911: goto end;*/
1.266 brouard 2912: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2913: }
2914:
1.218 brouard 2915: /*************** backward transition probabilities ***************/
2916:
2917: /* 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 ) */
2918: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2919: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2920: {
1.266 brouard 2921: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2922: * 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 2923: */
1.218 brouard 2924: int i, ii, j,k;
1.222 brouard 2925:
2926: double **out, **pmij();
2927: double sumnew=0.;
1.218 brouard 2928: double agefin;
1.268 brouard 2929: 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 2930: double **dnewm, **dsavm, **doldm;
2931: double **bbmij;
2932:
1.218 brouard 2933: doldm=ddoldms; /* global pointers */
1.222 brouard 2934: dnewm=ddnewms;
2935: dsavm=ddsavms;
2936:
2937: agefin=cov[2];
1.268 brouard 2938: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2939: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2940: the observed prevalence (with this covariate ij) at beginning of transition */
2941: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2942:
2943: /* P_x */
1.266 brouard 2944: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2945: /* outputs pmmij which is a stochastic matrix in row */
2946:
2947: /* Diag(w_x) */
2948: /* Problem with prevacurrent which can be zero */
2949: sumnew=0.;
1.269 brouard 2950: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2951: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2952: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2953: sumnew+=prevacurrent[(int)agefin][ii][ij];
2954: }
2955: if(sumnew >0.01){ /* At least some value in the prevalence */
2956: for (ii=1;ii<=nlstate+ndeath;ii++){
2957: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2958: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2959: }
2960: }else{
2961: for (ii=1;ii<=nlstate+ndeath;ii++){
2962: for (j=1;j<=nlstate+ndeath;j++)
2963: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
2964: }
2965: /* if(sumnew <0.9){ */
2966: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2967: /* } */
2968: }
2969: k3=0.0; /* We put the last diagonal to 0 */
2970: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
2971: doldm[ii][ii]= k3;
2972: }
2973: /* End doldm, At the end doldm is diag[(w_i)] */
2974:
2975: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
2976: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
2977:
2978: /* Diag(Sum_i w^i_x p^ij_x */
2979: /* 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 2980: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 2981: sumnew=0.;
1.222 brouard 2982: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 2983: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 2984: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 2985: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 2986: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 2987: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 2988: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2989: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 2990: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2991: /* }else */
1.268 brouard 2992: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2993: } /*End ii */
2994: } /* 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 */
2995:
2996: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
2997: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 2998: /* end bmij */
1.266 brouard 2999: return ps; /*pointer is unchanged */
1.218 brouard 3000: }
1.217 brouard 3001: /*************** transition probabilities ***************/
3002:
1.218 brouard 3003: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3004: {
3005: /* According to parameters values stored in x and the covariate's values stored in cov,
3006: computes the probability to be observed in state j being in state i by appying the
3007: model to the ncovmodel covariates (including constant and age).
3008: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3009: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3010: ncth covariate in the global vector x is given by the formula:
3011: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3012: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3013: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3014: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3015: Outputs ps[i][j] the probability to be observed in j being in j according to
3016: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3017: */
3018: double s1, lnpijopii;
3019: /*double t34;*/
3020: int i,j, nc, ii, jj;
3021:
1.234 brouard 3022: for(i=1; i<= nlstate; i++){
3023: for(j=1; j<i;j++){
3024: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3025: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3026: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3027: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3028: }
3029: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3030: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3031: }
3032: for(j=i+1; j<=nlstate+ndeath;j++){
3033: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3034: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3035: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3036: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3037: }
3038: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3039: }
3040: }
3041:
3042: for(i=1; i<= nlstate; i++){
3043: s1=0;
3044: for(j=1; j<i; j++){
3045: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3046: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3047: }
3048: for(j=i+1; j<=nlstate+ndeath; j++){
3049: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3050: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3051: }
3052: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3053: ps[i][i]=1./(s1+1.);
3054: /* Computing other pijs */
3055: for(j=1; j<i; j++)
3056: ps[i][j]= exp(ps[i][j])*ps[i][i];
3057: for(j=i+1; j<=nlstate+ndeath; j++)
3058: ps[i][j]= exp(ps[i][j])*ps[i][i];
3059: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3060: } /* end i */
3061:
3062: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3063: for(jj=1; jj<= nlstate+ndeath; jj++){
3064: ps[ii][jj]=0;
3065: ps[ii][ii]=1;
3066: }
3067: }
3068: /* Added for backcast */ /* Transposed matrix too */
3069: for(jj=1; jj<= nlstate+ndeath; jj++){
3070: s1=0.;
3071: for(ii=1; ii<= nlstate+ndeath; ii++){
3072: s1+=ps[ii][jj];
3073: }
3074: for(ii=1; ii<= nlstate; ii++){
3075: ps[ii][jj]=ps[ii][jj]/s1;
3076: }
3077: }
3078: /* Transposition */
3079: for(jj=1; jj<= nlstate+ndeath; jj++){
3080: for(ii=jj; ii<= nlstate+ndeath; ii++){
3081: s1=ps[ii][jj];
3082: ps[ii][jj]=ps[jj][ii];
3083: ps[jj][ii]=s1;
3084: }
3085: }
3086: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3087: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3088: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3089: /* } */
3090: /* printf("\n "); */
3091: /* } */
3092: /* printf("\n ");printf("%lf ",cov[2]);*/
3093: /*
3094: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3095: goto end;*/
3096: return ps;
1.217 brouard 3097: }
3098:
3099:
1.126 brouard 3100: /**************** Product of 2 matrices ******************/
3101:
1.145 brouard 3102: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3103: {
3104: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3105: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3106: /* in, b, out are matrice of pointers which should have been initialized
3107: before: only the contents of out is modified. The function returns
3108: a pointer to pointers identical to out */
1.145 brouard 3109: int i, j, k;
1.126 brouard 3110: for(i=nrl; i<= nrh; i++)
1.145 brouard 3111: for(k=ncolol; k<=ncoloh; k++){
3112: out[i][k]=0.;
3113: for(j=ncl; j<=nch; j++)
3114: out[i][k] +=in[i][j]*b[j][k];
3115: }
1.126 brouard 3116: return out;
3117: }
3118:
3119:
3120: /************* Higher Matrix Product ***************/
3121:
1.235 brouard 3122: 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 3123: {
1.218 brouard 3124: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3125: 'nhstepm*hstepm*stepm' months (i.e. until
3126: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3127: nhstepm*hstepm matrices.
3128: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3129: (typically every 2 years instead of every month which is too big
3130: for the memory).
3131: Model is determined by parameters x and covariates have to be
3132: included manually here.
3133:
3134: */
3135:
3136: int i, j, d, h, k;
1.131 brouard 3137: double **out, cov[NCOVMAX+1];
1.126 brouard 3138: double **newm;
1.187 brouard 3139: double agexact;
1.214 brouard 3140: double agebegin, ageend;
1.126 brouard 3141:
3142: /* Hstepm could be zero and should return the unit matrix */
3143: for (i=1;i<=nlstate+ndeath;i++)
3144: for (j=1;j<=nlstate+ndeath;j++){
3145: oldm[i][j]=(i==j ? 1.0 : 0.0);
3146: po[i][j][0]=(i==j ? 1.0 : 0.0);
3147: }
3148: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3149: for(h=1; h <=nhstepm; h++){
3150: for(d=1; d <=hstepm; d++){
3151: newm=savm;
3152: /* Covariates have to be included here again */
3153: cov[1]=1.;
1.214 brouard 3154: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3155: cov[2]=agexact;
3156: if(nagesqr==1)
1.227 brouard 3157: cov[3]= agexact*agexact;
1.235 brouard 3158: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3159: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3160: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3161: /* 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)); */
3162: }
3163: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3164: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3165: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3166: /* 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]); */
3167: }
3168: for (k=1; k<=cptcovage;k++){
3169: if(Dummy[Tvar[Tage[k]]]){
3170: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3171: } else{
3172: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3173: }
3174: /* 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]); */
3175: }
3176: for (k=1; k<=cptcovprod;k++){ /* */
3177: /* 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]); */
3178: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3179: }
3180: /* for (k=1; k<=cptcovn;k++) */
3181: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3182: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3183: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3184: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3185: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3186:
3187:
1.126 brouard 3188: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3189: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3190: /* right multiplication of oldm by the current matrix */
1.126 brouard 3191: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3192: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3193: /* if((int)age == 70){ */
3194: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3195: /* for(i=1; i<=nlstate+ndeath; i++) { */
3196: /* printf("%d pmmij ",i); */
3197: /* for(j=1;j<=nlstate+ndeath;j++) { */
3198: /* printf("%f ",pmmij[i][j]); */
3199: /* } */
3200: /* printf(" oldm "); */
3201: /* for(j=1;j<=nlstate+ndeath;j++) { */
3202: /* printf("%f ",oldm[i][j]); */
3203: /* } */
3204: /* printf("\n"); */
3205: /* } */
3206: /* } */
1.126 brouard 3207: savm=oldm;
3208: oldm=newm;
3209: }
3210: for(i=1; i<=nlstate+ndeath; i++)
3211: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3212: po[i][j][h]=newm[i][j];
3213: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3214: }
1.128 brouard 3215: /*printf("h=%d ",h);*/
1.126 brouard 3216: } /* end h */
1.267 brouard 3217: /* printf("\n H=%d \n",h); */
1.126 brouard 3218: return po;
3219: }
3220:
1.217 brouard 3221: /************* Higher Back Matrix Product ***************/
1.218 brouard 3222: /* 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 3223: 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 3224: {
1.266 brouard 3225: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3226: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3227: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3228: nhstepm*hstepm matrices.
3229: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3230: (typically every 2 years instead of every month which is too big
1.217 brouard 3231: for the memory).
1.218 brouard 3232: Model is determined by parameters x and covariates have to be
1.266 brouard 3233: included manually here. Then we use a call to bmij(x and cov)
3234: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3235: */
1.217 brouard 3236:
3237: int i, j, d, h, k;
1.266 brouard 3238: double **out, cov[NCOVMAX+1], **bmij();
3239: double **newm, ***newmm;
1.217 brouard 3240: double agexact;
3241: double agebegin, ageend;
1.222 brouard 3242: double **oldm, **savm;
1.217 brouard 3243:
1.266 brouard 3244: newmm=po; /* To be saved */
3245: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3246: /* Hstepm could be zero and should return the unit matrix */
3247: for (i=1;i<=nlstate+ndeath;i++)
3248: for (j=1;j<=nlstate+ndeath;j++){
3249: oldm[i][j]=(i==j ? 1.0 : 0.0);
3250: po[i][j][0]=(i==j ? 1.0 : 0.0);
3251: }
3252: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3253: for(h=1; h <=nhstepm; h++){
3254: for(d=1; d <=hstepm; d++){
3255: newm=savm;
3256: /* Covariates have to be included here again */
3257: cov[1]=1.;
1.271 brouard 3258: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3259: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3260: cov[2]=agexact;
3261: if(nagesqr==1)
1.222 brouard 3262: cov[3]= agexact*agexact;
1.266 brouard 3263: for (k=1; k<=cptcovn;k++){
3264: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3265: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3266: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3267: /* 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)); */
3268: }
1.267 brouard 3269: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3270: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3271: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3272: /* 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]); */
3273: }
3274: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3275: if(Dummy[Tvar[Tage[k]]]){
3276: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3277: } else{
3278: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3279: }
3280: /* 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]); */
3281: }
3282: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3283: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3284: }
1.217 brouard 3285: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3286: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3287:
1.218 brouard 3288: /* Careful transposed matrix */
1.266 brouard 3289: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3290: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3291: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3292: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3293: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3294: /* if((int)age == 70){ */
3295: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3296: /* for(i=1; i<=nlstate+ndeath; i++) { */
3297: /* printf("%d pmmij ",i); */
3298: /* for(j=1;j<=nlstate+ndeath;j++) { */
3299: /* printf("%f ",pmmij[i][j]); */
3300: /* } */
3301: /* printf(" oldm "); */
3302: /* for(j=1;j<=nlstate+ndeath;j++) { */
3303: /* printf("%f ",oldm[i][j]); */
3304: /* } */
3305: /* printf("\n"); */
3306: /* } */
3307: /* } */
3308: savm=oldm;
3309: oldm=newm;
3310: }
3311: for(i=1; i<=nlstate+ndeath; i++)
3312: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3313: po[i][j][h]=newm[i][j];
1.268 brouard 3314: /* if(h==nhstepm) */
3315: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3316: }
1.268 brouard 3317: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3318: } /* end h */
1.268 brouard 3319: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3320: return po;
3321: }
3322:
3323:
1.162 brouard 3324: #ifdef NLOPT
3325: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3326: double fret;
3327: double *xt;
3328: int j;
3329: myfunc_data *d2 = (myfunc_data *) pd;
3330: /* xt = (p1-1); */
3331: xt=vector(1,n);
3332: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3333:
3334: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3335: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3336: printf("Function = %.12lf ",fret);
3337: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3338: printf("\n");
3339: free_vector(xt,1,n);
3340: return fret;
3341: }
3342: #endif
1.126 brouard 3343:
3344: /*************** log-likelihood *************/
3345: double func( double *x)
3346: {
1.226 brouard 3347: int i, ii, j, k, mi, d, kk;
3348: int ioffset=0;
3349: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3350: double **out;
3351: double lli; /* Individual log likelihood */
3352: int s1, s2;
1.228 brouard 3353: 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 3354: double bbh, survp;
3355: long ipmx;
3356: double agexact;
3357: /*extern weight */
3358: /* We are differentiating ll according to initial status */
3359: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3360: /*for(i=1;i<imx;i++)
3361: printf(" %d\n",s[4][i]);
3362: */
1.162 brouard 3363:
1.226 brouard 3364: ++countcallfunc;
1.162 brouard 3365:
1.226 brouard 3366: cov[1]=1.;
1.126 brouard 3367:
1.226 brouard 3368: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3369: ioffset=0;
1.226 brouard 3370: if(mle==1){
3371: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3372: /* Computes the values of the ncovmodel covariates of the model
3373: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3374: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3375: to be observed in j being in i according to the model.
3376: */
1.243 brouard 3377: ioffset=2+nagesqr ;
1.233 brouard 3378: /* Fixed */
1.234 brouard 3379: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3380: 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)*/
3381: }
1.226 brouard 3382: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3383: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3384: has been calculated etc */
3385: /* For an individual i, wav[i] gives the number of effective waves */
3386: /* We compute the contribution to Likelihood of each effective transition
3387: mw[mi][i] is real wave of the mi th effectve wave */
3388: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3389: s2=s[mw[mi+1][i]][i];
3390: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3391: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3392: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3393: */
3394: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3395: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3396: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3397: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3398: }
3399: for (ii=1;ii<=nlstate+ndeath;ii++)
3400: for (j=1;j<=nlstate+ndeath;j++){
3401: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3402: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3403: }
3404: for(d=0; d<dh[mi][i]; d++){
3405: newm=savm;
3406: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3407: cov[2]=agexact;
3408: if(nagesqr==1)
3409: cov[3]= agexact*agexact; /* Should be changed here */
3410: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3411: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3412: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3413: else
3414: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3415: }
3416: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3417: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3418: savm=oldm;
3419: oldm=newm;
3420: } /* end mult */
3421:
3422: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3423: /* But now since version 0.9 we anticipate for bias at large stepm.
3424: * If stepm is larger than one month (smallest stepm) and if the exact delay
3425: * (in months) between two waves is not a multiple of stepm, we rounded to
3426: * the nearest (and in case of equal distance, to the lowest) interval but now
3427: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3428: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3429: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3430: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3431: * -stepm/2 to stepm/2 .
3432: * For stepm=1 the results are the same as for previous versions of Imach.
3433: * For stepm > 1 the results are less biased than in previous versions.
3434: */
1.234 brouard 3435: s1=s[mw[mi][i]][i];
3436: s2=s[mw[mi+1][i]][i];
3437: bbh=(double)bh[mi][i]/(double)stepm;
3438: /* bias bh is positive if real duration
3439: * is higher than the multiple of stepm and negative otherwise.
3440: */
3441: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3442: if( s2 > nlstate){
3443: /* i.e. if s2 is a death state and if the date of death is known
3444: then the contribution to the likelihood is the probability to
3445: die between last step unit time and current step unit time,
3446: which is also equal to probability to die before dh
3447: minus probability to die before dh-stepm .
3448: In version up to 0.92 likelihood was computed
3449: as if date of death was unknown. Death was treated as any other
3450: health state: the date of the interview describes the actual state
3451: and not the date of a change in health state. The former idea was
3452: to consider that at each interview the state was recorded
3453: (healthy, disable or death) and IMaCh was corrected; but when we
3454: introduced the exact date of death then we should have modified
3455: the contribution of an exact death to the likelihood. This new
3456: contribution is smaller and very dependent of the step unit
3457: stepm. It is no more the probability to die between last interview
3458: and month of death but the probability to survive from last
3459: interview up to one month before death multiplied by the
3460: probability to die within a month. Thanks to Chris
3461: Jackson for correcting this bug. Former versions increased
3462: mortality artificially. The bad side is that we add another loop
3463: which slows down the processing. The difference can be up to 10%
3464: lower mortality.
3465: */
3466: /* If, at the beginning of the maximization mostly, the
3467: cumulative probability or probability to be dead is
3468: constant (ie = 1) over time d, the difference is equal to
3469: 0. out[s1][3] = savm[s1][3]: probability, being at state
3470: s1 at precedent wave, to be dead a month before current
3471: wave is equal to probability, being at state s1 at
3472: precedent wave, to be dead at mont of the current
3473: wave. Then the observed probability (that this person died)
3474: is null according to current estimated parameter. In fact,
3475: it should be very low but not zero otherwise the log go to
3476: infinity.
3477: */
1.183 brouard 3478: /* #ifdef INFINITYORIGINAL */
3479: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3480: /* #else */
3481: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3482: /* lli=log(mytinydouble); */
3483: /* else */
3484: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3485: /* #endif */
1.226 brouard 3486: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3487:
1.226 brouard 3488: } else if ( s2==-1 ) { /* alive */
3489: for (j=1,survp=0. ; j<=nlstate; j++)
3490: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3491: /*survp += out[s1][j]; */
3492: lli= log(survp);
3493: }
3494: else if (s2==-4) {
3495: for (j=3,survp=0. ; j<=nlstate; j++)
3496: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3497: lli= log(survp);
3498: }
3499: else if (s2==-5) {
3500: for (j=1,survp=0. ; j<=2; j++)
3501: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3502: lli= log(survp);
3503: }
3504: else{
3505: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3506: /* 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 */
3507: }
3508: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3509: /*if(lli ==000.0)*/
3510: /*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); */
3511: ipmx +=1;
3512: sw += weight[i];
3513: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3514: /* if (lli < log(mytinydouble)){ */
3515: /* 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); */
3516: /* 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]); */
3517: /* } */
3518: } /* end of wave */
3519: } /* end of individual */
3520: } else if(mle==2){
3521: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3522: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3523: for(mi=1; mi<= wav[i]-1; mi++){
3524: for (ii=1;ii<=nlstate+ndeath;ii++)
3525: for (j=1;j<=nlstate+ndeath;j++){
3526: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3527: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3528: }
3529: for(d=0; d<=dh[mi][i]; d++){
3530: newm=savm;
3531: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3532: cov[2]=agexact;
3533: if(nagesqr==1)
3534: cov[3]= agexact*agexact;
3535: for (kk=1; kk<=cptcovage;kk++) {
3536: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3537: }
3538: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3539: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3540: savm=oldm;
3541: oldm=newm;
3542: } /* end mult */
3543:
3544: s1=s[mw[mi][i]][i];
3545: s2=s[mw[mi+1][i]][i];
3546: bbh=(double)bh[mi][i]/(double)stepm;
3547: 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 */
3548: ipmx +=1;
3549: sw += weight[i];
3550: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3551: } /* end of wave */
3552: } /* end of individual */
3553: } else if(mle==3){ /* exponential inter-extrapolation */
3554: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3555: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3556: for(mi=1; mi<= wav[i]-1; mi++){
3557: for (ii=1;ii<=nlstate+ndeath;ii++)
3558: for (j=1;j<=nlstate+ndeath;j++){
3559: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3560: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3561: }
3562: for(d=0; d<dh[mi][i]; d++){
3563: newm=savm;
3564: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3565: cov[2]=agexact;
3566: if(nagesqr==1)
3567: cov[3]= agexact*agexact;
3568: for (kk=1; kk<=cptcovage;kk++) {
3569: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3570: }
3571: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3572: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3573: savm=oldm;
3574: oldm=newm;
3575: } /* end mult */
3576:
3577: s1=s[mw[mi][i]][i];
3578: s2=s[mw[mi+1][i]][i];
3579: bbh=(double)bh[mi][i]/(double)stepm;
3580: 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 */
3581: ipmx +=1;
3582: sw += weight[i];
3583: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3584: } /* end of wave */
3585: } /* end of individual */
3586: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3587: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3588: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3589: for(mi=1; mi<= wav[i]-1; mi++){
3590: for (ii=1;ii<=nlstate+ndeath;ii++)
3591: for (j=1;j<=nlstate+ndeath;j++){
3592: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3593: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3594: }
3595: for(d=0; d<dh[mi][i]; d++){
3596: newm=savm;
3597: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3598: cov[2]=agexact;
3599: if(nagesqr==1)
3600: cov[3]= agexact*agexact;
3601: for (kk=1; kk<=cptcovage;kk++) {
3602: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3603: }
1.126 brouard 3604:
1.226 brouard 3605: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3606: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3607: savm=oldm;
3608: oldm=newm;
3609: } /* end mult */
3610:
3611: s1=s[mw[mi][i]][i];
3612: s2=s[mw[mi+1][i]][i];
3613: if( s2 > nlstate){
3614: lli=log(out[s1][s2] - savm[s1][s2]);
3615: } else if ( s2==-1 ) { /* alive */
3616: for (j=1,survp=0. ; j<=nlstate; j++)
3617: survp += out[s1][j];
3618: lli= log(survp);
3619: }else{
3620: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3621: }
3622: ipmx +=1;
3623: sw += weight[i];
3624: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3625: /* 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 3626: } /* end of wave */
3627: } /* end of individual */
3628: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3629: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3630: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3631: for(mi=1; mi<= wav[i]-1; mi++){
3632: for (ii=1;ii<=nlstate+ndeath;ii++)
3633: for (j=1;j<=nlstate+ndeath;j++){
3634: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3635: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3636: }
3637: for(d=0; d<dh[mi][i]; d++){
3638: newm=savm;
3639: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3640: cov[2]=agexact;
3641: if(nagesqr==1)
3642: cov[3]= agexact*agexact;
3643: for (kk=1; kk<=cptcovage;kk++) {
3644: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3645: }
1.126 brouard 3646:
1.226 brouard 3647: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3648: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3649: savm=oldm;
3650: oldm=newm;
3651: } /* end mult */
3652:
3653: s1=s[mw[mi][i]][i];
3654: s2=s[mw[mi+1][i]][i];
3655: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3656: ipmx +=1;
3657: sw += weight[i];
3658: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3659: /*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]);*/
3660: } /* end of wave */
3661: } /* end of individual */
3662: } /* End of if */
3663: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3664: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3665: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3666: return -l;
1.126 brouard 3667: }
3668:
3669: /*************** log-likelihood *************/
3670: double funcone( double *x)
3671: {
1.228 brouard 3672: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3673: int i, ii, j, k, mi, d, kk;
1.228 brouard 3674: int ioffset=0;
1.131 brouard 3675: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3676: double **out;
3677: double lli; /* Individual log likelihood */
3678: double llt;
3679: int s1, s2;
1.228 brouard 3680: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3681:
1.126 brouard 3682: double bbh, survp;
1.187 brouard 3683: double agexact;
1.214 brouard 3684: double agebegin, ageend;
1.126 brouard 3685: /*extern weight */
3686: /* We are differentiating ll according to initial status */
3687: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3688: /*for(i=1;i<imx;i++)
3689: printf(" %d\n",s[4][i]);
3690: */
3691: cov[1]=1.;
3692:
3693: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3694: ioffset=0;
3695: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3696: /* ioffset=2+nagesqr+cptcovage; */
3697: ioffset=2+nagesqr;
1.232 brouard 3698: /* Fixed */
1.224 brouard 3699: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3700: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3701: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3702: 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)*/
3703: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3704: /* cov[2+6]=covar[Tvar[6]][i]; */
3705: /* cov[2+6]=covar[2][i]; V2 */
3706: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3707: /* cov[2+7]=covar[Tvar[7]][i]; */
3708: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3709: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3710: /* cov[2+9]=covar[Tvar[9]][i]; */
3711: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3712: }
1.232 brouard 3713: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3714: /* 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?)*\/ */
3715: /* } */
1.231 brouard 3716: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3717: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3718: /* } */
1.225 brouard 3719:
1.233 brouard 3720:
3721: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3722: /* Wave varying (but not age varying) */
3723: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3724: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3725: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3726: }
1.232 brouard 3727: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3728: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3729: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3730: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3731: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3732: /* 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 3733: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3734: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3735: /* /\* 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]); *\/ */
3736: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3737: /* } */
1.126 brouard 3738: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3739: for (j=1;j<=nlstate+ndeath;j++){
3740: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3741: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3742: }
1.214 brouard 3743:
3744: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3745: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3746: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3747: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3748: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3749: and mw[mi+1][i]. dh depends on stepm.*/
3750: newm=savm;
1.247 brouard 3751: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3752: cov[2]=agexact;
3753: if(nagesqr==1)
3754: cov[3]= agexact*agexact;
3755: for (kk=1; kk<=cptcovage;kk++) {
3756: if(!FixedV[Tvar[Tage[kk]]])
3757: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3758: else
3759: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3760: }
3761: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3762: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3763: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3764: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3765: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3766: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3767: savm=oldm;
3768: oldm=newm;
1.126 brouard 3769: } /* end mult */
3770:
3771: s1=s[mw[mi][i]][i];
3772: s2=s[mw[mi+1][i]][i];
1.217 brouard 3773: /* if(s2==-1){ */
1.268 brouard 3774: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3775: /* /\* exit(1); *\/ */
3776: /* } */
1.126 brouard 3777: bbh=(double)bh[mi][i]/(double)stepm;
3778: /* bias is positive if real duration
3779: * is higher than the multiple of stepm and negative otherwise.
3780: */
3781: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3782: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3783: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3784: for (j=1,survp=0. ; j<=nlstate; j++)
3785: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3786: lli= log(survp);
1.126 brouard 3787: }else if (mle==1){
1.242 brouard 3788: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3789: } else if(mle==2){
1.242 brouard 3790: 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 3791: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3792: 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 3793: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3794: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3795: } else{ /* mle=0 back to 1 */
1.242 brouard 3796: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3797: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3798: } /* End of if */
3799: ipmx +=1;
3800: sw += weight[i];
3801: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3802: /*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 3803: if(globpr){
1.246 brouard 3804: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3805: %11.6f %11.6f %11.6f ", \
1.242 brouard 3806: 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 3807: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3808: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3809: llt +=ll[k]*gipmx/gsw;
3810: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3811: }
3812: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3813: }
1.232 brouard 3814: } /* end of wave */
3815: } /* end of individual */
3816: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3817: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3818: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3819: if(globpr==0){ /* First time we count the contributions and weights */
3820: gipmx=ipmx;
3821: gsw=sw;
3822: }
3823: return -l;
1.126 brouard 3824: }
3825:
3826:
3827: /*************** function likelione ***********/
3828: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3829: {
3830: /* This routine should help understanding what is done with
3831: the selection of individuals/waves and
3832: to check the exact contribution to the likelihood.
3833: Plotting could be done.
3834: */
3835: int k;
3836:
3837: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3838: strcpy(fileresilk,"ILK_");
1.202 brouard 3839: strcat(fileresilk,fileresu);
1.126 brouard 3840: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3841: printf("Problem with resultfile: %s\n", fileresilk);
3842: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3843: }
1.214 brouard 3844: 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");
3845: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3846: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3847: for(k=1; k<=nlstate; k++)
3848: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3849: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3850: }
3851:
3852: *fretone=(*funcone)(p);
3853: if(*globpri !=0){
3854: fclose(ficresilk);
1.205 brouard 3855: if (mle ==0)
3856: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3857: else if(mle >=1)
3858: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3859: 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 3860: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3861:
3862: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3863: 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 3864: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3865: }
1.207 brouard 3866: 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 3867: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3868: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3869: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3870: fflush(fichtm);
1.205 brouard 3871: }
1.126 brouard 3872: return;
3873: }
3874:
3875:
3876: /*********** Maximum Likelihood Estimation ***************/
3877:
3878: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3879: {
1.165 brouard 3880: int i,j, iter=0;
1.126 brouard 3881: double **xi;
3882: double fret;
3883: double fretone; /* Only one call to likelihood */
3884: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3885:
3886: #ifdef NLOPT
3887: int creturn;
3888: nlopt_opt opt;
3889: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3890: double *lb;
3891: double minf; /* the minimum objective value, upon return */
3892: double * p1; /* Shifted parameters from 0 instead of 1 */
3893: myfunc_data dinst, *d = &dinst;
3894: #endif
3895:
3896:
1.126 brouard 3897: xi=matrix(1,npar,1,npar);
3898: for (i=1;i<=npar;i++)
3899: for (j=1;j<=npar;j++)
3900: xi[i][j]=(i==j ? 1.0 : 0.0);
3901: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3902: strcpy(filerespow,"POW_");
1.126 brouard 3903: strcat(filerespow,fileres);
3904: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3905: printf("Problem with resultfile: %s\n", filerespow);
3906: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3907: }
3908: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3909: for (i=1;i<=nlstate;i++)
3910: for(j=1;j<=nlstate+ndeath;j++)
3911: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3912: fprintf(ficrespow,"\n");
1.162 brouard 3913: #ifdef POWELL
1.126 brouard 3914: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3915: #endif
1.126 brouard 3916:
1.162 brouard 3917: #ifdef NLOPT
3918: #ifdef NEWUOA
3919: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3920: #else
3921: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3922: #endif
3923: lb=vector(0,npar-1);
3924: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3925: nlopt_set_lower_bounds(opt, lb);
3926: nlopt_set_initial_step1(opt, 0.1);
3927:
3928: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3929: d->function = func;
3930: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3931: nlopt_set_min_objective(opt, myfunc, d);
3932: nlopt_set_xtol_rel(opt, ftol);
3933: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3934: printf("nlopt failed! %d\n",creturn);
3935: }
3936: else {
3937: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3938: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3939: iter=1; /* not equal */
3940: }
3941: nlopt_destroy(opt);
3942: #endif
1.126 brouard 3943: free_matrix(xi,1,npar,1,npar);
3944: fclose(ficrespow);
1.203 brouard 3945: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3946: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3947: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3948:
3949: }
3950:
3951: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3952: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3953: {
3954: double **a,**y,*x,pd;
1.203 brouard 3955: /* double **hess; */
1.164 brouard 3956: int i, j;
1.126 brouard 3957: int *indx;
3958:
3959: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3960: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3961: void lubksb(double **a, int npar, int *indx, double b[]) ;
3962: void ludcmp(double **a, int npar, int *indx, double *d) ;
3963: double gompertz(double p[]);
1.203 brouard 3964: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3965:
3966: printf("\nCalculation of the hessian matrix. Wait...\n");
3967: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3968: for (i=1;i<=npar;i++){
1.203 brouard 3969: printf("%d-",i);fflush(stdout);
3970: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3971:
3972: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3973:
3974: /* printf(" %f ",p[i]);
3975: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3976: }
3977:
3978: for (i=1;i<=npar;i++) {
3979: for (j=1;j<=npar;j++) {
3980: if (j>i) {
1.203 brouard 3981: printf(".%d-%d",i,j);fflush(stdout);
3982: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3983: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3984:
3985: hess[j][i]=hess[i][j];
3986: /*printf(" %lf ",hess[i][j]);*/
3987: }
3988: }
3989: }
3990: printf("\n");
3991: fprintf(ficlog,"\n");
3992:
3993: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3994: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3995:
3996: a=matrix(1,npar,1,npar);
3997: y=matrix(1,npar,1,npar);
3998: x=vector(1,npar);
3999: indx=ivector(1,npar);
4000: for (i=1;i<=npar;i++)
4001: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4002: ludcmp(a,npar,indx,&pd);
4003:
4004: for (j=1;j<=npar;j++) {
4005: for (i=1;i<=npar;i++) x[i]=0;
4006: x[j]=1;
4007: lubksb(a,npar,indx,x);
4008: for (i=1;i<=npar;i++){
4009: matcov[i][j]=x[i];
4010: }
4011: }
4012:
4013: printf("\n#Hessian matrix#\n");
4014: fprintf(ficlog,"\n#Hessian matrix#\n");
4015: for (i=1;i<=npar;i++) {
4016: for (j=1;j<=npar;j++) {
1.203 brouard 4017: printf("%.6e ",hess[i][j]);
4018: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4019: }
4020: printf("\n");
4021: fprintf(ficlog,"\n");
4022: }
4023:
1.203 brouard 4024: /* printf("\n#Covariance matrix#\n"); */
4025: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4026: /* for (i=1;i<=npar;i++) { */
4027: /* for (j=1;j<=npar;j++) { */
4028: /* printf("%.6e ",matcov[i][j]); */
4029: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4030: /* } */
4031: /* printf("\n"); */
4032: /* fprintf(ficlog,"\n"); */
4033: /* } */
4034:
1.126 brouard 4035: /* Recompute Inverse */
1.203 brouard 4036: /* for (i=1;i<=npar;i++) */
4037: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4038: /* ludcmp(a,npar,indx,&pd); */
4039:
4040: /* printf("\n#Hessian matrix recomputed#\n"); */
4041:
4042: /* for (j=1;j<=npar;j++) { */
4043: /* for (i=1;i<=npar;i++) x[i]=0; */
4044: /* x[j]=1; */
4045: /* lubksb(a,npar,indx,x); */
4046: /* for (i=1;i<=npar;i++){ */
4047: /* y[i][j]=x[i]; */
4048: /* printf("%.3e ",y[i][j]); */
4049: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4050: /* } */
4051: /* printf("\n"); */
4052: /* fprintf(ficlog,"\n"); */
4053: /* } */
4054:
4055: /* Verifying the inverse matrix */
4056: #ifdef DEBUGHESS
4057: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4058:
1.203 brouard 4059: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4060: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4061:
4062: for (j=1;j<=npar;j++) {
4063: for (i=1;i<=npar;i++){
1.203 brouard 4064: printf("%.2f ",y[i][j]);
4065: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4066: }
4067: printf("\n");
4068: fprintf(ficlog,"\n");
4069: }
1.203 brouard 4070: #endif
1.126 brouard 4071:
4072: free_matrix(a,1,npar,1,npar);
4073: free_matrix(y,1,npar,1,npar);
4074: free_vector(x,1,npar);
4075: free_ivector(indx,1,npar);
1.203 brouard 4076: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4077:
4078:
4079: }
4080:
4081: /*************** hessian matrix ****************/
4082: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4083: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4084: int i;
4085: int l=1, lmax=20;
1.203 brouard 4086: double k1,k2, res, fx;
1.132 brouard 4087: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4088: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4089: int k=0,kmax=10;
4090: double l1;
4091:
4092: fx=func(x);
4093: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4094: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4095: l1=pow(10,l);
4096: delts=delt;
4097: for(k=1 ; k <kmax; k=k+1){
4098: delt = delta*(l1*k);
4099: p2[theta]=x[theta] +delt;
1.145 brouard 4100: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4101: p2[theta]=x[theta]-delt;
4102: k2=func(p2)-fx;
4103: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4104: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4105:
1.203 brouard 4106: #ifdef DEBUGHESSII
1.126 brouard 4107: 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);
4108: 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);
4109: #endif
4110: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4111: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4112: k=kmax;
4113: }
4114: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4115: k=kmax; l=lmax*10;
1.126 brouard 4116: }
4117: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4118: delts=delt;
4119: }
1.203 brouard 4120: } /* End loop k */
1.126 brouard 4121: }
4122: delti[theta]=delts;
4123: return res;
4124:
4125: }
4126:
1.203 brouard 4127: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4128: {
4129: int i;
1.164 brouard 4130: int l=1, lmax=20;
1.126 brouard 4131: double k1,k2,k3,k4,res,fx;
1.132 brouard 4132: double p2[MAXPARM+1];
1.203 brouard 4133: int k, kmax=1;
4134: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4135:
4136: int firstime=0;
1.203 brouard 4137:
1.126 brouard 4138: fx=func(x);
1.203 brouard 4139: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4140: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4141: p2[thetai]=x[thetai]+delti[thetai]*k;
4142: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4143: k1=func(p2)-fx;
4144:
1.203 brouard 4145: p2[thetai]=x[thetai]+delti[thetai]*k;
4146: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4147: k2=func(p2)-fx;
4148:
1.203 brouard 4149: p2[thetai]=x[thetai]-delti[thetai]*k;
4150: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4151: k3=func(p2)-fx;
4152:
1.203 brouard 4153: p2[thetai]=x[thetai]-delti[thetai]*k;
4154: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4155: k4=func(p2)-fx;
1.203 brouard 4156: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4157: if(k1*k2*k3*k4 <0.){
1.208 brouard 4158: firstime=1;
1.203 brouard 4159: kmax=kmax+10;
1.208 brouard 4160: }
4161: if(kmax >=10 || firstime ==1){
1.246 brouard 4162: 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);
4163: 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 4164: 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);
4165: 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);
4166: }
4167: #ifdef DEBUGHESSIJ
4168: v1=hess[thetai][thetai];
4169: v2=hess[thetaj][thetaj];
4170: cv12=res;
4171: /* Computing eigen value of Hessian matrix */
4172: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4173: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4174: if ((lc2 <0) || (lc1 <0) ){
4175: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4176: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4177: 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);
4178: 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);
4179: }
1.126 brouard 4180: #endif
4181: }
4182: return res;
4183: }
4184:
1.203 brouard 4185: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4186: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4187: /* { */
4188: /* int i; */
4189: /* int l=1, lmax=20; */
4190: /* double k1,k2,k3,k4,res,fx; */
4191: /* double p2[MAXPARM+1]; */
4192: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4193: /* int k=0,kmax=10; */
4194: /* double l1; */
4195:
4196: /* fx=func(x); */
4197: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4198: /* l1=pow(10,l); */
4199: /* delts=delt; */
4200: /* for(k=1 ; k <kmax; k=k+1){ */
4201: /* delt = delti*(l1*k); */
4202: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4203: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4204: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4205: /* k1=func(p2)-fx; */
4206:
4207: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4208: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4209: /* k2=func(p2)-fx; */
4210:
4211: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4212: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4213: /* k3=func(p2)-fx; */
4214:
4215: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4216: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4217: /* k4=func(p2)-fx; */
4218: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4219: /* #ifdef DEBUGHESSIJ */
4220: /* 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); */
4221: /* 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); */
4222: /* #endif */
4223: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4224: /* k=kmax; */
4225: /* } */
4226: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4227: /* k=kmax; l=lmax*10; */
4228: /* } */
4229: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4230: /* delts=delt; */
4231: /* } */
4232: /* } /\* End loop k *\/ */
4233: /* } */
4234: /* delti[theta]=delts; */
4235: /* return res; */
4236: /* } */
4237:
4238:
1.126 brouard 4239: /************** Inverse of matrix **************/
4240: void ludcmp(double **a, int n, int *indx, double *d)
4241: {
4242: int i,imax,j,k;
4243: double big,dum,sum,temp;
4244: double *vv;
4245:
4246: vv=vector(1,n);
4247: *d=1.0;
4248: for (i=1;i<=n;i++) {
4249: big=0.0;
4250: for (j=1;j<=n;j++)
4251: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4252: if (big == 0.0){
4253: printf(" Singular Hessian matrix at row %d:\n",i);
4254: for (j=1;j<=n;j++) {
4255: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4256: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4257: }
4258: fflush(ficlog);
4259: fclose(ficlog);
4260: nrerror("Singular matrix in routine ludcmp");
4261: }
1.126 brouard 4262: vv[i]=1.0/big;
4263: }
4264: for (j=1;j<=n;j++) {
4265: for (i=1;i<j;i++) {
4266: sum=a[i][j];
4267: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4268: a[i][j]=sum;
4269: }
4270: big=0.0;
4271: for (i=j;i<=n;i++) {
4272: sum=a[i][j];
4273: for (k=1;k<j;k++)
4274: sum -= a[i][k]*a[k][j];
4275: a[i][j]=sum;
4276: if ( (dum=vv[i]*fabs(sum)) >= big) {
4277: big=dum;
4278: imax=i;
4279: }
4280: }
4281: if (j != imax) {
4282: for (k=1;k<=n;k++) {
4283: dum=a[imax][k];
4284: a[imax][k]=a[j][k];
4285: a[j][k]=dum;
4286: }
4287: *d = -(*d);
4288: vv[imax]=vv[j];
4289: }
4290: indx[j]=imax;
4291: if (a[j][j] == 0.0) a[j][j]=TINY;
4292: if (j != n) {
4293: dum=1.0/(a[j][j]);
4294: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4295: }
4296: }
4297: free_vector(vv,1,n); /* Doesn't work */
4298: ;
4299: }
4300:
4301: void lubksb(double **a, int n, int *indx, double b[])
4302: {
4303: int i,ii=0,ip,j;
4304: double sum;
4305:
4306: for (i=1;i<=n;i++) {
4307: ip=indx[i];
4308: sum=b[ip];
4309: b[ip]=b[i];
4310: if (ii)
4311: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4312: else if (sum) ii=i;
4313: b[i]=sum;
4314: }
4315: for (i=n;i>=1;i--) {
4316: sum=b[i];
4317: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4318: b[i]=sum/a[i][i];
4319: }
4320: }
4321:
4322: void pstamp(FILE *fichier)
4323: {
1.196 brouard 4324: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4325: }
4326:
1.253 brouard 4327:
4328:
1.126 brouard 4329: /************ Frequencies ********************/
1.251 brouard 4330: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4331: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4332: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4333: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4334:
1.265 brouard 4335: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4336: int iind=0, iage=0;
4337: int mi; /* Effective wave */
4338: int first;
4339: double ***freq; /* Frequencies */
1.268 brouard 4340: 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 */
4341: 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 4342: double *meanq;
4343: double **meanqt;
4344: double *pp, **prop, *posprop, *pospropt;
4345: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4346: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4347: double agebegin, ageend;
4348:
4349: pp=vector(1,nlstate);
1.251 brouard 4350: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4351: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4352: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4353: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4354: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4355: meanqt=matrix(1,lastpass,1,nqtveff);
4356: strcpy(fileresp,"P_");
4357: strcat(fileresp,fileresu);
4358: /*strcat(fileresphtm,fileresu);*/
4359: if((ficresp=fopen(fileresp,"w"))==NULL) {
4360: printf("Problem with prevalence resultfile: %s\n", fileresp);
4361: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4362: exit(0);
4363: }
1.240 brouard 4364:
1.226 brouard 4365: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4366: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4367: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4368: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4369: fflush(ficlog);
4370: exit(70);
4371: }
4372: else{
4373: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4374: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4375: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4376: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4377: }
1.237 brouard 4378: 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 4379:
1.226 brouard 4380: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4381: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4382: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4383: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4384: fflush(ficlog);
4385: exit(70);
1.240 brouard 4386: } else{
1.226 brouard 4387: 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 4388: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4389: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4390: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4391: }
1.240 brouard 4392: 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);
4393:
1.253 brouard 4394: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4395: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4396: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4397: j1=0;
1.126 brouard 4398:
1.227 brouard 4399: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4400: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4401: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4402:
4403:
1.226 brouard 4404: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4405: reference=low_education V1=0,V2=0
4406: med_educ V1=1 V2=0,
4407: high_educ V1=0 V2=1
4408: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4409: */
1.249 brouard 4410: dateintsum=0;
4411: k2cpt=0;
4412:
1.253 brouard 4413: if(cptcoveff == 0 )
1.265 brouard 4414: nl=1; /* Constant and age model only */
1.253 brouard 4415: else
4416: nl=2;
1.265 brouard 4417:
4418: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4419: /* Loop on nj=1 or 2 if dummy covariates j!=0
4420: * Loop on j1(1 to 2**cptcoveff) covariate combination
4421: * freq[s1][s2][iage] =0.
4422: * Loop on iind
4423: * ++freq[s1][s2][iage] weighted
4424: * end iind
4425: * if covariate and j!0
4426: * headers Variable on one line
4427: * endif cov j!=0
4428: * header of frequency table by age
4429: * Loop on age
4430: * pp[s1]+=freq[s1][s2][iage] weighted
4431: * pos+=freq[s1][s2][iage] weighted
4432: * Loop on s1 initial state
4433: * fprintf(ficresp
4434: * end s1
4435: * end age
4436: * if j!=0 computes starting values
4437: * end compute starting values
4438: * end j1
4439: * end nl
4440: */
1.253 brouard 4441: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4442: if(nj==1)
4443: j=0; /* First pass for the constant */
1.265 brouard 4444: else{
1.253 brouard 4445: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4446: }
1.251 brouard 4447: first=1;
1.265 brouard 4448: 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 4449: posproptt=0.;
4450: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4451: scanf("%d", i);*/
4452: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4453: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4454: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4455: freq[i][s2][m]=0;
1.251 brouard 4456:
4457: for (i=1; i<=nlstate; i++) {
1.240 brouard 4458: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4459: prop[i][m]=0;
4460: posprop[i]=0;
4461: pospropt[i]=0;
4462: }
4463: /* for (z1=1; z1<= nqfveff; z1++) { */
4464: /* meanq[z1]+=0.; */
4465: /* for(m=1;m<=lastpass;m++){ */
4466: /* meanqt[m][z1]=0.; */
4467: /* } */
4468: /* } */
4469:
4470: /* dateintsum=0; */
4471: /* k2cpt=0; */
4472:
1.265 brouard 4473: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4474: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4475: bool=1;
4476: if(j !=0){
4477: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4478: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4479: /* for (z1=1; z1<= nqfveff; z1++) { */
4480: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4481: /* } */
4482: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4483: /* if(Tvaraff[z1] ==-20){ */
4484: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4485: /* }else if(Tvaraff[z1] ==-10){ */
4486: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4487: /* }else */
4488: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4489: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4490: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4491: /* 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",
4492: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4493: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4494: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4495: } /* Onlyf fixed */
4496: } /* end z1 */
4497: } /* cptcovn > 0 */
4498: } /* end any */
4499: }/* end j==0 */
1.265 brouard 4500: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4501: /* for(m=firstpass; m<=lastpass; m++){ */
4502: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4503: m=mw[mi][iind];
4504: if(j!=0){
4505: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4506: for (z1=1; z1<=cptcoveff; z1++) {
4507: if( Fixed[Tmodelind[z1]]==1){
4508: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4509: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4510: value is -1, we don't select. It differs from the
4511: constant and age model which counts them. */
4512: bool=0; /* not selected */
4513: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4514: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4515: bool=0;
4516: }
4517: }
4518: }
4519: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4520: } /* end j==0 */
4521: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4522: if(bool==1){
4523: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4524: and mw[mi+1][iind]. dh depends on stepm. */
4525: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4526: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4527: if(m >=firstpass && m <=lastpass){
4528: k2=anint[m][iind]+(mint[m][iind]/12.);
4529: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4530: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4531: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4532: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4533: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4534: if (m<lastpass) {
4535: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4536: /* 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]); */
4537: if(s[m][iind]==-1)
4538: 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.));
4539: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4540: /* if((int)agev[m][iind] == 55) */
4541: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4542: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4543: 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 4544: }
1.251 brouard 4545: } /* end if between passes */
4546: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4547: dateintsum=dateintsum+k2; /* on all covariates ?*/
4548: k2cpt++;
4549: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4550: }
1.251 brouard 4551: }else{
4552: bool=1;
4553: }/* end bool 2 */
4554: } /* end m */
4555: } /* end bool */
4556: } /* end iind = 1 to imx */
4557: /* prop[s][age] is feeded for any initial and valid live state as well as
4558: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4559:
4560:
4561: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4562: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4563: pstamp(ficresp);
1.251 brouard 4564: if (cptcoveff>0 && j!=0){
1.265 brouard 4565: pstamp(ficresp);
1.251 brouard 4566: printf( "\n#********** Variable ");
4567: fprintf(ficresp, "\n#********** Variable ");
4568: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4569: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4570: fprintf(ficlog, "\n#********** Variable ");
4571: for (z1=1; z1<=cptcoveff; z1++){
4572: if(!FixedV[Tvaraff[z1]]){
4573: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4574: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4575: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4576: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4577: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4578: }else{
1.251 brouard 4579: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4580: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4581: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4582: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4583: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4584: }
4585: }
4586: printf( "**********\n#");
4587: fprintf(ficresp, "**********\n#");
4588: fprintf(ficresphtm, "**********</h3>\n");
4589: fprintf(ficresphtmfr, "**********</h3>\n");
4590: fprintf(ficlog, "**********\n");
4591: }
4592: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4593: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4594: fprintf(ficresp, " Age");
4595: 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 4596: for(i=1; i<=nlstate;i++) {
1.265 brouard 4597: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4598: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4599: }
1.265 brouard 4600: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4601: fprintf(ficresphtm, "\n");
4602:
4603: /* Header of frequency table by age */
4604: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4605: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4606: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4607: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4608: if(s2!=0 && m!=0)
4609: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4610: }
1.226 brouard 4611: }
1.251 brouard 4612: fprintf(ficresphtmfr, "\n");
4613:
4614: /* For each age */
4615: for(iage=iagemin; iage <= iagemax+3; iage++){
4616: fprintf(ficresphtm,"<tr>");
4617: if(iage==iagemax+1){
4618: fprintf(ficlog,"1");
4619: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4620: }else if(iage==iagemax+2){
4621: fprintf(ficlog,"0");
4622: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4623: }else if(iage==iagemax+3){
4624: fprintf(ficlog,"Total");
4625: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4626: }else{
1.240 brouard 4627: if(first==1){
1.251 brouard 4628: first=0;
4629: printf("See log file for details...\n");
4630: }
4631: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4632: fprintf(ficlog,"Age %d", iage);
4633: }
1.265 brouard 4634: for(s1=1; s1 <=nlstate ; s1++){
4635: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4636: pp[s1] += freq[s1][m][iage];
1.251 brouard 4637: }
1.265 brouard 4638: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4639: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4640: pos += freq[s1][m][iage];
4641: if(pp[s1]>=1.e-10){
1.251 brouard 4642: if(first==1){
1.265 brouard 4643: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4644: }
1.265 brouard 4645: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4646: }else{
4647: if(first==1)
1.265 brouard 4648: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4649: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4650: }
4651: }
4652:
1.265 brouard 4653: for(s1=1; s1 <=nlstate ; s1++){
4654: /* posprop[s1]=0; */
4655: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4656: pp[s1] += freq[s1][m][iage];
4657: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4658:
4659: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4660: pos += pp[s1]; /* pos is the total number of transitions until this age */
4661: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4662: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4663: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4664: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4665: }
4666:
4667: /* Writing ficresp */
4668: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4669: if( iage <= iagemax){
4670: fprintf(ficresp," %d",iage);
4671: }
4672: }else if( nj==2){
4673: if( iage <= iagemax){
4674: fprintf(ficresp," %d",iage);
4675: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4676: }
1.240 brouard 4677: }
1.265 brouard 4678: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4679: if(pos>=1.e-5){
1.251 brouard 4680: if(first==1)
1.265 brouard 4681: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4682: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4683: }else{
4684: if(first==1)
1.265 brouard 4685: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4686: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4687: }
4688: if( iage <= iagemax){
4689: if(pos>=1.e-5){
1.265 brouard 4690: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4691: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4692: }else if( nj==2){
4693: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4694: }
4695: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4696: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4697: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4698: } else{
4699: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4700: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4701: }
1.240 brouard 4702: }
1.265 brouard 4703: pospropt[s1] +=posprop[s1];
4704: } /* end loop s1 */
1.251 brouard 4705: /* pospropt=0.; */
1.265 brouard 4706: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4707: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4708: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4709: if(first==1){
1.265 brouard 4710: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4711: }
1.265 brouard 4712: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4713: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4714: }
1.265 brouard 4715: if(s1!=0 && m!=0)
4716: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4717: }
1.265 brouard 4718: } /* end loop s1 */
1.251 brouard 4719: posproptt=0.;
1.265 brouard 4720: for(s1=1; s1 <=nlstate; s1++){
4721: posproptt += pospropt[s1];
1.251 brouard 4722: }
4723: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4724: fprintf(ficresphtm,"</tr>\n");
4725: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4726: if(iage <= iagemax)
4727: fprintf(ficresp,"\n");
1.240 brouard 4728: }
1.251 brouard 4729: if(first==1)
4730: printf("Others in log...\n");
4731: fprintf(ficlog,"\n");
4732: } /* end loop age iage */
1.265 brouard 4733:
1.251 brouard 4734: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4735: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4736: if(posproptt < 1.e-5){
1.265 brouard 4737: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4738: }else{
1.265 brouard 4739: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4740: }
1.226 brouard 4741: }
1.251 brouard 4742: fprintf(ficresphtm,"</tr>\n");
4743: fprintf(ficresphtm,"</table>\n");
4744: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4745: if(posproptt < 1.e-5){
1.251 brouard 4746: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4747: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4748: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4749: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4750: invalidvarcomb[j1]=1;
1.226 brouard 4751: }else{
1.251 brouard 4752: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4753: invalidvarcomb[j1]=0;
1.226 brouard 4754: }
1.251 brouard 4755: fprintf(ficresphtmfr,"</table>\n");
4756: fprintf(ficlog,"\n");
4757: if(j!=0){
4758: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4759: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4760: for(k=1; k <=(nlstate+ndeath); k++){
4761: if (k != i) {
1.265 brouard 4762: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4763: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4764: if(j1==1){ /* All dummy covariates to zero */
4765: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4766: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4767: printf("%d%d ",i,k);
4768: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4769: 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]));
4770: 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]));
4771: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4772: }
1.253 brouard 4773: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4774: for(iage=iagemin; iage <= iagemax+3; iage++){
4775: x[iage]= (double)iage;
4776: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4777: /* 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 4778: }
1.268 brouard 4779: /* Some are not finite, but linreg will ignore these ages */
4780: no=0;
1.253 brouard 4781: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4782: pstart[s1]=b;
4783: pstart[s1-1]=a;
1.252 brouard 4784: }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 */
4785: 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]);
4786: 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 4787: 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 4788: printf("%d%d ",i,k);
4789: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4790: 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 4791: }else{ /* Other cases, like quantitative fixed or varying covariates */
4792: ;
4793: }
4794: /* printf("%12.7f )", param[i][jj][k]); */
4795: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4796: s1++;
1.251 brouard 4797: } /* end jj */
4798: } /* end k!= i */
4799: } /* end k */
1.265 brouard 4800: } /* end i, s1 */
1.251 brouard 4801: } /* end j !=0 */
4802: } /* end selected combination of covariate j1 */
4803: if(j==0){ /* We can estimate starting values from the occurences in each case */
4804: printf("#Freqsummary: Starting values for the constants:\n");
4805: fprintf(ficlog,"\n");
1.265 brouard 4806: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4807: for(k=1; k <=(nlstate+ndeath); k++){
4808: if (k != i) {
4809: printf("%d%d ",i,k);
4810: fprintf(ficlog,"%d%d ",i,k);
4811: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4812: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4813: if(jj==1){ /* Age has to be done */
1.265 brouard 4814: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4815: 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]));
4816: 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 4817: }
4818: /* printf("%12.7f )", param[i][jj][k]); */
4819: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4820: s1++;
1.250 brouard 4821: }
1.251 brouard 4822: printf("\n");
4823: fprintf(ficlog,"\n");
1.250 brouard 4824: }
4825: }
4826: }
1.251 brouard 4827: printf("#Freqsummary\n");
4828: fprintf(ficlog,"\n");
1.265 brouard 4829: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4830: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4831: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4832: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4833: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4834: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4835: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4836: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4837: /* } */
4838: }
1.265 brouard 4839: } /* end loop s1 */
1.251 brouard 4840:
4841: printf("\n");
4842: fprintf(ficlog,"\n");
4843: } /* end j=0 */
1.249 brouard 4844: } /* end j */
1.252 brouard 4845:
1.253 brouard 4846: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4847: for(i=1, jk=1; i <=nlstate; i++){
4848: for(j=1; j <=nlstate+ndeath; j++){
4849: if(j!=i){
4850: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4851: printf("%1d%1d",i,j);
4852: fprintf(ficparo,"%1d%1d",i,j);
4853: for(k=1; k<=ncovmodel;k++){
4854: /* printf(" %lf",param[i][j][k]); */
4855: /* fprintf(ficparo," %lf",param[i][j][k]); */
4856: p[jk]=pstart[jk];
4857: printf(" %f ",pstart[jk]);
4858: fprintf(ficparo," %f ",pstart[jk]);
4859: jk++;
4860: }
4861: printf("\n");
4862: fprintf(ficparo,"\n");
4863: }
4864: }
4865: }
4866: } /* end mle=-2 */
1.226 brouard 4867: dateintmean=dateintsum/k2cpt;
1.240 brouard 4868:
1.226 brouard 4869: fclose(ficresp);
4870: fclose(ficresphtm);
4871: fclose(ficresphtmfr);
4872: free_vector(meanq,1,nqfveff);
4873: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4874: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4875: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4876: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4877: free_vector(pospropt,1,nlstate);
4878: free_vector(posprop,1,nlstate);
1.251 brouard 4879: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4880: free_vector(pp,1,nlstate);
4881: /* End of freqsummary */
4882: }
1.126 brouard 4883:
1.268 brouard 4884: /* Simple linear regression */
4885: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4886:
4887: /* y=a+bx regression */
4888: double sumx = 0.0; /* sum of x */
4889: double sumx2 = 0.0; /* sum of x**2 */
4890: double sumxy = 0.0; /* sum of x * y */
4891: double sumy = 0.0; /* sum of y */
4892: double sumy2 = 0.0; /* sum of y**2 */
4893: double sume2 = 0.0; /* sum of square or residuals */
4894: double yhat;
4895:
4896: double denom=0;
4897: int i;
4898: int ne=*no;
4899:
4900: for ( i=ifi, ne=0;i<=ila;i++) {
4901: if(!isfinite(x[i]) || !isfinite(y[i])){
4902: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4903: continue;
4904: }
4905: ne=ne+1;
4906: sumx += x[i];
4907: sumx2 += x[i]*x[i];
4908: sumxy += x[i] * y[i];
4909: sumy += y[i];
4910: sumy2 += y[i]*y[i];
4911: denom = (ne * sumx2 - sumx*sumx);
4912: /* 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); */
4913: }
4914:
4915: denom = (ne * sumx2 - sumx*sumx);
4916: if (denom == 0) {
4917: // vertical, slope m is infinity
4918: *b = INFINITY;
4919: *a = 0;
4920: if (r) *r = 0;
4921: return 1;
4922: }
4923:
4924: *b = (ne * sumxy - sumx * sumy) / denom;
4925: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4926: if (r!=NULL) {
4927: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4928: sqrt((sumx2 - sumx*sumx/ne) *
4929: (sumy2 - sumy*sumy/ne));
4930: }
4931: *no=ne;
4932: for ( i=ifi, ne=0;i<=ila;i++) {
4933: if(!isfinite(x[i]) || !isfinite(y[i])){
4934: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4935: continue;
4936: }
4937: ne=ne+1;
4938: yhat = y[i] - *a -*b* x[i];
4939: sume2 += yhat * yhat ;
4940:
4941: denom = (ne * sumx2 - sumx*sumx);
4942: /* 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); */
4943: }
4944: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
4945: *sa= *sb * sqrt(sumx2/ne);
4946:
4947: return 0;
4948: }
4949:
1.126 brouard 4950: /************ Prevalence ********************/
1.227 brouard 4951: 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)
4952: {
4953: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4954: in each health status at the date of interview (if between dateprev1 and dateprev2).
4955: We still use firstpass and lastpass as another selection.
4956: */
1.126 brouard 4957:
1.227 brouard 4958: int i, m, jk, j1, bool, z1,j, iv;
4959: int mi; /* Effective wave */
4960: int iage;
4961: double agebegin, ageend;
4962:
4963: double **prop;
4964: double posprop;
4965: double y2; /* in fractional years */
4966: int iagemin, iagemax;
4967: int first; /** to stop verbosity which is redirected to log file */
4968:
4969: iagemin= (int) agemin;
4970: iagemax= (int) agemax;
4971: /*pp=vector(1,nlstate);*/
1.251 brouard 4972: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4973: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4974: j1=0;
1.222 brouard 4975:
1.227 brouard 4976: /*j=cptcoveff;*/
4977: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4978:
1.227 brouard 4979: first=1;
4980: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4981: for (i=1; i<=nlstate; i++)
1.251 brouard 4982: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4983: prop[i][iage]=0.0;
4984: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4985: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4986: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4987:
4988: for (i=1; i<=imx; i++) { /* Each individual */
4989: bool=1;
4990: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4991: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4992: m=mw[mi][i];
4993: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4994: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4995: for (z1=1; z1<=cptcoveff; z1++){
4996: if( Fixed[Tmodelind[z1]]==1){
4997: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4998: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4999: bool=0;
5000: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5001: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5002: bool=0;
5003: }
5004: }
5005: if(bool==1){ /* Otherwise we skip that wave/person */
5006: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5007: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5008: if(m >=firstpass && m <=lastpass){
5009: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5010: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5011: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5012: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5013: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5014: 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);
5015: exit(1);
5016: }
5017: if (s[m][i]>0 && s[m][i]<=nlstate) {
5018: /*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]]);*/
5019: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5020: prop[s[m][i]][iagemax+3] += weight[i];
5021: } /* end valid statuses */
5022: } /* end selection of dates */
5023: } /* end selection of waves */
5024: } /* end bool */
5025: } /* end wave */
5026: } /* end individual */
5027: for(i=iagemin; i <= iagemax+3; i++){
5028: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5029: posprop += prop[jk][i];
5030: }
5031:
5032: for(jk=1; jk <=nlstate ; jk++){
5033: if( i <= iagemax){
5034: if(posprop>=1.e-5){
5035: probs[i][jk][j1]= prop[jk][i]/posprop;
5036: } else{
5037: if(first==1){
5038: first=0;
1.266 brouard 5039: 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]);
5040: 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]);
5041: }else{
5042: 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 5043: }
5044: }
5045: }
5046: }/* end jk */
5047: }/* end i */
1.222 brouard 5048: /*} *//* end i1 */
1.227 brouard 5049: } /* end j1 */
1.222 brouard 5050:
1.227 brouard 5051: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5052: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5053: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5054: } /* End of prevalence */
1.126 brouard 5055:
5056: /************* Waves Concatenation ***************/
5057:
5058: 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)
5059: {
5060: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5061: Death is a valid wave (if date is known).
5062: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5063: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5064: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5065: */
1.126 brouard 5066:
1.224 brouard 5067: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5068: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5069: double sum=0., jmean=0.;*/
1.224 brouard 5070: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5071: int j, k=0,jk, ju, jl;
5072: double sum=0.;
5073: first=0;
1.214 brouard 5074: firstwo=0;
1.217 brouard 5075: firsthree=0;
1.218 brouard 5076: firstfour=0;
1.164 brouard 5077: jmin=100000;
1.126 brouard 5078: jmax=-1;
5079: jmean=0.;
1.224 brouard 5080:
5081: /* Treating live states */
1.214 brouard 5082: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5083: mi=0; /* First valid wave */
1.227 brouard 5084: mli=0; /* Last valid wave */
1.126 brouard 5085: m=firstpass;
1.214 brouard 5086: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5087: 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 */
5088: mli=m-1;/* mw[++mi][i]=m-1; */
5089: }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 */
5090: mw[++mi][i]=m;
5091: mli=m;
1.224 brouard 5092: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5093: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5094: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5095: }
1.227 brouard 5096: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5097: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5098: break;
1.224 brouard 5099: #else
1.227 brouard 5100: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5101: if(firsthree == 0){
1.262 brouard 5102: 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 5103: firsthree=1;
5104: }
1.262 brouard 5105: 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 5106: mw[++mi][i]=m;
5107: mli=m;
5108: }
5109: if(s[m][i]==-2){ /* Vital status is really unknown */
5110: nbwarn++;
5111: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5112: 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);
5113: 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);
5114: }
5115: break;
5116: }
5117: break;
1.224 brouard 5118: #endif
1.227 brouard 5119: }/* End m >= lastpass */
1.126 brouard 5120: }/* end while */
1.224 brouard 5121:
1.227 brouard 5122: /* 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 5123: /* After last pass */
1.224 brouard 5124: /* Treating death states */
1.214 brouard 5125: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5126: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5127: /* } */
1.126 brouard 5128: mi++; /* Death is another wave */
5129: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5130: /* Only death is a correct wave */
1.126 brouard 5131: mw[mi][i]=m;
1.257 brouard 5132: } /* else not in a death state */
1.224 brouard 5133: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5134: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5135: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5136: 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 */
5137: nbwarn++;
5138: if(firstfiv==0){
5139: 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 );
5140: firstfiv=1;
5141: }else{
5142: 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 );
5143: }
5144: }else{ /* Death occured afer last wave potential bias */
5145: nberr++;
5146: if(firstwo==0){
1.257 brouard 5147: 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 5148: firstwo=1;
5149: }
1.257 brouard 5150: 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 5151: }
1.257 brouard 5152: }else{ /* if date of interview is unknown */
1.227 brouard 5153: /* death is known but not confirmed by death status at any wave */
5154: if(firstfour==0){
5155: 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 );
5156: firstfour=1;
5157: }
5158: 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 5159: }
1.224 brouard 5160: } /* end if date of death is known */
5161: #endif
5162: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5163: /* wav[i]=mw[mi][i]; */
1.126 brouard 5164: if(mi==0){
5165: nbwarn++;
5166: if(first==0){
1.227 brouard 5167: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5168: first=1;
1.126 brouard 5169: }
5170: if(first==1){
1.227 brouard 5171: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5172: }
5173: } /* end mi==0 */
5174: } /* End individuals */
1.214 brouard 5175: /* wav and mw are no more changed */
1.223 brouard 5176:
1.214 brouard 5177:
1.126 brouard 5178: for(i=1; i<=imx; i++){
5179: for(mi=1; mi<wav[i];mi++){
5180: if (stepm <=0)
1.227 brouard 5181: dh[mi][i]=1;
1.126 brouard 5182: else{
1.260 brouard 5183: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5184: if (agedc[i] < 2*AGESUP) {
5185: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5186: if(j==0) j=1; /* Survives at least one month after exam */
5187: else if(j<0){
5188: nberr++;
5189: 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]);
5190: j=1; /* Temporary Dangerous patch */
5191: 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);
5192: 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]);
5193: 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);
5194: }
5195: k=k+1;
5196: if (j >= jmax){
5197: jmax=j;
5198: ijmax=i;
5199: }
5200: if (j <= jmin){
5201: jmin=j;
5202: ijmin=i;
5203: }
5204: sum=sum+j;
5205: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5206: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5207: }
5208: }
5209: else{
5210: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5211: /* 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 5212:
1.227 brouard 5213: k=k+1;
5214: if (j >= jmax) {
5215: jmax=j;
5216: ijmax=i;
5217: }
5218: else if (j <= jmin){
5219: jmin=j;
5220: ijmin=i;
5221: }
5222: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5223: /*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]);*/
5224: if(j<0){
5225: nberr++;
5226: 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]);
5227: 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]);
5228: }
5229: sum=sum+j;
5230: }
5231: jk= j/stepm;
5232: jl= j -jk*stepm;
5233: ju= j -(jk+1)*stepm;
5234: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5235: if(jl==0){
5236: dh[mi][i]=jk;
5237: bh[mi][i]=0;
5238: }else{ /* We want a negative bias in order to only have interpolation ie
5239: * to avoid the price of an extra matrix product in likelihood */
5240: dh[mi][i]=jk+1;
5241: bh[mi][i]=ju;
5242: }
5243: }else{
5244: if(jl <= -ju){
5245: dh[mi][i]=jk;
5246: bh[mi][i]=jl; /* bias is positive if real duration
5247: * is higher than the multiple of stepm and negative otherwise.
5248: */
5249: }
5250: else{
5251: dh[mi][i]=jk+1;
5252: bh[mi][i]=ju;
5253: }
5254: if(dh[mi][i]==0){
5255: dh[mi][i]=1; /* At least one step */
5256: bh[mi][i]=ju; /* At least one step */
5257: /* 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);*/
5258: }
5259: } /* end if mle */
1.126 brouard 5260: }
5261: } /* end wave */
5262: }
5263: jmean=sum/k;
5264: 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 5265: 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 5266: }
1.126 brouard 5267:
5268: /*********** Tricode ****************************/
1.220 brouard 5269: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5270: {
5271: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5272: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5273: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5274: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5275: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5276: */
1.130 brouard 5277:
1.242 brouard 5278: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5279: int modmaxcovj=0; /* Modality max of covariates j */
5280: int cptcode=0; /* Modality max of covariates j */
5281: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5282:
5283:
1.242 brouard 5284: /* cptcoveff=0; */
5285: /* *cptcov=0; */
1.126 brouard 5286:
1.242 brouard 5287: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5288:
1.242 brouard 5289: /* Loop on covariates without age and products and no quantitative variable */
5290: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5291: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5292: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5293: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5294: switch(Fixed[k]) {
5295: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5296: 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*/
5297: ij=(int)(covar[Tvar[k]][i]);
5298: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5299: * If product of Vn*Vm, still boolean *:
5300: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5301: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5302: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5303: modality of the nth covariate of individual i. */
5304: if (ij > modmaxcovj)
5305: modmaxcovj=ij;
5306: else if (ij < modmincovj)
5307: modmincovj=ij;
5308: if ((ij < -1) && (ij > NCOVMAX)){
5309: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5310: exit(1);
5311: }else
5312: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5313: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5314: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5315: /* getting the maximum value of the modality of the covariate
5316: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5317: female ies 1, then modmaxcovj=1.
5318: */
5319: } /* end for loop on individuals i */
5320: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5321: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5322: cptcode=modmaxcovj;
5323: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5324: /*for (i=0; i<=cptcode; i++) {*/
5325: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5326: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5327: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5328: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5329: if( j != -1){
5330: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5331: covariate for which somebody answered excluding
5332: undefined. Usually 2: 0 and 1. */
5333: }
5334: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5335: covariate for which somebody answered including
5336: undefined. Usually 3: -1, 0 and 1. */
5337: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5338: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5339: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5340:
1.242 brouard 5341: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5342: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5343: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5344: /* modmincovj=3; modmaxcovj = 7; */
5345: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5346: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5347: /* defining two dummy variables: variables V1_1 and V1_2.*/
5348: /* nbcode[Tvar[j]][ij]=k; */
5349: /* nbcode[Tvar[j]][1]=0; */
5350: /* nbcode[Tvar[j]][2]=1; */
5351: /* nbcode[Tvar[j]][3]=2; */
5352: /* To be continued (not working yet). */
5353: ij=0; /* ij is similar to i but can jump over null modalities */
5354: 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*/
5355: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5356: break;
5357: }
5358: ij++;
5359: 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*/
5360: cptcode = ij; /* New max modality for covar j */
5361: } /* end of loop on modality i=-1 to 1 or more */
5362: break;
5363: case 1: /* Testing on varying covariate, could be simple and
5364: * should look at waves or product of fixed *
5365: * varying. No time to test -1, assuming 0 and 1 only */
5366: ij=0;
5367: for(i=0; i<=1;i++){
5368: nbcode[Tvar[k]][++ij]=i;
5369: }
5370: break;
5371: default:
5372: break;
5373: } /* end switch */
5374: } /* end dummy test */
5375:
5376: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5377: /* /\*recode from 0 *\/ */
5378: /* k is a modality. If we have model=V1+V1*sex */
5379: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5380: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5381: /* } */
5382: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5383: /* if (ij > ncodemax[j]) { */
5384: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5385: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5386: /* break; */
5387: /* } */
5388: /* } /\* end of loop on modality k *\/ */
5389: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5390:
5391: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5392: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5393: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5394: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5395: 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 */
5396: 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 */
5397: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5398: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5399:
5400: ij=0;
5401: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5402: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5403: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5404: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5405: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5406: /* If product not in single variable we don't print results */
5407: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5408: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5409: 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*/
5410: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5411: 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 */
5412: if(Fixed[k]!=0)
5413: anyvaryingduminmodel=1;
5414: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5415: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5416: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5417: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5418: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5419: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5420: }
5421: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5422: /* ij--; */
5423: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5424: *cptcov=ij; /*Number of total real effective covariates: effective
5425: * because they can be excluded from the model and real
5426: * if in the model but excluded because missing values, but how to get k from ij?*/
5427: for(j=ij+1; j<= cptcovt; j++){
5428: Tvaraff[j]=0;
5429: Tmodelind[j]=0;
5430: }
5431: for(j=ntveff+1; j<= cptcovt; j++){
5432: TmodelInvind[j]=0;
5433: }
5434: /* To be sorted */
5435: ;
5436: }
1.126 brouard 5437:
1.145 brouard 5438:
1.126 brouard 5439: /*********** Health Expectancies ****************/
5440:
1.235 brouard 5441: 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 5442:
5443: {
5444: /* Health expectancies, no variances */
1.164 brouard 5445: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5446: int nhstepma, nstepma; /* Decreasing with age */
5447: double age, agelim, hf;
5448: double ***p3mat;
5449: double eip;
5450:
1.238 brouard 5451: /* pstamp(ficreseij); */
1.126 brouard 5452: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5453: fprintf(ficreseij,"# Age");
5454: for(i=1; i<=nlstate;i++){
5455: for(j=1; j<=nlstate;j++){
5456: fprintf(ficreseij," e%1d%1d ",i,j);
5457: }
5458: fprintf(ficreseij," e%1d. ",i);
5459: }
5460: fprintf(ficreseij,"\n");
5461:
5462:
5463: if(estepm < stepm){
5464: printf ("Problem %d lower than %d\n",estepm, stepm);
5465: }
5466: else hstepm=estepm;
5467: /* We compute the life expectancy from trapezoids spaced every estepm months
5468: * This is mainly to measure the difference between two models: for example
5469: * if stepm=24 months pijx are given only every 2 years and by summing them
5470: * we are calculating an estimate of the Life Expectancy assuming a linear
5471: * progression in between and thus overestimating or underestimating according
5472: * to the curvature of the survival function. If, for the same date, we
5473: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5474: * to compare the new estimate of Life expectancy with the same linear
5475: * hypothesis. A more precise result, taking into account a more precise
5476: * curvature will be obtained if estepm is as small as stepm. */
5477:
5478: /* For example we decided to compute the life expectancy with the smallest unit */
5479: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5480: nhstepm is the number of hstepm from age to agelim
5481: nstepm is the number of stepm from age to agelin.
1.270 brouard 5482: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5483: and note for a fixed period like estepm months */
5484: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5485: survival function given by stepm (the optimization length). Unfortunately it
5486: means that if the survival funtion is printed only each two years of age and if
5487: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5488: results. So we changed our mind and took the option of the best precision.
5489: */
5490: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5491:
5492: agelim=AGESUP;
5493: /* If stepm=6 months */
5494: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5495: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5496:
5497: /* nhstepm age range expressed in number of stepm */
5498: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5499: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5500: /* if (stepm >= YEARM) hstepm=1;*/
5501: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5502: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5503:
5504: for (age=bage; age<=fage; age ++){
5505: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5506: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5507: /* if (stepm >= YEARM) hstepm=1;*/
5508: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5509:
5510: /* If stepm=6 months */
5511: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5512: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5513:
1.235 brouard 5514: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5515:
5516: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5517:
5518: printf("%d|",(int)age);fflush(stdout);
5519: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5520:
5521: /* Computing expectancies */
5522: for(i=1; i<=nlstate;i++)
5523: for(j=1; j<=nlstate;j++)
5524: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5525: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5526:
5527: /* 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]);*/
5528:
5529: }
5530:
5531: fprintf(ficreseij,"%3.0f",age );
5532: for(i=1; i<=nlstate;i++){
5533: eip=0;
5534: for(j=1; j<=nlstate;j++){
5535: eip +=eij[i][j][(int)age];
5536: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5537: }
5538: fprintf(ficreseij,"%9.4f", eip );
5539: }
5540: fprintf(ficreseij,"\n");
5541:
5542: }
5543: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5544: printf("\n");
5545: fprintf(ficlog,"\n");
5546:
5547: }
5548:
1.235 brouard 5549: 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 5550:
5551: {
5552: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5553: to initial status i, ei. .
1.126 brouard 5554: */
5555: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5556: int nhstepma, nstepma; /* Decreasing with age */
5557: double age, agelim, hf;
5558: double ***p3matp, ***p3matm, ***varhe;
5559: double **dnewm,**doldm;
5560: double *xp, *xm;
5561: double **gp, **gm;
5562: double ***gradg, ***trgradg;
5563: int theta;
5564:
5565: double eip, vip;
5566:
5567: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5568: xp=vector(1,npar);
5569: xm=vector(1,npar);
5570: dnewm=matrix(1,nlstate*nlstate,1,npar);
5571: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5572:
5573: pstamp(ficresstdeij);
5574: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5575: fprintf(ficresstdeij,"# Age");
5576: for(i=1; i<=nlstate;i++){
5577: for(j=1; j<=nlstate;j++)
5578: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5579: fprintf(ficresstdeij," e%1d. ",i);
5580: }
5581: fprintf(ficresstdeij,"\n");
5582:
5583: pstamp(ficrescveij);
5584: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5585: fprintf(ficrescveij,"# Age");
5586: for(i=1; i<=nlstate;i++)
5587: for(j=1; j<=nlstate;j++){
5588: cptj= (j-1)*nlstate+i;
5589: for(i2=1; i2<=nlstate;i2++)
5590: for(j2=1; j2<=nlstate;j2++){
5591: cptj2= (j2-1)*nlstate+i2;
5592: if(cptj2 <= cptj)
5593: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5594: }
5595: }
5596: fprintf(ficrescveij,"\n");
5597:
5598: if(estepm < stepm){
5599: printf ("Problem %d lower than %d\n",estepm, stepm);
5600: }
5601: else hstepm=estepm;
5602: /* We compute the life expectancy from trapezoids spaced every estepm months
5603: * This is mainly to measure the difference between two models: for example
5604: * if stepm=24 months pijx are given only every 2 years and by summing them
5605: * we are calculating an estimate of the Life Expectancy assuming a linear
5606: * progression in between and thus overestimating or underestimating according
5607: * to the curvature of the survival function. If, for the same date, we
5608: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5609: * to compare the new estimate of Life expectancy with the same linear
5610: * hypothesis. A more precise result, taking into account a more precise
5611: * curvature will be obtained if estepm is as small as stepm. */
5612:
5613: /* For example we decided to compute the life expectancy with the smallest unit */
5614: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5615: nhstepm is the number of hstepm from age to agelim
5616: nstepm is the number of stepm from age to agelin.
5617: Look at hpijx to understand the reason of that which relies in memory size
5618: and note for a fixed period like estepm months */
5619: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5620: survival function given by stepm (the optimization length). Unfortunately it
5621: means that if the survival funtion is printed only each two years of age and if
5622: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5623: results. So we changed our mind and took the option of the best precision.
5624: */
5625: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5626:
5627: /* If stepm=6 months */
5628: /* nhstepm age range expressed in number of stepm */
5629: agelim=AGESUP;
5630: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5631: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5632: /* if (stepm >= YEARM) hstepm=1;*/
5633: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5634:
5635: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5636: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5637: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5638: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5639: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5640: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5641:
5642: for (age=bage; age<=fage; age ++){
5643: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5644: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5645: /* if (stepm >= YEARM) hstepm=1;*/
5646: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5647:
1.126 brouard 5648: /* If stepm=6 months */
5649: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5650: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5651:
5652: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5653:
1.126 brouard 5654: /* Computing Variances of health expectancies */
5655: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5656: decrease memory allocation */
5657: for(theta=1; theta <=npar; theta++){
5658: for(i=1; i<=npar; i++){
1.222 brouard 5659: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5660: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5661: }
1.235 brouard 5662: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5663: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5664:
1.126 brouard 5665: for(j=1; j<= nlstate; j++){
1.222 brouard 5666: for(i=1; i<=nlstate; i++){
5667: for(h=0; h<=nhstepm-1; h++){
5668: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5669: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5670: }
5671: }
1.126 brouard 5672: }
1.218 brouard 5673:
1.126 brouard 5674: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5675: for(h=0; h<=nhstepm-1; h++){
5676: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5677: }
1.126 brouard 5678: }/* End theta */
5679:
5680:
5681: for(h=0; h<=nhstepm-1; h++)
5682: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5683: for(theta=1; theta <=npar; theta++)
5684: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5685:
1.218 brouard 5686:
1.222 brouard 5687: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5688: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5689: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5690:
1.222 brouard 5691: printf("%d|",(int)age);fflush(stdout);
5692: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5693: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5694: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5695: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5696: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5697: for(ij=1;ij<=nlstate*nlstate;ij++)
5698: for(ji=1;ji<=nlstate*nlstate;ji++)
5699: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5700: }
5701: }
1.218 brouard 5702:
1.126 brouard 5703: /* Computing expectancies */
1.235 brouard 5704: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5705: for(i=1; i<=nlstate;i++)
5706: for(j=1; j<=nlstate;j++)
1.222 brouard 5707: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5708: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5709:
1.222 brouard 5710: /* 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 5711:
1.222 brouard 5712: }
1.269 brouard 5713:
5714: /* Standard deviation of expectancies ij */
1.126 brouard 5715: fprintf(ficresstdeij,"%3.0f",age );
5716: for(i=1; i<=nlstate;i++){
5717: eip=0.;
5718: vip=0.;
5719: for(j=1; j<=nlstate;j++){
1.222 brouard 5720: eip += eij[i][j][(int)age];
5721: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5722: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5723: 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 5724: }
5725: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5726: }
5727: fprintf(ficresstdeij,"\n");
1.218 brouard 5728:
1.269 brouard 5729: /* Variance of expectancies ij */
1.126 brouard 5730: fprintf(ficrescveij,"%3.0f",age );
5731: for(i=1; i<=nlstate;i++)
5732: for(j=1; j<=nlstate;j++){
1.222 brouard 5733: cptj= (j-1)*nlstate+i;
5734: for(i2=1; i2<=nlstate;i2++)
5735: for(j2=1; j2<=nlstate;j2++){
5736: cptj2= (j2-1)*nlstate+i2;
5737: if(cptj2 <= cptj)
5738: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5739: }
1.126 brouard 5740: }
5741: fprintf(ficrescveij,"\n");
1.218 brouard 5742:
1.126 brouard 5743: }
5744: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5745: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5746: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5747: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5748: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5749: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5750: printf("\n");
5751: fprintf(ficlog,"\n");
1.218 brouard 5752:
1.126 brouard 5753: free_vector(xm,1,npar);
5754: free_vector(xp,1,npar);
5755: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5756: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5757: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5758: }
1.218 brouard 5759:
1.126 brouard 5760: /************ Variance ******************/
1.235 brouard 5761: 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 5762: {
5763: /* Variance of health expectancies */
5764: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5765: /* double **newm;*/
5766: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5767:
5768: /* int movingaverage(); */
5769: double **dnewm,**doldm;
5770: double **dnewmp,**doldmp;
5771: int i, j, nhstepm, hstepm, h, nstepm ;
5772: int k;
5773: double *xp;
5774: double **gp, **gm; /* for var eij */
5775: double ***gradg, ***trgradg; /*for var eij */
5776: double **gradgp, **trgradgp; /* for var p point j */
5777: double *gpp, *gmp; /* for var p point j */
5778: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5779: double ***p3mat;
5780: double age,agelim, hf;
5781: /* double ***mobaverage; */
5782: int theta;
5783: char digit[4];
5784: char digitp[25];
5785:
5786: char fileresprobmorprev[FILENAMELENGTH];
5787:
5788: if(popbased==1){
5789: if(mobilav!=0)
5790: strcpy(digitp,"-POPULBASED-MOBILAV_");
5791: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5792: }
5793: else
5794: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5795:
1.218 brouard 5796: /* if (mobilav!=0) { */
5797: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5798: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5799: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5800: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5801: /* } */
5802: /* } */
5803:
5804: strcpy(fileresprobmorprev,"PRMORPREV-");
5805: sprintf(digit,"%-d",ij);
5806: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5807: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5808: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5809: strcat(fileresprobmorprev,fileresu);
5810: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5811: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5812: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5813: }
5814: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5815: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5816: pstamp(ficresprobmorprev);
5817: 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 5818: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5819: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5820: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5821: }
5822: for(j=1;j<=cptcoveff;j++)
5823: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5824: fprintf(ficresprobmorprev,"\n");
5825:
1.218 brouard 5826: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5827: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5828: fprintf(ficresprobmorprev," p.%-d SE",j);
5829: for(i=1; i<=nlstate;i++)
5830: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5831: }
5832: fprintf(ficresprobmorprev,"\n");
5833:
5834: fprintf(ficgp,"\n# Routine varevsij");
5835: fprintf(ficgp,"\nunset title \n");
5836: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5837: 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");
5838: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5839: /* } */
5840: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5841: pstamp(ficresvij);
5842: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5843: if(popbased==1)
5844: 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);
5845: else
5846: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5847: fprintf(ficresvij,"# Age");
5848: for(i=1; i<=nlstate;i++)
5849: for(j=1; j<=nlstate;j++)
5850: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5851: fprintf(ficresvij,"\n");
5852:
5853: xp=vector(1,npar);
5854: dnewm=matrix(1,nlstate,1,npar);
5855: doldm=matrix(1,nlstate,1,nlstate);
5856: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5857: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5858:
5859: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5860: gpp=vector(nlstate+1,nlstate+ndeath);
5861: gmp=vector(nlstate+1,nlstate+ndeath);
5862: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5863:
1.218 brouard 5864: if(estepm < stepm){
5865: printf ("Problem %d lower than %d\n",estepm, stepm);
5866: }
5867: else hstepm=estepm;
5868: /* For example we decided to compute the life expectancy with the smallest unit */
5869: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5870: nhstepm is the number of hstepm from age to agelim
5871: nstepm is the number of stepm from age to agelim.
5872: Look at function hpijx to understand why because of memory size limitations,
5873: we decided (b) to get a life expectancy respecting the most precise curvature of the
5874: survival function given by stepm (the optimization length). Unfortunately it
5875: means that if the survival funtion is printed every two years of age and if
5876: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5877: results. So we changed our mind and took the option of the best precision.
5878: */
5879: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5880: agelim = AGESUP;
5881: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5882: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5883: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5884: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5885: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5886: gp=matrix(0,nhstepm,1,nlstate);
5887: gm=matrix(0,nhstepm,1,nlstate);
5888:
5889:
5890: for(theta=1; theta <=npar; theta++){
5891: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5892: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5893: }
5894:
1.242 brouard 5895: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5896:
5897: if (popbased==1) {
5898: if(mobilav ==0){
5899: for(i=1; i<=nlstate;i++)
5900: prlim[i][i]=probs[(int)age][i][ij];
5901: }else{ /* mobilav */
5902: for(i=1; i<=nlstate;i++)
5903: prlim[i][i]=mobaverage[(int)age][i][ij];
5904: }
5905: }
5906:
1.235 brouard 5907: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=1 to nhstepm */
1.218 brouard 5908: for(j=1; j<= nlstate; j++){
5909: for(h=0; h<=nhstepm; h++){
5910: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5911: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5912: }
5913: }
5914: /* Next for computing probability of death (h=1 means
5915: computed over hstepm matrices product = hstepm*stepm months)
5916: as a weighted average of prlim.
5917: */
5918: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5919: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5920: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5921: }
5922: /* end probability of death */
5923:
5924: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5925: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5926:
1.242 brouard 5927: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5928:
5929: if (popbased==1) {
5930: if(mobilav ==0){
5931: for(i=1; i<=nlstate;i++)
5932: prlim[i][i]=probs[(int)age][i][ij];
5933: }else{ /* mobilav */
5934: for(i=1; i<=nlstate;i++)
5935: prlim[i][i]=mobaverage[(int)age][i][ij];
5936: }
5937: }
5938:
1.235 brouard 5939: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5940:
5941: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5942: for(h=0; h<=nhstepm; h++){
5943: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5944: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5945: }
5946: }
5947: /* This for computing probability of death (h=1 means
5948: computed over hstepm matrices product = hstepm*stepm months)
5949: as a weighted average of prlim.
5950: */
5951: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5952: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5953: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5954: }
5955: /* end probability of death */
5956:
5957: for(j=1; j<= nlstate; j++) /* vareij */
5958: for(h=0; h<=nhstepm; h++){
5959: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5960: }
5961:
5962: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5963: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5964: }
5965:
5966: } /* End theta */
5967:
5968: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5969:
5970: for(h=0; h<=nhstepm; h++) /* veij */
5971: for(j=1; j<=nlstate;j++)
5972: for(theta=1; theta <=npar; theta++)
5973: trgradg[h][j][theta]=gradg[h][theta][j];
5974:
5975: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5976: for(theta=1; theta <=npar; theta++)
5977: trgradgp[j][theta]=gradgp[theta][j];
5978:
5979:
5980: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5981: for(i=1;i<=nlstate;i++)
5982: for(j=1;j<=nlstate;j++)
5983: vareij[i][j][(int)age] =0.;
5984:
5985: for(h=0;h<=nhstepm;h++){
5986: for(k=0;k<=nhstepm;k++){
5987: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5988: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5989: for(i=1;i<=nlstate;i++)
5990: for(j=1;j<=nlstate;j++)
5991: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5992: }
5993: }
5994:
5995: /* pptj */
5996: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5997: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5998: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5999: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6000: varppt[j][i]=doldmp[j][i];
6001: /* end ppptj */
6002: /* x centered again */
6003:
1.242 brouard 6004: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6005:
6006: if (popbased==1) {
6007: if(mobilav ==0){
6008: for(i=1; i<=nlstate;i++)
6009: prlim[i][i]=probs[(int)age][i][ij];
6010: }else{ /* mobilav */
6011: for(i=1; i<=nlstate;i++)
6012: prlim[i][i]=mobaverage[(int)age][i][ij];
6013: }
6014: }
6015:
6016: /* This for computing probability of death (h=1 means
6017: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6018: as a weighted average of prlim.
6019: */
1.235 brouard 6020: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6021: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6022: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6023: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6024: }
6025: /* end probability of death */
6026:
6027: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6028: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6029: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6030: for(i=1; i<=nlstate;i++){
6031: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6032: }
6033: }
6034: fprintf(ficresprobmorprev,"\n");
6035:
6036: fprintf(ficresvij,"%.0f ",age );
6037: for(i=1; i<=nlstate;i++)
6038: for(j=1; j<=nlstate;j++){
6039: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6040: }
6041: fprintf(ficresvij,"\n");
6042: free_matrix(gp,0,nhstepm,1,nlstate);
6043: free_matrix(gm,0,nhstepm,1,nlstate);
6044: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6045: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6046: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6047: } /* End age */
6048: free_vector(gpp,nlstate+1,nlstate+ndeath);
6049: free_vector(gmp,nlstate+1,nlstate+ndeath);
6050: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6051: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6052: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6053: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6054: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6055: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6056: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6057: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6058: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6059: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6060: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6061: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6062: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6063: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6064: 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);
6065: /* 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 6066: */
1.218 brouard 6067: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6068: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6069:
1.218 brouard 6070: free_vector(xp,1,npar);
6071: free_matrix(doldm,1,nlstate,1,nlstate);
6072: free_matrix(dnewm,1,nlstate,1,npar);
6073: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6074: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6075: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6076: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6077: fclose(ficresprobmorprev);
6078: fflush(ficgp);
6079: fflush(fichtm);
6080: } /* end varevsij */
1.126 brouard 6081:
6082: /************ Variance of prevlim ******************/
1.269 brouard 6083: 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 6084: {
1.205 brouard 6085: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6086: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6087:
1.268 brouard 6088: double **dnewmpar,**doldm;
1.126 brouard 6089: int i, j, nhstepm, hstepm;
6090: double *xp;
6091: double *gp, *gm;
6092: double **gradg, **trgradg;
1.208 brouard 6093: double **mgm, **mgp;
1.126 brouard 6094: double age,agelim;
6095: int theta;
6096:
6097: pstamp(ficresvpl);
6098: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6099: fprintf(ficresvpl,"# Age ");
6100: if(nresult >=1)
6101: fprintf(ficresvpl," Result# ");
1.126 brouard 6102: for(i=1; i<=nlstate;i++)
6103: fprintf(ficresvpl," %1d-%1d",i,i);
6104: fprintf(ficresvpl,"\n");
6105:
6106: xp=vector(1,npar);
1.268 brouard 6107: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6108: doldm=matrix(1,nlstate,1,nlstate);
6109:
6110: hstepm=1*YEARM; /* Every year of age */
6111: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6112: agelim = AGESUP;
6113: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6114: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6115: if (stepm >= YEARM) hstepm=1;
6116: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6117: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6118: mgp=matrix(1,npar,1,nlstate);
6119: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6120: gp=vector(1,nlstate);
6121: gm=vector(1,nlstate);
6122:
6123: for(theta=1; theta <=npar; theta++){
6124: for(i=1; i<=npar; i++){ /* Computes gradient */
6125: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6126: }
1.209 brouard 6127: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6128: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6129: else
1.235 brouard 6130: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6131: for(i=1;i<=nlstate;i++){
1.126 brouard 6132: gp[i] = prlim[i][i];
1.208 brouard 6133: mgp[theta][i] = prlim[i][i];
6134: }
1.126 brouard 6135: for(i=1; i<=npar; i++) /* Computes gradient */
6136: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6137: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6138: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6139: else
1.235 brouard 6140: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6141: for(i=1;i<=nlstate;i++){
1.126 brouard 6142: gm[i] = prlim[i][i];
1.208 brouard 6143: mgm[theta][i] = prlim[i][i];
6144: }
1.126 brouard 6145: for(i=1;i<=nlstate;i++)
6146: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6147: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6148: } /* End theta */
6149:
6150: trgradg =matrix(1,nlstate,1,npar);
6151:
6152: for(j=1; j<=nlstate;j++)
6153: for(theta=1; theta <=npar; theta++)
6154: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6155: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6156: /* printf("\nmgm mgp %d ",(int)age); */
6157: /* for(j=1; j<=nlstate;j++){ */
6158: /* printf(" %d ",j); */
6159: /* for(theta=1; theta <=npar; theta++) */
6160: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6161: /* printf("\n "); */
6162: /* } */
6163: /* } */
6164: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6165: /* printf("\n gradg %d ",(int)age); */
6166: /* for(j=1; j<=nlstate;j++){ */
6167: /* printf("%d ",j); */
6168: /* for(theta=1; theta <=npar; theta++) */
6169: /* printf("%d %lf ",theta,gradg[theta][j]); */
6170: /* printf("\n "); */
6171: /* } */
6172: /* } */
1.126 brouard 6173:
6174: for(i=1;i<=nlstate;i++)
6175: varpl[i][(int)age] =0.;
1.209 brouard 6176: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6177: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6178: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6179: }else{
1.268 brouard 6180: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6181: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6182: }
1.126 brouard 6183: for(i=1;i<=nlstate;i++)
6184: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6185:
6186: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6187: if(nresult >=1)
6188: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6189: for(i=1; i<=nlstate;i++)
6190: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6191: fprintf(ficresvpl,"\n");
6192: free_vector(gp,1,nlstate);
6193: free_vector(gm,1,nlstate);
1.208 brouard 6194: free_matrix(mgm,1,npar,1,nlstate);
6195: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6196: free_matrix(gradg,1,npar,1,nlstate);
6197: free_matrix(trgradg,1,nlstate,1,npar);
6198: } /* End age */
6199:
6200: free_vector(xp,1,npar);
6201: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6202: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6203:
6204: }
6205:
6206:
6207: /************ Variance of backprevalence limit ******************/
1.269 brouard 6208: 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 6209: {
6210: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6211: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6212:
6213: double **dnewmpar,**doldm;
6214: int i, j, nhstepm, hstepm;
6215: double *xp;
6216: double *gp, *gm;
6217: double **gradg, **trgradg;
6218: double **mgm, **mgp;
6219: double age,agelim;
6220: int theta;
6221:
6222: pstamp(ficresvbl);
6223: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6224: fprintf(ficresvbl,"# Age ");
6225: if(nresult >=1)
6226: fprintf(ficresvbl," Result# ");
6227: for(i=1; i<=nlstate;i++)
6228: fprintf(ficresvbl," %1d-%1d",i,i);
6229: fprintf(ficresvbl,"\n");
6230:
6231: xp=vector(1,npar);
6232: dnewmpar=matrix(1,nlstate,1,npar);
6233: doldm=matrix(1,nlstate,1,nlstate);
6234:
6235: hstepm=1*YEARM; /* Every year of age */
6236: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6237: agelim = AGEINF;
6238: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6239: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6240: if (stepm >= YEARM) hstepm=1;
6241: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6242: gradg=matrix(1,npar,1,nlstate);
6243: mgp=matrix(1,npar,1,nlstate);
6244: mgm=matrix(1,npar,1,nlstate);
6245: gp=vector(1,nlstate);
6246: gm=vector(1,nlstate);
6247:
6248: for(theta=1; theta <=npar; theta++){
6249: for(i=1; i<=npar; i++){ /* Computes gradient */
6250: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6251: }
6252: if(mobilavproj > 0 )
6253: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6254: else
6255: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6256: for(i=1;i<=nlstate;i++){
6257: gp[i] = bprlim[i][i];
6258: mgp[theta][i] = bprlim[i][i];
6259: }
6260: for(i=1; i<=npar; i++) /* Computes gradient */
6261: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6262: if(mobilavproj > 0 )
6263: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6264: else
6265: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6266: for(i=1;i<=nlstate;i++){
6267: gm[i] = bprlim[i][i];
6268: mgm[theta][i] = bprlim[i][i];
6269: }
6270: for(i=1;i<=nlstate;i++)
6271: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6272: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6273: } /* End theta */
6274:
6275: trgradg =matrix(1,nlstate,1,npar);
6276:
6277: for(j=1; j<=nlstate;j++)
6278: for(theta=1; theta <=npar; theta++)
6279: trgradg[j][theta]=gradg[theta][j];
6280: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6281: /* printf("\nmgm mgp %d ",(int)age); */
6282: /* for(j=1; j<=nlstate;j++){ */
6283: /* printf(" %d ",j); */
6284: /* for(theta=1; theta <=npar; theta++) */
6285: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6286: /* printf("\n "); */
6287: /* } */
6288: /* } */
6289: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6290: /* printf("\n gradg %d ",(int)age); */
6291: /* for(j=1; j<=nlstate;j++){ */
6292: /* printf("%d ",j); */
6293: /* for(theta=1; theta <=npar; theta++) */
6294: /* printf("%d %lf ",theta,gradg[theta][j]); */
6295: /* printf("\n "); */
6296: /* } */
6297: /* } */
6298:
6299: for(i=1;i<=nlstate;i++)
6300: varbpl[i][(int)age] =0.;
6301: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6302: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6303: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6304: }else{
6305: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6306: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6307: }
6308: for(i=1;i<=nlstate;i++)
6309: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6310:
6311: fprintf(ficresvbl,"%.0f ",age );
6312: if(nresult >=1)
6313: fprintf(ficresvbl,"%d ",nres );
6314: for(i=1; i<=nlstate;i++)
6315: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6316: fprintf(ficresvbl,"\n");
6317: free_vector(gp,1,nlstate);
6318: free_vector(gm,1,nlstate);
6319: free_matrix(mgm,1,npar,1,nlstate);
6320: free_matrix(mgp,1,npar,1,nlstate);
6321: free_matrix(gradg,1,npar,1,nlstate);
6322: free_matrix(trgradg,1,nlstate,1,npar);
6323: } /* End age */
6324:
6325: free_vector(xp,1,npar);
6326: free_matrix(doldm,1,nlstate,1,npar);
6327: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6328:
6329: }
6330:
6331: /************ Variance of one-step probabilities ******************/
6332: 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 6333: {
6334: int i, j=0, k1, l1, tj;
6335: int k2, l2, j1, z1;
6336: int k=0, l;
6337: int first=1, first1, first2;
6338: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6339: double **dnewm,**doldm;
6340: double *xp;
6341: double *gp, *gm;
6342: double **gradg, **trgradg;
6343: double **mu;
6344: double age, cov[NCOVMAX+1];
6345: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6346: int theta;
6347: char fileresprob[FILENAMELENGTH];
6348: char fileresprobcov[FILENAMELENGTH];
6349: char fileresprobcor[FILENAMELENGTH];
6350: double ***varpij;
6351:
6352: strcpy(fileresprob,"PROB_");
6353: strcat(fileresprob,fileres);
6354: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6355: printf("Problem with resultfile: %s\n", fileresprob);
6356: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6357: }
6358: strcpy(fileresprobcov,"PROBCOV_");
6359: strcat(fileresprobcov,fileresu);
6360: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6361: printf("Problem with resultfile: %s\n", fileresprobcov);
6362: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6363: }
6364: strcpy(fileresprobcor,"PROBCOR_");
6365: strcat(fileresprobcor,fileresu);
6366: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6367: printf("Problem with resultfile: %s\n", fileresprobcor);
6368: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6369: }
6370: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6371: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6372: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6373: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6374: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6375: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6376: pstamp(ficresprob);
6377: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6378: fprintf(ficresprob,"# Age");
6379: pstamp(ficresprobcov);
6380: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6381: fprintf(ficresprobcov,"# Age");
6382: pstamp(ficresprobcor);
6383: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6384: fprintf(ficresprobcor,"# Age");
1.126 brouard 6385:
6386:
1.222 brouard 6387: for(i=1; i<=nlstate;i++)
6388: for(j=1; j<=(nlstate+ndeath);j++){
6389: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6390: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6391: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6392: }
6393: /* fprintf(ficresprob,"\n");
6394: fprintf(ficresprobcov,"\n");
6395: fprintf(ficresprobcor,"\n");
6396: */
6397: xp=vector(1,npar);
6398: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6399: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6400: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6401: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6402: first=1;
6403: fprintf(ficgp,"\n# Routine varprob");
6404: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6405: fprintf(fichtm,"\n");
6406:
1.266 brouard 6407: 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 6408: 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);
6409: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6410: and drawn. It helps understanding how is the covariance between two incidences.\
6411: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6412: 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 6413: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6414: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6415: standard deviations wide on each axis. <br>\
6416: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6417: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6418: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6419:
1.222 brouard 6420: cov[1]=1;
6421: /* tj=cptcoveff; */
1.225 brouard 6422: tj = (int) pow(2,cptcoveff);
1.222 brouard 6423: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6424: j1=0;
1.224 brouard 6425: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6426: if (cptcovn>0) {
6427: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6428: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6429: fprintf(ficresprob, "**********\n#\n");
6430: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6431: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6432: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6433:
1.222 brouard 6434: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6435: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6436: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6437:
6438:
1.222 brouard 6439: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6440: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6441: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6442:
1.222 brouard 6443: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6444: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6445: fprintf(ficresprobcor, "**********\n#");
6446: if(invalidvarcomb[j1]){
6447: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6448: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6449: continue;
6450: }
6451: }
6452: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6453: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6454: gp=vector(1,(nlstate)*(nlstate+ndeath));
6455: gm=vector(1,(nlstate)*(nlstate+ndeath));
6456: for (age=bage; age<=fage; age ++){
6457: cov[2]=age;
6458: if(nagesqr==1)
6459: cov[3]= age*age;
6460: for (k=1; k<=cptcovn;k++) {
6461: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6462: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6463: * 1 1 1 1 1
6464: * 2 2 1 1 1
6465: * 3 1 2 1 1
6466: */
6467: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6468: }
6469: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6470: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6471: for (k=1; k<=cptcovprod;k++)
6472: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6473:
6474:
1.222 brouard 6475: for(theta=1; theta <=npar; theta++){
6476: for(i=1; i<=npar; i++)
6477: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6478:
1.222 brouard 6479: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6480:
1.222 brouard 6481: k=0;
6482: for(i=1; i<= (nlstate); i++){
6483: for(j=1; j<=(nlstate+ndeath);j++){
6484: k=k+1;
6485: gp[k]=pmmij[i][j];
6486: }
6487: }
1.220 brouard 6488:
1.222 brouard 6489: for(i=1; i<=npar; i++)
6490: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6491:
1.222 brouard 6492: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6493: k=0;
6494: for(i=1; i<=(nlstate); i++){
6495: for(j=1; j<=(nlstate+ndeath);j++){
6496: k=k+1;
6497: gm[k]=pmmij[i][j];
6498: }
6499: }
1.220 brouard 6500:
1.222 brouard 6501: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6502: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6503: }
1.126 brouard 6504:
1.222 brouard 6505: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6506: for(theta=1; theta <=npar; theta++)
6507: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6508:
1.222 brouard 6509: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6510: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6511:
1.222 brouard 6512: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6513:
1.222 brouard 6514: k=0;
6515: for(i=1; i<=(nlstate); i++){
6516: for(j=1; j<=(nlstate+ndeath);j++){
6517: k=k+1;
6518: mu[k][(int) age]=pmmij[i][j];
6519: }
6520: }
6521: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6522: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6523: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6524:
1.222 brouard 6525: /*printf("\n%d ",(int)age);
6526: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6527: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6528: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6529: }*/
1.220 brouard 6530:
1.222 brouard 6531: fprintf(ficresprob,"\n%d ",(int)age);
6532: fprintf(ficresprobcov,"\n%d ",(int)age);
6533: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6534:
1.222 brouard 6535: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6536: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6537: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6538: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6539: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6540: }
6541: i=0;
6542: for (k=1; k<=(nlstate);k++){
6543: for (l=1; l<=(nlstate+ndeath);l++){
6544: i++;
6545: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6546: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6547: for (j=1; j<=i;j++){
6548: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6549: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6550: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6551: }
6552: }
6553: }/* end of loop for state */
6554: } /* end of loop for age */
6555: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6556: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6557: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6558: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6559:
6560: /* Confidence intervalle of pij */
6561: /*
6562: fprintf(ficgp,"\nunset parametric;unset label");
6563: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6564: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6565: 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);
6566: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6567: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6568: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6569: */
6570:
6571: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6572: first1=1;first2=2;
6573: for (k2=1; k2<=(nlstate);k2++){
6574: for (l2=1; l2<=(nlstate+ndeath);l2++){
6575: if(l2==k2) continue;
6576: j=(k2-1)*(nlstate+ndeath)+l2;
6577: for (k1=1; k1<=(nlstate);k1++){
6578: for (l1=1; l1<=(nlstate+ndeath);l1++){
6579: if(l1==k1) continue;
6580: i=(k1-1)*(nlstate+ndeath)+l1;
6581: if(i<=j) continue;
6582: for (age=bage; age<=fage; age ++){
6583: if ((int)age %5==0){
6584: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6585: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6586: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6587: mu1=mu[i][(int) age]/stepm*YEARM ;
6588: mu2=mu[j][(int) age]/stepm*YEARM;
6589: c12=cv12/sqrt(v1*v2);
6590: /* Computing eigen value of matrix of covariance */
6591: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6592: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6593: if ((lc2 <0) || (lc1 <0) ){
6594: if(first2==1){
6595: first1=0;
6596: 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);
6597: }
6598: 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);
6599: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6600: /* lc2=fabs(lc2); */
6601: }
1.220 brouard 6602:
1.222 brouard 6603: /* Eigen vectors */
6604: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6605: /*v21=sqrt(1.-v11*v11); *//* error */
6606: v21=(lc1-v1)/cv12*v11;
6607: v12=-v21;
6608: v22=v11;
6609: tnalp=v21/v11;
6610: if(first1==1){
6611: first1=0;
6612: 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);
6613: }
6614: 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);
6615: /*printf(fignu*/
6616: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6617: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6618: if(first==1){
6619: first=0;
6620: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6621: fprintf(ficgp,"\nset parametric;unset label");
6622: 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);
6623: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6624: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6625: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6626: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6627: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6628: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6629: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6630: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6631: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6632: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6633: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6634: fprintf(ficgp,"\nplot [-pi:pi] %11.3e+ %.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)), %11.3e +%.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)) not", \
1.266 brouard 6635: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6636: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6637: }else{
6638: first=0;
6639: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6640: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6641: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6642: 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 6643: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6644: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6645: }/* if first */
6646: } /* age mod 5 */
6647: } /* end loop age */
6648: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6649: first=1;
6650: } /*l12 */
6651: } /* k12 */
6652: } /*l1 */
6653: }/* k1 */
6654: } /* loop on combination of covariates j1 */
6655: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6656: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6657: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6658: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6659: free_vector(xp,1,npar);
6660: fclose(ficresprob);
6661: fclose(ficresprobcov);
6662: fclose(ficresprobcor);
6663: fflush(ficgp);
6664: fflush(fichtmcov);
6665: }
1.126 brouard 6666:
6667:
6668: /******************* Printing html file ***********/
1.201 brouard 6669: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6670: int lastpass, int stepm, int weightopt, char model[],\
6671: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6672: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6673: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6674: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6675: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6676:
6677: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6678: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6679: </ul>");
1.237 brouard 6680: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6681: </ul>", model);
1.214 brouard 6682: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6683: 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",
6684: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6685: 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 6686: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6687: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6688: fprintf(fichtm,"\
6689: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6690: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6691: fprintf(fichtm,"\
1.217 brouard 6692: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6693: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6694: fprintf(fichtm,"\
1.126 brouard 6695: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6696: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6697: fprintf(fichtm,"\
1.217 brouard 6698: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6699: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6700: fprintf(fichtm,"\
1.211 brouard 6701: - (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 6702: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6703: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6704: if(prevfcast==1){
6705: fprintf(fichtm,"\
6706: - Prevalence projections by age and states: \
1.201 brouard 6707: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6708: }
1.126 brouard 6709:
6710:
1.225 brouard 6711: m=pow(2,cptcoveff);
1.222 brouard 6712: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6713:
1.264 brouard 6714: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6715:
6716: jj1=0;
6717:
6718: fprintf(fichtm," \n<ul>");
6719: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6720: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6721: if(m != 1 && TKresult[nres]!= k1)
6722: continue;
6723: jj1++;
6724: if (cptcovn > 0) {
6725: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6726: for (cpt=1; cpt<=cptcoveff;cpt++){
6727: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6728: }
6729: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6730: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6731: }
6732: fprintf(fichtm,"\">");
6733:
6734: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6735: fprintf(fichtm,"************ Results for covariates");
6736: for (cpt=1; cpt<=cptcoveff;cpt++){
6737: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6738: }
6739: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6740: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6741: }
6742: if(invalidvarcomb[k1]){
6743: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6744: continue;
6745: }
6746: fprintf(fichtm,"</a></li>");
6747: } /* cptcovn >0 */
6748: }
6749: fprintf(fichtm," \n</ul>");
6750:
1.222 brouard 6751: jj1=0;
1.237 brouard 6752:
6753: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6754: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6755: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6756: continue;
1.220 brouard 6757:
1.222 brouard 6758: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6759: jj1++;
6760: if (cptcovn > 0) {
1.264 brouard 6761: fprintf(fichtm,"\n<p><a name=\"rescov");
6762: for (cpt=1; cpt<=cptcoveff;cpt++){
6763: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6764: }
6765: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6766: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6767: }
6768: fprintf(fichtm,"\"</a>");
6769:
1.222 brouard 6770: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6771: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6772: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6773: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6774: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6775: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6776: }
1.237 brouard 6777: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6778: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6779: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6780: }
6781:
1.230 brouard 6782: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6783: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6784: if(invalidvarcomb[k1]){
6785: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6786: printf("\nCombination (%d) ignored because no cases \n",k1);
6787: continue;
6788: }
6789: }
6790: /* aij, bij */
1.259 brouard 6791: 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 6792: <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 6793: /* Pij */
1.241 brouard 6794: 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> \
6795: <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 6796: /* Quasi-incidences */
6797: 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 6798: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6799: 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 6800: 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> \
6801: <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 6802: /* Survival functions (period) in state j */
6803: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6804: 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> \
6805: <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 6806: }
6807: /* State specific survival functions (period) */
6808: for(cpt=1; cpt<=nlstate;cpt++){
6809: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6810: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6811: <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 6812: }
6813: /* Period (stable) prevalence in each health state */
6814: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6815: 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> \
6816: <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 6817: }
6818: if(backcast==1){
6819: /* Period (stable) back prevalence in each health state */
6820: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6821: 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 6822: <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 6823: }
1.217 brouard 6824: }
1.222 brouard 6825: if(prevfcast==1){
6826: /* Projection of prevalence up to period (stable) prevalence in each health state */
6827: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6828: 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> \
6829: <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 6830: }
6831: }
1.268 brouard 6832: if(backcast==1){
6833: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6834: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6835: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6836: 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 \
6837: 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) \
6838: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6839: <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 6840: }
6841: }
1.220 brouard 6842:
1.222 brouard 6843: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6844: 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> \
6845: <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 6846: }
6847: /* } /\* end i1 *\/ */
6848: }/* End k1 */
6849: fprintf(fichtm,"</ul>");
1.126 brouard 6850:
1.222 brouard 6851: fprintf(fichtm,"\
1.126 brouard 6852: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6853: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6854: - 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 6855: But because parameters are usually highly correlated (a higher incidence of disability \
6856: and a higher incidence of recovery can give very close observed transition) it might \
6857: be very useful to look not only at linear confidence intervals estimated from the \
6858: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6859: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6860: covariance matrix of the one-step probabilities. \
6861: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6862:
1.222 brouard 6863: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6864: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6865: fprintf(fichtm,"\
1.126 brouard 6866: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6867: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6868:
1.222 brouard 6869: fprintf(fichtm,"\
1.126 brouard 6870: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6871: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6872: fprintf(fichtm,"\
1.126 brouard 6873: - 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): \
6874: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6875: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6876: fprintf(fichtm,"\
1.126 brouard 6877: - (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): \
6878: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6879: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6880: fprintf(fichtm,"\
1.128 brouard 6881: - 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 6882: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6883: fprintf(fichtm,"\
1.128 brouard 6884: - 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 6885: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6886: fprintf(fichtm,"\
1.126 brouard 6887: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6888: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6889:
6890: /* if(popforecast==1) fprintf(fichtm,"\n */
6891: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6892: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6893: /* <br>",fileres,fileres,fileres,fileres); */
6894: /* else */
6895: /* 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 6896: fflush(fichtm);
6897: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6898:
1.225 brouard 6899: m=pow(2,cptcoveff);
1.222 brouard 6900: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6901:
1.222 brouard 6902: jj1=0;
1.237 brouard 6903:
1.241 brouard 6904: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6905: for(k1=1; k1<=m;k1++){
1.253 brouard 6906: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6907: continue;
1.222 brouard 6908: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6909: jj1++;
1.126 brouard 6910: if (cptcovn > 0) {
6911: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6912: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6913: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6914: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6915: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6916: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6917: }
6918:
1.126 brouard 6919: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6920:
1.222 brouard 6921: if(invalidvarcomb[k1]){
6922: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6923: continue;
6924: }
1.126 brouard 6925: }
6926: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6927: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6928: 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 6929: <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 6930: }
6931: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6932: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6933: true period expectancies (those weighted with period prevalences are also\
6934: drawn in addition to the population based expectancies computed using\
1.241 brouard 6935: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6936: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6937: /* } /\* end i1 *\/ */
6938: }/* End k1 */
1.241 brouard 6939: }/* End nres */
1.222 brouard 6940: fprintf(fichtm,"</ul>");
6941: fflush(fichtm);
1.126 brouard 6942: }
6943:
6944: /******************* Gnuplot file **************/
1.270 brouard 6945: 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 6946:
6947: char dirfileres[132],optfileres[132];
1.264 brouard 6948: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6949: 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 6950: int lv=0, vlv=0, kl=0;
1.130 brouard 6951: int ng=0;
1.201 brouard 6952: int vpopbased;
1.223 brouard 6953: int ioffset; /* variable offset for columns */
1.270 brouard 6954: int iyearc=1; /* variable column for year of projection */
6955: int iagec=1; /* variable column for age of projection */
1.235 brouard 6956: int nres=0; /* Index of resultline */
1.266 brouard 6957: int istart=1; /* For starting graphs in projections */
1.219 brouard 6958:
1.126 brouard 6959: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6960: /* printf("Problem with file %s",optionfilegnuplot); */
6961: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6962: /* } */
6963:
6964: /*#ifdef windows */
6965: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6966: /*#endif */
1.225 brouard 6967: m=pow(2,cptcoveff);
1.126 brouard 6968:
1.274 brouard 6969: /* diagram of the model */
6970: fprintf(ficgp,"\n#Diagram of the model \n");
6971: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
6972: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
6973: 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);
6974:
6975: 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);
6976: fprintf(ficgp,"\n#show arrow\nunset label\n");
6977: 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);
6978: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
6979: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
6980: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
6981: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
6982:
1.202 brouard 6983: /* Contribution to likelihood */
6984: /* Plot the probability implied in the likelihood */
1.223 brouard 6985: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6986: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6987: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6988: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6989: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6990: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6991: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6992: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6993: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6994: 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));
6995: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6996: 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));
6997: for (i=1; i<= nlstate ; i ++) {
6998: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6999: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7000: 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);
7001: for (j=2; j<= nlstate+ndeath ; j ++) {
7002: 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);
7003: }
7004: fprintf(ficgp,";\nset out; unset ylabel;\n");
7005: }
7006: /* 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 */
7007: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7008: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7009: fprintf(ficgp,"\nset out;unset log\n");
7010: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7011:
1.126 brouard 7012: strcpy(dirfileres,optionfilefiname);
7013: strcpy(optfileres,"vpl");
1.223 brouard 7014: /* 1eme*/
1.238 brouard 7015: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7016: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7017: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7018: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7019: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7020: continue;
7021: /* We are interested in selected combination by the resultline */
1.246 brouard 7022: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7023: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7024: strcpy(gplotlabel,"(");
1.238 brouard 7025: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7026: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7027: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7028: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7029: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7030: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7031: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7032: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7033: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7034: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7035: }
7036: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7037: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7038: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7039: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7040: }
7041: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7042: /* printf("\n#\n"); */
1.238 brouard 7043: fprintf(ficgp,"\n#\n");
7044: if(invalidvarcomb[k1]){
1.260 brouard 7045: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7046: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7047: continue;
7048: }
1.235 brouard 7049:
1.241 brouard 7050: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7051: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.264 brouard 7052: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7053: 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);
7054: /* 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); */
7055: /* k1-1 error should be nres-1*/
1.238 brouard 7056: for (i=1; i<= nlstate ; i ++) {
7057: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7058: else fprintf(ficgp," %%*lf (%%*lf)");
7059: }
1.260 brouard 7060: 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 7061: for (i=1; i<= nlstate ; i ++) {
7062: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7063: else fprintf(ficgp," %%*lf (%%*lf)");
7064: }
1.260 brouard 7065: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7066: for (i=1; i<= nlstate ; i ++) {
7067: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7068: else fprintf(ficgp," %%*lf (%%*lf)");
7069: }
1.265 brouard 7070: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7071:
7072: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7073: if(cptcoveff ==0){
1.271 brouard 7074: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7075: }else{
7076: kl=0;
7077: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7078: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7079: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7080: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7081: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7082: vlv= nbcode[Tvaraff[k]][lv];
7083: kl++;
7084: /* 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 *\/ */
7085: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7086: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7087: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7088: if(k==cptcoveff){
7089: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7090: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7091: }else{
7092: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7093: kl++;
7094: }
7095: } /* end covariate */
7096: } /* end if no covariate */
7097:
1.238 brouard 7098: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7099: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7100: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7101: if(cptcoveff ==0){
1.245 brouard 7102: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7103: }else{
7104: kl=0;
7105: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7106: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7107: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7108: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7109: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7110: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7111: kl++;
1.238 brouard 7112: /* 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 *\/ */
7113: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7114: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7115: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7116: if(k==cptcoveff){
1.245 brouard 7117: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7118: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7119: }else{
7120: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7121: kl++;
7122: }
7123: } /* end covariate */
7124: } /* end if no covariate */
1.268 brouard 7125: if(backcast == 1){
7126: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7127: /* k1-1 error should be nres-1*/
7128: for (i=1; i<= nlstate ; i ++) {
7129: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7130: else fprintf(ficgp," %%*lf (%%*lf)");
7131: }
1.271 brouard 7132: 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 7133: for (i=1; i<= nlstate ; i ++) {
7134: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7135: else fprintf(ficgp," %%*lf (%%*lf)");
7136: }
1.272 brouard 7137: fprintf(ficgp,"\" t\"95%% CI\" w l lt 5,\"%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 7138: for (i=1; i<= nlstate ; i ++) {
7139: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7140: else fprintf(ficgp," %%*lf (%%*lf)");
7141: }
1.274 brouard 7142: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7143: } /* end if backprojcast */
1.238 brouard 7144: } /* end if backcast */
1.264 brouard 7145: fprintf(ficgp,"\nset out ;unset label;\n");
1.238 brouard 7146: } /* nres */
1.201 brouard 7147: } /* k1 */
7148: } /* cpt */
1.235 brouard 7149:
7150:
1.126 brouard 7151: /*2 eme*/
1.238 brouard 7152: for (k1=1; k1<= m ; k1 ++){
7153: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7154: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7155: continue;
7156: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7157: strcpy(gplotlabel,"(");
1.238 brouard 7158: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7159: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7160: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7161: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7162: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7163: vlv= nbcode[Tvaraff[k]][lv];
7164: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7165: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7166: }
1.237 brouard 7167: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7168: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7169: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7170: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7171: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7172: }
1.264 brouard 7173: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7174: fprintf(ficgp,"\n#\n");
1.223 brouard 7175: if(invalidvarcomb[k1]){
7176: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7177: continue;
7178: }
1.219 brouard 7179:
1.241 brouard 7180: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7181: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7182: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7183: if(vpopbased==0){
1.238 brouard 7184: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7185: }else
1.238 brouard 7186: fprintf(ficgp,"\nreplot ");
7187: for (i=1; i<= nlstate+1 ; i ++) {
7188: k=2*i;
1.261 brouard 7189: 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 7190: for (j=1; j<= nlstate+1 ; j ++) {
7191: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7192: else fprintf(ficgp," %%*lf (%%*lf)");
7193: }
7194: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7195: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7196: 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 7197: for (j=1; j<= nlstate+1 ; j ++) {
7198: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7199: else fprintf(ficgp," %%*lf (%%*lf)");
7200: }
7201: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7202: 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 7203: for (j=1; j<= nlstate+1 ; j ++) {
7204: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7205: else fprintf(ficgp," %%*lf (%%*lf)");
7206: }
7207: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7208: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7209: } /* state */
7210: } /* vpopbased */
1.264 brouard 7211: 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 7212: } /* end nres */
7213: } /* k1 end 2 eme*/
7214:
7215:
7216: /*3eme*/
7217: for (k1=1; k1<= m ; k1 ++){
7218: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7219: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7220: continue;
7221:
7222: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7223: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7224: strcpy(gplotlabel,"(");
1.238 brouard 7225: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7226: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7227: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7228: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7229: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7230: vlv= nbcode[Tvaraff[k]][lv];
7231: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7232: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7233: }
7234: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7235: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7236: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7237: }
1.264 brouard 7238: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7239: fprintf(ficgp,"\n#\n");
7240: if(invalidvarcomb[k1]){
7241: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7242: continue;
7243: }
7244:
7245: /* k=2+nlstate*(2*cpt-2); */
7246: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7247: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7248: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7249: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7250: 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 7251: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7252: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7253: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7254: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7255: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7256: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7257:
1.238 brouard 7258: */
7259: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7260: 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 7261: /* 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 7262:
1.238 brouard 7263: }
1.261 brouard 7264: 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 7265: }
1.264 brouard 7266: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7267: } /* end nres */
7268: } /* end kl 3eme */
1.126 brouard 7269:
1.223 brouard 7270: /* 4eme */
1.201 brouard 7271: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7272: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7273: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7274: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7275: continue;
1.238 brouard 7276: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7277: strcpy(gplotlabel,"(");
1.238 brouard 7278: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7279: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7280: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7281: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7282: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7283: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7284: vlv= nbcode[Tvaraff[k]][lv];
7285: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7286: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7287: }
7288: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7289: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7290: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7291: }
1.264 brouard 7292: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7293: fprintf(ficgp,"\n#\n");
7294: if(invalidvarcomb[k1]){
7295: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7296: continue;
1.223 brouard 7297: }
1.238 brouard 7298:
1.241 brouard 7299: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7300: 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 7301: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7302: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7303: k=3;
7304: for (i=1; i<= nlstate ; i ++){
7305: if(i==1){
7306: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7307: }else{
7308: fprintf(ficgp,", '' ");
7309: }
7310: l=(nlstate+ndeath)*(i-1)+1;
7311: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7312: for (j=2; j<= nlstate+ndeath ; j ++)
7313: fprintf(ficgp,"+$%d",k+l+j-1);
7314: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7315: } /* nlstate */
1.264 brouard 7316: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7317: } /* end cpt state*/
7318: } /* end nres */
7319: } /* end covariate k1 */
7320:
1.220 brouard 7321: /* 5eme */
1.201 brouard 7322: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7323: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7324: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7325: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7326: continue;
1.238 brouard 7327: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7328: strcpy(gplotlabel,"(");
1.238 brouard 7329: 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);
7330: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7331: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7332: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7333: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7334: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7335: vlv= nbcode[Tvaraff[k]][lv];
7336: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7337: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7338: }
7339: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7340: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7341: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7342: }
1.264 brouard 7343: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7344: fprintf(ficgp,"\n#\n");
7345: if(invalidvarcomb[k1]){
7346: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7347: continue;
7348: }
1.227 brouard 7349:
1.241 brouard 7350: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7351: 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 7352: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7353: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7354: k=3;
7355: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7356: if(j==1)
7357: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7358: else
7359: fprintf(ficgp,", '' ");
7360: l=(nlstate+ndeath)*(cpt-1) +j;
7361: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7362: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7363: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7364: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7365: } /* nlstate */
7366: fprintf(ficgp,", '' ");
7367: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7368: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7369: l=(nlstate+ndeath)*(cpt-1) +j;
7370: if(j < nlstate)
7371: fprintf(ficgp,"$%d +",k+l);
7372: else
7373: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7374: }
1.264 brouard 7375: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7376: } /* end cpt state*/
7377: } /* end covariate */
7378: } /* end nres */
1.227 brouard 7379:
1.220 brouard 7380: /* 6eme */
1.202 brouard 7381: /* CV preval stable (period) for each covariate */
1.237 brouard 7382: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7383: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7384: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7385: continue;
1.255 brouard 7386: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7387: strcpy(gplotlabel,"(");
1.211 brouard 7388: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7389: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7390: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7391: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7392: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7393: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7394: vlv= nbcode[Tvaraff[k]][lv];
7395: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7396: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7397: }
1.237 brouard 7398: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7399: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7400: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7401: }
1.264 brouard 7402: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7403: fprintf(ficgp,"\n#\n");
1.223 brouard 7404: if(invalidvarcomb[k1]){
1.227 brouard 7405: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7406: continue;
1.223 brouard 7407: }
1.227 brouard 7408:
1.241 brouard 7409: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7410: 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 7411: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7412: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7413: k=3; /* Offset */
1.255 brouard 7414: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7415: if(i==1)
7416: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7417: else
7418: fprintf(ficgp,", '' ");
1.255 brouard 7419: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7420: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7421: for (j=2; j<= nlstate ; j ++)
7422: fprintf(ficgp,"+$%d",k+l+j-1);
7423: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7424: } /* nlstate */
1.264 brouard 7425: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7426: } /* end cpt state*/
7427: } /* end covariate */
1.227 brouard 7428:
7429:
1.220 brouard 7430: /* 7eme */
1.218 brouard 7431: if(backcast == 1){
1.217 brouard 7432: /* CV back preval stable (period) for each covariate */
1.237 brouard 7433: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7434: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7435: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7436: continue;
1.268 brouard 7437: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7438: strcpy(gplotlabel,"(");
7439: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7440: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7441: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7442: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7443: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7444: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7445: vlv= nbcode[Tvaraff[k]][lv];
7446: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7447: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7448: }
1.237 brouard 7449: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7450: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7451: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7452: }
1.264 brouard 7453: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7454: fprintf(ficgp,"\n#\n");
7455: if(invalidvarcomb[k1]){
7456: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7457: continue;
7458: }
7459:
1.241 brouard 7460: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7461: 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 7462: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7463: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7464: k=3; /* Offset */
1.268 brouard 7465: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7466: if(i==1)
7467: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7468: else
7469: fprintf(ficgp,", '' ");
7470: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7471: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7472: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7473: /* 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 7474: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7475: /* for (j=2; j<= nlstate ; j ++) */
7476: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7477: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7478: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7479: } /* nlstate */
1.264 brouard 7480: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7481: } /* end cpt state*/
7482: } /* end covariate */
7483: } /* End if backcast */
7484:
1.223 brouard 7485: /* 8eme */
1.218 brouard 7486: if(prevfcast==1){
7487: /* Projection from cross-sectional to stable (period) for each covariate */
7488:
1.237 brouard 7489: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7490: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7491: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7492: continue;
1.211 brouard 7493: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7494: strcpy(gplotlabel,"(");
1.227 brouard 7495: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7496: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7497: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7498: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7499: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7500: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7501: vlv= nbcode[Tvaraff[k]][lv];
7502: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7503: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7504: }
1.237 brouard 7505: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7506: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7507: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7508: }
1.264 brouard 7509: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7510: fprintf(ficgp,"\n#\n");
7511: if(invalidvarcomb[k1]){
7512: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7513: continue;
7514: }
7515:
7516: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7517: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7518: 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 7519: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7520: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7521:
7522: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7523: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7524: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7525: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7526: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7527: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7528: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7529: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7530: if(i==istart){
1.227 brouard 7531: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7532: }else{
7533: fprintf(ficgp,",\\\n '' ");
7534: }
7535: if(cptcoveff ==0){ /* No covariate */
7536: ioffset=2; /* Age is in 2 */
7537: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7538: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7539: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7540: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7541: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7542: if(i==nlstate+1){
1.270 brouard 7543: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7544: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7545: fprintf(ficgp,",\\\n '' ");
7546: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7547: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7548: offyear, \
1.268 brouard 7549: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7550: }else
1.227 brouard 7551: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7552: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7553: }else{ /* more than 2 covariates */
1.270 brouard 7554: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7555: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7556: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7557: iyearc=ioffset-1;
7558: iagec=ioffset;
1.227 brouard 7559: fprintf(ficgp," u %d:(",ioffset);
7560: kl=0;
7561: strcpy(gplotcondition,"(");
7562: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7563: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7564: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7565: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7566: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7567: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7568: kl++;
7569: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7570: kl++;
7571: if(k <cptcoveff && cptcoveff>1)
7572: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7573: }
7574: strcpy(gplotcondition+strlen(gplotcondition),")");
7575: /* 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 *\/ */
7576: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7577: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7578: /* '' 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*/
7579: if(i==nlstate+1){
1.270 brouard 7580: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7581: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7582: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7583: fprintf(ficgp," u %d:(",iagec);
7584: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7585: iyearc, iagec, offyear, \
7586: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7587: /* '' 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 7588: }else{
7589: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7590: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7591: }
7592: } /* end if covariate */
7593: } /* nlstate */
1.264 brouard 7594: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7595: } /* end cpt state*/
7596: } /* end covariate */
7597: } /* End if prevfcast */
1.227 brouard 7598:
1.268 brouard 7599: if(backcast==1){
7600: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7601:
7602: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7603: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7604: if(m != 1 && TKresult[nres]!= k1)
7605: continue;
7606: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7607: strcpy(gplotlabel,"(");
7608: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7609: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7610: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7611: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7612: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7613: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7614: vlv= nbcode[Tvaraff[k]][lv];
7615: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7616: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7617: }
7618: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7619: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7620: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7621: }
7622: strcpy(gplotlabel+strlen(gplotlabel),")");
7623: fprintf(ficgp,"\n#\n");
7624: if(invalidvarcomb[k1]){
7625: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7626: continue;
7627: }
7628:
7629: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7630: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7631: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7632: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7633: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7634:
7635: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7636: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7637: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7638: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7639: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7640: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7641: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7642: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7643: if(i==istart){
7644: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7645: }else{
7646: fprintf(ficgp,",\\\n '' ");
7647: }
7648: if(cptcoveff ==0){ /* No covariate */
7649: ioffset=2; /* Age is in 2 */
7650: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7651: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7652: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7653: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7654: fprintf(ficgp," u %d:(", ioffset);
7655: if(i==nlstate+1){
1.270 brouard 7656: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7657: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7658: fprintf(ficgp,",\\\n '' ");
7659: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7660: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7661: offbyear, \
7662: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7663: }else
7664: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7665: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7666: }else{ /* more than 2 covariates */
1.270 brouard 7667: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7668: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7669: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7670: iyearc=ioffset-1;
7671: iagec=ioffset;
1.268 brouard 7672: fprintf(ficgp," u %d:(",ioffset);
7673: kl=0;
7674: strcpy(gplotcondition,"(");
7675: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7676: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7677: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7678: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7679: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7680: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7681: kl++;
7682: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7683: kl++;
7684: if(k <cptcoveff && cptcoveff>1)
7685: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7686: }
7687: strcpy(gplotcondition+strlen(gplotcondition),")");
7688: /* 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 *\/ */
7689: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7690: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7691: /* '' 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*/
7692: if(i==nlstate+1){
1.270 brouard 7693: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7694: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7695: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7696: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7697: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7698: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7699: iyearc,iagec,offbyear, \
7700: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7701: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7702: }else{
7703: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7704: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7705: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7706: }
7707: } /* end if covariate */
7708: } /* nlstate */
7709: fprintf(ficgp,"\nset out; unset label;\n");
7710: } /* end cpt state*/
7711: } /* end covariate */
7712: } /* End if backcast */
7713:
1.227 brouard 7714:
1.238 brouard 7715: /* 9eme writing MLE parameters */
7716: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7717: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7718: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7719: for(k=1; k <=(nlstate+ndeath); k++){
7720: if (k != i) {
1.227 brouard 7721: fprintf(ficgp,"# current state %d\n",k);
7722: for(j=1; j <=ncovmodel; j++){
7723: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7724: jk++;
7725: }
7726: fprintf(ficgp,"\n");
1.126 brouard 7727: }
7728: }
1.223 brouard 7729: }
1.187 brouard 7730: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7731:
1.145 brouard 7732: /*goto avoid;*/
1.238 brouard 7733: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7734: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7735: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7736: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7737: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7738: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7739: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7740: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7741: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7742: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7743: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7744: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7745: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7746: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7747: fprintf(ficgp,"#\n");
1.223 brouard 7748: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7749: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7750: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7751: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7752: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7753: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7754: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7755: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7756: continue;
1.264 brouard 7757: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7758: strcpy(gplotlabel,"(");
7759: sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);
7760: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7761: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7762: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7763: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7764: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7765: vlv= nbcode[Tvaraff[k]][lv];
7766: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7767: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7768: }
1.237 brouard 7769: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7770: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7771: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7772: }
1.264 brouard 7773: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7774: fprintf(ficgp,"\n#\n");
1.264 brouard 7775: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
7776: fprintf(ficgp,"\nset label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7777: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7778: if (ng==1){
7779: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7780: fprintf(ficgp,"\nunset log y");
7781: }else if (ng==2){
7782: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7783: fprintf(ficgp,"\nset log y");
7784: }else if (ng==3){
7785: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7786: fprintf(ficgp,"\nset log y");
7787: }else
7788: fprintf(ficgp,"\nunset title ");
7789: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7790: i=1;
7791: for(k2=1; k2<=nlstate; k2++) {
7792: k3=i;
7793: for(k=1; k<=(nlstate+ndeath); k++) {
7794: if (k != k2){
7795: switch( ng) {
7796: case 1:
7797: if(nagesqr==0)
7798: fprintf(ficgp," p%d+p%d*x",i,i+1);
7799: else /* nagesqr =1 */
7800: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7801: break;
7802: case 2: /* ng=2 */
7803: if(nagesqr==0)
7804: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7805: else /* nagesqr =1 */
7806: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7807: break;
7808: case 3:
7809: if(nagesqr==0)
7810: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7811: else /* nagesqr =1 */
7812: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7813: break;
7814: }
7815: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7816: ijp=1; /* product no age */
7817: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7818: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7819: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7820: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7821: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7822: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7823: if(DummyV[j]==0){
7824: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7825: }else{ /* quantitative */
7826: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7827: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7828: }
7829: ij++;
1.237 brouard 7830: }
1.268 brouard 7831: }
7832: }else if(cptcovprod >0){
7833: if(j==Tprod[ijp]) { /* */
7834: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7835: if(ijp <=cptcovprod) { /* Product */
7836: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7837: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7838: /* 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)]); */
7839: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7840: }else{ /* Vn is dummy and Vm is quanti */
7841: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7842: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7843: }
7844: }else{ /* Vn*Vm Vn is quanti */
7845: if(DummyV[Tvard[ijp][2]]==0){
7846: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7847: }else{ /* Both quanti */
7848: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7849: }
1.237 brouard 7850: }
1.268 brouard 7851: ijp++;
1.237 brouard 7852: }
1.268 brouard 7853: } /* end Tprod */
1.237 brouard 7854: } else{ /* simple covariate */
1.264 brouard 7855: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7856: if(Dummy[j]==0){
7857: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7858: }else{ /* quantitative */
7859: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7860: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7861: }
1.237 brouard 7862: } /* end simple */
7863: } /* end j */
1.223 brouard 7864: }else{
7865: i=i-ncovmodel;
7866: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7867: fprintf(ficgp," (1.");
7868: }
1.227 brouard 7869:
1.223 brouard 7870: if(ng != 1){
7871: fprintf(ficgp,")/(1");
1.227 brouard 7872:
1.264 brouard 7873: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7874: if(nagesqr==0)
1.264 brouard 7875: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7876: else /* nagesqr =1 */
1.264 brouard 7877: 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 7878:
1.223 brouard 7879: ij=1;
7880: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7881: if(cptcovage >0){
7882: if((j-2)==Tage[ij]) { /* Bug valgrind */
7883: if(ij <=cptcovage) { /* Bug valgrind */
7884: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7885: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7886: ij++;
7887: }
7888: }
7889: }else
7890: 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 7891: }
7892: fprintf(ficgp,")");
7893: }
7894: fprintf(ficgp,")");
7895: if(ng ==2)
1.275 ! brouard 7896: fprintf(ficgp," w l lt (%d*%d+%d)%%%d+1 dt %d t \"p%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 7897: else /* ng= 3 */
1.275 ! brouard 7898: fprintf(ficgp," w l lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 7899: }else{ /* end ng <> 1 */
7900: if( k !=k2) /* logit p11 is hard to draw */
1.275 ! brouard 7901: fprintf(ficgp," w l 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 7902: }
7903: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7904: fprintf(ficgp,",");
7905: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7906: fprintf(ficgp,",");
7907: i=i+ncovmodel;
7908: } /* end k */
7909: } /* end k2 */
1.264 brouard 7910: fprintf(ficgp,"\n set out; unset label;\n");
7911: } /* end k1 */
1.223 brouard 7912: } /* end ng */
7913: /* avoid: */
7914: fflush(ficgp);
1.126 brouard 7915: } /* end gnuplot */
7916:
7917:
7918: /*************** Moving average **************/
1.219 brouard 7919: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7920: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7921:
1.222 brouard 7922: int i, cpt, cptcod;
7923: int modcovmax =1;
7924: int mobilavrange, mob;
7925: int iage=0;
7926:
1.266 brouard 7927: double sum=0., sumr=0.;
1.222 brouard 7928: double age;
1.266 brouard 7929: double *sumnewp, *sumnewm, *sumnewmr;
7930: double *agemingood, *agemaxgood;
7931: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7932:
7933:
1.225 brouard 7934: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7935: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7936:
7937: sumnewp = vector(1,ncovcombmax);
7938: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7939: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7940: agemingood = vector(1,ncovcombmax);
1.266 brouard 7941: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7942: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7943: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7944:
7945: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7946: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7947: sumnewp[cptcod]=0.;
1.266 brouard 7948: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
7949: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7950: }
7951: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7952:
1.266 brouard 7953: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7954: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7955: else mobilavrange=mobilav;
7956: for (age=bage; age<=fage; age++)
7957: for (i=1; i<=nlstate;i++)
7958: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7959: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7960: /* We keep the original values on the extreme ages bage, fage and for
7961: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7962: we use a 5 terms etc. until the borders are no more concerned.
7963: */
7964: for (mob=3;mob <=mobilavrange;mob=mob+2){
7965: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 7966: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7967: sumnewm[cptcod]=0.;
7968: for (i=1; i<=nlstate;i++){
1.222 brouard 7969: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7970: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7971: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7972: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7973: }
7974: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 7975: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7976: } /* end i */
7977: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
7978: } /* end cptcod */
1.222 brouard 7979: }/* end age */
7980: }/* end mob */
1.266 brouard 7981: }else{
7982: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 7983: return -1;
1.266 brouard 7984: }
7985:
7986: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 7987: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7988: if(invalidvarcomb[cptcod]){
7989: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7990: continue;
7991: }
1.219 brouard 7992:
1.266 brouard 7993: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
7994: sumnewm[cptcod]=0.;
7995: sumnewmr[cptcod]=0.;
7996: for (i=1; i<=nlstate;i++){
7997: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7998: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
7999: }
8000: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8001: agemingoodr[cptcod]=age;
8002: }
8003: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8004: agemingood[cptcod]=age;
8005: }
8006: } /* age */
8007: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8008: sumnewm[cptcod]=0.;
1.266 brouard 8009: sumnewmr[cptcod]=0.;
1.222 brouard 8010: for (i=1; i<=nlstate;i++){
8011: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8012: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8013: }
8014: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8015: agemaxgoodr[cptcod]=age;
1.222 brouard 8016: }
8017: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8018: agemaxgood[cptcod]=age;
8019: }
8020: } /* age */
8021: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8022: /* but they will change */
8023: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8024: sumnewm[cptcod]=0.;
8025: sumnewmr[cptcod]=0.;
8026: for (i=1; i<=nlstate;i++){
8027: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8028: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8029: }
8030: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8031: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8032: agemaxgoodr[cptcod]=age; /* age min */
8033: for (i=1; i<=nlstate;i++)
8034: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8035: }else{ /* bad we change the value with the values of good ages */
8036: for (i=1; i<=nlstate;i++){
8037: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8038: } /* i */
8039: } /* end bad */
8040: }else{
8041: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8042: agemaxgood[cptcod]=age;
8043: }else{ /* bad we change the value with the values of good ages */
8044: for (i=1; i<=nlstate;i++){
8045: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8046: } /* i */
8047: } /* end bad */
8048: }/* end else */
8049: sum=0.;sumr=0.;
8050: for (i=1; i<=nlstate;i++){
8051: sum+=mobaverage[(int)age][i][cptcod];
8052: sumr+=probs[(int)age][i][cptcod];
8053: }
8054: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8055: 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 8056: } /* end bad */
8057: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8058: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8059: 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 8060: } /* end bad */
8061: }/* age */
1.266 brouard 8062:
8063: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8064: sumnewm[cptcod]=0.;
1.266 brouard 8065: sumnewmr[cptcod]=0.;
1.222 brouard 8066: for (i=1; i<=nlstate;i++){
8067: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8068: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8069: }
8070: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8071: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8072: agemingoodr[cptcod]=age;
8073: for (i=1; i<=nlstate;i++)
8074: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8075: }else{ /* bad we change the value with the values of good ages */
8076: for (i=1; i<=nlstate;i++){
8077: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8078: } /* i */
8079: } /* end bad */
8080: }else{
8081: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8082: agemingood[cptcod]=age;
8083: }else{ /* bad */
8084: for (i=1; i<=nlstate;i++){
8085: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8086: } /* i */
8087: } /* end bad */
8088: }/* end else */
8089: sum=0.;sumr=0.;
8090: for (i=1; i<=nlstate;i++){
8091: sum+=mobaverage[(int)age][i][cptcod];
8092: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8093: }
1.266 brouard 8094: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8095: 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 8096: } /* end bad */
8097: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8098: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8099: 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 8100: } /* end bad */
8101: }/* age */
1.266 brouard 8102:
1.222 brouard 8103:
8104: for (age=bage; age<=fage; age++){
1.235 brouard 8105: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8106: sumnewp[cptcod]=0.;
8107: sumnewm[cptcod]=0.;
8108: for (i=1; i<=nlstate;i++){
8109: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8110: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8111: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8112: }
8113: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8114: }
8115: /* printf("\n"); */
8116: /* } */
1.266 brouard 8117:
1.222 brouard 8118: /* brutal averaging */
1.266 brouard 8119: /* for (i=1; i<=nlstate;i++){ */
8120: /* for (age=1; age<=bage; age++){ */
8121: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8122: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8123: /* } */
8124: /* for (age=fage; age<=AGESUP; age++){ */
8125: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8126: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8127: /* } */
8128: /* } /\* end i status *\/ */
8129: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8130: /* for (age=1; age<=AGESUP; age++){ */
8131: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8132: /* mobaverage[(int)age][i][cptcod]=0.; */
8133: /* } */
8134: /* } */
1.222 brouard 8135: }/* end cptcod */
1.266 brouard 8136: free_vector(agemaxgoodr,1, ncovcombmax);
8137: free_vector(agemaxgood,1, ncovcombmax);
8138: free_vector(agemingood,1, ncovcombmax);
8139: free_vector(agemingoodr,1, ncovcombmax);
8140: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8141: free_vector(sumnewm,1, ncovcombmax);
8142: free_vector(sumnewp,1, ncovcombmax);
8143: return 0;
8144: }/* End movingaverage */
1.218 brouard 8145:
1.126 brouard 8146:
8147: /************** Forecasting ******************/
1.269 brouard 8148: 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 8149: /* proj1, year, month, day of starting projection
8150: agemin, agemax range of age
8151: dateprev1 dateprev2 range of dates during which prevalence is computed
8152: anproj2 year of en of projection (same day and month as proj1).
8153: */
1.267 brouard 8154: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8155: double agec; /* generic age */
8156: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8157: double *popeffectif,*popcount;
8158: double ***p3mat;
1.218 brouard 8159: /* double ***mobaverage; */
1.126 brouard 8160: char fileresf[FILENAMELENGTH];
8161:
8162: agelim=AGESUP;
1.211 brouard 8163: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8164: in each health status at the date of interview (if between dateprev1 and dateprev2).
8165: We still use firstpass and lastpass as another selection.
8166: */
1.214 brouard 8167: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8168: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8169:
1.201 brouard 8170: strcpy(fileresf,"F_");
8171: strcat(fileresf,fileresu);
1.126 brouard 8172: if((ficresf=fopen(fileresf,"w"))==NULL) {
8173: printf("Problem with forecast resultfile: %s\n", fileresf);
8174: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8175: }
1.235 brouard 8176: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8177: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8178:
1.225 brouard 8179: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8180:
8181:
8182: stepsize=(int) (stepm+YEARM-1)/YEARM;
8183: if (stepm<=12) stepsize=1;
8184: if(estepm < stepm){
8185: printf ("Problem %d lower than %d\n",estepm, stepm);
8186: }
1.270 brouard 8187: else{
8188: hstepm=estepm;
8189: }
8190: if(estepm > stepm){ /* Yes every two year */
8191: stepsize=2;
8192: }
1.126 brouard 8193:
8194: hstepm=hstepm/stepm;
8195: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8196: fractional in yp1 */
8197: anprojmean=yp;
8198: yp2=modf((yp1*12),&yp);
8199: mprojmean=yp;
8200: yp1=modf((yp2*30.5),&yp);
8201: jprojmean=yp;
8202: if(jprojmean==0) jprojmean=1;
8203: if(mprojmean==0) jprojmean=1;
8204:
1.227 brouard 8205: i1=pow(2,cptcoveff);
1.126 brouard 8206: if (cptcovn < 1){i1=1;}
8207:
8208: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8209:
8210: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8211:
1.126 brouard 8212: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8213: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8214: for(k=1; k<=i1;k++){
1.253 brouard 8215: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8216: continue;
1.227 brouard 8217: if(invalidvarcomb[k]){
8218: printf("\nCombination (%d) projection ignored because no cases \n",k);
8219: continue;
8220: }
8221: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8222: for(j=1;j<=cptcoveff;j++) {
8223: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8224: }
1.235 brouard 8225: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8226: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8227: }
1.227 brouard 8228: fprintf(ficresf," yearproj age");
8229: for(j=1; j<=nlstate+ndeath;j++){
8230: for(i=1; i<=nlstate;i++)
8231: fprintf(ficresf," p%d%d",i,j);
8232: fprintf(ficresf," wp.%d",j);
8233: }
8234: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8235: fprintf(ficresf,"\n");
8236: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8237: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8238: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8239: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8240: nhstepm = nhstepm/hstepm;
8241: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8242: oldm=oldms;savm=savms;
1.268 brouard 8243: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8244: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8245: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8246: for (h=0; h<=nhstepm; h++){
8247: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8248: break;
8249: }
8250: }
8251: fprintf(ficresf,"\n");
8252: for(j=1;j<=cptcoveff;j++)
8253: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8254: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8255:
8256: for(j=1; j<=nlstate+ndeath;j++) {
8257: ppij=0.;
8258: for(i=1; i<=nlstate;i++) {
8259: /* if (mobilav>=1) */
1.269 brouard 8260: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
1.268 brouard 8261: /* else { */ /* even if mobilav==-1 we use mobaverage */
8262: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8263: /* } */
8264: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8265: } /* end i */
8266: fprintf(ficresf," %.3f", ppij);
8267: }/* end j */
1.227 brouard 8268: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8269: } /* end agec */
1.266 brouard 8270: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8271: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8272: } /* end yearp */
8273: } /* end k */
1.219 brouard 8274:
1.126 brouard 8275: fclose(ficresf);
1.215 brouard 8276: printf("End of Computing forecasting \n");
8277: fprintf(ficlog,"End of Computing forecasting\n");
8278:
1.126 brouard 8279: }
8280:
1.269 brouard 8281: /************** Back Forecasting ******************/
8282: 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 8283: /* back1, year, month, day of starting backection
8284: agemin, agemax range of age
8285: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8286: anback2 year of end of backprojection (same day and month as back1).
8287: prevacurrent and prev are prevalences.
1.267 brouard 8288: */
8289: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8290: double agec; /* generic age */
1.268 brouard 8291: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8292: double *popeffectif,*popcount;
8293: double ***p3mat;
8294: /* double ***mobaverage; */
8295: char fileresfb[FILENAMELENGTH];
8296:
1.268 brouard 8297: agelim=AGEINF;
1.267 brouard 8298: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8299: in each health status at the date of interview (if between dateprev1 and dateprev2).
8300: We still use firstpass and lastpass as another selection.
8301: */
8302: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8303: /* firstpass, lastpass, stepm, weightopt, model); */
8304:
8305: /*Do we need to compute prevalence again?*/
8306:
8307: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8308:
8309: strcpy(fileresfb,"FB_");
8310: strcat(fileresfb,fileresu);
8311: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8312: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8313: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8314: }
8315: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8316: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8317:
8318: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8319:
8320:
8321: stepsize=(int) (stepm+YEARM-1)/YEARM;
8322: if (stepm<=12) stepsize=1;
8323: if(estepm < stepm){
8324: printf ("Problem %d lower than %d\n",estepm, stepm);
8325: }
1.270 brouard 8326: else{
8327: hstepm=estepm;
8328: }
8329: if(estepm >= stepm){ /* Yes every two year */
8330: stepsize=2;
8331: }
1.267 brouard 8332:
8333: hstepm=hstepm/stepm;
8334: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8335: fractional in yp1 */
8336: anprojmean=yp;
8337: yp2=modf((yp1*12),&yp);
8338: mprojmean=yp;
8339: yp1=modf((yp2*30.5),&yp);
8340: jprojmean=yp;
8341: if(jprojmean==0) jprojmean=1;
8342: if(mprojmean==0) jprojmean=1;
8343:
8344: i1=pow(2,cptcoveff);
8345: if (cptcovn < 1){i1=1;}
8346:
8347: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8348: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8349:
8350: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8351:
8352: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8353: for(k=1; k<=i1;k++){
8354: if(i1 != 1 && TKresult[nres]!= k)
8355: continue;
8356: if(invalidvarcomb[k]){
8357: printf("\nCombination (%d) projection ignored because no cases \n",k);
8358: continue;
8359: }
1.268 brouard 8360: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8361: for(j=1;j<=cptcoveff;j++) {
8362: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8363: }
8364: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8365: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8366: }
8367: fprintf(ficresfb," yearbproj age");
8368: for(j=1; j<=nlstate+ndeath;j++){
8369: for(i=1; i<=nlstate;i++)
1.268 brouard 8370: fprintf(ficresfb," b%d%d",i,j);
8371: fprintf(ficresfb," b.%d",j);
1.267 brouard 8372: }
8373: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8374: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8375: fprintf(ficresfb,"\n");
8376: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8377: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8378: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8379: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8380: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8381: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8382: nhstepm = nhstepm/hstepm;
8383: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8384: oldm=oldms;savm=savms;
1.268 brouard 8385: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8386: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8387: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8388: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8389: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8390: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8391: for (h=0; h<=nhstepm; h++){
1.268 brouard 8392: if (h*hstepm/YEARM*stepm ==-yearp) {
8393: break;
8394: }
8395: }
8396: fprintf(ficresfb,"\n");
8397: for(j=1;j<=cptcoveff;j++)
8398: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8399: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8400: for(i=1; i<=nlstate+ndeath;i++) {
8401: ppij=0.;ppi=0.;
8402: for(j=1; j<=nlstate;j++) {
8403: /* if (mobilav==1) */
1.269 brouard 8404: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8405: ppi=ppi+prevacurrent[(int)agec][j][k];
8406: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8407: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8408: /* else { */
8409: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8410: /* } */
1.268 brouard 8411: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8412: } /* end j */
8413: if(ppi <0.99){
8414: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8415: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8416: }
8417: fprintf(ficresfb," %.3f", ppij);
8418: }/* end j */
1.267 brouard 8419: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8420: } /* end agec */
8421: } /* end yearp */
8422: } /* end k */
1.217 brouard 8423:
1.267 brouard 8424: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8425:
1.267 brouard 8426: fclose(ficresfb);
8427: printf("End of Computing Back forecasting \n");
8428: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8429:
1.267 brouard 8430: }
1.217 brouard 8431:
1.269 brouard 8432: /* Variance of prevalence limit: varprlim */
8433: 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){
8434: /*------- Variance of period (stable) prevalence------*/
8435:
8436: char fileresvpl[FILENAMELENGTH];
8437: FILE *ficresvpl;
8438: double **oldm, **savm;
8439: double **varpl; /* Variances of prevalence limits by age */
8440: int i1, k, nres, j ;
8441:
8442: strcpy(fileresvpl,"VPL_");
8443: strcat(fileresvpl,fileresu);
8444: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8445: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8446: exit(0);
8447: }
8448: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8449: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8450:
8451: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8452: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8453:
8454: i1=pow(2,cptcoveff);
8455: if (cptcovn < 1){i1=1;}
8456:
8457: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8458: for(k=1; k<=i1;k++){
8459: if(i1 != 1 && TKresult[nres]!= k)
8460: continue;
8461: fprintf(ficresvpl,"\n#****** ");
8462: printf("\n#****** ");
8463: fprintf(ficlog,"\n#****** ");
8464: for(j=1;j<=cptcoveff;j++) {
8465: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8466: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8467: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8468: }
8469: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8470: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8471: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8472: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8473: }
8474: fprintf(ficresvpl,"******\n");
8475: printf("******\n");
8476: fprintf(ficlog,"******\n");
8477:
8478: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8479: oldm=oldms;savm=savms;
8480: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8481: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8482: /*}*/
8483: }
8484:
8485: fclose(ficresvpl);
8486: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8487: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8488:
8489: }
8490: /* Variance of back prevalence: varbprlim */
8491: 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){
8492: /*------- Variance of back (stable) prevalence------*/
8493:
8494: char fileresvbl[FILENAMELENGTH];
8495: FILE *ficresvbl;
8496:
8497: double **oldm, **savm;
8498: double **varbpl; /* Variances of back prevalence limits by age */
8499: int i1, k, nres, j ;
8500:
8501: strcpy(fileresvbl,"VBL_");
8502: strcat(fileresvbl,fileresu);
8503: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8504: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8505: exit(0);
8506: }
8507: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8508: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8509:
8510:
8511: i1=pow(2,cptcoveff);
8512: if (cptcovn < 1){i1=1;}
8513:
8514: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8515: for(k=1; k<=i1;k++){
8516: if(i1 != 1 && TKresult[nres]!= k)
8517: continue;
8518: fprintf(ficresvbl,"\n#****** ");
8519: printf("\n#****** ");
8520: fprintf(ficlog,"\n#****** ");
8521: for(j=1;j<=cptcoveff;j++) {
8522: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8523: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8524: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8525: }
8526: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8527: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8528: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8529: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8530: }
8531: fprintf(ficresvbl,"******\n");
8532: printf("******\n");
8533: fprintf(ficlog,"******\n");
8534:
8535: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8536: oldm=oldms;savm=savms;
8537:
8538: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8539: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8540: /*}*/
8541: }
8542:
8543: fclose(ficresvbl);
8544: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8545: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8546:
8547: } /* End of varbprlim */
8548:
1.126 brouard 8549: /************** Forecasting *****not tested NB*************/
1.227 brouard 8550: /* 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 8551:
1.227 brouard 8552: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8553: /* int *popage; */
8554: /* double calagedatem, agelim, kk1, kk2; */
8555: /* double *popeffectif,*popcount; */
8556: /* double ***p3mat,***tabpop,***tabpopprev; */
8557: /* /\* double ***mobaverage; *\/ */
8558: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8559:
1.227 brouard 8560: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8561: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8562: /* agelim=AGESUP; */
8563: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8564:
1.227 brouard 8565: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8566:
8567:
1.227 brouard 8568: /* strcpy(filerespop,"POP_"); */
8569: /* strcat(filerespop,fileresu); */
8570: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8571: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8572: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8573: /* } */
8574: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8575: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8576:
1.227 brouard 8577: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8578:
1.227 brouard 8579: /* /\* if (mobilav!=0) { *\/ */
8580: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8581: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8582: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8583: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8584: /* /\* } *\/ */
8585: /* /\* } *\/ */
1.126 brouard 8586:
1.227 brouard 8587: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8588: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8589:
1.227 brouard 8590: /* agelim=AGESUP; */
1.126 brouard 8591:
1.227 brouard 8592: /* hstepm=1; */
8593: /* hstepm=hstepm/stepm; */
1.218 brouard 8594:
1.227 brouard 8595: /* if (popforecast==1) { */
8596: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8597: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8598: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8599: /* } */
8600: /* popage=ivector(0,AGESUP); */
8601: /* popeffectif=vector(0,AGESUP); */
8602: /* popcount=vector(0,AGESUP); */
1.126 brouard 8603:
1.227 brouard 8604: /* i=1; */
8605: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8606:
1.227 brouard 8607: /* imx=i; */
8608: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8609: /* } */
1.218 brouard 8610:
1.227 brouard 8611: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8612: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8613: /* k=k+1; */
8614: /* fprintf(ficrespop,"\n#******"); */
8615: /* for(j=1;j<=cptcoveff;j++) { */
8616: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8617: /* } */
8618: /* fprintf(ficrespop,"******\n"); */
8619: /* fprintf(ficrespop,"# Age"); */
8620: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8621: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8622:
1.227 brouard 8623: /* for (cpt=0; cpt<=0;cpt++) { */
8624: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8625:
1.227 brouard 8626: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8627: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8628: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8629:
1.227 brouard 8630: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8631: /* oldm=oldms;savm=savms; */
8632: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8633:
1.227 brouard 8634: /* for (h=0; h<=nhstepm; h++){ */
8635: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8636: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8637: /* } */
8638: /* for(j=1; j<=nlstate+ndeath;j++) { */
8639: /* kk1=0.;kk2=0; */
8640: /* for(i=1; i<=nlstate;i++) { */
8641: /* if (mobilav==1) */
8642: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8643: /* else { */
8644: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8645: /* } */
8646: /* } */
8647: /* if (h==(int)(calagedatem+12*cpt)){ */
8648: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8649: /* /\*fprintf(ficrespop," %.3f", kk1); */
8650: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8651: /* } */
8652: /* } */
8653: /* for(i=1; i<=nlstate;i++){ */
8654: /* kk1=0.; */
8655: /* for(j=1; j<=nlstate;j++){ */
8656: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8657: /* } */
8658: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8659: /* } */
1.218 brouard 8660:
1.227 brouard 8661: /* if (h==(int)(calagedatem+12*cpt)) */
8662: /* for(j=1; j<=nlstate;j++) */
8663: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8664: /* } */
8665: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8666: /* } */
8667: /* } */
1.218 brouard 8668:
1.227 brouard 8669: /* /\******\/ */
1.218 brouard 8670:
1.227 brouard 8671: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8672: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8673: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8674: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8675: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8676:
1.227 brouard 8677: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8678: /* oldm=oldms;savm=savms; */
8679: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8680: /* for (h=0; h<=nhstepm; h++){ */
8681: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8682: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8683: /* } */
8684: /* for(j=1; j<=nlstate+ndeath;j++) { */
8685: /* kk1=0.;kk2=0; */
8686: /* for(i=1; i<=nlstate;i++) { */
8687: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8688: /* } */
8689: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8690: /* } */
8691: /* } */
8692: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8693: /* } */
8694: /* } */
8695: /* } */
8696: /* } */
1.218 brouard 8697:
1.227 brouard 8698: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8699:
1.227 brouard 8700: /* if (popforecast==1) { */
8701: /* free_ivector(popage,0,AGESUP); */
8702: /* free_vector(popeffectif,0,AGESUP); */
8703: /* free_vector(popcount,0,AGESUP); */
8704: /* } */
8705: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8706: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8707: /* fclose(ficrespop); */
8708: /* } /\* End of popforecast *\/ */
1.218 brouard 8709:
1.126 brouard 8710: int fileappend(FILE *fichier, char *optionfich)
8711: {
8712: if((fichier=fopen(optionfich,"a"))==NULL) {
8713: printf("Problem with file: %s\n", optionfich);
8714: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8715: return (0);
8716: }
8717: fflush(fichier);
8718: return (1);
8719: }
8720:
8721:
8722: /**************** function prwizard **********************/
8723: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8724: {
8725:
8726: /* Wizard to print covariance matrix template */
8727:
1.164 brouard 8728: char ca[32], cb[32];
8729: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8730: int numlinepar;
8731:
8732: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8733: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8734: for(i=1; i <=nlstate; i++){
8735: jj=0;
8736: for(j=1; j <=nlstate+ndeath; j++){
8737: if(j==i) continue;
8738: jj++;
8739: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8740: printf("%1d%1d",i,j);
8741: fprintf(ficparo,"%1d%1d",i,j);
8742: for(k=1; k<=ncovmodel;k++){
8743: /* printf(" %lf",param[i][j][k]); */
8744: /* fprintf(ficparo," %lf",param[i][j][k]); */
8745: printf(" 0.");
8746: fprintf(ficparo," 0.");
8747: }
8748: printf("\n");
8749: fprintf(ficparo,"\n");
8750: }
8751: }
8752: printf("# Scales (for hessian or gradient estimation)\n");
8753: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8754: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8755: for(i=1; i <=nlstate; i++){
8756: jj=0;
8757: for(j=1; j <=nlstate+ndeath; j++){
8758: if(j==i) continue;
8759: jj++;
8760: fprintf(ficparo,"%1d%1d",i,j);
8761: printf("%1d%1d",i,j);
8762: fflush(stdout);
8763: for(k=1; k<=ncovmodel;k++){
8764: /* printf(" %le",delti3[i][j][k]); */
8765: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8766: printf(" 0.");
8767: fprintf(ficparo," 0.");
8768: }
8769: numlinepar++;
8770: printf("\n");
8771: fprintf(ficparo,"\n");
8772: }
8773: }
8774: printf("# Covariance matrix\n");
8775: /* # 121 Var(a12)\n\ */
8776: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8777: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8778: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8779: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8780: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8781: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8782: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8783: fflush(stdout);
8784: fprintf(ficparo,"# Covariance matrix\n");
8785: /* # 121 Var(a12)\n\ */
8786: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8787: /* # ...\n\ */
8788: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8789:
8790: for(itimes=1;itimes<=2;itimes++){
8791: jj=0;
8792: for(i=1; i <=nlstate; i++){
8793: for(j=1; j <=nlstate+ndeath; j++){
8794: if(j==i) continue;
8795: for(k=1; k<=ncovmodel;k++){
8796: jj++;
8797: ca[0]= k+'a'-1;ca[1]='\0';
8798: if(itimes==1){
8799: printf("#%1d%1d%d",i,j,k);
8800: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8801: }else{
8802: printf("%1d%1d%d",i,j,k);
8803: fprintf(ficparo,"%1d%1d%d",i,j,k);
8804: /* printf(" %.5le",matcov[i][j]); */
8805: }
8806: ll=0;
8807: for(li=1;li <=nlstate; li++){
8808: for(lj=1;lj <=nlstate+ndeath; lj++){
8809: if(lj==li) continue;
8810: for(lk=1;lk<=ncovmodel;lk++){
8811: ll++;
8812: if(ll<=jj){
8813: cb[0]= lk +'a'-1;cb[1]='\0';
8814: if(ll<jj){
8815: if(itimes==1){
8816: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8817: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8818: }else{
8819: printf(" 0.");
8820: fprintf(ficparo," 0.");
8821: }
8822: }else{
8823: if(itimes==1){
8824: printf(" Var(%s%1d%1d)",ca,i,j);
8825: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8826: }else{
8827: printf(" 0.");
8828: fprintf(ficparo," 0.");
8829: }
8830: }
8831: }
8832: } /* end lk */
8833: } /* end lj */
8834: } /* end li */
8835: printf("\n");
8836: fprintf(ficparo,"\n");
8837: numlinepar++;
8838: } /* end k*/
8839: } /*end j */
8840: } /* end i */
8841: } /* end itimes */
8842:
8843: } /* end of prwizard */
8844: /******************* Gompertz Likelihood ******************************/
8845: double gompertz(double x[])
8846: {
8847: double A,B,L=0.0,sump=0.,num=0.;
8848: int i,n=0; /* n is the size of the sample */
8849:
1.220 brouard 8850: for (i=1;i<=imx ; i++) {
1.126 brouard 8851: sump=sump+weight[i];
8852: /* sump=sump+1;*/
8853: num=num+1;
8854: }
8855:
8856:
8857: /* for (i=0; i<=imx; i++)
8858: 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]);*/
8859:
8860: for (i=1;i<=imx ; i++)
8861: {
8862: if (cens[i] == 1 && wav[i]>1)
8863: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8864:
8865: if (cens[i] == 0 && wav[i]>1)
8866: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8867: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8868:
8869: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8870: if (wav[i] > 1 ) { /* ??? */
8871: L=L+A*weight[i];
8872: /* 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]);*/
8873: }
8874: }
8875:
8876: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8877:
8878: return -2*L*num/sump;
8879: }
8880:
1.136 brouard 8881: #ifdef GSL
8882: /******************* Gompertz_f Likelihood ******************************/
8883: double gompertz_f(const gsl_vector *v, void *params)
8884: {
8885: double A,B,LL=0.0,sump=0.,num=0.;
8886: double *x= (double *) v->data;
8887: int i,n=0; /* n is the size of the sample */
8888:
8889: for (i=0;i<=imx-1 ; i++) {
8890: sump=sump+weight[i];
8891: /* sump=sump+1;*/
8892: num=num+1;
8893: }
8894:
8895:
8896: /* for (i=0; i<=imx; i++)
8897: 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]);*/
8898: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8899: for (i=1;i<=imx ; i++)
8900: {
8901: if (cens[i] == 1 && wav[i]>1)
8902: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8903:
8904: if (cens[i] == 0 && wav[i]>1)
8905: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8906: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8907:
8908: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8909: if (wav[i] > 1 ) { /* ??? */
8910: LL=LL+A*weight[i];
8911: /* 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]);*/
8912: }
8913: }
8914:
8915: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8916: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8917:
8918: return -2*LL*num/sump;
8919: }
8920: #endif
8921:
1.126 brouard 8922: /******************* Printing html file ***********/
1.201 brouard 8923: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8924: int lastpass, int stepm, int weightopt, char model[],\
8925: int imx, double p[],double **matcov,double agemortsup){
8926: int i,k;
8927:
8928: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8929: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8930: for (i=1;i<=2;i++)
8931: 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 8932: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8933: fprintf(fichtm,"</ul>");
8934:
8935: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8936:
8937: 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>");
8938:
8939: for (k=agegomp;k<(agemortsup-2);k++)
8940: 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]);
8941:
8942:
8943: fflush(fichtm);
8944: }
8945:
8946: /******************* Gnuplot file **************/
1.201 brouard 8947: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8948:
8949: char dirfileres[132],optfileres[132];
1.164 brouard 8950:
1.126 brouard 8951: int ng;
8952:
8953:
8954: /*#ifdef windows */
8955: fprintf(ficgp,"cd \"%s\" \n",pathc);
8956: /*#endif */
8957:
8958:
8959: strcpy(dirfileres,optionfilefiname);
8960: strcpy(optfileres,"vpl");
1.199 brouard 8961: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8962: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8963: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8964: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8965: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8966:
8967: }
8968:
1.136 brouard 8969: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8970: {
1.126 brouard 8971:
1.136 brouard 8972: /*-------- data file ----------*/
8973: FILE *fic;
8974: char dummy[]=" ";
1.240 brouard 8975: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8976: int lstra;
1.136 brouard 8977: int linei, month, year,iout;
8978: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8979: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8980: char *stratrunc;
1.223 brouard 8981:
1.240 brouard 8982: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8983: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8984:
1.240 brouard 8985: for(v=1; v <=ncovcol;v++){
8986: DummyV[v]=0;
8987: FixedV[v]=0;
8988: }
8989: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8990: DummyV[v]=1;
8991: FixedV[v]=0;
8992: }
8993: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8994: DummyV[v]=0;
8995: FixedV[v]=1;
8996: }
8997: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8998: DummyV[v]=1;
8999: FixedV[v]=1;
9000: }
9001: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9002: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9003: 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]);
9004: }
1.126 brouard 9005:
1.136 brouard 9006: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9007: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9008: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9009: }
1.126 brouard 9010:
1.136 brouard 9011: i=1;
9012: linei=0;
9013: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9014: linei=linei+1;
9015: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9016: if(line[j] == '\t')
9017: line[j] = ' ';
9018: }
9019: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9020: ;
9021: };
9022: line[j+1]=0; /* Trims blanks at end of line */
9023: if(line[0]=='#'){
9024: fprintf(ficlog,"Comment line\n%s\n",line);
9025: printf("Comment line\n%s\n",line);
9026: continue;
9027: }
9028: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9029: strcpy(line, linetmp);
1.223 brouard 9030:
9031: /* Loops on waves */
9032: for (j=maxwav;j>=1;j--){
9033: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9034: cutv(stra, strb, line, ' ');
9035: if(strb[0]=='.') { /* Missing value */
9036: lval=-1;
9037: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9038: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9039: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9040: 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);
9041: 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);
9042: return 1;
9043: }
9044: }else{
9045: errno=0;
9046: /* what_kind_of_number(strb); */
9047: dval=strtod(strb,&endptr);
9048: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9049: /* if(strb != endptr && *endptr == '\0') */
9050: /* dval=dlval; */
9051: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9052: if( strb[0]=='\0' || (*endptr != '\0')){
9053: 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);
9054: 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);
9055: return 1;
9056: }
9057: cotqvar[j][iv][i]=dval;
9058: cotvar[j][ntv+iv][i]=dval;
9059: }
9060: strcpy(line,stra);
1.223 brouard 9061: }/* end loop ntqv */
1.225 brouard 9062:
1.223 brouard 9063: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9064: cutv(stra, strb, line, ' ');
9065: if(strb[0]=='.') { /* Missing value */
9066: lval=-1;
9067: }else{
9068: errno=0;
9069: lval=strtol(strb,&endptr,10);
9070: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9071: if( strb[0]=='\0' || (*endptr != '\0')){
9072: 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);
9073: 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);
9074: return 1;
9075: }
9076: }
9077: if(lval <-1 || lval >1){
9078: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9079: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9080: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9081: For example, for multinomial values like 1, 2 and 3,\n \
9082: build V1=0 V2=0 for the reference value (1),\n \
9083: V1=1 V2=0 for (2) \n \
1.223 brouard 9084: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9085: output of IMaCh is often meaningless.\n \
1.223 brouard 9086: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9087: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9088: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9089: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9090: For example, for multinomial values like 1, 2 and 3,\n \
9091: build V1=0 V2=0 for the reference value (1),\n \
9092: V1=1 V2=0 for (2) \n \
1.223 brouard 9093: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9094: output of IMaCh is often meaningless.\n \
1.223 brouard 9095: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9096: return 1;
9097: }
9098: cotvar[j][iv][i]=(double)(lval);
9099: strcpy(line,stra);
1.223 brouard 9100: }/* end loop ntv */
1.225 brouard 9101:
1.223 brouard 9102: /* Statuses at wave */
1.137 brouard 9103: cutv(stra, strb, line, ' ');
1.223 brouard 9104: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9105: lval=-1;
1.136 brouard 9106: }else{
1.238 brouard 9107: errno=0;
9108: lval=strtol(strb,&endptr,10);
9109: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9110: if( strb[0]=='\0' || (*endptr != '\0')){
9111: 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);
9112: 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);
9113: return 1;
9114: }
1.136 brouard 9115: }
1.225 brouard 9116:
1.136 brouard 9117: s[j][i]=lval;
1.225 brouard 9118:
1.223 brouard 9119: /* Date of Interview */
1.136 brouard 9120: strcpy(line,stra);
9121: cutv(stra, strb,line,' ');
1.169 brouard 9122: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9123: }
1.169 brouard 9124: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9125: month=99;
9126: year=9999;
1.136 brouard 9127: }else{
1.225 brouard 9128: 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);
9129: 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);
9130: return 1;
1.136 brouard 9131: }
9132: anint[j][i]= (double) year;
9133: mint[j][i]= (double)month;
9134: strcpy(line,stra);
1.223 brouard 9135: } /* End loop on waves */
1.225 brouard 9136:
1.223 brouard 9137: /* Date of death */
1.136 brouard 9138: cutv(stra, strb,line,' ');
1.169 brouard 9139: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9140: }
1.169 brouard 9141: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9142: month=99;
9143: year=9999;
9144: }else{
1.141 brouard 9145: 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 9146: 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);
9147: return 1;
1.136 brouard 9148: }
9149: andc[i]=(double) year;
9150: moisdc[i]=(double) month;
9151: strcpy(line,stra);
9152:
1.223 brouard 9153: /* Date of birth */
1.136 brouard 9154: cutv(stra, strb,line,' ');
1.169 brouard 9155: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9156: }
1.169 brouard 9157: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9158: month=99;
9159: year=9999;
9160: }else{
1.141 brouard 9161: 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);
9162: 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 9163: return 1;
1.136 brouard 9164: }
9165: if (year==9999) {
1.141 brouard 9166: 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);
9167: 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 9168: return 1;
9169:
1.136 brouard 9170: }
9171: annais[i]=(double)(year);
9172: moisnais[i]=(double)(month);
9173: strcpy(line,stra);
1.225 brouard 9174:
1.223 brouard 9175: /* Sample weight */
1.136 brouard 9176: cutv(stra, strb,line,' ');
9177: errno=0;
9178: dval=strtod(strb,&endptr);
9179: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9180: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9181: 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 9182: fflush(ficlog);
9183: return 1;
9184: }
9185: weight[i]=dval;
9186: strcpy(line,stra);
1.225 brouard 9187:
1.223 brouard 9188: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9189: cutv(stra, strb, line, ' ');
9190: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9191: lval=-1;
1.223 brouard 9192: }else{
1.225 brouard 9193: errno=0;
9194: /* what_kind_of_number(strb); */
9195: dval=strtod(strb,&endptr);
9196: /* if(strb != endptr && *endptr == '\0') */
9197: /* dval=dlval; */
9198: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9199: if( strb[0]=='\0' || (*endptr != '\0')){
9200: 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);
9201: 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);
9202: return 1;
9203: }
9204: coqvar[iv][i]=dval;
1.226 brouard 9205: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9206: }
9207: strcpy(line,stra);
9208: }/* end loop nqv */
1.136 brouard 9209:
1.223 brouard 9210: /* Covariate values */
1.136 brouard 9211: for (j=ncovcol;j>=1;j--){
9212: cutv(stra, strb,line,' ');
1.223 brouard 9213: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9214: lval=-1;
1.136 brouard 9215: }else{
1.225 brouard 9216: errno=0;
9217: lval=strtol(strb,&endptr,10);
9218: if( strb[0]=='\0' || (*endptr != '\0')){
9219: 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);
9220: 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);
9221: return 1;
9222: }
1.136 brouard 9223: }
9224: if(lval <-1 || lval >1){
1.225 brouard 9225: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9226: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9227: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9228: For example, for multinomial values like 1, 2 and 3,\n \
9229: build V1=0 V2=0 for the reference value (1),\n \
9230: V1=1 V2=0 for (2) \n \
1.136 brouard 9231: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9232: output of IMaCh is often meaningless.\n \
1.136 brouard 9233: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9234: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9235: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9236: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9237: For example, for multinomial values like 1, 2 and 3,\n \
9238: build V1=0 V2=0 for the reference value (1),\n \
9239: V1=1 V2=0 for (2) \n \
1.136 brouard 9240: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9241: output of IMaCh is often meaningless.\n \
1.136 brouard 9242: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9243: return 1;
1.136 brouard 9244: }
9245: covar[j][i]=(double)(lval);
9246: strcpy(line,stra);
9247: }
9248: lstra=strlen(stra);
1.225 brouard 9249:
1.136 brouard 9250: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9251: stratrunc = &(stra[lstra-9]);
9252: num[i]=atol(stratrunc);
9253: }
9254: else
9255: num[i]=atol(stra);
9256: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9257: 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;}*/
9258:
9259: i=i+1;
9260: } /* End loop reading data */
1.225 brouard 9261:
1.136 brouard 9262: *imax=i-1; /* Number of individuals */
9263: fclose(fic);
1.225 brouard 9264:
1.136 brouard 9265: return (0);
1.164 brouard 9266: /* endread: */
1.225 brouard 9267: printf("Exiting readdata: ");
9268: fclose(fic);
9269: return (1);
1.223 brouard 9270: }
1.126 brouard 9271:
1.234 brouard 9272: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9273: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9274: while (*p2 == ' ')
1.234 brouard 9275: p2++;
9276: /* while ((*p1++ = *p2++) !=0) */
9277: /* ; */
9278: /* do */
9279: /* while (*p2 == ' ') */
9280: /* p2++; */
9281: /* while (*p1++ == *p2++); */
9282: *stri=p2;
1.145 brouard 9283: }
9284:
1.235 brouard 9285: int decoderesult ( char resultline[], int nres)
1.230 brouard 9286: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9287: {
1.235 brouard 9288: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9289: char resultsav[MAXLINE];
1.234 brouard 9290: int resultmodel[MAXLINE];
9291: int modelresult[MAXLINE];
1.230 brouard 9292: char stra[80], strb[80], strc[80], strd[80],stre[80];
9293:
1.234 brouard 9294: removefirstspace(&resultline);
1.233 brouard 9295: printf("decoderesult:%s\n",resultline);
1.230 brouard 9296:
9297: if (strstr(resultline,"v") !=0){
9298: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9299: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9300: return 1;
9301: }
9302: trimbb(resultsav, resultline);
9303: if (strlen(resultsav) >1){
9304: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9305: }
1.253 brouard 9306: if(j == 0){ /* Resultline but no = */
9307: TKresult[nres]=0; /* Combination for the nresult and the model */
9308: return (0);
9309: }
9310:
1.234 brouard 9311: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9312: 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);
9313: 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);
9314: }
9315: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9316: if(nbocc(resultsav,'=') >1){
9317: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9318: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9319: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9320: }else
9321: cutl(strc,strd,resultsav,'=');
1.230 brouard 9322: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9323:
1.230 brouard 9324: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9325: Tvarsel[k]=atoi(strc);
9326: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9327: /* cptcovsel++; */
9328: if (nbocc(stra,'=') >0)
9329: strcpy(resultsav,stra); /* and analyzes it */
9330: }
1.235 brouard 9331: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9332: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9333: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9334: match=0;
1.236 brouard 9335: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9336: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9337: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9338: match=1;
9339: break;
9340: }
9341: }
9342: if(match == 0){
9343: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9344: }
9345: }
9346: }
1.235 brouard 9347: /* Checking for missing or useless values in comparison of current model needs */
9348: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9349: match=0;
1.235 brouard 9350: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9351: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9352: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9353: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9354: ++match;
9355: }
9356: }
9357: }
9358: if(match == 0){
9359: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9360: }else if(match > 1){
9361: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9362: }
9363: }
1.235 brouard 9364:
1.234 brouard 9365: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9366: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9367: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9368: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9369: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9370: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9371: /* 1 0 0 0 */
9372: /* 2 1 0 0 */
9373: /* 3 0 1 0 */
9374: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9375: /* 5 0 0 1 */
9376: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9377: /* 7 0 1 1 */
9378: /* 8 1 1 1 */
1.237 brouard 9379: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9380: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9381: /* V5*age V5 known which value for nres? */
9382: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9383: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9384: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9385: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9386: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9387: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9388: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9389: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9390: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9391: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9392: k4++;;
9393: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9394: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9395: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9396: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9397: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9398: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9399: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9400: k4q++;;
9401: }
9402: }
1.234 brouard 9403:
1.235 brouard 9404: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9405: return (0);
9406: }
1.235 brouard 9407:
1.230 brouard 9408: int decodemodel( char model[], int lastobs)
9409: /**< This routine decodes the model and returns:
1.224 brouard 9410: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9411: * - nagesqr = 1 if age*age in the model, otherwise 0.
9412: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9413: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9414: * - cptcovage number of covariates with age*products =2
9415: * - cptcovs number of simple covariates
9416: * - 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
9417: * which is a new column after the 9 (ncovcol) variables.
9418: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9419: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9420: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9421: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9422: */
1.136 brouard 9423: {
1.238 brouard 9424: int i, j, k, ks, v;
1.227 brouard 9425: int j1, k1, k2, k3, k4;
1.136 brouard 9426: char modelsav[80];
1.145 brouard 9427: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9428: char *strpt;
1.136 brouard 9429:
1.145 brouard 9430: /*removespace(model);*/
1.136 brouard 9431: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9432: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9433: if (strstr(model,"AGE") !=0){
1.192 brouard 9434: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9435: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9436: return 1;
9437: }
1.141 brouard 9438: if (strstr(model,"v") !=0){
9439: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9440: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9441: return 1;
9442: }
1.187 brouard 9443: strcpy(modelsav,model);
9444: if ((strpt=strstr(model,"age*age")) !=0){
9445: printf(" strpt=%s, model=%s\n",strpt, model);
9446: if(strpt != model){
1.234 brouard 9447: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9448: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9449: corresponding column of parameters.\n",model);
1.234 brouard 9450: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9451: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9452: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9453: return 1;
1.225 brouard 9454: }
1.187 brouard 9455: nagesqr=1;
9456: if (strstr(model,"+age*age") !=0)
1.234 brouard 9457: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9458: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9459: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9460: else
1.234 brouard 9461: substrchaine(modelsav, model, "age*age");
1.187 brouard 9462: }else
9463: nagesqr=0;
9464: if (strlen(modelsav) >1){
9465: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9466: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9467: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9468: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9469: * cst, age and age*age
9470: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9471: /* including age products which are counted in cptcovage.
9472: * but the covariates which are products must be treated
9473: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9474: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9475: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9476:
9477:
1.187 brouard 9478: /* Design
9479: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9480: * < ncovcol=8 >
9481: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9482: * k= 1 2 3 4 5 6 7 8
9483: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9484: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9485: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9486: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9487: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9488: * Tage[++cptcovage]=k
9489: * if products, new covar are created after ncovcol with k1
9490: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9491: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9492: * 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
9493: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9494: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9495: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9496: * < ncovcol=8 >
9497: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9498: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9499: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9500: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9501: * p Tprod[1]@2={ 6, 5}
9502: *p Tvard[1][1]@4= {7, 8, 5, 6}
9503: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9504: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9505: *How to reorganize?
9506: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9507: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9508: * {2, 1, 4, 8, 5, 6, 3, 7}
9509: * Struct []
9510: */
1.225 brouard 9511:
1.187 brouard 9512: /* This loop fills the array Tvar from the string 'model'.*/
9513: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9514: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9515: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9516: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9517: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9518: /* k=1 Tvar[1]=2 (from V2) */
9519: /* k=5 Tvar[5] */
9520: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9521: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9522: /* } */
1.198 brouard 9523: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9524: /*
9525: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9526: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9527: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9528: }
1.187 brouard 9529: cptcovage=0;
9530: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9531: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9532: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9533: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9534: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9535: /*scanf("%d",i);*/
9536: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9537: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9538: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9539: /* covar is not filled and then is empty */
9540: cptcovprod--;
9541: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9542: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9543: Typevar[k]=1; /* 1 for age product */
9544: cptcovage++; /* Sums the number of covariates which include age as a product */
9545: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9546: /*printf("stre=%s ", stre);*/
9547: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9548: cptcovprod--;
9549: cutl(stre,strb,strc,'V');
9550: Tvar[k]=atoi(stre);
9551: Typevar[k]=1; /* 1 for age product */
9552: cptcovage++;
9553: Tage[cptcovage]=k;
9554: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9555: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9556: cptcovn++;
9557: cptcovprodnoage++;k1++;
9558: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9559: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9560: because this model-covariate is a construction we invent a new column
9561: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9562: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9563: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9564: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9565: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9566: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9567: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9568: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9569: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9570: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9571: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9572: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9573: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9574: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9575: for (i=1; i<=lastobs;i++){
9576: /* Computes the new covariate which is a product of
9577: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9578: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9579: }
9580: } /* End age is not in the model */
9581: } /* End if model includes a product */
9582: else { /* no more sum */
9583: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9584: /* scanf("%d",i);*/
9585: cutl(strd,strc,strb,'V');
9586: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9587: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9588: Tvar[k]=atoi(strd);
9589: Typevar[k]=0; /* 0 for simple covariates */
9590: }
9591: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9592: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9593: scanf("%d",i);*/
1.187 brouard 9594: } /* end of loop + on total covariates */
9595: } /* end if strlen(modelsave == 0) age*age might exist */
9596: } /* end if strlen(model == 0) */
1.136 brouard 9597:
9598: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9599: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9600:
1.136 brouard 9601: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9602: printf("cptcovprod=%d ", cptcovprod);
9603: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9604: scanf("%d ",i);*/
9605:
9606:
1.230 brouard 9607: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9608: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9609: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9610: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9611: k = 1 2 3 4 5 6 7 8 9
9612: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9613: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9614: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9615: Dummy[k] 1 0 0 0 3 1 1 2 3
9616: Tmodelind[combination of covar]=k;
1.225 brouard 9617: */
9618: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9619: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9620: /* 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 9621: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9622: printf("Model=%s\n\
9623: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9624: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9625: 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);
9626: fprintf(ficlog,"Model=%s\n\
9627: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9628: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9629: 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 9630: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9631: 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 */
9632: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9633: Fixed[k]= 0;
9634: Dummy[k]= 0;
1.225 brouard 9635: ncoveff++;
1.232 brouard 9636: ncovf++;
1.234 brouard 9637: nsd++;
9638: modell[k].maintype= FTYPE;
9639: TvarsD[nsd]=Tvar[k];
9640: TvarsDind[nsd]=k;
9641: TvarF[ncovf]=Tvar[k];
9642: TvarFind[ncovf]=k;
9643: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9644: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9645: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9646: Fixed[k]= 0;
9647: Dummy[k]= 0;
9648: ncoveff++;
9649: ncovf++;
9650: modell[k].maintype= FTYPE;
9651: TvarF[ncovf]=Tvar[k];
9652: TvarFind[ncovf]=k;
1.230 brouard 9653: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9654: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9655: }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 9656: Fixed[k]= 0;
9657: Dummy[k]= 1;
1.230 brouard 9658: nqfveff++;
1.234 brouard 9659: modell[k].maintype= FTYPE;
9660: modell[k].subtype= FQ;
9661: nsq++;
9662: TvarsQ[nsq]=Tvar[k];
9663: TvarsQind[nsq]=k;
1.232 brouard 9664: ncovf++;
1.234 brouard 9665: TvarF[ncovf]=Tvar[k];
9666: TvarFind[ncovf]=k;
1.231 brouard 9667: 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 9668: 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 9669: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9670: Fixed[k]= 1;
9671: Dummy[k]= 0;
1.225 brouard 9672: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9673: modell[k].maintype= VTYPE;
9674: modell[k].subtype= VD;
9675: nsd++;
9676: TvarsD[nsd]=Tvar[k];
9677: TvarsDind[nsd]=k;
9678: ncovv++; /* Only simple time varying variables */
9679: TvarV[ncovv]=Tvar[k];
1.242 brouard 9680: 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 9681: 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 */
9682: 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 9683: 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);
9684: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9685: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9686: Fixed[k]= 1;
9687: Dummy[k]= 1;
9688: nqtveff++;
9689: modell[k].maintype= VTYPE;
9690: modell[k].subtype= VQ;
9691: ncovv++; /* Only simple time varying variables */
9692: nsq++;
9693: TvarsQ[nsq]=Tvar[k];
9694: TvarsQind[nsq]=k;
9695: TvarV[ncovv]=Tvar[k];
1.242 brouard 9696: 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 9697: 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 */
9698: 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 9699: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9700: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9701: 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 9702: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9703: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9704: ncova++;
9705: TvarA[ncova]=Tvar[k];
9706: TvarAind[ncova]=k;
1.231 brouard 9707: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9708: Fixed[k]= 2;
9709: Dummy[k]= 2;
9710: modell[k].maintype= ATYPE;
9711: modell[k].subtype= APFD;
9712: /* ncoveff++; */
1.227 brouard 9713: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9714: Fixed[k]= 2;
9715: Dummy[k]= 3;
9716: modell[k].maintype= ATYPE;
9717: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9718: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9719: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9720: Fixed[k]= 3;
9721: Dummy[k]= 2;
9722: modell[k].maintype= ATYPE;
9723: modell[k].subtype= APVD; /* Product age * varying dummy */
9724: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9725: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9726: Fixed[k]= 3;
9727: Dummy[k]= 3;
9728: modell[k].maintype= ATYPE;
9729: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9730: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9731: }
9732: }else if (Typevar[k] == 2) { /* product without age */
9733: k1=Tposprod[k];
9734: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9735: if(Tvard[k1][2] <=ncovcol){
9736: Fixed[k]= 1;
9737: Dummy[k]= 0;
9738: modell[k].maintype= FTYPE;
9739: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9740: ncovf++; /* Fixed variables without age */
9741: TvarF[ncovf]=Tvar[k];
9742: TvarFind[ncovf]=k;
9743: }else if(Tvard[k1][2] <=ncovcol+nqv){
9744: Fixed[k]= 0; /* or 2 ?*/
9745: Dummy[k]= 1;
9746: modell[k].maintype= FTYPE;
9747: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9748: ncovf++; /* Varying variables without age */
9749: TvarF[ncovf]=Tvar[k];
9750: TvarFind[ncovf]=k;
9751: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9752: Fixed[k]= 1;
9753: Dummy[k]= 0;
9754: modell[k].maintype= VTYPE;
9755: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9756: ncovv++; /* Varying variables without age */
9757: TvarV[ncovv]=Tvar[k];
9758: TvarVind[ncovv]=k;
9759: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9760: Fixed[k]= 1;
9761: Dummy[k]= 1;
9762: modell[k].maintype= VTYPE;
9763: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9764: ncovv++; /* Varying variables without age */
9765: TvarV[ncovv]=Tvar[k];
9766: TvarVind[ncovv]=k;
9767: }
1.227 brouard 9768: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9769: if(Tvard[k1][2] <=ncovcol){
9770: Fixed[k]= 0; /* or 2 ?*/
9771: Dummy[k]= 1;
9772: modell[k].maintype= FTYPE;
9773: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9774: ncovf++; /* Fixed variables without age */
9775: TvarF[ncovf]=Tvar[k];
9776: TvarFind[ncovf]=k;
9777: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9778: Fixed[k]= 1;
9779: Dummy[k]= 1;
9780: modell[k].maintype= VTYPE;
9781: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9782: ncovv++; /* Varying variables without age */
9783: TvarV[ncovv]=Tvar[k];
9784: TvarVind[ncovv]=k;
9785: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9786: Fixed[k]= 1;
9787: Dummy[k]= 1;
9788: modell[k].maintype= VTYPE;
9789: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9790: ncovv++; /* Varying variables without age */
9791: TvarV[ncovv]=Tvar[k];
9792: TvarVind[ncovv]=k;
9793: ncovv++; /* Varying variables without age */
9794: TvarV[ncovv]=Tvar[k];
9795: TvarVind[ncovv]=k;
9796: }
1.227 brouard 9797: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9798: if(Tvard[k1][2] <=ncovcol){
9799: Fixed[k]= 1;
9800: Dummy[k]= 1;
9801: modell[k].maintype= VTYPE;
9802: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9803: ncovv++; /* Varying variables without age */
9804: TvarV[ncovv]=Tvar[k];
9805: TvarVind[ncovv]=k;
9806: }else if(Tvard[k1][2] <=ncovcol+nqv){
9807: Fixed[k]= 1;
9808: Dummy[k]= 1;
9809: modell[k].maintype= VTYPE;
9810: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9811: ncovv++; /* Varying variables without age */
9812: TvarV[ncovv]=Tvar[k];
9813: TvarVind[ncovv]=k;
9814: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9815: Fixed[k]= 1;
9816: Dummy[k]= 0;
9817: modell[k].maintype= VTYPE;
9818: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9819: ncovv++; /* Varying variables without age */
9820: TvarV[ncovv]=Tvar[k];
9821: TvarVind[ncovv]=k;
9822: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9823: Fixed[k]= 1;
9824: Dummy[k]= 1;
9825: modell[k].maintype= VTYPE;
9826: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9827: ncovv++; /* Varying variables without age */
9828: TvarV[ncovv]=Tvar[k];
9829: TvarVind[ncovv]=k;
9830: }
1.227 brouard 9831: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9832: if(Tvard[k1][2] <=ncovcol){
9833: Fixed[k]= 1;
9834: Dummy[k]= 1;
9835: modell[k].maintype= VTYPE;
9836: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9837: ncovv++; /* Varying variables without age */
9838: TvarV[ncovv]=Tvar[k];
9839: TvarVind[ncovv]=k;
9840: }else if(Tvard[k1][2] <=ncovcol+nqv){
9841: Fixed[k]= 1;
9842: Dummy[k]= 1;
9843: modell[k].maintype= VTYPE;
9844: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9845: ncovv++; /* Varying variables without age */
9846: TvarV[ncovv]=Tvar[k];
9847: TvarVind[ncovv]=k;
9848: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9849: Fixed[k]= 1;
9850: Dummy[k]= 1;
9851: modell[k].maintype= VTYPE;
9852: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9853: ncovv++; /* Varying variables without age */
9854: TvarV[ncovv]=Tvar[k];
9855: TvarVind[ncovv]=k;
9856: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9857: Fixed[k]= 1;
9858: Dummy[k]= 1;
9859: modell[k].maintype= VTYPE;
9860: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9861: ncovv++; /* Varying variables without age */
9862: TvarV[ncovv]=Tvar[k];
9863: TvarVind[ncovv]=k;
9864: }
1.227 brouard 9865: }else{
1.240 brouard 9866: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9867: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9868: } /*end k1*/
1.225 brouard 9869: }else{
1.226 brouard 9870: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9871: 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 9872: }
1.227 brouard 9873: 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 9874: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9875: 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]);
9876: }
9877: /* Searching for doublons in the model */
9878: for(k1=1; k1<= cptcovt;k1++){
9879: for(k2=1; k2 <k1;k2++){
9880: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9881: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9882: if(Tvar[k1]==Tvar[k2]){
9883: 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]]);
9884: 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);
9885: return(1);
9886: }
9887: }else if (Typevar[k1] ==2){
9888: k3=Tposprod[k1];
9889: k4=Tposprod[k2];
9890: 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])) ){
9891: 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]]);
9892: 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);
9893: return(1);
9894: }
9895: }
1.227 brouard 9896: }
9897: }
1.225 brouard 9898: }
9899: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9900: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9901: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9902: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9903: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9904: /*endread:*/
1.225 brouard 9905: printf("Exiting decodemodel: ");
9906: return (1);
1.136 brouard 9907: }
9908:
1.169 brouard 9909: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9910: {/* Check ages at death */
1.136 brouard 9911: int i, m;
1.218 brouard 9912: int firstone=0;
9913:
1.136 brouard 9914: for (i=1; i<=imx; i++) {
9915: for(m=2; (m<= maxwav); m++) {
9916: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9917: anint[m][i]=9999;
1.216 brouard 9918: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9919: s[m][i]=-1;
1.136 brouard 9920: }
9921: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9922: *nberr = *nberr + 1;
1.218 brouard 9923: if(firstone == 0){
9924: firstone=1;
1.260 brouard 9925: 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 9926: }
1.262 brouard 9927: 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 9928: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9929: }
9930: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9931: (*nberr)++;
1.259 brouard 9932: 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 9933: 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 9934: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9935: }
9936: }
9937: }
9938:
9939: for (i=1; i<=imx; i++) {
9940: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9941: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9942: 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 9943: if (s[m][i] >= nlstate+1) {
1.169 brouard 9944: if(agedc[i]>0){
9945: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9946: agev[m][i]=agedc[i];
1.214 brouard 9947: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9948: }else {
1.136 brouard 9949: if ((int)andc[i]!=9999){
9950: nbwarn++;
9951: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9952: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9953: agev[m][i]=-1;
9954: }
9955: }
1.169 brouard 9956: } /* agedc > 0 */
1.214 brouard 9957: } /* end if */
1.136 brouard 9958: else if(s[m][i] !=9){ /* Standard case, age in fractional
9959: years but with the precision of a month */
9960: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9961: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9962: agev[m][i]=1;
9963: else if(agev[m][i] < *agemin){
9964: *agemin=agev[m][i];
9965: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9966: }
9967: else if(agev[m][i] >*agemax){
9968: *agemax=agev[m][i];
1.156 brouard 9969: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9970: }
9971: /*agev[m][i]=anint[m][i]-annais[i];*/
9972: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9973: } /* en if 9*/
1.136 brouard 9974: else { /* =9 */
1.214 brouard 9975: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9976: agev[m][i]=1;
9977: s[m][i]=-1;
9978: }
9979: }
1.214 brouard 9980: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9981: agev[m][i]=1;
1.214 brouard 9982: else{
9983: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9984: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9985: agev[m][i]=0;
9986: }
9987: } /* End for lastpass */
9988: }
1.136 brouard 9989:
9990: for (i=1; i<=imx; i++) {
9991: for(m=firstpass; (m<=lastpass); m++){
9992: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9993: (*nberr)++;
1.136 brouard 9994: 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);
9995: 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);
9996: return 1;
9997: }
9998: }
9999: }
10000:
10001: /*for (i=1; i<=imx; i++){
10002: for (m=firstpass; (m<lastpass); m++){
10003: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10004: }
10005:
10006: }*/
10007:
10008:
1.139 brouard 10009: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10010: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10011:
10012: return (0);
1.164 brouard 10013: /* endread:*/
1.136 brouard 10014: printf("Exiting calandcheckages: ");
10015: return (1);
10016: }
10017:
1.172 brouard 10018: #if defined(_MSC_VER)
10019: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10020: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10021: //#include "stdafx.h"
10022: //#include <stdio.h>
10023: //#include <tchar.h>
10024: //#include <windows.h>
10025: //#include <iostream>
10026: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10027:
10028: LPFN_ISWOW64PROCESS fnIsWow64Process;
10029:
10030: BOOL IsWow64()
10031: {
10032: BOOL bIsWow64 = FALSE;
10033:
10034: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10035: // (HANDLE, PBOOL);
10036:
10037: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10038:
10039: HMODULE module = GetModuleHandle(_T("kernel32"));
10040: const char funcName[] = "IsWow64Process";
10041: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10042: GetProcAddress(module, funcName);
10043:
10044: if (NULL != fnIsWow64Process)
10045: {
10046: if (!fnIsWow64Process(GetCurrentProcess(),
10047: &bIsWow64))
10048: //throw std::exception("Unknown error");
10049: printf("Unknown error\n");
10050: }
10051: return bIsWow64 != FALSE;
10052: }
10053: #endif
1.177 brouard 10054:
1.191 brouard 10055: void syscompilerinfo(int logged)
1.167 brouard 10056: {
10057: /* #include "syscompilerinfo.h"*/
1.185 brouard 10058: /* command line Intel compiler 32bit windows, XP compatible:*/
10059: /* /GS /W3 /Gy
10060: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10061: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10062: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10063: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10064: */
10065: /* 64 bits */
1.185 brouard 10066: /*
10067: /GS /W3 /Gy
10068: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10069: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10070: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10071: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10072: /* Optimization are useless and O3 is slower than O2 */
10073: /*
10074: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10075: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10076: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10077: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10078: */
1.186 brouard 10079: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10080: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10081: /PDB:"visual studio
10082: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10083: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10084: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10085: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10086: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10087: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10088: uiAccess='false'"
10089: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10090: /NOLOGO /TLBID:1
10091: */
1.177 brouard 10092: #if defined __INTEL_COMPILER
1.178 brouard 10093: #if defined(__GNUC__)
10094: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10095: #endif
1.177 brouard 10096: #elif defined(__GNUC__)
1.179 brouard 10097: #ifndef __APPLE__
1.174 brouard 10098: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10099: #endif
1.177 brouard 10100: struct utsname sysInfo;
1.178 brouard 10101: int cross = CROSS;
10102: if (cross){
10103: printf("Cross-");
1.191 brouard 10104: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10105: }
1.174 brouard 10106: #endif
10107:
1.171 brouard 10108: #include <stdint.h>
1.178 brouard 10109:
1.191 brouard 10110: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10111: #if defined(__clang__)
1.191 brouard 10112: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10113: #endif
10114: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10115: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10116: #endif
10117: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10118: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10119: #endif
10120: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10121: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10122: #endif
10123: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10124: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10125: #endif
10126: #if defined(_MSC_VER)
1.191 brouard 10127: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10128: #endif
10129: #if defined(__PGI)
1.191 brouard 10130: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10131: #endif
10132: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10133: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10134: #endif
1.191 brouard 10135: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10136:
1.167 brouard 10137: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10138: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10139: // Windows (x64 and x86)
1.191 brouard 10140: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10141: #elif __unix__ // all unices, not all compilers
10142: // Unix
1.191 brouard 10143: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10144: #elif __linux__
10145: // linux
1.191 brouard 10146: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10147: #elif __APPLE__
1.174 brouard 10148: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10149: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10150: #endif
10151:
10152: /* __MINGW32__ */
10153: /* __CYGWIN__ */
10154: /* __MINGW64__ */
10155: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10156: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10157: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10158: /* _WIN64 // Defined for applications for Win64. */
10159: /* _M_X64 // Defined for compilations that target x64 processors. */
10160: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10161:
1.167 brouard 10162: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10163: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10164: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10165: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10166: #else
1.191 brouard 10167: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10168: #endif
10169:
1.169 brouard 10170: #if defined(__GNUC__)
10171: # if defined(__GNUC_PATCHLEVEL__)
10172: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10173: + __GNUC_MINOR__ * 100 \
10174: + __GNUC_PATCHLEVEL__)
10175: # else
10176: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10177: + __GNUC_MINOR__ * 100)
10178: # endif
1.174 brouard 10179: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10180: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10181:
10182: if (uname(&sysInfo) != -1) {
10183: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10184: 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 10185: }
10186: else
10187: perror("uname() error");
1.179 brouard 10188: //#ifndef __INTEL_COMPILER
10189: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10190: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10191: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10192: #endif
1.169 brouard 10193: #endif
1.172 brouard 10194:
10195: // void main()
10196: // {
1.169 brouard 10197: #if defined(_MSC_VER)
1.174 brouard 10198: if (IsWow64()){
1.191 brouard 10199: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10200: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10201: }
10202: else{
1.191 brouard 10203: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10204: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10205: }
1.172 brouard 10206: // printf("\nPress Enter to continue...");
10207: // getchar();
10208: // }
10209:
1.169 brouard 10210: #endif
10211:
1.167 brouard 10212:
1.219 brouard 10213: }
1.136 brouard 10214:
1.219 brouard 10215: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10216: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10217: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10218: /* double ftolpl = 1.e-10; */
1.180 brouard 10219: double age, agebase, agelim;
1.203 brouard 10220: double tot;
1.180 brouard 10221:
1.202 brouard 10222: strcpy(filerespl,"PL_");
10223: strcat(filerespl,fileresu);
10224: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10225: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10226: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10227: }
1.227 brouard 10228: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10229: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10230: pstamp(ficrespl);
1.203 brouard 10231: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10232: fprintf(ficrespl,"#Age ");
10233: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10234: fprintf(ficrespl,"\n");
1.180 brouard 10235:
1.219 brouard 10236: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10237:
1.219 brouard 10238: agebase=ageminpar;
10239: agelim=agemaxpar;
1.180 brouard 10240:
1.227 brouard 10241: /* i1=pow(2,ncoveff); */
1.234 brouard 10242: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10243: if (cptcovn < 1){i1=1;}
1.180 brouard 10244:
1.238 brouard 10245: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10246: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10247: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10248: continue;
1.235 brouard 10249:
1.238 brouard 10250: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10251: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10252: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10253: /* k=k+1; */
10254: /* to clean */
10255: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10256: fprintf(ficrespl,"#******");
10257: printf("#******");
10258: fprintf(ficlog,"#******");
10259: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10260: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10261: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10262: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10263: }
10264: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10265: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10266: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10267: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10268: }
10269: fprintf(ficrespl,"******\n");
10270: printf("******\n");
10271: fprintf(ficlog,"******\n");
10272: if(invalidvarcomb[k]){
10273: printf("\nCombination (%d) ignored because no case \n",k);
10274: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10275: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10276: continue;
10277: }
1.219 brouard 10278:
1.238 brouard 10279: fprintf(ficrespl,"#Age ");
10280: for(j=1;j<=cptcoveff;j++) {
10281: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10282: }
10283: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10284: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10285:
1.238 brouard 10286: for (age=agebase; age<=agelim; age++){
10287: /* for (age=agebase; age<=agebase; age++){ */
10288: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10289: fprintf(ficrespl,"%.0f ",age );
10290: for(j=1;j<=cptcoveff;j++)
10291: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10292: tot=0.;
10293: for(i=1; i<=nlstate;i++){
10294: tot += prlim[i][i];
10295: fprintf(ficrespl," %.5f", prlim[i][i]);
10296: }
10297: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10298: } /* Age */
10299: /* was end of cptcod */
10300: } /* cptcov */
10301: } /* nres */
1.219 brouard 10302: return 0;
1.180 brouard 10303: }
10304:
1.218 brouard 10305: 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){
10306: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10307:
10308: /* Computes the back prevalence limit for any combination of covariate values
10309: * at any age between ageminpar and agemaxpar
10310: */
1.235 brouard 10311: int i, j, k, i1, nres=0 ;
1.217 brouard 10312: /* double ftolpl = 1.e-10; */
10313: double age, agebase, agelim;
10314: double tot;
1.218 brouard 10315: /* double ***mobaverage; */
10316: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10317:
10318: strcpy(fileresplb,"PLB_");
10319: strcat(fileresplb,fileresu);
10320: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10321: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10322: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10323: }
10324: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10325: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10326: pstamp(ficresplb);
10327: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10328: fprintf(ficresplb,"#Age ");
10329: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10330: fprintf(ficresplb,"\n");
10331:
1.218 brouard 10332:
10333: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10334:
10335: agebase=ageminpar;
10336: agelim=agemaxpar;
10337:
10338:
1.227 brouard 10339: i1=pow(2,cptcoveff);
1.218 brouard 10340: if (cptcovn < 1){i1=1;}
1.227 brouard 10341:
1.238 brouard 10342: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10343: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10344: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10345: continue;
10346: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10347: fprintf(ficresplb,"#******");
10348: printf("#******");
10349: fprintf(ficlog,"#******");
10350: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10351: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10352: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10353: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10354: }
10355: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10356: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10357: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10358: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10359: }
10360: fprintf(ficresplb,"******\n");
10361: printf("******\n");
10362: fprintf(ficlog,"******\n");
10363: if(invalidvarcomb[k]){
10364: printf("\nCombination (%d) ignored because no cases \n",k);
10365: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10366: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10367: continue;
10368: }
1.218 brouard 10369:
1.238 brouard 10370: fprintf(ficresplb,"#Age ");
10371: for(j=1;j<=cptcoveff;j++) {
10372: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10373: }
10374: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10375: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10376:
10377:
1.238 brouard 10378: for (age=agebase; age<=agelim; age++){
10379: /* for (age=agebase; age<=agebase; age++){ */
10380: if(mobilavproj > 0){
10381: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10382: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10383: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10384: }else if (mobilavproj == 0){
10385: 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);
10386: 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);
10387: exit(1);
10388: }else{
10389: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10390: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10391: /* printf("TOTOT\n"); */
10392: /* exit(1); */
1.238 brouard 10393: }
10394: fprintf(ficresplb,"%.0f ",age );
10395: for(j=1;j<=cptcoveff;j++)
10396: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10397: tot=0.;
10398: for(i=1; i<=nlstate;i++){
10399: tot += bprlim[i][i];
10400: fprintf(ficresplb," %.5f", bprlim[i][i]);
10401: }
10402: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10403: } /* Age */
10404: /* was end of cptcod */
1.255 brouard 10405: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10406: } /* end of any combination */
10407: } /* end of nres */
1.218 brouard 10408: /* hBijx(p, bage, fage); */
10409: /* fclose(ficrespijb); */
10410:
10411: return 0;
1.217 brouard 10412: }
1.218 brouard 10413:
1.180 brouard 10414: int hPijx(double *p, int bage, int fage){
10415: /*------------- h Pij x at various ages ------------*/
10416:
10417: int stepsize;
10418: int agelim;
10419: int hstepm;
10420: int nhstepm;
1.235 brouard 10421: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10422:
10423: double agedeb;
10424: double ***p3mat;
10425:
1.201 brouard 10426: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10427: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10428: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10429: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10430: }
10431: printf("Computing pij: result on file '%s' \n", filerespij);
10432: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10433:
10434: stepsize=(int) (stepm+YEARM-1)/YEARM;
10435: /*if (stepm<=24) stepsize=2;*/
10436:
10437: agelim=AGESUP;
10438: hstepm=stepsize*YEARM; /* Every year of age */
10439: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10440:
1.180 brouard 10441: /* hstepm=1; aff par mois*/
10442: pstamp(ficrespij);
10443: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10444: i1= pow(2,cptcoveff);
1.218 brouard 10445: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10446: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10447: /* k=k+1; */
1.235 brouard 10448: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10449: for(k=1; k<=i1;k++){
1.253 brouard 10450: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10451: continue;
1.183 brouard 10452: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10453: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10454: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10455: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10456: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10457: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10458: }
1.183 brouard 10459: fprintf(ficrespij,"******\n");
10460:
10461: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10462: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10463: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10464:
10465: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10466:
1.183 brouard 10467: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10468: oldm=oldms;savm=savms;
1.235 brouard 10469: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10470: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10471: for(i=1; i<=nlstate;i++)
10472: for(j=1; j<=nlstate+ndeath;j++)
10473: fprintf(ficrespij," %1d-%1d",i,j);
10474: fprintf(ficrespij,"\n");
10475: for (h=0; h<=nhstepm; h++){
10476: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10477: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10478: for(i=1; i<=nlstate;i++)
10479: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10480: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10481: fprintf(ficrespij,"\n");
10482: }
1.183 brouard 10483: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10484: fprintf(ficrespij,"\n");
10485: }
1.180 brouard 10486: /*}*/
10487: }
1.218 brouard 10488: return 0;
1.180 brouard 10489: }
1.218 brouard 10490:
10491: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10492: /*------------- h Bij x at various ages ------------*/
10493:
10494: int stepsize;
1.218 brouard 10495: /* int agelim; */
10496: int ageminl;
1.217 brouard 10497: int hstepm;
10498: int nhstepm;
1.238 brouard 10499: int h, i, i1, j, k, nres;
1.218 brouard 10500:
1.217 brouard 10501: double agedeb;
10502: double ***p3mat;
1.218 brouard 10503:
10504: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10505: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10506: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10507: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10508: }
10509: printf("Computing pij back: result on file '%s' \n", filerespijb);
10510: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10511:
10512: stepsize=(int) (stepm+YEARM-1)/YEARM;
10513: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10514:
1.218 brouard 10515: /* agelim=AGESUP; */
10516: ageminl=30;
10517: hstepm=stepsize*YEARM; /* Every year of age */
10518: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10519:
10520: /* hstepm=1; aff par mois*/
10521: pstamp(ficrespijb);
1.255 brouard 10522: 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 10523: i1= pow(2,cptcoveff);
1.218 brouard 10524: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10525: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10526: /* k=k+1; */
1.238 brouard 10527: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10528: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10529: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10530: continue;
10531: fprintf(ficrespijb,"\n#****** ");
10532: for(j=1;j<=cptcoveff;j++)
10533: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10534: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10535: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10536: }
10537: fprintf(ficrespijb,"******\n");
1.264 brouard 10538: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10539: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10540: continue;
10541: }
10542:
10543: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10544: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10545: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10546: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10547: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10548:
10549: /* nhstepm=nhstepm*YEARM; aff par mois*/
10550:
1.266 brouard 10551: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10552: /* and memory limitations if stepm is small */
10553:
1.238 brouard 10554: /* oldm=oldms;savm=savms; */
10555: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10556: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10557: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10558: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10559: for(i=1; i<=nlstate;i++)
10560: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10561: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10562: fprintf(ficrespijb,"\n");
1.238 brouard 10563: for (h=0; h<=nhstepm; h++){
10564: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10565: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10566: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10567: for(i=1; i<=nlstate;i++)
10568: for(j=1; j<=nlstate+ndeath;j++)
10569: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10570: fprintf(ficrespijb,"\n");
10571: }
10572: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10573: fprintf(ficrespijb,"\n");
10574: } /* end age deb */
10575: } /* end combination */
10576: } /* end nres */
1.218 brouard 10577: return 0;
10578: } /* hBijx */
1.217 brouard 10579:
1.180 brouard 10580:
1.136 brouard 10581: /***********************************************/
10582: /**************** Main Program *****************/
10583: /***********************************************/
10584:
10585: int main(int argc, char *argv[])
10586: {
10587: #ifdef GSL
10588: const gsl_multimin_fminimizer_type *T;
10589: size_t iteri = 0, it;
10590: int rval = GSL_CONTINUE;
10591: int status = GSL_SUCCESS;
10592: double ssval;
10593: #endif
10594: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10595: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10596: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10597: int jj, ll, li, lj, lk;
1.136 brouard 10598: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10599: int num_filled;
1.136 brouard 10600: int itimes;
10601: int NDIM=2;
10602: int vpopbased=0;
1.235 brouard 10603: int nres=0;
1.258 brouard 10604: int endishere=0;
1.136 brouard 10605:
1.274 brouard 10606: int ncurrv=0; /* Temporary variable */
10607:
1.164 brouard 10608: char ca[32], cb[32];
1.136 brouard 10609: /* FILE *fichtm; *//* Html File */
10610: /* FILE *ficgp;*/ /*Gnuplot File */
10611: struct stat info;
1.191 brouard 10612: double agedeb=0.;
1.194 brouard 10613:
10614: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10615: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10616:
1.165 brouard 10617: double fret;
1.191 brouard 10618: double dum=0.; /* Dummy variable */
1.136 brouard 10619: double ***p3mat;
1.218 brouard 10620: /* double ***mobaverage; */
1.164 brouard 10621:
10622: char line[MAXLINE];
1.197 brouard 10623: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10624:
1.234 brouard 10625: char modeltemp[MAXLINE];
1.230 brouard 10626: char resultline[MAXLINE];
10627:
1.136 brouard 10628: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10629: char *tok, *val; /* pathtot */
1.136 brouard 10630: int firstobs=1, lastobs=10;
1.195 brouard 10631: int c, h , cpt, c2;
1.191 brouard 10632: int jl=0;
10633: int i1, j1, jk, stepsize=0;
1.194 brouard 10634: int count=0;
10635:
1.164 brouard 10636: int *tab;
1.136 brouard 10637: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10638: int backcast=0;
1.136 brouard 10639: int mobilav=0,popforecast=0;
1.191 brouard 10640: int hstepm=0, nhstepm=0;
1.136 brouard 10641: int agemortsup;
10642: float sumlpop=0.;
10643: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10644: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10645:
1.191 brouard 10646: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10647: double ftolpl=FTOL;
10648: double **prlim;
1.217 brouard 10649: double **bprlim;
1.136 brouard 10650: double ***param; /* Matrix of parameters */
1.251 brouard 10651: double ***paramstart; /* Matrix of starting parameter values */
10652: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10653: double **matcov; /* Matrix of covariance */
1.203 brouard 10654: double **hess; /* Hessian matrix */
1.136 brouard 10655: double ***delti3; /* Scale */
10656: double *delti; /* Scale */
10657: double ***eij, ***vareij;
10658: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10659:
1.136 brouard 10660: double *epj, vepp;
1.164 brouard 10661:
1.273 brouard 10662: double dateprev1, dateprev2;
10663: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10664: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10665:
1.136 brouard 10666: double **ximort;
1.145 brouard 10667: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10668: int *dcwave;
10669:
1.164 brouard 10670: char z[1]="c";
1.136 brouard 10671:
10672: /*char *strt;*/
10673: char strtend[80];
1.126 brouard 10674:
1.164 brouard 10675:
1.126 brouard 10676: /* setlocale (LC_ALL, ""); */
10677: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10678: /* textdomain (PACKAGE); */
10679: /* setlocale (LC_CTYPE, ""); */
10680: /* setlocale (LC_MESSAGES, ""); */
10681:
10682: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10683: rstart_time = time(NULL);
10684: /* (void) gettimeofday(&start_time,&tzp);*/
10685: start_time = *localtime(&rstart_time);
1.126 brouard 10686: curr_time=start_time;
1.157 brouard 10687: /*tml = *localtime(&start_time.tm_sec);*/
10688: /* strcpy(strstart,asctime(&tml)); */
10689: strcpy(strstart,asctime(&start_time));
1.126 brouard 10690:
10691: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10692: /* tp.tm_sec = tp.tm_sec +86400; */
10693: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10694: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10695: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10696: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10697: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10698: /* strt=asctime(&tmg); */
10699: /* printf("Time(after) =%s",strstart); */
10700: /* (void) time (&time_value);
10701: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10702: * tm = *localtime(&time_value);
10703: * strstart=asctime(&tm);
10704: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10705: */
10706:
10707: nberr=0; /* Number of errors and warnings */
10708: nbwarn=0;
1.184 brouard 10709: #ifdef WIN32
10710: _getcwd(pathcd, size);
10711: #else
1.126 brouard 10712: getcwd(pathcd, size);
1.184 brouard 10713: #endif
1.191 brouard 10714: syscompilerinfo(0);
1.196 brouard 10715: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10716: if(argc <=1){
10717: printf("\nEnter the parameter file name: ");
1.205 brouard 10718: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10719: printf("ERROR Empty parameter file name\n");
10720: goto end;
10721: }
1.126 brouard 10722: i=strlen(pathr);
10723: if(pathr[i-1]=='\n')
10724: pathr[i-1]='\0';
1.156 brouard 10725: i=strlen(pathr);
1.205 brouard 10726: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10727: pathr[i-1]='\0';
1.205 brouard 10728: }
10729: i=strlen(pathr);
10730: if( i==0 ){
10731: printf("ERROR Empty parameter file name\n");
10732: goto end;
10733: }
10734: for (tok = pathr; tok != NULL; ){
1.126 brouard 10735: printf("Pathr |%s|\n",pathr);
10736: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10737: printf("val= |%s| pathr=%s\n",val,pathr);
10738: strcpy (pathtot, val);
10739: if(pathr[0] == '\0') break; /* Dirty */
10740: }
10741: }
10742: else{
10743: strcpy(pathtot,argv[1]);
10744: }
10745: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10746: /*cygwin_split_path(pathtot,path,optionfile);
10747: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10748: /* cutv(path,optionfile,pathtot,'\\');*/
10749:
10750: /* Split argv[0], imach program to get pathimach */
10751: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10752: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10753: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10754: /* strcpy(pathimach,argv[0]); */
10755: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10756: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10757: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10758: #ifdef WIN32
10759: _chdir(path); /* Can be a relative path */
10760: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10761: #else
1.126 brouard 10762: chdir(path); /* Can be a relative path */
1.184 brouard 10763: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10764: #endif
10765: printf("Current directory %s!\n",pathcd);
1.126 brouard 10766: strcpy(command,"mkdir ");
10767: strcat(command,optionfilefiname);
10768: if((outcmd=system(command)) != 0){
1.169 brouard 10769: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10770: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10771: /* fclose(ficlog); */
10772: /* exit(1); */
10773: }
10774: /* if((imk=mkdir(optionfilefiname))<0){ */
10775: /* perror("mkdir"); */
10776: /* } */
10777:
10778: /*-------- arguments in the command line --------*/
10779:
1.186 brouard 10780: /* Main Log file */
1.126 brouard 10781: strcat(filelog, optionfilefiname);
10782: strcat(filelog,".log"); /* */
10783: if((ficlog=fopen(filelog,"w"))==NULL) {
10784: printf("Problem with logfile %s\n",filelog);
10785: goto end;
10786: }
10787: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10788: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10789: fprintf(ficlog,"\nEnter the parameter file name: \n");
10790: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10791: path=%s \n\
10792: optionfile=%s\n\
10793: optionfilext=%s\n\
1.156 brouard 10794: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10795:
1.197 brouard 10796: syscompilerinfo(1);
1.167 brouard 10797:
1.126 brouard 10798: printf("Local time (at start):%s",strstart);
10799: fprintf(ficlog,"Local time (at start): %s",strstart);
10800: fflush(ficlog);
10801: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10802: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10803:
10804: /* */
10805: strcpy(fileres,"r");
10806: strcat(fileres, optionfilefiname);
1.201 brouard 10807: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10808: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10809: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10810:
1.186 brouard 10811: /* Main ---------arguments file --------*/
1.126 brouard 10812:
10813: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10814: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10815: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10816: fflush(ficlog);
1.149 brouard 10817: /* goto end; */
10818: exit(70);
1.126 brouard 10819: }
10820:
10821:
10822:
10823: strcpy(filereso,"o");
1.201 brouard 10824: strcat(filereso,fileresu);
1.126 brouard 10825: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10826: printf("Problem with Output resultfile: %s\n", filereso);
10827: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10828: fflush(ficlog);
10829: goto end;
10830: }
10831:
10832: /* Reads comments: lines beginning with '#' */
10833: numlinepar=0;
1.197 brouard 10834:
10835: /* First parameter line */
10836: while(fgets(line, MAXLINE, ficpar)) {
10837: /* If line starts with a # it is a comment */
10838: if (line[0] == '#') {
10839: numlinepar++;
10840: fputs(line,stdout);
10841: fputs(line,ficparo);
10842: fputs(line,ficlog);
10843: continue;
10844: }else
10845: break;
10846: }
10847: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10848: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10849: if (num_filled != 5) {
10850: printf("Should be 5 parameters\n");
10851: }
1.126 brouard 10852: numlinepar++;
1.197 brouard 10853: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10854: }
10855: /* Second parameter line */
10856: while(fgets(line, MAXLINE, ficpar)) {
10857: /* If line starts with a # it is a comment */
10858: if (line[0] == '#') {
10859: numlinepar++;
10860: fputs(line,stdout);
10861: fputs(line,ficparo);
10862: fputs(line,ficlog);
10863: continue;
10864: }else
10865: break;
10866: }
1.223 brouard 10867: 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", \
10868: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10869: if (num_filled != 11) {
10870: 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 10871: printf("but line=%s\n",line);
1.197 brouard 10872: }
1.223 brouard 10873: 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 10874: }
1.203 brouard 10875: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10876: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10877: /* Third parameter line */
10878: while(fgets(line, MAXLINE, ficpar)) {
10879: /* If line starts with a # it is a comment */
10880: if (line[0] == '#') {
10881: numlinepar++;
10882: fputs(line,stdout);
10883: fputs(line,ficparo);
10884: fputs(line,ficlog);
10885: continue;
10886: }else
10887: break;
10888: }
1.201 brouard 10889: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 brouard 10890: if (num_filled == 0){
10891: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10892: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10893: model[0]='\0';
10894: goto end;
10895: } else if (num_filled != 1){
1.197 brouard 10896: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10897: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10898: model[0]='\0';
10899: goto end;
10900: }
10901: else{
10902: if (model[0]=='+'){
10903: for(i=1; i<=strlen(model);i++)
10904: modeltemp[i-1]=model[i];
1.201 brouard 10905: strcpy(model,modeltemp);
1.197 brouard 10906: }
10907: }
1.199 brouard 10908: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10909: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10910: }
10911: /* 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); */
10912: /* numlinepar=numlinepar+3; /\* In general *\/ */
10913: /* 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 10914: 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);
10915: 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 10916: fflush(ficlog);
1.190 brouard 10917: /* if(model[0]=='#'|| model[0]== '\0'){ */
10918: if(model[0]=='#'){
1.187 brouard 10919: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10920: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10921: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10922: if(mle != -1){
10923: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10924: exit(1);
10925: }
10926: }
1.126 brouard 10927: while((c=getc(ficpar))=='#' && c!= EOF){
10928: ungetc(c,ficpar);
10929: fgets(line, MAXLINE, ficpar);
10930: numlinepar++;
1.195 brouard 10931: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10932: z[0]=line[1];
10933: }
10934: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10935: fputs(line, stdout);
10936: //puts(line);
1.126 brouard 10937: fputs(line,ficparo);
10938: fputs(line,ficlog);
10939: }
10940: ungetc(c,ficpar);
10941:
10942:
1.145 brouard 10943: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 10944: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
10945: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
10946: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 10947: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10948: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10949: v1+v2*age+v2*v3 makes cptcovn = 3
10950: */
10951: if (strlen(model)>1)
1.187 brouard 10952: 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 10953: else
1.187 brouard 10954: ncovmodel=2; /* Constant and age */
1.133 brouard 10955: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10956: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10957: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10958: 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);
10959: 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);
10960: fflush(stdout);
10961: fclose (ficlog);
10962: goto end;
10963: }
1.126 brouard 10964: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10965: delti=delti3[1][1];
10966: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10967: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10968: /* We could also provide initial parameters values giving by simple logistic regression
10969: * only one way, that is without matrix product. We will have nlstate maximizations */
10970: /* for(i=1;i<nlstate;i++){ */
10971: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10972: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10973: /* } */
1.126 brouard 10974: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10975: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10976: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10977: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10978: fclose (ficparo);
10979: fclose (ficlog);
10980: goto end;
10981: exit(0);
1.220 brouard 10982: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10983: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10984: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10985: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10986: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10987: matcov=matrix(1,npar,1,npar);
1.203 brouard 10988: hess=matrix(1,npar,1,npar);
1.220 brouard 10989: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10990: /* Read guessed parameters */
1.126 brouard 10991: /* Reads comments: lines beginning with '#' */
10992: while((c=getc(ficpar))=='#' && c!= EOF){
10993: ungetc(c,ficpar);
10994: fgets(line, MAXLINE, ficpar);
10995: numlinepar++;
1.141 brouard 10996: fputs(line,stdout);
1.126 brouard 10997: fputs(line,ficparo);
10998: fputs(line,ficlog);
10999: }
11000: ungetc(c,ficpar);
11001:
11002: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11003: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11004: for(i=1; i <=nlstate; i++){
1.234 brouard 11005: j=0;
1.126 brouard 11006: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11007: if(jj==i) continue;
11008: j++;
11009: fscanf(ficpar,"%1d%1d",&i1,&j1);
11010: if ((i1 != i) || (j1 != jj)){
11011: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11012: It might be a problem of design; if ncovcol and the model are correct\n \
11013: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11014: exit(1);
11015: }
11016: fprintf(ficparo,"%1d%1d",i1,j1);
11017: if(mle==1)
11018: printf("%1d%1d",i,jj);
11019: fprintf(ficlog,"%1d%1d",i,jj);
11020: for(k=1; k<=ncovmodel;k++){
11021: fscanf(ficpar," %lf",¶m[i][j][k]);
11022: if(mle==1){
11023: printf(" %lf",param[i][j][k]);
11024: fprintf(ficlog," %lf",param[i][j][k]);
11025: }
11026: else
11027: fprintf(ficlog," %lf",param[i][j][k]);
11028: fprintf(ficparo," %lf",param[i][j][k]);
11029: }
11030: fscanf(ficpar,"\n");
11031: numlinepar++;
11032: if(mle==1)
11033: printf("\n");
11034: fprintf(ficlog,"\n");
11035: fprintf(ficparo,"\n");
1.126 brouard 11036: }
11037: }
11038: fflush(ficlog);
1.234 brouard 11039:
1.251 brouard 11040: /* Reads parameters values */
1.126 brouard 11041: p=param[1][1];
1.251 brouard 11042: pstart=paramstart[1][1];
1.126 brouard 11043:
11044: /* Reads comments: lines beginning with '#' */
11045: while((c=getc(ficpar))=='#' && c!= EOF){
11046: ungetc(c,ficpar);
11047: fgets(line, MAXLINE, ficpar);
11048: numlinepar++;
1.141 brouard 11049: fputs(line,stdout);
1.126 brouard 11050: fputs(line,ficparo);
11051: fputs(line,ficlog);
11052: }
11053: ungetc(c,ficpar);
11054:
11055: for(i=1; i <=nlstate; i++){
11056: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11057: fscanf(ficpar,"%1d%1d",&i1,&j1);
11058: if ( (i1-i) * (j1-j) != 0){
11059: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11060: exit(1);
11061: }
11062: printf("%1d%1d",i,j);
11063: fprintf(ficparo,"%1d%1d",i1,j1);
11064: fprintf(ficlog,"%1d%1d",i1,j1);
11065: for(k=1; k<=ncovmodel;k++){
11066: fscanf(ficpar,"%le",&delti3[i][j][k]);
11067: printf(" %le",delti3[i][j][k]);
11068: fprintf(ficparo," %le",delti3[i][j][k]);
11069: fprintf(ficlog," %le",delti3[i][j][k]);
11070: }
11071: fscanf(ficpar,"\n");
11072: numlinepar++;
11073: printf("\n");
11074: fprintf(ficparo,"\n");
11075: fprintf(ficlog,"\n");
1.126 brouard 11076: }
11077: }
11078: fflush(ficlog);
1.234 brouard 11079:
1.145 brouard 11080: /* Reads covariance matrix */
1.126 brouard 11081: delti=delti3[1][1];
1.220 brouard 11082:
11083:
1.126 brouard 11084: /* 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 11085:
1.126 brouard 11086: /* Reads comments: lines beginning with '#' */
11087: while((c=getc(ficpar))=='#' && c!= EOF){
11088: ungetc(c,ficpar);
11089: fgets(line, MAXLINE, ficpar);
11090: numlinepar++;
1.141 brouard 11091: fputs(line,stdout);
1.126 brouard 11092: fputs(line,ficparo);
11093: fputs(line,ficlog);
11094: }
11095: ungetc(c,ficpar);
1.220 brouard 11096:
1.126 brouard 11097: matcov=matrix(1,npar,1,npar);
1.203 brouard 11098: hess=matrix(1,npar,1,npar);
1.131 brouard 11099: for(i=1; i <=npar; i++)
11100: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11101:
1.194 brouard 11102: /* Scans npar lines */
1.126 brouard 11103: for(i=1; i <=npar; i++){
1.226 brouard 11104: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11105: if(count != 3){
1.226 brouard 11106: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11107: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11108: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11109: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11110: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11111: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11112: exit(1);
1.220 brouard 11113: }else{
1.226 brouard 11114: if(mle==1)
11115: printf("%1d%1d%d",i1,j1,jk);
11116: }
11117: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11118: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11119: for(j=1; j <=i; j++){
1.226 brouard 11120: fscanf(ficpar," %le",&matcov[i][j]);
11121: if(mle==1){
11122: printf(" %.5le",matcov[i][j]);
11123: }
11124: fprintf(ficlog," %.5le",matcov[i][j]);
11125: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11126: }
11127: fscanf(ficpar,"\n");
11128: numlinepar++;
11129: if(mle==1)
1.220 brouard 11130: printf("\n");
1.126 brouard 11131: fprintf(ficlog,"\n");
11132: fprintf(ficparo,"\n");
11133: }
1.194 brouard 11134: /* End of read covariance matrix npar lines */
1.126 brouard 11135: for(i=1; i <=npar; i++)
11136: for(j=i+1;j<=npar;j++)
1.226 brouard 11137: matcov[i][j]=matcov[j][i];
1.126 brouard 11138:
11139: if(mle==1)
11140: printf("\n");
11141: fprintf(ficlog,"\n");
11142:
11143: fflush(ficlog);
11144:
11145: /*-------- Rewriting parameter file ----------*/
11146: strcpy(rfileres,"r"); /* "Rparameterfile */
11147: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
11148: strcat(rfileres,"."); /* */
11149: strcat(rfileres,optionfilext); /* Other files have txt extension */
11150: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 11151: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11152: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 11153: }
11154: fprintf(ficres,"#%s\n",version);
11155: } /* End of mle != -3 */
1.218 brouard 11156:
1.186 brouard 11157: /* Main data
11158: */
1.126 brouard 11159: n= lastobs;
11160: num=lvector(1,n);
11161: moisnais=vector(1,n);
11162: annais=vector(1,n);
11163: moisdc=vector(1,n);
11164: andc=vector(1,n);
1.220 brouard 11165: weight=vector(1,n);
1.126 brouard 11166: agedc=vector(1,n);
11167: cod=ivector(1,n);
1.220 brouard 11168: for(i=1;i<=n;i++){
1.234 brouard 11169: num[i]=0;
11170: moisnais[i]=0;
11171: annais[i]=0;
11172: moisdc[i]=0;
11173: andc[i]=0;
11174: agedc[i]=0;
11175: cod[i]=0;
11176: weight[i]=1.0; /* Equal weights, 1 by default */
11177: }
1.126 brouard 11178: mint=matrix(1,maxwav,1,n);
11179: anint=matrix(1,maxwav,1,n);
1.131 brouard 11180: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11181: tab=ivector(1,NCOVMAX);
1.144 brouard 11182: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11183: 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 11184:
1.136 brouard 11185: /* Reads data from file datafile */
11186: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11187: goto end;
11188:
11189: /* Calculation of the number of parameters from char model */
1.234 brouard 11190: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11191: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11192: k=3 V4 Tvar[k=3]= 4 (from V4)
11193: k=2 V1 Tvar[k=2]= 1 (from V1)
11194: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11195: */
11196:
11197: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11198: TvarsDind=ivector(1,NCOVMAX); /* */
11199: TvarsD=ivector(1,NCOVMAX); /* */
11200: TvarsQind=ivector(1,NCOVMAX); /* */
11201: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11202: TvarF=ivector(1,NCOVMAX); /* */
11203: TvarFind=ivector(1,NCOVMAX); /* */
11204: TvarV=ivector(1,NCOVMAX); /* */
11205: TvarVind=ivector(1,NCOVMAX); /* */
11206: TvarA=ivector(1,NCOVMAX); /* */
11207: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11208: TvarFD=ivector(1,NCOVMAX); /* */
11209: TvarFDind=ivector(1,NCOVMAX); /* */
11210: TvarFQ=ivector(1,NCOVMAX); /* */
11211: TvarFQind=ivector(1,NCOVMAX); /* */
11212: TvarVD=ivector(1,NCOVMAX); /* */
11213: TvarVDind=ivector(1,NCOVMAX); /* */
11214: TvarVQ=ivector(1,NCOVMAX); /* */
11215: TvarVQind=ivector(1,NCOVMAX); /* */
11216:
1.230 brouard 11217: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11218: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11219: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11220: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11221: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11222: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11223: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11224: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11225: */
11226: /* For model-covariate k tells which data-covariate to use but
11227: because this model-covariate is a construction we invent a new column
11228: ncovcol + k1
11229: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11230: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11231: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11232: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11233: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11234: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11235: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11236: */
1.145 brouard 11237: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11238: 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 11239: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11240: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11241: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11242: 4 covariates (3 plus signs)
11243: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11244: */
1.230 brouard 11245: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11246: * individual dummy, fixed or varying:
11247: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11248: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11249: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11250: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11251: * Tmodelind[1]@9={9,0,3,2,}*/
11252: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11253: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11254: * individual quantitative, fixed or varying:
11255: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11256: * 3, 1, 0, 0, 0, 0, 0, 0},
11257: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11258: /* Main decodemodel */
11259:
1.187 brouard 11260:
1.223 brouard 11261: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11262: goto end;
11263:
1.137 brouard 11264: if((double)(lastobs-imx)/(double)imx > 1.10){
11265: nbwarn++;
11266: 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);
11267: 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);
11268: }
1.136 brouard 11269: /* if(mle==1){*/
1.137 brouard 11270: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11271: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11272: }
11273:
11274: /*-calculation of age at interview from date of interview and age at death -*/
11275: agev=matrix(1,maxwav,1,imx);
11276:
11277: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11278: goto end;
11279:
1.126 brouard 11280:
1.136 brouard 11281: agegomp=(int)agemin;
11282: free_vector(moisnais,1,n);
11283: free_vector(annais,1,n);
1.126 brouard 11284: /* free_matrix(mint,1,maxwav,1,n);
11285: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11286: /* free_vector(moisdc,1,n); */
11287: /* free_vector(andc,1,n); */
1.145 brouard 11288: /* */
11289:
1.126 brouard 11290: wav=ivector(1,imx);
1.214 brouard 11291: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11292: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11293: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11294: 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.*/
11295: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11296: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11297:
11298: /* Concatenates waves */
1.214 brouard 11299: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11300: Death is a valid wave (if date is known).
11301: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11302: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11303: and mw[mi+1][i]. dh depends on stepm.
11304: */
11305:
1.126 brouard 11306: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11307: /* Concatenates waves */
1.145 brouard 11308:
1.215 brouard 11309: free_vector(moisdc,1,n);
11310: free_vector(andc,1,n);
11311:
1.126 brouard 11312: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11313: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11314: ncodemax[1]=1;
1.145 brouard 11315: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11316: cptcoveff=0;
1.220 brouard 11317: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11318: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11319: }
11320:
11321: ncovcombmax=pow(2,cptcoveff);
11322: invalidvarcomb=ivector(1, ncovcombmax);
11323: for(i=1;i<ncovcombmax;i++)
11324: invalidvarcomb[i]=0;
11325:
1.211 brouard 11326: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11327: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11328: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11329:
1.200 brouard 11330: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11331: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11332: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11333: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11334: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11335: * (currently 0 or 1) in the data.
11336: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11337: * corresponding modality (h,j).
11338: */
11339:
1.145 brouard 11340: h=0;
11341: /*if (cptcovn > 0) */
1.126 brouard 11342: m=pow(2,cptcoveff);
11343:
1.144 brouard 11344: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11345: * For k=4 covariates, h goes from 1 to m=2**k
11346: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11347: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11348: * h\k 1 2 3 4
1.143 brouard 11349: *______________________________
11350: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11351: * 2 2 1 1 1
11352: * 3 i=2 1 2 1 1
11353: * 4 2 2 1 1
11354: * 5 i=3 1 i=2 1 2 1
11355: * 6 2 1 2 1
11356: * 7 i=4 1 2 2 1
11357: * 8 2 2 2 1
1.197 brouard 11358: * 9 i=5 1 i=3 1 i=2 1 2
11359: * 10 2 1 1 2
11360: * 11 i=6 1 2 1 2
11361: * 12 2 2 1 2
11362: * 13 i=7 1 i=4 1 2 2
11363: * 14 2 1 2 2
11364: * 15 i=8 1 2 2 2
11365: * 16 2 2 2 2
1.143 brouard 11366: */
1.212 brouard 11367: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11368: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11369: * and the value of each covariate?
11370: * V1=1, V2=1, V3=2, V4=1 ?
11371: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11372: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11373: * In order to get the real value in the data, we use nbcode
11374: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11375: * We are keeping this crazy system in order to be able (in the future?)
11376: * to have more than 2 values (0 or 1) for a covariate.
11377: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11378: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11379: * bbbbbbbb
11380: * 76543210
11381: * h-1 00000101 (6-1=5)
1.219 brouard 11382: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11383: * &
11384: * 1 00000001 (1)
1.219 brouard 11385: * 00000000 = 1 & ((h-1) >> (k-1))
11386: * +1= 00000001 =1
1.211 brouard 11387: *
11388: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11389: * h' 1101 =2^3+2^2+0x2^1+2^0
11390: * >>k' 11
11391: * & 00000001
11392: * = 00000001
11393: * +1 = 00000010=2 = codtabm(14,3)
11394: * Reverse h=6 and m=16?
11395: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11396: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11397: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11398: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11399: * V3=decodtabm(14,3,2**4)=2
11400: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11401: *(h-1) >> (j-1) 0011 =13 >> 2
11402: * &1 000000001
11403: * = 000000001
11404: * +1= 000000010 =2
11405: * 2211
11406: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11407: * V3=2
1.220 brouard 11408: * codtabm and decodtabm are identical
1.211 brouard 11409: */
11410:
1.145 brouard 11411:
11412: free_ivector(Ndum,-1,NCOVMAX);
11413:
11414:
1.126 brouard 11415:
1.186 brouard 11416: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11417: strcpy(optionfilegnuplot,optionfilefiname);
11418: if(mle==-3)
1.201 brouard 11419: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11420: strcat(optionfilegnuplot,".gp");
11421:
11422: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11423: printf("Problem with file %s",optionfilegnuplot);
11424: }
11425: else{
1.204 brouard 11426: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11427: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11428: //fprintf(ficgp,"set missing 'NaNq'\n");
11429: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11430: }
11431: /* fclose(ficgp);*/
1.186 brouard 11432:
11433:
11434: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11435:
11436: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11437: if(mle==-3)
1.201 brouard 11438: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11439: strcat(optionfilehtm,".htm");
11440: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11441: printf("Problem with %s \n",optionfilehtm);
11442: exit(0);
1.126 brouard 11443: }
11444:
11445: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11446: strcat(optionfilehtmcov,"-cov.htm");
11447: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11448: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11449: }
11450: else{
11451: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11452: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11453: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11454: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11455: }
11456:
1.213 brouard 11457: 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 11458: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11459: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11460: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11461: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11462: \n\
11463: <hr size=\"2\" color=\"#EC5E5E\">\
11464: <ul><li><h4>Parameter files</h4>\n\
11465: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11466: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11467: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11468: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11469: - Date and time at start: %s</ul>\n",\
11470: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11471: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11472: fileres,fileres,\
11473: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11474: fflush(fichtm);
11475:
11476: strcpy(pathr,path);
11477: strcat(pathr,optionfilefiname);
1.184 brouard 11478: #ifdef WIN32
11479: _chdir(optionfilefiname); /* Move to directory named optionfile */
11480: #else
1.126 brouard 11481: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11482: #endif
11483:
1.126 brouard 11484:
1.220 brouard 11485: /* Calculates basic frequencies. Computes observed prevalence at single age
11486: and for any valid combination of covariates
1.126 brouard 11487: and prints on file fileres'p'. */
1.251 brouard 11488: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11489: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11490:
11491: fprintf(fichtm,"\n");
1.274 brouard 11492: 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",\
11493: ftol, stepm);
11494: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11495: ncurrv=1;
11496: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11497: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11498: ncurrv=i;
11499: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
11500: fprintf(fichtm,"\n<li> Number of time varying (wave varying) covariates: ntv=%d ", ntv);
11501: ncurrv=i;
11502: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
11503: fprintf(fichtm,"\n<li>Number of quantitative time varying covariates: nqtv=%d ", nqtv);
11504: ncurrv=i;
11505: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11506: 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", \
11507: nlstate, ndeath, maxwav, mle, weightopt);
11508:
11509: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11510: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11511:
11512:
11513: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11514: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11515: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11516: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11517: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11518: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11519: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11520: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11521: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11522:
1.126 brouard 11523: /* For Powell, parameters are in a vector p[] starting at p[1]
11524: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11525: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11526:
11527: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11528: /* For mortality only */
1.126 brouard 11529: if (mle==-3){
1.136 brouard 11530: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11531: for(i=1;i<=NDIM;i++)
11532: for(j=1;j<=NDIM;j++)
11533: ximort[i][j]=0.;
1.186 brouard 11534: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11535: cens=ivector(1,n);
11536: ageexmed=vector(1,n);
11537: agecens=vector(1,n);
11538: dcwave=ivector(1,n);
1.223 brouard 11539:
1.126 brouard 11540: for (i=1; i<=imx; i++){
11541: dcwave[i]=-1;
11542: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11543: if (s[m][i]>nlstate) {
11544: dcwave[i]=m;
11545: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11546: break;
11547: }
1.126 brouard 11548: }
1.226 brouard 11549:
1.126 brouard 11550: for (i=1; i<=imx; i++) {
11551: if (wav[i]>0){
1.226 brouard 11552: ageexmed[i]=agev[mw[1][i]][i];
11553: j=wav[i];
11554: agecens[i]=1.;
11555:
11556: if (ageexmed[i]> 1 && wav[i] > 0){
11557: agecens[i]=agev[mw[j][i]][i];
11558: cens[i]= 1;
11559: }else if (ageexmed[i]< 1)
11560: cens[i]= -1;
11561: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11562: cens[i]=0 ;
1.126 brouard 11563: }
11564: else cens[i]=-1;
11565: }
11566:
11567: for (i=1;i<=NDIM;i++) {
11568: for (j=1;j<=NDIM;j++)
1.226 brouard 11569: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11570: }
11571:
1.145 brouard 11572: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11573: /*printf("%lf %lf", p[1], p[2]);*/
11574:
11575:
1.136 brouard 11576: #ifdef GSL
11577: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11578: #else
1.126 brouard 11579: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11580: #endif
1.201 brouard 11581: strcpy(filerespow,"POW-MORT_");
11582: strcat(filerespow,fileresu);
1.126 brouard 11583: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11584: printf("Problem with resultfile: %s\n", filerespow);
11585: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11586: }
1.136 brouard 11587: #ifdef GSL
11588: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11589: #else
1.126 brouard 11590: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11591: #endif
1.126 brouard 11592: /* for (i=1;i<=nlstate;i++)
11593: for(j=1;j<=nlstate+ndeath;j++)
11594: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11595: */
11596: fprintf(ficrespow,"\n");
1.136 brouard 11597: #ifdef GSL
11598: /* gsl starts here */
11599: T = gsl_multimin_fminimizer_nmsimplex;
11600: gsl_multimin_fminimizer *sfm = NULL;
11601: gsl_vector *ss, *x;
11602: gsl_multimin_function minex_func;
11603:
11604: /* Initial vertex size vector */
11605: ss = gsl_vector_alloc (NDIM);
11606:
11607: if (ss == NULL){
11608: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11609: }
11610: /* Set all step sizes to 1 */
11611: gsl_vector_set_all (ss, 0.001);
11612:
11613: /* Starting point */
1.126 brouard 11614:
1.136 brouard 11615: x = gsl_vector_alloc (NDIM);
11616:
11617: if (x == NULL){
11618: gsl_vector_free(ss);
11619: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11620: }
11621:
11622: /* Initialize method and iterate */
11623: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11624: /* gsl_vector_set(x, 0, 0.0268); */
11625: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11626: gsl_vector_set(x, 0, p[1]);
11627: gsl_vector_set(x, 1, p[2]);
11628:
11629: minex_func.f = &gompertz_f;
11630: minex_func.n = NDIM;
11631: minex_func.params = (void *)&p; /* ??? */
11632:
11633: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11634: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11635:
11636: printf("Iterations beginning .....\n\n");
11637: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11638:
11639: iteri=0;
11640: while (rval == GSL_CONTINUE){
11641: iteri++;
11642: status = gsl_multimin_fminimizer_iterate(sfm);
11643:
11644: if (status) printf("error: %s\n", gsl_strerror (status));
11645: fflush(0);
11646:
11647: if (status)
11648: break;
11649:
11650: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11651: ssval = gsl_multimin_fminimizer_size (sfm);
11652:
11653: if (rval == GSL_SUCCESS)
11654: printf ("converged to a local maximum at\n");
11655:
11656: printf("%5d ", iteri);
11657: for (it = 0; it < NDIM; it++){
11658: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11659: }
11660: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11661: }
11662:
11663: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11664:
11665: gsl_vector_free(x); /* initial values */
11666: gsl_vector_free(ss); /* inital step size */
11667: for (it=0; it<NDIM; it++){
11668: p[it+1]=gsl_vector_get(sfm->x,it);
11669: fprintf(ficrespow," %.12lf", p[it]);
11670: }
11671: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11672: #endif
11673: #ifdef POWELL
11674: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11675: #endif
1.126 brouard 11676: fclose(ficrespow);
11677:
1.203 brouard 11678: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11679:
11680: for(i=1; i <=NDIM; i++)
11681: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11682: matcov[i][j]=matcov[j][i];
1.126 brouard 11683:
11684: printf("\nCovariance matrix\n ");
1.203 brouard 11685: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11686: for(i=1; i <=NDIM; i++) {
11687: for(j=1;j<=NDIM;j++){
1.220 brouard 11688: printf("%f ",matcov[i][j]);
11689: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11690: }
1.203 brouard 11691: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11692: }
11693:
11694: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11695: for (i=1;i<=NDIM;i++) {
1.126 brouard 11696: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11697: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11698: }
1.126 brouard 11699: lsurv=vector(1,AGESUP);
11700: lpop=vector(1,AGESUP);
11701: tpop=vector(1,AGESUP);
11702: lsurv[agegomp]=100000;
11703:
11704: for (k=agegomp;k<=AGESUP;k++) {
11705: agemortsup=k;
11706: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11707: }
11708:
11709: for (k=agegomp;k<agemortsup;k++)
11710: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11711:
11712: for (k=agegomp;k<agemortsup;k++){
11713: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11714: sumlpop=sumlpop+lpop[k];
11715: }
11716:
11717: tpop[agegomp]=sumlpop;
11718: for (k=agegomp;k<(agemortsup-3);k++){
11719: /* tpop[k+1]=2;*/
11720: tpop[k+1]=tpop[k]-lpop[k];
11721: }
11722:
11723:
11724: printf("\nAge lx qx dx Lx Tx e(x)\n");
11725: for (k=agegomp;k<(agemortsup-2);k++)
11726: 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]);
11727:
11728:
11729: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11730: ageminpar=50;
11731: agemaxpar=100;
1.194 brouard 11732: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11733: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11734: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11735: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11736: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11737: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11738: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11739: }else{
11740: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11741: 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 11742: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11743: }
1.201 brouard 11744: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11745: stepm, weightopt,\
11746: model,imx,p,matcov,agemortsup);
11747:
11748: free_vector(lsurv,1,AGESUP);
11749: free_vector(lpop,1,AGESUP);
11750: free_vector(tpop,1,AGESUP);
1.220 brouard 11751: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11752: free_ivector(cens,1,n);
11753: free_vector(agecens,1,n);
11754: free_ivector(dcwave,1,n);
1.220 brouard 11755: #ifdef GSL
1.136 brouard 11756: #endif
1.186 brouard 11757: } /* Endof if mle==-3 mortality only */
1.205 brouard 11758: /* Standard */
11759: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11760: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11761: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11762: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11763: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11764: for (k=1; k<=npar;k++)
11765: printf(" %d %8.5f",k,p[k]);
11766: printf("\n");
1.205 brouard 11767: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11768: /* mlikeli uses func not funcone */
1.247 brouard 11769: /* for(i=1;i<nlstate;i++){ */
11770: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11771: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11772: /* } */
1.205 brouard 11773: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11774: }
11775: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11776: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11777: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11778: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11779: }
11780: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11781: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11782: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11783: for (k=1; k<=npar;k++)
11784: printf(" %d %8.5f",k,p[k]);
11785: printf("\n");
11786:
11787: /*--------- results files --------------*/
1.224 brouard 11788: 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 11789:
11790:
11791: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11792: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11793: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11794: for(i=1,jk=1; i <=nlstate; i++){
11795: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11796: if (k != i) {
11797: printf("%d%d ",i,k);
11798: fprintf(ficlog,"%d%d ",i,k);
11799: fprintf(ficres,"%1d%1d ",i,k);
11800: for(j=1; j <=ncovmodel; j++){
11801: printf("%12.7f ",p[jk]);
11802: fprintf(ficlog,"%12.7f ",p[jk]);
11803: fprintf(ficres,"%12.7f ",p[jk]);
11804: jk++;
11805: }
11806: printf("\n");
11807: fprintf(ficlog,"\n");
11808: fprintf(ficres,"\n");
11809: }
1.126 brouard 11810: }
11811: }
1.203 brouard 11812: if(mle != 0){
11813: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11814: ftolhess=ftol; /* Usually correct */
1.203 brouard 11815: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11816: 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");
11817: 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");
11818: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11819: for(k=1; k <=(nlstate+ndeath); k++){
11820: if (k != i) {
11821: printf("%d%d ",i,k);
11822: fprintf(ficlog,"%d%d ",i,k);
11823: for(j=1; j <=ncovmodel; j++){
11824: 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]));
11825: 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]));
11826: jk++;
11827: }
11828: printf("\n");
11829: fprintf(ficlog,"\n");
11830: }
11831: }
1.193 brouard 11832: }
1.203 brouard 11833: } /* end of hesscov and Wald tests */
1.225 brouard 11834:
1.203 brouard 11835: /* */
1.126 brouard 11836: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11837: printf("# Scales (for hessian or gradient estimation)\n");
11838: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11839: for(i=1,jk=1; i <=nlstate; i++){
11840: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11841: if (j!=i) {
11842: fprintf(ficres,"%1d%1d",i,j);
11843: printf("%1d%1d",i,j);
11844: fprintf(ficlog,"%1d%1d",i,j);
11845: for(k=1; k<=ncovmodel;k++){
11846: printf(" %.5e",delti[jk]);
11847: fprintf(ficlog," %.5e",delti[jk]);
11848: fprintf(ficres," %.5e",delti[jk]);
11849: jk++;
11850: }
11851: printf("\n");
11852: fprintf(ficlog,"\n");
11853: fprintf(ficres,"\n");
11854: }
1.126 brouard 11855: }
11856: }
11857:
11858: 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 11859: if(mle >= 1) /* To big for the screen */
1.126 brouard 11860: 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");
11861: 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");
11862: /* # 121 Var(a12)\n\ */
11863: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11864: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11865: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11866: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11867: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11868: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11869: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11870:
11871:
11872: /* Just to have a covariance matrix which will be more understandable
11873: even is we still don't want to manage dictionary of variables
11874: */
11875: for(itimes=1;itimes<=2;itimes++){
11876: jj=0;
11877: for(i=1; i <=nlstate; i++){
1.225 brouard 11878: for(j=1; j <=nlstate+ndeath; j++){
11879: if(j==i) continue;
11880: for(k=1; k<=ncovmodel;k++){
11881: jj++;
11882: ca[0]= k+'a'-1;ca[1]='\0';
11883: if(itimes==1){
11884: if(mle>=1)
11885: printf("#%1d%1d%d",i,j,k);
11886: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11887: fprintf(ficres,"#%1d%1d%d",i,j,k);
11888: }else{
11889: if(mle>=1)
11890: printf("%1d%1d%d",i,j,k);
11891: fprintf(ficlog,"%1d%1d%d",i,j,k);
11892: fprintf(ficres,"%1d%1d%d",i,j,k);
11893: }
11894: ll=0;
11895: for(li=1;li <=nlstate; li++){
11896: for(lj=1;lj <=nlstate+ndeath; lj++){
11897: if(lj==li) continue;
11898: for(lk=1;lk<=ncovmodel;lk++){
11899: ll++;
11900: if(ll<=jj){
11901: cb[0]= lk +'a'-1;cb[1]='\0';
11902: if(ll<jj){
11903: if(itimes==1){
11904: if(mle>=1)
11905: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11906: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11907: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11908: }else{
11909: if(mle>=1)
11910: printf(" %.5e",matcov[jj][ll]);
11911: fprintf(ficlog," %.5e",matcov[jj][ll]);
11912: fprintf(ficres," %.5e",matcov[jj][ll]);
11913: }
11914: }else{
11915: if(itimes==1){
11916: if(mle>=1)
11917: printf(" Var(%s%1d%1d)",ca,i,j);
11918: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11919: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11920: }else{
11921: if(mle>=1)
11922: printf(" %.7e",matcov[jj][ll]);
11923: fprintf(ficlog," %.7e",matcov[jj][ll]);
11924: fprintf(ficres," %.7e",matcov[jj][ll]);
11925: }
11926: }
11927: }
11928: } /* end lk */
11929: } /* end lj */
11930: } /* end li */
11931: if(mle>=1)
11932: printf("\n");
11933: fprintf(ficlog,"\n");
11934: fprintf(ficres,"\n");
11935: numlinepar++;
11936: } /* end k*/
11937: } /*end j */
1.126 brouard 11938: } /* end i */
11939: } /* end itimes */
11940:
11941: fflush(ficlog);
11942: fflush(ficres);
1.225 brouard 11943: while(fgets(line, MAXLINE, ficpar)) {
11944: /* If line starts with a # it is a comment */
11945: if (line[0] == '#') {
11946: numlinepar++;
11947: fputs(line,stdout);
11948: fputs(line,ficparo);
11949: fputs(line,ficlog);
11950: continue;
11951: }else
11952: break;
11953: }
11954:
1.209 brouard 11955: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11956: /* ungetc(c,ficpar); */
11957: /* fgets(line, MAXLINE, ficpar); */
11958: /* fputs(line,stdout); */
11959: /* fputs(line,ficparo); */
11960: /* } */
11961: /* ungetc(c,ficpar); */
1.126 brouard 11962:
11963: estepm=0;
1.209 brouard 11964: 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 11965:
11966: if (num_filled != 6) {
11967: 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);
11968: 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);
11969: goto end;
11970: }
11971: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11972: }
11973: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11974: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11975:
1.209 brouard 11976: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11977: if (estepm==0 || estepm < stepm) estepm=stepm;
11978: if (fage <= 2) {
11979: bage = ageminpar;
11980: fage = agemaxpar;
11981: }
11982:
11983: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11984: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11985: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11986:
1.186 brouard 11987: /* Other stuffs, more or less useful */
1.254 brouard 11988: while(fgets(line, MAXLINE, ficpar)) {
11989: /* If line starts with a # it is a comment */
11990: if (line[0] == '#') {
11991: numlinepar++;
11992: fputs(line,stdout);
11993: fputs(line,ficparo);
11994: fputs(line,ficlog);
11995: continue;
11996: }else
11997: break;
11998: }
11999:
12000: 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){
12001:
12002: if (num_filled != 7) {
12003: 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);
12004: 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);
12005: goto end;
12006: }
12007: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12008: 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);
12009: 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);
12010: 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 12011: }
1.254 brouard 12012:
12013: while(fgets(line, MAXLINE, ficpar)) {
12014: /* If line starts with a # it is a comment */
12015: if (line[0] == '#') {
12016: numlinepar++;
12017: fputs(line,stdout);
12018: fputs(line,ficparo);
12019: fputs(line,ficlog);
12020: continue;
12021: }else
12022: break;
1.126 brouard 12023: }
12024:
12025:
12026: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12027: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12028:
1.254 brouard 12029: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12030: if (num_filled != 1) {
12031: 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);
12032: 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);
12033: goto end;
12034: }
12035: printf("pop_based=%d\n",popbased);
12036: fprintf(ficlog,"pop_based=%d\n",popbased);
12037: fprintf(ficparo,"pop_based=%d\n",popbased);
12038: fprintf(ficres,"pop_based=%d\n",popbased);
12039: }
12040:
1.258 brouard 12041: /* Results */
12042: nresult=0;
12043: do{
12044: if(!fgets(line, MAXLINE, ficpar)){
12045: endishere=1;
12046: parameterline=14;
12047: }else if (line[0] == '#') {
12048: /* If line starts with a # it is a comment */
1.254 brouard 12049: numlinepar++;
12050: fputs(line,stdout);
12051: fputs(line,ficparo);
12052: fputs(line,ficlog);
12053: continue;
1.258 brouard 12054: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12055: parameterline=11;
12056: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12057: parameterline=12;
12058: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12059: parameterline=13;
12060: else{
12061: parameterline=14;
1.254 brouard 12062: }
1.258 brouard 12063: switch (parameterline){
12064: case 11:
12065: 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){
12066: if (num_filled != 8) {
12067: 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);
12068: 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);
12069: goto end;
12070: }
12071: 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);
12072: 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);
12073: 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);
12074: 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);
12075: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12076: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12077: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12078:
1.258 brouard 12079: }
1.254 brouard 12080: break;
1.258 brouard 12081: case 12:
12082: /*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);*/
12083: 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){
12084: if (num_filled != 8) {
1.262 brouard 12085: 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);
12086: 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 12087: goto end;
12088: }
12089: 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);
12090: 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);
12091: 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);
12092: 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);
12093: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12094: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12095: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12096: }
1.230 brouard 12097: break;
1.258 brouard 12098: case 13:
12099: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12100: if (num_filled == 0){
12101: resultline[0]='\0';
12102: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12103: 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);
12104: break;
12105: } else if (num_filled != 1){
12106: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12107: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12108: }
12109: nresult++; /* Sum of resultlines */
12110: printf("Result %d: result=%s\n",nresult, resultline);
12111: if(nresult > MAXRESULTLINES){
12112: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12113: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12114: goto end;
12115: }
12116: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12117: fprintf(ficparo,"result: %s\n",resultline);
12118: fprintf(ficres,"result: %s\n",resultline);
12119: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12120: break;
1.258 brouard 12121: case 14:
1.259 brouard 12122: if(ncovmodel >2 && nresult==0 ){
12123: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12124: goto end;
12125: }
1.259 brouard 12126: break;
1.258 brouard 12127: default:
12128: nresult=1;
12129: decoderesult(".",nresult ); /* No covariate */
12130: }
12131: } /* End switch parameterline */
12132: }while(endishere==0); /* End do */
1.126 brouard 12133:
1.230 brouard 12134: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12135: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12136:
12137: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12138: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12139: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12140: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12141: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12142: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12143: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12144: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12145: }else{
1.270 brouard 12146: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12147: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12148: }
12149: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12150: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12151: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12152:
1.225 brouard 12153: /*------------ free_vector -------------*/
12154: /* chdir(path); */
1.220 brouard 12155:
1.215 brouard 12156: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12157: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12158: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12159: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12160: free_lvector(num,1,n);
12161: free_vector(agedc,1,n);
12162: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12163: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12164: fclose(ficparo);
12165: fclose(ficres);
1.220 brouard 12166:
12167:
1.186 brouard 12168: /* Other results (useful)*/
1.220 brouard 12169:
12170:
1.126 brouard 12171: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12172: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12173: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12174: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12175: fclose(ficrespl);
12176:
12177: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12178: /*#include "hpijx.h"*/
12179: hPijx(p, bage, fage);
1.145 brouard 12180: fclose(ficrespij);
1.227 brouard 12181:
1.220 brouard 12182: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12183: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12184: k=1;
1.126 brouard 12185: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12186:
1.269 brouard 12187: /* Prevalence for each covariate combination in probs[age][status][cov] */
12188: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12189: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12190: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12191: for(k=1;k<=ncovcombmax;k++)
12192: probs[i][j][k]=0.;
1.269 brouard 12193: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12194: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12195: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12196: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12197: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12198: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12199: for(k=1;k<=ncovcombmax;k++)
12200: mobaverages[i][j][k]=0.;
1.219 brouard 12201: mobaverage=mobaverages;
12202: if (mobilav!=0) {
1.235 brouard 12203: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12204: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12205: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12206: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12207: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12208: }
1.269 brouard 12209: } else if (mobilavproj !=0) {
1.235 brouard 12210: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12211: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12212: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12213: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12214: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12215: }
1.269 brouard 12216: }else{
12217: printf("Internal error moving average\n");
12218: fflush(stdout);
12219: exit(1);
1.219 brouard 12220: }
12221: }/* end if moving average */
1.227 brouard 12222:
1.126 brouard 12223: /*---------- Forecasting ------------------*/
12224: if(prevfcast==1){
12225: /* if(stepm ==1){*/
1.269 brouard 12226: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12227: }
1.269 brouard 12228:
12229: /* Backcasting */
1.217 brouard 12230: if(backcast==1){
1.219 brouard 12231: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12232: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12233: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12234:
12235: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12236:
12237: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12238:
1.219 brouard 12239: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12240: fclose(ficresplb);
12241:
1.222 brouard 12242: hBijx(p, bage, fage, mobaverage);
12243: fclose(ficrespijb);
1.219 brouard 12244:
1.269 brouard 12245: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12246: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12247: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12248:
12249:
1.269 brouard 12250: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12251: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12252: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12253: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12254: } /* end Backcasting */
1.268 brouard 12255:
1.186 brouard 12256:
12257: /* ------ Other prevalence ratios------------ */
1.126 brouard 12258:
1.215 brouard 12259: free_ivector(wav,1,imx);
12260: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12261: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12262: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12263:
12264:
1.127 brouard 12265: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12266:
1.201 brouard 12267: strcpy(filerese,"E_");
12268: strcat(filerese,fileresu);
1.126 brouard 12269: if((ficreseij=fopen(filerese,"w"))==NULL) {
12270: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12271: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12272: }
1.208 brouard 12273: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12274: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12275:
12276: pstamp(ficreseij);
1.219 brouard 12277:
1.235 brouard 12278: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12279: if (cptcovn < 1){i1=1;}
12280:
12281: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12282: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12283: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12284: continue;
1.219 brouard 12285: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12286: printf("\n#****** ");
1.225 brouard 12287: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12288: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12289: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12290: }
12291: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12292: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12293: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12294: }
12295: fprintf(ficreseij,"******\n");
1.235 brouard 12296: printf("******\n");
1.219 brouard 12297:
12298: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12299: oldm=oldms;savm=savms;
1.235 brouard 12300: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12301:
1.219 brouard 12302: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12303: }
12304: fclose(ficreseij);
1.208 brouard 12305: printf("done evsij\n");fflush(stdout);
12306: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12307:
1.218 brouard 12308:
1.227 brouard 12309: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12310:
1.201 brouard 12311: strcpy(filerest,"T_");
12312: strcat(filerest,fileresu);
1.127 brouard 12313: if((ficrest=fopen(filerest,"w"))==NULL) {
12314: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12315: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12316: }
1.208 brouard 12317: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12318: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12319: strcpy(fileresstde,"STDE_");
12320: strcat(fileresstde,fileresu);
1.126 brouard 12321: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12322: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12323: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12324: }
1.227 brouard 12325: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12326: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12327:
1.201 brouard 12328: strcpy(filerescve,"CVE_");
12329: strcat(filerescve,fileresu);
1.126 brouard 12330: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12331: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12332: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12333: }
1.227 brouard 12334: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12335: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12336:
1.201 brouard 12337: strcpy(fileresv,"V_");
12338: strcat(fileresv,fileresu);
1.126 brouard 12339: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12340: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12341: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12342: }
1.227 brouard 12343: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12344: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12345:
1.235 brouard 12346: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12347: if (cptcovn < 1){i1=1;}
12348:
12349: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12350: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12351: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12352: continue;
1.242 brouard 12353: printf("\n#****** Result for:");
12354: fprintf(ficrest,"\n#****** Result for:");
12355: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12356: for(j=1;j<=cptcoveff;j++){
12357: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12358: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12359: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12360: }
1.235 brouard 12361: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12362: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12363: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12364: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12365: }
1.208 brouard 12366: fprintf(ficrest,"******\n");
1.227 brouard 12367: fprintf(ficlog,"******\n");
12368: printf("******\n");
1.208 brouard 12369:
12370: fprintf(ficresstdeij,"\n#****** ");
12371: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12372: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12373: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12374: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12375: }
1.235 brouard 12376: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12377: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12378: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12379: }
1.208 brouard 12380: fprintf(ficresstdeij,"******\n");
12381: fprintf(ficrescveij,"******\n");
12382:
12383: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12384: /* pstamp(ficresvij); */
1.225 brouard 12385: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12386: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12387: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12388: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12389: }
1.208 brouard 12390: fprintf(ficresvij,"******\n");
12391:
12392: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12393: oldm=oldms;savm=savms;
1.235 brouard 12394: printf(" cvevsij ");
12395: fprintf(ficlog, " cvevsij ");
12396: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12397: printf(" end cvevsij \n ");
12398: fprintf(ficlog, " end cvevsij \n ");
12399:
12400: /*
12401: */
12402: /* goto endfree; */
12403:
12404: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12405: pstamp(ficrest);
12406:
1.269 brouard 12407: epj=vector(1,nlstate+1);
1.208 brouard 12408: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12409: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12410: cptcod= 0; /* To be deleted */
12411: printf("varevsij vpopbased=%d \n",vpopbased);
12412: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12413: 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 12414: 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 ");
12415: if(vpopbased==1)
12416: 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);
12417: else
12418: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12419: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12420: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12421: fprintf(ficrest,"\n");
12422: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12423: printf("Computing age specific period (stable) prevalences in each health state \n");
12424: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12425: for(age=bage; age <=fage ;age++){
1.235 brouard 12426: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12427: if (vpopbased==1) {
12428: if(mobilav ==0){
12429: for(i=1; i<=nlstate;i++)
12430: prlim[i][i]=probs[(int)age][i][k];
12431: }else{ /* mobilav */
12432: for(i=1; i<=nlstate;i++)
12433: prlim[i][i]=mobaverage[(int)age][i][k];
12434: }
12435: }
1.219 brouard 12436:
1.227 brouard 12437: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12438: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12439: /* printf(" age %4.0f ",age); */
12440: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12441: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12442: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12443: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12444: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12445: }
12446: epj[nlstate+1] +=epj[j];
12447: }
12448: /* printf(" age %4.0f \n",age); */
1.219 brouard 12449:
1.227 brouard 12450: for(i=1, vepp=0.;i <=nlstate;i++)
12451: for(j=1;j <=nlstate;j++)
12452: vepp += vareij[i][j][(int)age];
12453: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12454: for(j=1;j <=nlstate;j++){
12455: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12456: }
12457: fprintf(ficrest,"\n");
12458: }
1.208 brouard 12459: } /* End vpopbased */
1.269 brouard 12460: free_vector(epj,1,nlstate+1);
1.208 brouard 12461: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12462: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12463: printf("done selection\n");fflush(stdout);
12464: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12465:
1.235 brouard 12466: } /* End k selection */
1.227 brouard 12467:
12468: printf("done State-specific expectancies\n");fflush(stdout);
12469: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12470:
1.269 brouard 12471: /* variance-covariance of period prevalence*/
12472: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12473:
1.227 brouard 12474:
12475: free_vector(weight,1,n);
12476: free_imatrix(Tvard,1,NCOVMAX,1,2);
12477: free_imatrix(s,1,maxwav+1,1,n);
12478: free_matrix(anint,1,maxwav,1,n);
12479: free_matrix(mint,1,maxwav,1,n);
12480: free_ivector(cod,1,n);
12481: free_ivector(tab,1,NCOVMAX);
12482: fclose(ficresstdeij);
12483: fclose(ficrescveij);
12484: fclose(ficresvij);
12485: fclose(ficrest);
12486: fclose(ficpar);
12487:
12488:
1.126 brouard 12489: /*---------- End : free ----------------*/
1.219 brouard 12490: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12491: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12492: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12493: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12494: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12495: } /* mle==-3 arrives here for freeing */
1.227 brouard 12496: /* endfree:*/
12497: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12498: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12499: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12500: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12501: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12502: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12503: free_matrix(covar,0,NCOVMAX,1,n);
12504: free_matrix(matcov,1,npar,1,npar);
12505: free_matrix(hess,1,npar,1,npar);
12506: /*free_vector(delti,1,npar);*/
12507: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12508: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12509: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12510: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12511:
12512: free_ivector(ncodemax,1,NCOVMAX);
12513: free_ivector(ncodemaxwundef,1,NCOVMAX);
12514: free_ivector(Dummy,-1,NCOVMAX);
12515: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12516: free_ivector(DummyV,1,NCOVMAX);
12517: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12518: free_ivector(Typevar,-1,NCOVMAX);
12519: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12520: free_ivector(TvarsQ,1,NCOVMAX);
12521: free_ivector(TvarsQind,1,NCOVMAX);
12522: free_ivector(TvarsD,1,NCOVMAX);
12523: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12524: free_ivector(TvarFD,1,NCOVMAX);
12525: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12526: free_ivector(TvarF,1,NCOVMAX);
12527: free_ivector(TvarFind,1,NCOVMAX);
12528: free_ivector(TvarV,1,NCOVMAX);
12529: free_ivector(TvarVind,1,NCOVMAX);
12530: free_ivector(TvarA,1,NCOVMAX);
12531: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12532: free_ivector(TvarFQ,1,NCOVMAX);
12533: free_ivector(TvarFQind,1,NCOVMAX);
12534: free_ivector(TvarVD,1,NCOVMAX);
12535: free_ivector(TvarVDind,1,NCOVMAX);
12536: free_ivector(TvarVQ,1,NCOVMAX);
12537: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12538: free_ivector(Tvarsel,1,NCOVMAX);
12539: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12540: free_ivector(Tposprod,1,NCOVMAX);
12541: free_ivector(Tprod,1,NCOVMAX);
12542: free_ivector(Tvaraff,1,NCOVMAX);
12543: free_ivector(invalidvarcomb,1,ncovcombmax);
12544: free_ivector(Tage,1,NCOVMAX);
12545: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12546: free_ivector(TmodelInvind,1,NCOVMAX);
12547: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12548:
12549: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12550: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12551: fflush(fichtm);
12552: fflush(ficgp);
12553:
1.227 brouard 12554:
1.126 brouard 12555: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12556: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12557: 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 12558: }else{
12559: printf("End of Imach\n");
12560: fprintf(ficlog,"End of Imach\n");
12561: }
12562: printf("See log file on %s\n",filelog);
12563: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12564: /*(void) gettimeofday(&end_time,&tzp);*/
12565: rend_time = time(NULL);
12566: end_time = *localtime(&rend_time);
12567: /* tml = *localtime(&end_time.tm_sec); */
12568: strcpy(strtend,asctime(&end_time));
1.126 brouard 12569: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12570: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12571: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12572:
1.157 brouard 12573: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12574: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12575: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12576: /* printf("Total time was %d uSec.\n", total_usecs);*/
12577: /* if(fileappend(fichtm,optionfilehtm)){ */
12578: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12579: fclose(fichtm);
12580: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12581: fclose(fichtmcov);
12582: fclose(ficgp);
12583: fclose(ficlog);
12584: /*------ End -----------*/
1.227 brouard 12585:
12586:
12587: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12588: #ifdef WIN32
1.227 brouard 12589: if (_chdir(pathcd) != 0)
12590: printf("Can't move to directory %s!\n",path);
12591: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12592: #else
1.227 brouard 12593: if(chdir(pathcd) != 0)
12594: printf("Can't move to directory %s!\n", path);
12595: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12596: #endif
1.126 brouard 12597: printf("Current directory %s!\n",pathcd);
12598: /*strcat(plotcmd,CHARSEPARATOR);*/
12599: sprintf(plotcmd,"gnuplot");
1.157 brouard 12600: #ifdef _WIN32
1.126 brouard 12601: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12602: #endif
12603: if(!stat(plotcmd,&info)){
1.158 brouard 12604: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12605: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12606: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12607: }else
12608: strcpy(pplotcmd,plotcmd);
1.157 brouard 12609: #ifdef __unix
1.126 brouard 12610: strcpy(plotcmd,GNUPLOTPROGRAM);
12611: if(!stat(plotcmd,&info)){
1.158 brouard 12612: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12613: }else
12614: strcpy(pplotcmd,plotcmd);
12615: #endif
12616: }else
12617: strcpy(pplotcmd,plotcmd);
12618:
12619: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12620: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12621:
1.126 brouard 12622: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12623: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12624: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12625: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12626: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12627: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12628: }
1.158 brouard 12629: printf(" Successful, please wait...");
1.126 brouard 12630: while (z[0] != 'q') {
12631: /* chdir(path); */
1.154 brouard 12632: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12633: scanf("%s",z);
12634: /* if (z[0] == 'c') system("./imach"); */
12635: if (z[0] == 'e') {
1.158 brouard 12636: #ifdef __APPLE__
1.152 brouard 12637: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12638: #elif __linux
12639: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12640: #else
1.152 brouard 12641: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12642: #endif
12643: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12644: system(pplotcmd);
1.126 brouard 12645: }
12646: else if (z[0] == 'g') system(plotcmd);
12647: else if (z[0] == 'q') exit(0);
12648: }
1.227 brouard 12649: end:
1.126 brouard 12650: while (z[0] != 'q') {
1.195 brouard 12651: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12652: scanf("%s",z);
12653: }
12654: }
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