Annotation of imach/src/imach.c, revision 1.266
1.266 ! brouard 1: /* $Id: imach.c,v 1.265 2017/04/26 16:22:11 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.266 ! brouard 4: Revision 1.265 2017/04/26 16:22:11 brouard
! 5: Summary: imach 0.99r13 Some bugs fixed
! 6:
1.265 brouard 7: Revision 1.264 2017/04/26 06:01:29 brouard
8: Summary: Labels in graphs
9:
1.264 brouard 10: Revision 1.263 2017/04/24 15:23:15 brouard
11: Summary: to save
12:
1.263 brouard 13: Revision 1.262 2017/04/18 16:48:12 brouard
14: *** empty log message ***
15:
1.262 brouard 16: Revision 1.261 2017/04/05 10:14:09 brouard
17: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
18:
1.261 brouard 19: Revision 1.260 2017/04/04 17:46:59 brouard
20: Summary: Gnuplot indexations fixed (humm)
21:
1.260 brouard 22: Revision 1.259 2017/04/04 13:01:16 brouard
23: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
24:
1.259 brouard 25: Revision 1.258 2017/04/03 10:17:47 brouard
26: Summary: Version 0.99r12
27:
28: Some cleanings, conformed with updated documentation.
29:
1.258 brouard 30: Revision 1.257 2017/03/29 16:53:30 brouard
31: Summary: Temp
32:
1.257 brouard 33: Revision 1.256 2017/03/27 05:50:23 brouard
34: Summary: Temporary
35:
1.256 brouard 36: Revision 1.255 2017/03/08 16:02:28 brouard
37: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
38:
1.255 brouard 39: Revision 1.254 2017/03/08 07:13:00 brouard
40: Summary: Fixing data parameter line
41:
1.254 brouard 42: Revision 1.253 2016/12/15 11:59:41 brouard
43: Summary: 0.99 in progress
44:
1.253 brouard 45: Revision 1.252 2016/09/15 21:15:37 brouard
46: *** empty log message ***
47:
1.252 brouard 48: Revision 1.251 2016/09/15 15:01:13 brouard
49: Summary: not working
50:
1.251 brouard 51: Revision 1.250 2016/09/08 16:07:27 brouard
52: Summary: continue
53:
1.250 brouard 54: Revision 1.249 2016/09/07 17:14:18 brouard
55: Summary: Starting values from frequencies
56:
1.249 brouard 57: Revision 1.248 2016/09/07 14:10:18 brouard
58: *** empty log message ***
59:
1.248 brouard 60: Revision 1.247 2016/09/02 11:11:21 brouard
61: *** empty log message ***
62:
1.247 brouard 63: Revision 1.246 2016/09/02 08:49:22 brouard
64: *** empty log message ***
65:
1.246 brouard 66: Revision 1.245 2016/09/02 07:25:01 brouard
67: *** empty log message ***
68:
1.245 brouard 69: Revision 1.244 2016/09/02 07:17:34 brouard
70: *** empty log message ***
71:
1.244 brouard 72: Revision 1.243 2016/09/02 06:45:35 brouard
73: *** empty log message ***
74:
1.243 brouard 75: Revision 1.242 2016/08/30 15:01:20 brouard
76: Summary: Fixing a lots
77:
1.242 brouard 78: Revision 1.241 2016/08/29 17:17:25 brouard
79: Summary: gnuplot problem in Back projection to fix
80:
1.241 brouard 81: Revision 1.240 2016/08/29 07:53:18 brouard
82: Summary: Better
83:
1.240 brouard 84: Revision 1.239 2016/08/26 15:51:03 brouard
85: Summary: Improvement in Powell output in order to copy and paste
86:
87: Author:
88:
1.239 brouard 89: Revision 1.238 2016/08/26 14:23:35 brouard
90: Summary: Starting tests of 0.99
91:
1.238 brouard 92: Revision 1.237 2016/08/26 09:20:19 brouard
93: Summary: to valgrind
94:
1.237 brouard 95: Revision 1.236 2016/08/25 10:50:18 brouard
96: *** empty log message ***
97:
1.236 brouard 98: Revision 1.235 2016/08/25 06:59:23 brouard
99: *** empty log message ***
100:
1.235 brouard 101: Revision 1.234 2016/08/23 16:51:20 brouard
102: *** empty log message ***
103:
1.234 brouard 104: Revision 1.233 2016/08/23 07:40:50 brouard
105: Summary: not working
106:
1.233 brouard 107: Revision 1.232 2016/08/22 14:20:21 brouard
108: Summary: not working
109:
1.232 brouard 110: Revision 1.231 2016/08/22 07:17:15 brouard
111: Summary: not working
112:
1.231 brouard 113: Revision 1.230 2016/08/22 06:55:53 brouard
114: Summary: Not working
115:
1.230 brouard 116: Revision 1.229 2016/07/23 09:45:53 brouard
117: Summary: Completing for func too
118:
1.229 brouard 119: Revision 1.228 2016/07/22 17:45:30 brouard
120: Summary: Fixing some arrays, still debugging
121:
1.227 brouard 122: Revision 1.226 2016/07/12 18:42:34 brouard
123: Summary: temp
124:
1.226 brouard 125: Revision 1.225 2016/07/12 08:40:03 brouard
126: Summary: saving but not running
127:
1.225 brouard 128: Revision 1.224 2016/07/01 13:16:01 brouard
129: Summary: Fixes
130:
1.224 brouard 131: Revision 1.223 2016/02/19 09:23:35 brouard
132: Summary: temporary
133:
1.223 brouard 134: Revision 1.222 2016/02/17 08:14:50 brouard
135: Summary: Probably last 0.98 stable version 0.98r6
136:
1.222 brouard 137: Revision 1.221 2016/02/15 23:35:36 brouard
138: Summary: minor bug
139:
1.220 brouard 140: Revision 1.219 2016/02/15 00:48:12 brouard
141: *** empty log message ***
142:
1.219 brouard 143: Revision 1.218 2016/02/12 11:29:23 brouard
144: Summary: 0.99 Back projections
145:
1.218 brouard 146: Revision 1.217 2015/12/23 17:18:31 brouard
147: Summary: Experimental backcast
148:
1.217 brouard 149: Revision 1.216 2015/12/18 17:32:11 brouard
150: Summary: 0.98r4 Warning and status=-2
151:
152: Version 0.98r4 is now:
153: - displaying an error when status is -1, date of interview unknown and date of death known;
154: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
155: Older changes concerning s=-2, dating from 2005 have been supersed.
156:
1.216 brouard 157: Revision 1.215 2015/12/16 08:52:24 brouard
158: Summary: 0.98r4 working
159:
1.215 brouard 160: Revision 1.214 2015/12/16 06:57:54 brouard
161: Summary: temporary not working
162:
1.214 brouard 163: Revision 1.213 2015/12/11 18:22:17 brouard
164: Summary: 0.98r4
165:
1.213 brouard 166: Revision 1.212 2015/11/21 12:47:24 brouard
167: Summary: minor typo
168:
1.212 brouard 169: Revision 1.211 2015/11/21 12:41:11 brouard
170: Summary: 0.98r3 with some graph of projected cross-sectional
171:
172: Author: Nicolas Brouard
173:
1.211 brouard 174: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 175: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 176: Summary: Adding ftolpl parameter
177: Author: N Brouard
178:
179: We had difficulties to get smoothed confidence intervals. It was due
180: to the period prevalence which wasn't computed accurately. The inner
181: parameter ftolpl is now an outer parameter of the .imach parameter
182: file after estepm. If ftolpl is small 1.e-4 and estepm too,
183: computation are long.
184:
1.209 brouard 185: Revision 1.208 2015/11/17 14:31:57 brouard
186: Summary: temporary
187:
1.208 brouard 188: Revision 1.207 2015/10/27 17:36:57 brouard
189: *** empty log message ***
190:
1.207 brouard 191: Revision 1.206 2015/10/24 07:14:11 brouard
192: *** empty log message ***
193:
1.206 brouard 194: Revision 1.205 2015/10/23 15:50:53 brouard
195: Summary: 0.98r3 some clarification for graphs on likelihood contributions
196:
1.205 brouard 197: Revision 1.204 2015/10/01 16:20:26 brouard
198: Summary: Some new graphs of contribution to likelihood
199:
1.204 brouard 200: Revision 1.203 2015/09/30 17:45:14 brouard
201: Summary: looking at better estimation of the hessian
202:
203: Also a better criteria for convergence to the period prevalence And
204: therefore adding the number of years needed to converge. (The
205: prevalence in any alive state shold sum to one
206:
1.203 brouard 207: Revision 1.202 2015/09/22 19:45:16 brouard
208: Summary: Adding some overall graph on contribution to likelihood. Might change
209:
1.202 brouard 210: Revision 1.201 2015/09/15 17:34:58 brouard
211: Summary: 0.98r0
212:
213: - Some new graphs like suvival functions
214: - Some bugs fixed like model=1+age+V2.
215:
1.201 brouard 216: Revision 1.200 2015/09/09 16:53:55 brouard
217: Summary: Big bug thanks to Flavia
218:
219: Even model=1+age+V2. did not work anymore
220:
1.200 brouard 221: Revision 1.199 2015/09/07 14:09:23 brouard
222: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
223:
1.199 brouard 224: Revision 1.198 2015/09/03 07:14:39 brouard
225: Summary: 0.98q5 Flavia
226:
1.198 brouard 227: Revision 1.197 2015/09/01 18:24:39 brouard
228: *** empty log message ***
229:
1.197 brouard 230: Revision 1.196 2015/08/18 23:17:52 brouard
231: Summary: 0.98q5
232:
1.196 brouard 233: Revision 1.195 2015/08/18 16:28:39 brouard
234: Summary: Adding a hack for testing purpose
235:
236: After reading the title, ftol and model lines, if the comment line has
237: a q, starting with #q, the answer at the end of the run is quit. It
238: permits to run test files in batch with ctest. The former workaround was
239: $ echo q | imach foo.imach
240:
1.195 brouard 241: Revision 1.194 2015/08/18 13:32:00 brouard
242: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
243:
1.194 brouard 244: Revision 1.193 2015/08/04 07:17:42 brouard
245: Summary: 0.98q4
246:
1.193 brouard 247: Revision 1.192 2015/07/16 16:49:02 brouard
248: Summary: Fixing some outputs
249:
1.192 brouard 250: Revision 1.191 2015/07/14 10:00:33 brouard
251: Summary: Some fixes
252:
1.191 brouard 253: Revision 1.190 2015/05/05 08:51:13 brouard
254: Summary: Adding digits in output parameters (7 digits instead of 6)
255:
256: Fix 1+age+.
257:
1.190 brouard 258: Revision 1.189 2015/04/30 14:45:16 brouard
259: Summary: 0.98q2
260:
1.189 brouard 261: Revision 1.188 2015/04/30 08:27:53 brouard
262: *** empty log message ***
263:
1.188 brouard 264: Revision 1.187 2015/04/29 09:11:15 brouard
265: *** empty log message ***
266:
1.187 brouard 267: Revision 1.186 2015/04/23 12:01:52 brouard
268: Summary: V1*age is working now, version 0.98q1
269:
270: Some codes had been disabled in order to simplify and Vn*age was
271: working in the optimization phase, ie, giving correct MLE parameters,
272: but, as usual, outputs were not correct and program core dumped.
273:
1.186 brouard 274: Revision 1.185 2015/03/11 13:26:42 brouard
275: Summary: Inclusion of compile and links command line for Intel Compiler
276:
1.185 brouard 277: Revision 1.184 2015/03/11 11:52:39 brouard
278: Summary: Back from Windows 8. Intel Compiler
279:
1.184 brouard 280: Revision 1.183 2015/03/10 20:34:32 brouard
281: Summary: 0.98q0, trying with directest, mnbrak fixed
282:
283: We use directest instead of original Powell test; probably no
284: incidence on the results, but better justifications;
285: We fixed Numerical Recipes mnbrak routine which was wrong and gave
286: wrong results.
287:
1.183 brouard 288: Revision 1.182 2015/02/12 08:19:57 brouard
289: Summary: Trying to keep directest which seems simpler and more general
290: Author: Nicolas Brouard
291:
1.182 brouard 292: Revision 1.181 2015/02/11 23:22:24 brouard
293: Summary: Comments on Powell added
294:
295: Author:
296:
1.181 brouard 297: Revision 1.180 2015/02/11 17:33:45 brouard
298: Summary: Finishing move from main to function (hpijx and prevalence_limit)
299:
1.180 brouard 300: Revision 1.179 2015/01/04 09:57:06 brouard
301: Summary: back to OS/X
302:
1.179 brouard 303: Revision 1.178 2015/01/04 09:35:48 brouard
304: *** empty log message ***
305:
1.178 brouard 306: Revision 1.177 2015/01/03 18:40:56 brouard
307: Summary: Still testing ilc32 on OSX
308:
1.177 brouard 309: Revision 1.176 2015/01/03 16:45:04 brouard
310: *** empty log message ***
311:
1.176 brouard 312: Revision 1.175 2015/01/03 16:33:42 brouard
313: *** empty log message ***
314:
1.175 brouard 315: Revision 1.174 2015/01/03 16:15:49 brouard
316: Summary: Still in cross-compilation
317:
1.174 brouard 318: Revision 1.173 2015/01/03 12:06:26 brouard
319: Summary: trying to detect cross-compilation
320:
1.173 brouard 321: Revision 1.172 2014/12/27 12:07:47 brouard
322: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
323:
1.172 brouard 324: Revision 1.171 2014/12/23 13:26:59 brouard
325: Summary: Back from Visual C
326:
327: Still problem with utsname.h on Windows
328:
1.171 brouard 329: Revision 1.170 2014/12/23 11:17:12 brouard
330: Summary: Cleaning some \%% back to %%
331:
332: The escape was mandatory for a specific compiler (which one?), but too many warnings.
333:
1.170 brouard 334: Revision 1.169 2014/12/22 23:08:31 brouard
335: Summary: 0.98p
336:
337: Outputs some informations on compiler used, OS etc. Testing on different platforms.
338:
1.169 brouard 339: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 340: Summary: update
1.169 brouard 341:
1.168 brouard 342: Revision 1.167 2014/12/22 13:50:56 brouard
343: Summary: Testing uname and compiler version and if compiled 32 or 64
344:
345: Testing on Linux 64
346:
1.167 brouard 347: Revision 1.166 2014/12/22 11:40:47 brouard
348: *** empty log message ***
349:
1.166 brouard 350: Revision 1.165 2014/12/16 11:20:36 brouard
351: Summary: After compiling on Visual C
352:
353: * imach.c (Module): Merging 1.61 to 1.162
354:
1.165 brouard 355: Revision 1.164 2014/12/16 10:52:11 brouard
356: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
357:
358: * imach.c (Module): Merging 1.61 to 1.162
359:
1.164 brouard 360: Revision 1.163 2014/12/16 10:30:11 brouard
361: * imach.c (Module): Merging 1.61 to 1.162
362:
1.163 brouard 363: Revision 1.162 2014/09/25 11:43:39 brouard
364: Summary: temporary backup 0.99!
365:
1.162 brouard 366: Revision 1.1 2014/09/16 11:06:58 brouard
367: Summary: With some code (wrong) for nlopt
368:
369: Author:
370:
371: Revision 1.161 2014/09/15 20:41:41 brouard
372: Summary: Problem with macro SQR on Intel compiler
373:
1.161 brouard 374: Revision 1.160 2014/09/02 09:24:05 brouard
375: *** empty log message ***
376:
1.160 brouard 377: Revision 1.159 2014/09/01 10:34:10 brouard
378: Summary: WIN32
379: Author: Brouard
380:
1.159 brouard 381: Revision 1.158 2014/08/27 17:11:51 brouard
382: *** empty log message ***
383:
1.158 brouard 384: Revision 1.157 2014/08/27 16:26:55 brouard
385: Summary: Preparing windows Visual studio version
386: Author: Brouard
387:
388: In order to compile on Visual studio, time.h is now correct and time_t
389: and tm struct should be used. difftime should be used but sometimes I
390: just make the differences in raw time format (time(&now).
391: Trying to suppress #ifdef LINUX
392: Add xdg-open for __linux in order to open default browser.
393:
1.157 brouard 394: Revision 1.156 2014/08/25 20:10:10 brouard
395: *** empty log message ***
396:
1.156 brouard 397: Revision 1.155 2014/08/25 18:32:34 brouard
398: Summary: New compile, minor changes
399: Author: Brouard
400:
1.155 brouard 401: Revision 1.154 2014/06/20 17:32:08 brouard
402: Summary: Outputs now all graphs of convergence to period prevalence
403:
1.154 brouard 404: Revision 1.153 2014/06/20 16:45:46 brouard
405: Summary: If 3 live state, convergence to period prevalence on same graph
406: Author: Brouard
407:
1.153 brouard 408: Revision 1.152 2014/06/18 17:54:09 brouard
409: Summary: open browser, use gnuplot on same dir than imach if not found in the path
410:
1.152 brouard 411: Revision 1.151 2014/06/18 16:43:30 brouard
412: *** empty log message ***
413:
1.151 brouard 414: Revision 1.150 2014/06/18 16:42:35 brouard
415: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
416: Author: brouard
417:
1.150 brouard 418: Revision 1.149 2014/06/18 15:51:14 brouard
419: Summary: Some fixes in parameter files errors
420: Author: Nicolas Brouard
421:
1.149 brouard 422: Revision 1.148 2014/06/17 17:38:48 brouard
423: Summary: Nothing new
424: Author: Brouard
425:
426: Just a new packaging for OS/X version 0.98nS
427:
1.148 brouard 428: Revision 1.147 2014/06/16 10:33:11 brouard
429: *** empty log message ***
430:
1.147 brouard 431: Revision 1.146 2014/06/16 10:20:28 brouard
432: Summary: Merge
433: Author: Brouard
434:
435: Merge, before building revised version.
436:
1.146 brouard 437: Revision 1.145 2014/06/10 21:23:15 brouard
438: Summary: Debugging with valgrind
439: Author: Nicolas Brouard
440:
441: Lot of changes in order to output the results with some covariates
442: After the Edimburgh REVES conference 2014, it seems mandatory to
443: improve the code.
444: No more memory valgrind error but a lot has to be done in order to
445: continue the work of splitting the code into subroutines.
446: Also, decodemodel has been improved. Tricode is still not
447: optimal. nbcode should be improved. Documentation has been added in
448: the source code.
449:
1.144 brouard 450: Revision 1.143 2014/01/26 09:45:38 brouard
451: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
452:
453: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
454: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
455:
1.143 brouard 456: Revision 1.142 2014/01/26 03:57:36 brouard
457: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
458:
459: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
460:
1.142 brouard 461: Revision 1.141 2014/01/26 02:42:01 brouard
462: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
463:
1.141 brouard 464: Revision 1.140 2011/09/02 10:37:54 brouard
465: Summary: times.h is ok with mingw32 now.
466:
1.140 brouard 467: Revision 1.139 2010/06/14 07:50:17 brouard
468: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
469: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
470:
1.139 brouard 471: Revision 1.138 2010/04/30 18:19:40 brouard
472: *** empty log message ***
473:
1.138 brouard 474: Revision 1.137 2010/04/29 18:11:38 brouard
475: (Module): Checking covariates for more complex models
476: than V1+V2. A lot of change to be done. Unstable.
477:
1.137 brouard 478: Revision 1.136 2010/04/26 20:30:53 brouard
479: (Module): merging some libgsl code. Fixing computation
480: of likelione (using inter/intrapolation if mle = 0) in order to
481: get same likelihood as if mle=1.
482: Some cleaning of code and comments added.
483:
1.136 brouard 484: Revision 1.135 2009/10/29 15:33:14 brouard
485: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
486:
1.135 brouard 487: Revision 1.134 2009/10/29 13:18:53 brouard
488: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
489:
1.134 brouard 490: Revision 1.133 2009/07/06 10:21:25 brouard
491: just nforces
492:
1.133 brouard 493: Revision 1.132 2009/07/06 08:22:05 brouard
494: Many tings
495:
1.132 brouard 496: Revision 1.131 2009/06/20 16:22:47 brouard
497: Some dimensions resccaled
498:
1.131 brouard 499: Revision 1.130 2009/05/26 06:44:34 brouard
500: (Module): Max Covariate is now set to 20 instead of 8. A
501: lot of cleaning with variables initialized to 0. Trying to make
502: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
503:
1.130 brouard 504: Revision 1.129 2007/08/31 13:49:27 lievre
505: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
506:
1.129 lievre 507: Revision 1.128 2006/06/30 13:02:05 brouard
508: (Module): Clarifications on computing e.j
509:
1.128 brouard 510: Revision 1.127 2006/04/28 18:11:50 brouard
511: (Module): Yes the sum of survivors was wrong since
512: imach-114 because nhstepm was no more computed in the age
513: loop. Now we define nhstepma in the age loop.
514: (Module): In order to speed up (in case of numerous covariates) we
515: compute health expectancies (without variances) in a first step
516: and then all the health expectancies with variances or standard
517: deviation (needs data from the Hessian matrices) which slows the
518: computation.
519: In the future we should be able to stop the program is only health
520: expectancies and graph are needed without standard deviations.
521:
1.127 brouard 522: Revision 1.126 2006/04/28 17:23:28 brouard
523: (Module): Yes the sum of survivors was wrong since
524: imach-114 because nhstepm was no more computed in the age
525: loop. Now we define nhstepma in the age loop.
526: Version 0.98h
527:
1.126 brouard 528: Revision 1.125 2006/04/04 15:20:31 lievre
529: Errors in calculation of health expectancies. Age was not initialized.
530: Forecasting file added.
531:
532: Revision 1.124 2006/03/22 17:13:53 lievre
533: Parameters are printed with %lf instead of %f (more numbers after the comma).
534: The log-likelihood is printed in the log file
535:
536: Revision 1.123 2006/03/20 10:52:43 brouard
537: * imach.c (Module): <title> changed, corresponds to .htm file
538: name. <head> headers where missing.
539:
540: * imach.c (Module): Weights can have a decimal point as for
541: English (a comma might work with a correct LC_NUMERIC environment,
542: otherwise the weight is truncated).
543: Modification of warning when the covariates values are not 0 or
544: 1.
545: Version 0.98g
546:
547: Revision 1.122 2006/03/20 09:45:41 brouard
548: (Module): Weights can have a decimal point as for
549: English (a comma might work with a correct LC_NUMERIC environment,
550: otherwise the weight is truncated).
551: Modification of warning when the covariates values are not 0 or
552: 1.
553: Version 0.98g
554:
555: Revision 1.121 2006/03/16 17:45:01 lievre
556: * imach.c (Module): Comments concerning covariates added
557:
558: * imach.c (Module): refinements in the computation of lli if
559: status=-2 in order to have more reliable computation if stepm is
560: not 1 month. Version 0.98f
561:
562: Revision 1.120 2006/03/16 15:10:38 lievre
563: (Module): refinements in the computation of lli if
564: status=-2 in order to have more reliable computation if stepm is
565: not 1 month. Version 0.98f
566:
567: Revision 1.119 2006/03/15 17:42:26 brouard
568: (Module): Bug if status = -2, the loglikelihood was
569: computed as likelihood omitting the logarithm. Version O.98e
570:
571: Revision 1.118 2006/03/14 18:20:07 brouard
572: (Module): varevsij Comments added explaining the second
573: table of variances if popbased=1 .
574: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
575: (Module): Function pstamp added
576: (Module): Version 0.98d
577:
578: Revision 1.117 2006/03/14 17:16:22 brouard
579: (Module): varevsij Comments added explaining the second
580: table of variances if popbased=1 .
581: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
582: (Module): Function pstamp added
583: (Module): Version 0.98d
584:
585: Revision 1.116 2006/03/06 10:29:27 brouard
586: (Module): Variance-covariance wrong links and
587: varian-covariance of ej. is needed (Saito).
588:
589: Revision 1.115 2006/02/27 12:17:45 brouard
590: (Module): One freematrix added in mlikeli! 0.98c
591:
592: Revision 1.114 2006/02/26 12:57:58 brouard
593: (Module): Some improvements in processing parameter
594: filename with strsep.
595:
596: Revision 1.113 2006/02/24 14:20:24 brouard
597: (Module): Memory leaks checks with valgrind and:
598: datafile was not closed, some imatrix were not freed and on matrix
599: allocation too.
600:
601: Revision 1.112 2006/01/30 09:55:26 brouard
602: (Module): Back to gnuplot.exe instead of wgnuplot.exe
603:
604: Revision 1.111 2006/01/25 20:38:18 brouard
605: (Module): Lots of cleaning and bugs added (Gompertz)
606: (Module): Comments can be added in data file. Missing date values
607: can be a simple dot '.'.
608:
609: Revision 1.110 2006/01/25 00:51:50 brouard
610: (Module): Lots of cleaning and bugs added (Gompertz)
611:
612: Revision 1.109 2006/01/24 19:37:15 brouard
613: (Module): Comments (lines starting with a #) are allowed in data.
614:
615: Revision 1.108 2006/01/19 18:05:42 lievre
616: Gnuplot problem appeared...
617: To be fixed
618:
619: Revision 1.107 2006/01/19 16:20:37 brouard
620: Test existence of gnuplot in imach path
621:
622: Revision 1.106 2006/01/19 13:24:36 brouard
623: Some cleaning and links added in html output
624:
625: Revision 1.105 2006/01/05 20:23:19 lievre
626: *** empty log message ***
627:
628: Revision 1.104 2005/09/30 16:11:43 lievre
629: (Module): sump fixed, loop imx fixed, and simplifications.
630: (Module): If the status is missing at the last wave but we know
631: that the person is alive, then we can code his/her status as -2
632: (instead of missing=-1 in earlier versions) and his/her
633: contributions to the likelihood is 1 - Prob of dying from last
634: health status (= 1-p13= p11+p12 in the easiest case of somebody in
635: the healthy state at last known wave). Version is 0.98
636:
637: Revision 1.103 2005/09/30 15:54:49 lievre
638: (Module): sump fixed, loop imx fixed, and simplifications.
639:
640: Revision 1.102 2004/09/15 17:31:30 brouard
641: Add the possibility to read data file including tab characters.
642:
643: Revision 1.101 2004/09/15 10:38:38 brouard
644: Fix on curr_time
645:
646: Revision 1.100 2004/07/12 18:29:06 brouard
647: Add version for Mac OS X. Just define UNIX in Makefile
648:
649: Revision 1.99 2004/06/05 08:57:40 brouard
650: *** empty log message ***
651:
652: Revision 1.98 2004/05/16 15:05:56 brouard
653: New version 0.97 . First attempt to estimate force of mortality
654: directly from the data i.e. without the need of knowing the health
655: state at each age, but using a Gompertz model: log u =a + b*age .
656: This is the basic analysis of mortality and should be done before any
657: other analysis, in order to test if the mortality estimated from the
658: cross-longitudinal survey is different from the mortality estimated
659: from other sources like vital statistic data.
660:
661: The same imach parameter file can be used but the option for mle should be -3.
662:
1.133 brouard 663: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 664: former routines in order to include the new code within the former code.
665:
666: The output is very simple: only an estimate of the intercept and of
667: the slope with 95% confident intervals.
668:
669: Current limitations:
670: A) Even if you enter covariates, i.e. with the
671: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
672: B) There is no computation of Life Expectancy nor Life Table.
673:
674: Revision 1.97 2004/02/20 13:25:42 lievre
675: Version 0.96d. Population forecasting command line is (temporarily)
676: suppressed.
677:
678: Revision 1.96 2003/07/15 15:38:55 brouard
679: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
680: rewritten within the same printf. Workaround: many printfs.
681:
682: Revision 1.95 2003/07/08 07:54:34 brouard
683: * imach.c (Repository):
684: (Repository): Using imachwizard code to output a more meaningful covariance
685: matrix (cov(a12,c31) instead of numbers.
686:
687: Revision 1.94 2003/06/27 13:00:02 brouard
688: Just cleaning
689:
690: Revision 1.93 2003/06/25 16:33:55 brouard
691: (Module): On windows (cygwin) function asctime_r doesn't
692: exist so I changed back to asctime which exists.
693: (Module): Version 0.96b
694:
695: Revision 1.92 2003/06/25 16:30:45 brouard
696: (Module): On windows (cygwin) function asctime_r doesn't
697: exist so I changed back to asctime which exists.
698:
699: Revision 1.91 2003/06/25 15:30:29 brouard
700: * imach.c (Repository): Duplicated warning errors corrected.
701: (Repository): Elapsed time after each iteration is now output. It
702: helps to forecast when convergence will be reached. Elapsed time
703: is stamped in powell. We created a new html file for the graphs
704: concerning matrix of covariance. It has extension -cov.htm.
705:
706: Revision 1.90 2003/06/24 12:34:15 brouard
707: (Module): Some bugs corrected for windows. Also, when
708: mle=-1 a template is output in file "or"mypar.txt with the design
709: of the covariance matrix to be input.
710:
711: Revision 1.89 2003/06/24 12:30:52 brouard
712: (Module): Some bugs corrected for windows. Also, when
713: mle=-1 a template is output in file "or"mypar.txt with the design
714: of the covariance matrix to be input.
715:
716: Revision 1.88 2003/06/23 17:54:56 brouard
717: * 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.
718:
719: Revision 1.87 2003/06/18 12:26:01 brouard
720: Version 0.96
721:
722: Revision 1.86 2003/06/17 20:04:08 brouard
723: (Module): Change position of html and gnuplot routines and added
724: routine fileappend.
725:
726: Revision 1.85 2003/06/17 13:12:43 brouard
727: * imach.c (Repository): Check when date of death was earlier that
728: current date of interview. It may happen when the death was just
729: prior to the death. In this case, dh was negative and likelihood
730: was wrong (infinity). We still send an "Error" but patch by
731: assuming that the date of death was just one stepm after the
732: interview.
733: (Repository): Because some people have very long ID (first column)
734: we changed int to long in num[] and we added a new lvector for
735: memory allocation. But we also truncated to 8 characters (left
736: truncation)
737: (Repository): No more line truncation errors.
738:
739: Revision 1.84 2003/06/13 21:44:43 brouard
740: * imach.c (Repository): Replace "freqsummary" at a correct
741: place. It differs from routine "prevalence" which may be called
742: many times. Probs is memory consuming and must be used with
743: parcimony.
744: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
745:
746: Revision 1.83 2003/06/10 13:39:11 lievre
747: *** empty log message ***
748:
749: Revision 1.82 2003/06/05 15:57:20 brouard
750: Add log in imach.c and fullversion number is now printed.
751:
752: */
753: /*
754: Interpolated Markov Chain
755:
756: Short summary of the programme:
757:
1.227 brouard 758: This program computes Healthy Life Expectancies or State-specific
759: (if states aren't health statuses) Expectancies from
760: cross-longitudinal data. Cross-longitudinal data consist in:
761:
762: -1- a first survey ("cross") where individuals from different ages
763: are interviewed on their health status or degree of disability (in
764: the case of a health survey which is our main interest)
765:
766: -2- at least a second wave of interviews ("longitudinal") which
767: measure each change (if any) in individual health status. Health
768: expectancies are computed from the time spent in each health state
769: according to a model. More health states you consider, more time is
770: necessary to reach the Maximum Likelihood of the parameters involved
771: in the model. The simplest model is the multinomial logistic model
772: where pij is the probability to be observed in state j at the second
773: wave conditional to be observed in state i at the first
774: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
775: etc , where 'age' is age and 'sex' is a covariate. If you want to
776: have a more complex model than "constant and age", you should modify
777: the program where the markup *Covariates have to be included here
778: again* invites you to do it. More covariates you add, slower the
1.126 brouard 779: convergence.
780:
781: The advantage of this computer programme, compared to a simple
782: multinomial logistic model, is clear when the delay between waves is not
783: identical for each individual. Also, if a individual missed an
784: intermediate interview, the information is lost, but taken into
785: account using an interpolation or extrapolation.
786:
787: hPijx is the probability to be observed in state i at age x+h
788: conditional to the observed state i at age x. The delay 'h' can be
789: split into an exact number (nh*stepm) of unobserved intermediate
790: states. This elementary transition (by month, quarter,
791: semester or year) is modelled as a multinomial logistic. The hPx
792: matrix is simply the matrix product of nh*stepm elementary matrices
793: and the contribution of each individual to the likelihood is simply
794: hPijx.
795:
796: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 797: of the life expectancies. It also computes the period (stable) prevalence.
798:
799: Back prevalence and projections:
1.227 brouard 800:
801: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
802: double agemaxpar, double ftolpl, int *ncvyearp, double
803: dateprev1,double dateprev2, int firstpass, int lastpass, int
804: mobilavproj)
805:
806: Computes the back prevalence limit for any combination of
807: covariate values k at any age between ageminpar and agemaxpar and
808: returns it in **bprlim. In the loops,
809:
810: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
811: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
812:
813: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 814: Computes for any combination of covariates k and any age between bage and fage
815: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
816: oldm=oldms;savm=savms;
1.227 brouard 817:
818: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 819: Computes the transition matrix starting at age 'age' over
820: 'nhstepm*hstepm*stepm' months (i.e. until
821: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 822: nhstepm*hstepm matrices.
823:
824: Returns p3mat[i][j][h] after calling
825: p3mat[i][j][h]=matprod2(newm,
826: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
827: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
828: oldm);
1.226 brouard 829:
830: Important routines
831:
832: - func (or funcone), computes logit (pij) distinguishing
833: o fixed variables (single or product dummies or quantitative);
834: o varying variables by:
835: (1) wave (single, product dummies, quantitative),
836: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
837: % fixed dummy (treated) or quantitative (not done because time-consuming);
838: % varying dummy (not done) or quantitative (not done);
839: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
840: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
841: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
842: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
843: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 844:
1.226 brouard 845:
846:
1.133 brouard 847: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
848: Institut national d'études démographiques, Paris.
1.126 brouard 849: This software have been partly granted by Euro-REVES, a concerted action
850: from the European Union.
851: It is copyrighted identically to a GNU software product, ie programme and
852: software can be distributed freely for non commercial use. Latest version
853: can be accessed at http://euroreves.ined.fr/imach .
854:
855: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
856: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
857:
858: **********************************************************************/
859: /*
860: main
861: read parameterfile
862: read datafile
863: concatwav
864: freqsummary
865: if (mle >= 1)
866: mlikeli
867: print results files
868: if mle==1
869: computes hessian
870: read end of parameter file: agemin, agemax, bage, fage, estepm
871: begin-prev-date,...
872: open gnuplot file
873: open html file
1.145 brouard 874: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
875: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
876: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
877: freexexit2 possible for memory heap.
878:
879: h Pij x | pij_nom ficrestpij
880: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
881: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
882: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
883:
884: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
885: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
886: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
887: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
888: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
889:
1.126 brouard 890: forecasting if prevfcast==1 prevforecast call prevalence()
891: health expectancies
892: Variance-covariance of DFLE
893: prevalence()
894: movingaverage()
895: varevsij()
896: if popbased==1 varevsij(,popbased)
897: total life expectancies
898: Variance of period (stable) prevalence
899: end
900: */
901:
1.187 brouard 902: /* #define DEBUG */
903: /* #define DEBUGBRENT */
1.203 brouard 904: /* #define DEBUGLINMIN */
905: /* #define DEBUGHESS */
906: #define DEBUGHESSIJ
1.224 brouard 907: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 908: #define POWELL /* Instead of NLOPT */
1.224 brouard 909: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 910: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
911: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 912:
913: #include <math.h>
914: #include <stdio.h>
915: #include <stdlib.h>
916: #include <string.h>
1.226 brouard 917: #include <ctype.h>
1.159 brouard 918:
919: #ifdef _WIN32
920: #include <io.h>
1.172 brouard 921: #include <windows.h>
922: #include <tchar.h>
1.159 brouard 923: #else
1.126 brouard 924: #include <unistd.h>
1.159 brouard 925: #endif
1.126 brouard 926:
927: #include <limits.h>
928: #include <sys/types.h>
1.171 brouard 929:
930: #if defined(__GNUC__)
931: #include <sys/utsname.h> /* Doesn't work on Windows */
932: #endif
933:
1.126 brouard 934: #include <sys/stat.h>
935: #include <errno.h>
1.159 brouard 936: /* extern int errno; */
1.126 brouard 937:
1.157 brouard 938: /* #ifdef LINUX */
939: /* #include <time.h> */
940: /* #include "timeval.h" */
941: /* #else */
942: /* #include <sys/time.h> */
943: /* #endif */
944:
1.126 brouard 945: #include <time.h>
946:
1.136 brouard 947: #ifdef GSL
948: #include <gsl/gsl_errno.h>
949: #include <gsl/gsl_multimin.h>
950: #endif
951:
1.167 brouard 952:
1.162 brouard 953: #ifdef NLOPT
954: #include <nlopt.h>
955: typedef struct {
956: double (* function)(double [] );
957: } myfunc_data ;
958: #endif
959:
1.126 brouard 960: /* #include <libintl.h> */
961: /* #define _(String) gettext (String) */
962:
1.251 brouard 963: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 964:
965: #define GNUPLOTPROGRAM "gnuplot"
966: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
967: #define FILENAMELENGTH 132
968:
969: #define GLOCK_ERROR_NOPATH -1 /* empty path */
970: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
971:
1.144 brouard 972: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
973: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 974:
975: #define NINTERVMAX 8
1.144 brouard 976: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
977: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
978: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 979: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 980: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
981: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 982: #define MAXN 20000
1.144 brouard 983: #define YEARM 12. /**< Number of months per year */
1.218 brouard 984: /* #define AGESUP 130 */
985: #define AGESUP 150
986: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 987: #define AGEBASE 40
1.194 brouard 988: #define AGEOVERFLOW 1.e20
1.164 brouard 989: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 990: #ifdef _WIN32
991: #define DIRSEPARATOR '\\'
992: #define CHARSEPARATOR "\\"
993: #define ODIRSEPARATOR '/'
994: #else
1.126 brouard 995: #define DIRSEPARATOR '/'
996: #define CHARSEPARATOR "/"
997: #define ODIRSEPARATOR '\\'
998: #endif
999:
1.266 ! brouard 1000: /* $Id: imach.c,v 1.265 2017/04/26 16:22:11 brouard Exp $ */
1.126 brouard 1001: /* $State: Exp $ */
1.196 brouard 1002: #include "version.h"
1003: char version[]=__IMACH_VERSION__;
1.224 brouard 1004: 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";
1.266 ! brouard 1005: char fullversion[]="$Revision: 1.265 $ $Date: 2017/04/26 16:22:11 $";
1.126 brouard 1006: char strstart[80];
1007: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1008: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1009: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1010: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1011: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1012: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1013: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1014: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1015: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1016: int cptcovprodnoage=0; /**< Number of covariate products without age */
1017: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1018: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1019: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1020: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1021: int nsd=0; /**< Total number of single dummy variables (output) */
1022: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1023: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1024: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1025: int ntveff=0; /**< ntveff number of effective time varying variables */
1026: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1027: int cptcov=0; /* Working variable */
1.218 brouard 1028: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1029: int npar=NPARMAX;
1030: int nlstate=2; /* Number of live states */
1031: int ndeath=1; /* Number of dead states */
1.130 brouard 1032: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1033: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1034: int popbased=0;
1035:
1036: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1037: int maxwav=0; /* Maxim number of waves */
1038: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1039: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1040: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1041: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1042: int mle=1, weightopt=0;
1.126 brouard 1043: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1044: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1045: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1046: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1047: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1048: int selected(int kvar); /* Is covariate kvar selected for printing results */
1049:
1.130 brouard 1050: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1051: double **matprod2(); /* test */
1.126 brouard 1052: double **oldm, **newm, **savm; /* Working pointers to matrices */
1053: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1054: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1055:
1.136 brouard 1056: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1057: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1058: FILE *ficlog, *ficrespow;
1.130 brouard 1059: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1060: double fretone; /* Only one call to likelihood */
1.130 brouard 1061: long ipmx=0; /* Number of contributions */
1.126 brouard 1062: double sw; /* Sum of weights */
1063: char filerespow[FILENAMELENGTH];
1064: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1065: FILE *ficresilk;
1066: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1067: FILE *ficresprobmorprev;
1068: FILE *fichtm, *fichtmcov; /* Html File */
1069: FILE *ficreseij;
1070: char filerese[FILENAMELENGTH];
1071: FILE *ficresstdeij;
1072: char fileresstde[FILENAMELENGTH];
1073: FILE *ficrescveij;
1074: char filerescve[FILENAMELENGTH];
1075: FILE *ficresvij;
1076: char fileresv[FILENAMELENGTH];
1077: FILE *ficresvpl;
1078: char fileresvpl[FILENAMELENGTH];
1079: char title[MAXLINE];
1.234 brouard 1080: char model[MAXLINE]; /**< The model line */
1.217 brouard 1081: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1082: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1083: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1084: char command[FILENAMELENGTH];
1085: int outcmd=0;
1086:
1.217 brouard 1087: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1088: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1089: char filelog[FILENAMELENGTH]; /* Log file */
1090: char filerest[FILENAMELENGTH];
1091: char fileregp[FILENAMELENGTH];
1092: char popfile[FILENAMELENGTH];
1093:
1094: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1095:
1.157 brouard 1096: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1097: /* struct timezone tzp; */
1098: /* extern int gettimeofday(); */
1099: struct tm tml, *gmtime(), *localtime();
1100:
1101: extern time_t time();
1102:
1103: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1104: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1105: struct tm tm;
1106:
1.126 brouard 1107: char strcurr[80], strfor[80];
1108:
1109: char *endptr;
1110: long lval;
1111: double dval;
1112:
1113: #define NR_END 1
1114: #define FREE_ARG char*
1115: #define FTOL 1.0e-10
1116:
1117: #define NRANSI
1.240 brouard 1118: #define ITMAX 200
1119: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1120:
1121: #define TOL 2.0e-4
1122:
1123: #define CGOLD 0.3819660
1124: #define ZEPS 1.0e-10
1125: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1126:
1127: #define GOLD 1.618034
1128: #define GLIMIT 100.0
1129: #define TINY 1.0e-20
1130:
1131: static double maxarg1,maxarg2;
1132: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1133: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1134:
1135: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1136: #define rint(a) floor(a+0.5)
1.166 brouard 1137: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1138: #define mytinydouble 1.0e-16
1.166 brouard 1139: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1140: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1141: /* static double dsqrarg; */
1142: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1143: static double sqrarg;
1144: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1145: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1146: int agegomp= AGEGOMP;
1147:
1148: int imx;
1149: int stepm=1;
1150: /* Stepm, step in month: minimum step interpolation*/
1151:
1152: int estepm;
1153: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1154:
1155: int m,nb;
1156: long *num;
1.197 brouard 1157: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1158: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1159: covariate for which somebody answered excluding
1160: undefined. Usually 2: 0 and 1. */
1161: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1162: covariate for which somebody answered including
1163: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1164: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1165: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1166: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1167: double *ageexmed,*agecens;
1168: double dateintmean=0;
1169:
1170: double *weight;
1171: int **s; /* Status */
1.141 brouard 1172: double *agedc;
1.145 brouard 1173: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1174: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1175: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1176: double **coqvar; /* Fixed quantitative covariate iqv */
1177: double ***cotvar; /* Time varying covariate itv */
1178: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1179: double idx;
1180: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1181: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1182: /*k 1 2 3 4 5 6 7 8 9 */
1183: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1184: /* Tndvar[k] 1 2 3 4 5 */
1185: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1186: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1187: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1188: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1189: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1190: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1191: /* Tprod[i]=k 4 7 */
1192: /* Tage[i]=k 5 8 */
1193: /* */
1194: /* Type */
1195: /* V 1 2 3 4 5 */
1196: /* F F V V V */
1197: /* D Q D D Q */
1198: /* */
1199: int *TvarsD;
1200: int *TvarsDind;
1201: int *TvarsQ;
1202: int *TvarsQind;
1203:
1.235 brouard 1204: #define MAXRESULTLINES 10
1205: int nresult=0;
1.258 brouard 1206: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1207: int TKresult[MAXRESULTLINES];
1.237 brouard 1208: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1209: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1210: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1211: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1212: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1213: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1214:
1.234 brouard 1215: /* 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 1216: 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 */
1217: 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 */
1218: 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 */
1219: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1220: 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 */
1221: 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 1222: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1223: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1224: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1225: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1226: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1227: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1228: 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 */
1229: 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 */
1230:
1.230 brouard 1231: int *Tvarsel; /**< Selected covariates for output */
1232: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1233: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1234: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1235: 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 1236: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1237: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1238: int *Tage;
1.227 brouard 1239: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1240: 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 1241: 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*/
1242: 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 1243: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1244: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1245: int **Tvard;
1246: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1247: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1248: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1249: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1250: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1251: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1252: double *lsurv, *lpop, *tpop;
1253:
1.231 brouard 1254: #define FD 1; /* Fixed dummy covariate */
1255: #define FQ 2; /* Fixed quantitative covariate */
1256: #define FP 3; /* Fixed product covariate */
1257: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1258: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1259: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1260: #define VD 10; /* Varying dummy covariate */
1261: #define VQ 11; /* Varying quantitative covariate */
1262: #define VP 12; /* Varying product covariate */
1263: #define VPDD 13; /* Varying product dummy*dummy covariate */
1264: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1265: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1266: #define APFD 16; /* Age product * fixed dummy covariate */
1267: #define APFQ 17; /* Age product * fixed quantitative covariate */
1268: #define APVD 18; /* Age product * varying dummy covariate */
1269: #define APVQ 19; /* Age product * varying quantitative covariate */
1270:
1271: #define FTYPE 1; /* Fixed covariate */
1272: #define VTYPE 2; /* Varying covariate (loop in wave) */
1273: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1274:
1275: struct kmodel{
1276: int maintype; /* main type */
1277: int subtype; /* subtype */
1278: };
1279: struct kmodel modell[NCOVMAX];
1280:
1.143 brouard 1281: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1282: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1283:
1284: /**************** split *************************/
1285: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1286: {
1287: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1288: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1289: */
1290: char *ss; /* pointer */
1.186 brouard 1291: int l1=0, l2=0; /* length counters */
1.126 brouard 1292:
1293: l1 = strlen(path ); /* length of path */
1294: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1295: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1296: if ( ss == NULL ) { /* no directory, so determine current directory */
1297: strcpy( name, path ); /* we got the fullname name because no directory */
1298: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1299: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1300: /* get current working directory */
1301: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1302: #ifdef WIN32
1303: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1304: #else
1305: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1306: #endif
1.126 brouard 1307: return( GLOCK_ERROR_GETCWD );
1308: }
1309: /* got dirc from getcwd*/
1310: printf(" DIRC = %s \n",dirc);
1.205 brouard 1311: } else { /* strip directory from path */
1.126 brouard 1312: ss++; /* after this, the filename */
1313: l2 = strlen( ss ); /* length of filename */
1314: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1315: strcpy( name, ss ); /* save file name */
1316: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1317: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1318: printf(" DIRC2 = %s \n",dirc);
1319: }
1320: /* We add a separator at the end of dirc if not exists */
1321: l1 = strlen( dirc ); /* length of directory */
1322: if( dirc[l1-1] != DIRSEPARATOR ){
1323: dirc[l1] = DIRSEPARATOR;
1324: dirc[l1+1] = 0;
1325: printf(" DIRC3 = %s \n",dirc);
1326: }
1327: ss = strrchr( name, '.' ); /* find last / */
1328: if (ss >0){
1329: ss++;
1330: strcpy(ext,ss); /* save extension */
1331: l1= strlen( name);
1332: l2= strlen(ss)+1;
1333: strncpy( finame, name, l1-l2);
1334: finame[l1-l2]= 0;
1335: }
1336:
1337: return( 0 ); /* we're done */
1338: }
1339:
1340:
1341: /******************************************/
1342:
1343: void replace_back_to_slash(char *s, char*t)
1344: {
1345: int i;
1346: int lg=0;
1347: i=0;
1348: lg=strlen(t);
1349: for(i=0; i<= lg; i++) {
1350: (s[i] = t[i]);
1351: if (t[i]== '\\') s[i]='/';
1352: }
1353: }
1354:
1.132 brouard 1355: char *trimbb(char *out, char *in)
1.137 brouard 1356: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1357: char *s;
1358: s=out;
1359: while (*in != '\0'){
1.137 brouard 1360: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1361: in++;
1362: }
1363: *out++ = *in++;
1364: }
1365: *out='\0';
1366: return s;
1367: }
1368:
1.187 brouard 1369: /* char *substrchaine(char *out, char *in, char *chain) */
1370: /* { */
1371: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1372: /* char *s, *t; */
1373: /* t=in;s=out; */
1374: /* while ((*in != *chain) && (*in != '\0')){ */
1375: /* *out++ = *in++; */
1376: /* } */
1377:
1378: /* /\* *in matches *chain *\/ */
1379: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1380: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1381: /* } */
1382: /* in--; chain--; */
1383: /* while ( (*in != '\0')){ */
1384: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1385: /* *out++ = *in++; */
1386: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1387: /* } */
1388: /* *out='\0'; */
1389: /* out=s; */
1390: /* return out; */
1391: /* } */
1392: char *substrchaine(char *out, char *in, char *chain)
1393: {
1394: /* Substract chain 'chain' from 'in', return and output 'out' */
1395: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1396:
1397: char *strloc;
1398:
1399: strcpy (out, in);
1400: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1401: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1402: if(strloc != NULL){
1403: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1404: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1405: /* strcpy (strloc, strloc +strlen(chain));*/
1406: }
1407: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1408: return out;
1409: }
1410:
1411:
1.145 brouard 1412: char *cutl(char *blocc, char *alocc, char *in, char occ)
1413: {
1.187 brouard 1414: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1415: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1416: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1417: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1418: */
1.160 brouard 1419: char *s, *t;
1.145 brouard 1420: t=in;s=in;
1421: while ((*in != occ) && (*in != '\0')){
1422: *alocc++ = *in++;
1423: }
1424: if( *in == occ){
1425: *(alocc)='\0';
1426: s=++in;
1427: }
1428:
1429: if (s == t) {/* occ not found */
1430: *(alocc-(in-s))='\0';
1431: in=s;
1432: }
1433: while ( *in != '\0'){
1434: *blocc++ = *in++;
1435: }
1436:
1437: *blocc='\0';
1438: return t;
1439: }
1.137 brouard 1440: char *cutv(char *blocc, char *alocc, char *in, char occ)
1441: {
1.187 brouard 1442: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1443: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1444: gives blocc="abcdef2ghi" and alocc="j".
1445: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1446: */
1447: char *s, *t;
1448: t=in;s=in;
1449: while (*in != '\0'){
1450: while( *in == occ){
1451: *blocc++ = *in++;
1452: s=in;
1453: }
1454: *blocc++ = *in++;
1455: }
1456: if (s == t) /* occ not found */
1457: *(blocc-(in-s))='\0';
1458: else
1459: *(blocc-(in-s)-1)='\0';
1460: in=s;
1461: while ( *in != '\0'){
1462: *alocc++ = *in++;
1463: }
1464:
1465: *alocc='\0';
1466: return s;
1467: }
1468:
1.126 brouard 1469: int nbocc(char *s, char occ)
1470: {
1471: int i,j=0;
1472: int lg=20;
1473: i=0;
1474: lg=strlen(s);
1475: for(i=0; i<= lg; i++) {
1.234 brouard 1476: if (s[i] == occ ) j++;
1.126 brouard 1477: }
1478: return j;
1479: }
1480:
1.137 brouard 1481: /* void cutv(char *u,char *v, char*t, char occ) */
1482: /* { */
1483: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1484: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1485: /* gives u="abcdef2ghi" and v="j" *\/ */
1486: /* int i,lg,j,p=0; */
1487: /* i=0; */
1488: /* lg=strlen(t); */
1489: /* for(j=0; j<=lg-1; j++) { */
1490: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1491: /* } */
1.126 brouard 1492:
1.137 brouard 1493: /* for(j=0; j<p; j++) { */
1494: /* (u[j] = t[j]); */
1495: /* } */
1496: /* u[p]='\0'; */
1.126 brouard 1497:
1.137 brouard 1498: /* for(j=0; j<= lg; j++) { */
1499: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1500: /* } */
1501: /* } */
1.126 brouard 1502:
1.160 brouard 1503: #ifdef _WIN32
1504: char * strsep(char **pp, const char *delim)
1505: {
1506: char *p, *q;
1507:
1508: if ((p = *pp) == NULL)
1509: return 0;
1510: if ((q = strpbrk (p, delim)) != NULL)
1511: {
1512: *pp = q + 1;
1513: *q = '\0';
1514: }
1515: else
1516: *pp = 0;
1517: return p;
1518: }
1519: #endif
1520:
1.126 brouard 1521: /********************** nrerror ********************/
1522:
1523: void nrerror(char error_text[])
1524: {
1525: fprintf(stderr,"ERREUR ...\n");
1526: fprintf(stderr,"%s\n",error_text);
1527: exit(EXIT_FAILURE);
1528: }
1529: /*********************** vector *******************/
1530: double *vector(int nl, int nh)
1531: {
1532: double *v;
1533: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1534: if (!v) nrerror("allocation failure in vector");
1535: return v-nl+NR_END;
1536: }
1537:
1538: /************************ free vector ******************/
1539: void free_vector(double*v, int nl, int nh)
1540: {
1541: free((FREE_ARG)(v+nl-NR_END));
1542: }
1543:
1544: /************************ivector *******************************/
1545: int *ivector(long nl,long nh)
1546: {
1547: int *v;
1548: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1549: if (!v) nrerror("allocation failure in ivector");
1550: return v-nl+NR_END;
1551: }
1552:
1553: /******************free ivector **************************/
1554: void free_ivector(int *v, long nl, long nh)
1555: {
1556: free((FREE_ARG)(v+nl-NR_END));
1557: }
1558:
1559: /************************lvector *******************************/
1560: long *lvector(long nl,long nh)
1561: {
1562: long *v;
1563: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1564: if (!v) nrerror("allocation failure in ivector");
1565: return v-nl+NR_END;
1566: }
1567:
1568: /******************free lvector **************************/
1569: void free_lvector(long *v, long nl, long nh)
1570: {
1571: free((FREE_ARG)(v+nl-NR_END));
1572: }
1573:
1574: /******************* imatrix *******************************/
1575: int **imatrix(long nrl, long nrh, long ncl, long nch)
1576: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1577: {
1578: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1579: int **m;
1580:
1581: /* allocate pointers to rows */
1582: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1583: if (!m) nrerror("allocation failure 1 in matrix()");
1584: m += NR_END;
1585: m -= nrl;
1586:
1587:
1588: /* allocate rows and set pointers to them */
1589: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1590: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1591: m[nrl] += NR_END;
1592: m[nrl] -= ncl;
1593:
1594: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1595:
1596: /* return pointer to array of pointers to rows */
1597: return m;
1598: }
1599:
1600: /****************** free_imatrix *************************/
1601: void free_imatrix(m,nrl,nrh,ncl,nch)
1602: int **m;
1603: long nch,ncl,nrh,nrl;
1604: /* free an int matrix allocated by imatrix() */
1605: {
1606: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1607: free((FREE_ARG) (m+nrl-NR_END));
1608: }
1609:
1610: /******************* matrix *******************************/
1611: double **matrix(long nrl, long nrh, long ncl, long nch)
1612: {
1613: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1614: double **m;
1615:
1616: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1617: if (!m) nrerror("allocation failure 1 in matrix()");
1618: m += NR_END;
1619: m -= nrl;
1620:
1621: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1622: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1623: m[nrl] += NR_END;
1624: m[nrl] -= ncl;
1625:
1626: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1627: return m;
1.145 brouard 1628: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1629: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1630: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1631: */
1632: }
1633:
1634: /*************************free matrix ************************/
1635: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1636: {
1637: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1638: free((FREE_ARG)(m+nrl-NR_END));
1639: }
1640:
1641: /******************* ma3x *******************************/
1642: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1643: {
1644: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1645: double ***m;
1646:
1647: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1648: if (!m) nrerror("allocation failure 1 in matrix()");
1649: m += NR_END;
1650: m -= nrl;
1651:
1652: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1653: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1654: m[nrl] += NR_END;
1655: m[nrl] -= ncl;
1656:
1657: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1658:
1659: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1660: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1661: m[nrl][ncl] += NR_END;
1662: m[nrl][ncl] -= nll;
1663: for (j=ncl+1; j<=nch; j++)
1664: m[nrl][j]=m[nrl][j-1]+nlay;
1665:
1666: for (i=nrl+1; i<=nrh; i++) {
1667: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1668: for (j=ncl+1; j<=nch; j++)
1669: m[i][j]=m[i][j-1]+nlay;
1670: }
1671: return m;
1672: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1673: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1674: */
1675: }
1676:
1677: /*************************free ma3x ************************/
1678: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1679: {
1680: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1681: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1682: free((FREE_ARG)(m+nrl-NR_END));
1683: }
1684:
1685: /*************** function subdirf ***********/
1686: char *subdirf(char fileres[])
1687: {
1688: /* Caution optionfilefiname is hidden */
1689: strcpy(tmpout,optionfilefiname);
1690: strcat(tmpout,"/"); /* Add to the right */
1691: strcat(tmpout,fileres);
1692: return tmpout;
1693: }
1694:
1695: /*************** function subdirf2 ***********/
1696: char *subdirf2(char fileres[], char *preop)
1697: {
1698:
1699: /* Caution optionfilefiname is hidden */
1700: strcpy(tmpout,optionfilefiname);
1701: strcat(tmpout,"/");
1702: strcat(tmpout,preop);
1703: strcat(tmpout,fileres);
1704: return tmpout;
1705: }
1706:
1707: /*************** function subdirf3 ***********/
1708: char *subdirf3(char fileres[], char *preop, char *preop2)
1709: {
1710:
1711: /* Caution optionfilefiname is hidden */
1712: strcpy(tmpout,optionfilefiname);
1713: strcat(tmpout,"/");
1714: strcat(tmpout,preop);
1715: strcat(tmpout,preop2);
1716: strcat(tmpout,fileres);
1717: return tmpout;
1718: }
1.213 brouard 1719:
1720: /*************** function subdirfext ***********/
1721: char *subdirfext(char fileres[], char *preop, char *postop)
1722: {
1723:
1724: strcpy(tmpout,preop);
1725: strcat(tmpout,fileres);
1726: strcat(tmpout,postop);
1727: return tmpout;
1728: }
1.126 brouard 1729:
1.213 brouard 1730: /*************** function subdirfext3 ***********/
1731: char *subdirfext3(char fileres[], char *preop, char *postop)
1732: {
1733:
1734: /* Caution optionfilefiname is hidden */
1735: strcpy(tmpout,optionfilefiname);
1736: strcat(tmpout,"/");
1737: strcat(tmpout,preop);
1738: strcat(tmpout,fileres);
1739: strcat(tmpout,postop);
1740: return tmpout;
1741: }
1742:
1.162 brouard 1743: char *asc_diff_time(long time_sec, char ascdiff[])
1744: {
1745: long sec_left, days, hours, minutes;
1746: days = (time_sec) / (60*60*24);
1747: sec_left = (time_sec) % (60*60*24);
1748: hours = (sec_left) / (60*60) ;
1749: sec_left = (sec_left) %(60*60);
1750: minutes = (sec_left) /60;
1751: sec_left = (sec_left) % (60);
1752: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1753: return ascdiff;
1754: }
1755:
1.126 brouard 1756: /***************** f1dim *************************/
1757: extern int ncom;
1758: extern double *pcom,*xicom;
1759: extern double (*nrfunc)(double []);
1760:
1761: double f1dim(double x)
1762: {
1763: int j;
1764: double f;
1765: double *xt;
1766:
1767: xt=vector(1,ncom);
1768: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1769: f=(*nrfunc)(xt);
1770: free_vector(xt,1,ncom);
1771: return f;
1772: }
1773:
1774: /*****************brent *************************/
1775: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1776: {
1777: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1778: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1779: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1780: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1781: * returned function value.
1782: */
1.126 brouard 1783: int iter;
1784: double a,b,d,etemp;
1.159 brouard 1785: double fu=0,fv,fw,fx;
1.164 brouard 1786: double ftemp=0.;
1.126 brouard 1787: double p,q,r,tol1,tol2,u,v,w,x,xm;
1788: double e=0.0;
1789:
1790: a=(ax < cx ? ax : cx);
1791: b=(ax > cx ? ax : cx);
1792: x=w=v=bx;
1793: fw=fv=fx=(*f)(x);
1794: for (iter=1;iter<=ITMAX;iter++) {
1795: xm=0.5*(a+b);
1796: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1797: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1798: printf(".");fflush(stdout);
1799: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1800: #ifdef DEBUGBRENT
1.126 brouard 1801: 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);
1802: 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);
1803: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1804: #endif
1805: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1806: *xmin=x;
1807: return fx;
1808: }
1809: ftemp=fu;
1810: if (fabs(e) > tol1) {
1811: r=(x-w)*(fx-fv);
1812: q=(x-v)*(fx-fw);
1813: p=(x-v)*q-(x-w)*r;
1814: q=2.0*(q-r);
1815: if (q > 0.0) p = -p;
1816: q=fabs(q);
1817: etemp=e;
1818: e=d;
1819: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1820: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1821: else {
1.224 brouard 1822: d=p/q;
1823: u=x+d;
1824: if (u-a < tol2 || b-u < tol2)
1825: d=SIGN(tol1,xm-x);
1.126 brouard 1826: }
1827: } else {
1828: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1829: }
1830: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1831: fu=(*f)(u);
1832: if (fu <= fx) {
1833: if (u >= x) a=x; else b=x;
1834: SHFT(v,w,x,u)
1.183 brouard 1835: SHFT(fv,fw,fx,fu)
1836: } else {
1837: if (u < x) a=u; else b=u;
1838: if (fu <= fw || w == x) {
1.224 brouard 1839: v=w;
1840: w=u;
1841: fv=fw;
1842: fw=fu;
1.183 brouard 1843: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1844: v=u;
1845: fv=fu;
1.183 brouard 1846: }
1847: }
1.126 brouard 1848: }
1849: nrerror("Too many iterations in brent");
1850: *xmin=x;
1851: return fx;
1852: }
1853:
1854: /****************** mnbrak ***********************/
1855:
1856: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1857: double (*func)(double))
1.183 brouard 1858: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1859: the downhill direction (defined by the function as evaluated at the initial points) and returns
1860: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1861: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1862: */
1.126 brouard 1863: double ulim,u,r,q, dum;
1864: double fu;
1.187 brouard 1865:
1866: double scale=10.;
1867: int iterscale=0;
1868:
1869: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1870: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1871:
1872:
1873: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1874: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1875: /* *bx = *ax - (*ax - *bx)/scale; */
1876: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1877: /* } */
1878:
1.126 brouard 1879: if (*fb > *fa) {
1880: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1881: SHFT(dum,*fb,*fa,dum)
1882: }
1.126 brouard 1883: *cx=(*bx)+GOLD*(*bx-*ax);
1884: *fc=(*func)(*cx);
1.183 brouard 1885: #ifdef DEBUG
1.224 brouard 1886: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1887: 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 1888: #endif
1.224 brouard 1889: 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 1890: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1891: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1892: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1893: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1894: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1895: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1896: fu=(*func)(u);
1.163 brouard 1897: #ifdef DEBUG
1898: /* f(x)=A(x-u)**2+f(u) */
1899: double A, fparabu;
1900: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1901: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1902: 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);
1903: 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 1904: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1905: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1906: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1907: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1908: #endif
1.184 brouard 1909: #ifdef MNBRAKORIGINAL
1.183 brouard 1910: #else
1.191 brouard 1911: /* if (fu > *fc) { */
1912: /* #ifdef DEBUG */
1913: /* printf("mnbrak4 fu > fc \n"); */
1914: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1915: /* #endif */
1916: /* /\* 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 *\\/ *\/ */
1917: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1918: /* dum=u; /\* Shifting c and u *\/ */
1919: /* u = *cx; */
1920: /* *cx = dum; */
1921: /* dum = fu; */
1922: /* fu = *fc; */
1923: /* *fc =dum; */
1924: /* } else { /\* end *\/ */
1925: /* #ifdef DEBUG */
1926: /* printf("mnbrak3 fu < fc \n"); */
1927: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1928: /* #endif */
1929: /* dum=u; /\* Shifting c and u *\/ */
1930: /* u = *cx; */
1931: /* *cx = dum; */
1932: /* dum = fu; */
1933: /* fu = *fc; */
1934: /* *fc =dum; */
1935: /* } */
1.224 brouard 1936: #ifdef DEBUGMNBRAK
1937: double A, fparabu;
1938: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1939: fparabu= *fa - A*(*ax-u)*(*ax-u);
1940: 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);
1941: 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 1942: #endif
1.191 brouard 1943: dum=u; /* Shifting c and u */
1944: u = *cx;
1945: *cx = dum;
1946: dum = fu;
1947: fu = *fc;
1948: *fc =dum;
1.183 brouard 1949: #endif
1.162 brouard 1950: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1951: #ifdef DEBUG
1.224 brouard 1952: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1953: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1954: #endif
1.126 brouard 1955: fu=(*func)(u);
1956: if (fu < *fc) {
1.183 brouard 1957: #ifdef DEBUG
1.224 brouard 1958: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1959: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1960: #endif
1961: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1962: SHFT(*fb,*fc,fu,(*func)(u))
1963: #ifdef DEBUG
1964: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1965: #endif
1966: }
1.162 brouard 1967: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1968: #ifdef DEBUG
1.224 brouard 1969: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1970: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1971: #endif
1.126 brouard 1972: u=ulim;
1973: fu=(*func)(u);
1.183 brouard 1974: } else { /* u could be left to b (if r > q parabola has a maximum) */
1975: #ifdef DEBUG
1.224 brouard 1976: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1977: 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 1978: #endif
1.126 brouard 1979: u=(*cx)+GOLD*(*cx-*bx);
1980: fu=(*func)(u);
1.224 brouard 1981: #ifdef DEBUG
1982: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1983: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1984: #endif
1.183 brouard 1985: } /* end tests */
1.126 brouard 1986: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1987: SHFT(*fa,*fb,*fc,fu)
1988: #ifdef DEBUG
1.224 brouard 1989: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1990: 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 1991: #endif
1992: } /* 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 1993: }
1994:
1995: /*************** linmin ************************/
1.162 brouard 1996: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1997: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1998: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1999: the value of func at the returned location p . This is actually all accomplished by calling the
2000: routines mnbrak and brent .*/
1.126 brouard 2001: int ncom;
2002: double *pcom,*xicom;
2003: double (*nrfunc)(double []);
2004:
1.224 brouard 2005: #ifdef LINMINORIGINAL
1.126 brouard 2006: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2007: #else
2008: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2009: #endif
1.126 brouard 2010: {
2011: double brent(double ax, double bx, double cx,
2012: double (*f)(double), double tol, double *xmin);
2013: double f1dim(double x);
2014: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2015: double *fc, double (*func)(double));
2016: int j;
2017: double xx,xmin,bx,ax;
2018: double fx,fb,fa;
1.187 brouard 2019:
1.203 brouard 2020: #ifdef LINMINORIGINAL
2021: #else
2022: double scale=10., axs, xxs; /* Scale added for infinity */
2023: #endif
2024:
1.126 brouard 2025: ncom=n;
2026: pcom=vector(1,n);
2027: xicom=vector(1,n);
2028: nrfunc=func;
2029: for (j=1;j<=n;j++) {
2030: pcom[j]=p[j];
1.202 brouard 2031: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2032: }
1.187 brouard 2033:
1.203 brouard 2034: #ifdef LINMINORIGINAL
2035: xx=1.;
2036: #else
2037: axs=0.0;
2038: xxs=1.;
2039: do{
2040: xx= xxs;
2041: #endif
1.187 brouard 2042: ax=0.;
2043: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2044: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2045: /* 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)) */
2046: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2047: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2048: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2049: /* 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 2050: #ifdef LINMINORIGINAL
2051: #else
2052: if (fx != fx){
1.224 brouard 2053: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2054: printf("|");
2055: fprintf(ficlog,"|");
1.203 brouard 2056: #ifdef DEBUGLINMIN
1.224 brouard 2057: 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 2058: #endif
2059: }
1.224 brouard 2060: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2061: #endif
2062:
1.191 brouard 2063: #ifdef DEBUGLINMIN
2064: 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 2065: 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 2066: #endif
1.224 brouard 2067: #ifdef LINMINORIGINAL
2068: #else
2069: if(fb == fx){ /* Flat function in the direction */
2070: xmin=xx;
2071: *flat=1;
2072: }else{
2073: *flat=0;
2074: #endif
2075: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2076: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2077: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2078: /* fmin = f(p[j] + xmin * xi[j]) */
2079: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2080: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2081: #ifdef DEBUG
1.224 brouard 2082: 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);
2083: 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);
2084: #endif
2085: #ifdef LINMINORIGINAL
2086: #else
2087: }
1.126 brouard 2088: #endif
1.191 brouard 2089: #ifdef DEBUGLINMIN
2090: printf("linmin end ");
1.202 brouard 2091: fprintf(ficlog,"linmin end ");
1.191 brouard 2092: #endif
1.126 brouard 2093: for (j=1;j<=n;j++) {
1.203 brouard 2094: #ifdef LINMINORIGINAL
2095: xi[j] *= xmin;
2096: #else
2097: #ifdef DEBUGLINMIN
2098: if(xxs <1.0)
2099: printf(" before xi[%d]=%12.8f", j,xi[j]);
2100: #endif
2101: 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) */
2102: #ifdef DEBUGLINMIN
2103: if(xxs <1.0)
2104: 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 );
2105: #endif
2106: #endif
1.187 brouard 2107: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2108: }
1.191 brouard 2109: #ifdef DEBUGLINMIN
1.203 brouard 2110: printf("\n");
1.191 brouard 2111: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2112: 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 2113: for (j=1;j<=n;j++) {
1.202 brouard 2114: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2115: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2116: if(j % ncovmodel == 0){
1.191 brouard 2117: printf("\n");
1.202 brouard 2118: fprintf(ficlog,"\n");
2119: }
1.191 brouard 2120: }
1.203 brouard 2121: #else
1.191 brouard 2122: #endif
1.126 brouard 2123: free_vector(xicom,1,n);
2124: free_vector(pcom,1,n);
2125: }
2126:
2127:
2128: /*************** powell ************************/
1.162 brouard 2129: /*
2130: Minimization of a function func of n variables. Input consists of an initial starting point
2131: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2132: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2133: such that failure to decrease by more than this amount on one iteration signals doneness. On
2134: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2135: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2136: */
1.224 brouard 2137: #ifdef LINMINORIGINAL
2138: #else
2139: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2140: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2141: #endif
1.126 brouard 2142: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2143: double (*func)(double []))
2144: {
1.224 brouard 2145: #ifdef LINMINORIGINAL
2146: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2147: double (*func)(double []));
1.224 brouard 2148: #else
1.241 brouard 2149: void linmin(double p[], double xi[], int n, double *fret,
2150: double (*func)(double []),int *flat);
1.224 brouard 2151: #endif
1.239 brouard 2152: int i,ibig,j,jk,k;
1.126 brouard 2153: double del,t,*pt,*ptt,*xit;
1.181 brouard 2154: double directest;
1.126 brouard 2155: double fp,fptt;
2156: double *xits;
2157: int niterf, itmp;
1.224 brouard 2158: #ifdef LINMINORIGINAL
2159: #else
2160:
2161: flatdir=ivector(1,n);
2162: for (j=1;j<=n;j++) flatdir[j]=0;
2163: #endif
1.126 brouard 2164:
2165: pt=vector(1,n);
2166: ptt=vector(1,n);
2167: xit=vector(1,n);
2168: xits=vector(1,n);
2169: *fret=(*func)(p);
2170: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2171: rcurr_time = time(NULL);
1.126 brouard 2172: for (*iter=1;;++(*iter)) {
1.187 brouard 2173: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2174: ibig=0;
2175: del=0.0;
1.157 brouard 2176: rlast_time=rcurr_time;
2177: /* (void) gettimeofday(&curr_time,&tzp); */
2178: rcurr_time = time(NULL);
2179: curr_time = *localtime(&rcurr_time);
2180: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2181: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2182: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2183: for (i=1;i<=n;i++) {
1.126 brouard 2184: fprintf(ficrespow," %.12lf", p[i]);
2185: }
1.239 brouard 2186: fprintf(ficrespow,"\n");fflush(ficrespow);
2187: printf("\n#model= 1 + age ");
2188: fprintf(ficlog,"\n#model= 1 + age ");
2189: if(nagesqr==1){
1.241 brouard 2190: printf(" + age*age ");
2191: fprintf(ficlog," + age*age ");
1.239 brouard 2192: }
2193: for(j=1;j <=ncovmodel-2;j++){
2194: if(Typevar[j]==0) {
2195: printf(" + V%d ",Tvar[j]);
2196: fprintf(ficlog," + V%d ",Tvar[j]);
2197: }else if(Typevar[j]==1) {
2198: printf(" + V%d*age ",Tvar[j]);
2199: fprintf(ficlog," + V%d*age ",Tvar[j]);
2200: }else if(Typevar[j]==2) {
2201: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2202: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2203: }
2204: }
1.126 brouard 2205: printf("\n");
1.239 brouard 2206: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2207: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2208: fprintf(ficlog,"\n");
1.239 brouard 2209: for(i=1,jk=1; i <=nlstate; i++){
2210: for(k=1; k <=(nlstate+ndeath); k++){
2211: if (k != i) {
2212: printf("%d%d ",i,k);
2213: fprintf(ficlog,"%d%d ",i,k);
2214: for(j=1; j <=ncovmodel; j++){
2215: printf("%12.7f ",p[jk]);
2216: fprintf(ficlog,"%12.7f ",p[jk]);
2217: jk++;
2218: }
2219: printf("\n");
2220: fprintf(ficlog,"\n");
2221: }
2222: }
2223: }
1.241 brouard 2224: if(*iter <=3 && *iter >1){
1.157 brouard 2225: tml = *localtime(&rcurr_time);
2226: strcpy(strcurr,asctime(&tml));
2227: rforecast_time=rcurr_time;
1.126 brouard 2228: itmp = strlen(strcurr);
2229: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2230: strcurr[itmp-1]='\0';
1.162 brouard 2231: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2232: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2233: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2234: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2235: forecast_time = *localtime(&rforecast_time);
2236: strcpy(strfor,asctime(&forecast_time));
2237: itmp = strlen(strfor);
2238: if(strfor[itmp-1]=='\n')
2239: strfor[itmp-1]='\0';
2240: 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);
2241: 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 2242: }
2243: }
1.187 brouard 2244: for (i=1;i<=n;i++) { /* For each direction i */
2245: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2246: fptt=(*fret);
2247: #ifdef DEBUG
1.203 brouard 2248: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2249: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2250: #endif
1.203 brouard 2251: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2252: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2253: #ifdef LINMINORIGINAL
1.188 brouard 2254: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2255: #else
2256: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2257: flatdir[i]=flat; /* Function is vanishing in that direction i */
2258: #endif
2259: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2260: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2261: /* because that direction will be replaced unless the gain del is small */
2262: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2263: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2264: /* with the new direction. */
2265: del=fabs(fptt-(*fret));
2266: ibig=i;
1.126 brouard 2267: }
2268: #ifdef DEBUG
2269: printf("%d %.12e",i,(*fret));
2270: fprintf(ficlog,"%d %.12e",i,(*fret));
2271: for (j=1;j<=n;j++) {
1.224 brouard 2272: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2273: printf(" x(%d)=%.12e",j,xit[j]);
2274: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2275: }
2276: for(j=1;j<=n;j++) {
1.225 brouard 2277: printf(" p(%d)=%.12e",j,p[j]);
2278: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2279: }
2280: printf("\n");
2281: fprintf(ficlog,"\n");
2282: #endif
1.187 brouard 2283: } /* end loop on each direction i */
2284: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2285: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2286: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2287: for(j=1;j<=n;j++) {
1.225 brouard 2288: if(flatdir[j] >0){
2289: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2290: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2291: }
2292: /* printf("\n"); */
2293: /* fprintf(ficlog,"\n"); */
2294: }
1.243 brouard 2295: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2296: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2297: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2298: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2299: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2300: /* decreased of more than 3.84 */
2301: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2302: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2303: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2304:
1.188 brouard 2305: /* Starting the program with initial values given by a former maximization will simply change */
2306: /* the scales of the directions and the directions, because the are reset to canonical directions */
2307: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2308: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2309: #ifdef DEBUG
2310: int k[2],l;
2311: k[0]=1;
2312: k[1]=-1;
2313: printf("Max: %.12e",(*func)(p));
2314: fprintf(ficlog,"Max: %.12e",(*func)(p));
2315: for (j=1;j<=n;j++) {
2316: printf(" %.12e",p[j]);
2317: fprintf(ficlog," %.12e",p[j]);
2318: }
2319: printf("\n");
2320: fprintf(ficlog,"\n");
2321: for(l=0;l<=1;l++) {
2322: for (j=1;j<=n;j++) {
2323: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2324: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2325: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2326: }
2327: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2328: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2329: }
2330: #endif
2331:
1.224 brouard 2332: #ifdef LINMINORIGINAL
2333: #else
2334: free_ivector(flatdir,1,n);
2335: #endif
1.126 brouard 2336: free_vector(xit,1,n);
2337: free_vector(xits,1,n);
2338: free_vector(ptt,1,n);
2339: free_vector(pt,1,n);
2340: return;
1.192 brouard 2341: } /* enough precision */
1.240 brouard 2342: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2343: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2344: ptt[j]=2.0*p[j]-pt[j];
2345: xit[j]=p[j]-pt[j];
2346: pt[j]=p[j];
2347: }
1.181 brouard 2348: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2349: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2350: if (*iter <=4) {
1.225 brouard 2351: #else
2352: #endif
1.224 brouard 2353: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2354: #else
1.161 brouard 2355: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2356: #endif
1.162 brouard 2357: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2358: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2359: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2360: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2361: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2362: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2363: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2364: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2365: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2366: /* Even if f3 <f1, directest can be negative and t >0 */
2367: /* mu² and del² are equal when f3=f1 */
2368: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2369: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2370: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2371: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2372: #ifdef NRCORIGINAL
2373: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2374: #else
2375: 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 2376: t= t- del*SQR(fp-fptt);
1.183 brouard 2377: #endif
1.202 brouard 2378: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2379: #ifdef DEBUG
1.181 brouard 2380: 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);
2381: 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 2382: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2383: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2384: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2385: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2386: 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);
2387: 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);
2388: #endif
1.183 brouard 2389: #ifdef POWELLORIGINAL
2390: if (t < 0.0) { /* Then we use it for new direction */
2391: #else
1.182 brouard 2392: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2393: 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 2394: 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 2395: 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 2396: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2397: }
1.181 brouard 2398: if (directest < 0.0) { /* Then we use it for new direction */
2399: #endif
1.191 brouard 2400: #ifdef DEBUGLINMIN
1.234 brouard 2401: printf("Before linmin in direction P%d-P0\n",n);
2402: for (j=1;j<=n;j++) {
2403: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2404: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2405: if(j % ncovmodel == 0){
2406: printf("\n");
2407: fprintf(ficlog,"\n");
2408: }
2409: }
1.224 brouard 2410: #endif
2411: #ifdef LINMINORIGINAL
1.234 brouard 2412: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2413: #else
1.234 brouard 2414: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2415: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2416: #endif
1.234 brouard 2417:
1.191 brouard 2418: #ifdef DEBUGLINMIN
1.234 brouard 2419: for (j=1;j<=n;j++) {
2420: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2421: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2422: if(j % ncovmodel == 0){
2423: printf("\n");
2424: fprintf(ficlog,"\n");
2425: }
2426: }
1.224 brouard 2427: #endif
1.234 brouard 2428: for (j=1;j<=n;j++) {
2429: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2430: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2431: }
1.224 brouard 2432: #ifdef LINMINORIGINAL
2433: #else
1.234 brouard 2434: for (j=1, flatd=0;j<=n;j++) {
2435: if(flatdir[j]>0)
2436: flatd++;
2437: }
2438: if(flatd >0){
1.255 brouard 2439: printf("%d flat directions: ",flatd);
2440: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2441: for (j=1;j<=n;j++) {
2442: if(flatdir[j]>0){
2443: printf("%d ",j);
2444: fprintf(ficlog,"%d ",j);
2445: }
2446: }
2447: printf("\n");
2448: fprintf(ficlog,"\n");
2449: }
1.191 brouard 2450: #endif
1.234 brouard 2451: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2452: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2453:
1.126 brouard 2454: #ifdef DEBUG
1.234 brouard 2455: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2456: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2457: for(j=1;j<=n;j++){
2458: printf(" %lf",xit[j]);
2459: fprintf(ficlog," %lf",xit[j]);
2460: }
2461: printf("\n");
2462: fprintf(ficlog,"\n");
1.126 brouard 2463: #endif
1.192 brouard 2464: } /* end of t or directest negative */
1.224 brouard 2465: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2466: #else
1.234 brouard 2467: } /* end if (fptt < fp) */
1.192 brouard 2468: #endif
1.225 brouard 2469: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2470: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2471: #else
1.224 brouard 2472: #endif
1.234 brouard 2473: } /* loop iteration */
1.126 brouard 2474: }
1.234 brouard 2475:
1.126 brouard 2476: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2477:
1.235 brouard 2478: 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 2479: {
1.235 brouard 2480: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2481: (and selected quantitative values in nres)
2482: by left multiplying the unit
1.234 brouard 2483: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2484: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2485: /* Wx is row vector: population in state 1, population in state 2, population dead */
2486: /* or prevalence in state 1, prevalence in state 2, 0 */
2487: /* newm is the matrix after multiplications, its rows are identical at a factor */
2488: /* Initial matrix pimij */
2489: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2490: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2491: /* 0, 0 , 1} */
2492: /*
2493: * and after some iteration: */
2494: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2495: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2496: /* 0, 0 , 1} */
2497: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2498: /* {0.51571254859325999, 0.4842874514067399, */
2499: /* 0.51326036147820708, 0.48673963852179264} */
2500: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2501:
1.126 brouard 2502: int i, ii,j,k;
1.209 brouard 2503: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2504: /* double **matprod2(); */ /* test */
1.218 brouard 2505: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2506: double **newm;
1.209 brouard 2507: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2508: int ncvloop=0;
1.169 brouard 2509:
1.209 brouard 2510: min=vector(1,nlstate);
2511: max=vector(1,nlstate);
2512: meandiff=vector(1,nlstate);
2513:
1.218 brouard 2514: /* Starting with matrix unity */
1.126 brouard 2515: for (ii=1;ii<=nlstate+ndeath;ii++)
2516: for (j=1;j<=nlstate+ndeath;j++){
2517: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2518: }
1.169 brouard 2519:
2520: cov[1]=1.;
2521:
2522: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2523: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2524: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2525: ncvloop++;
1.126 brouard 2526: newm=savm;
2527: /* Covariates have to be included here again */
1.138 brouard 2528: cov[2]=agefin;
1.187 brouard 2529: if(nagesqr==1)
2530: cov[3]= agefin*agefin;;
1.234 brouard 2531: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2532: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2533: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2534: /* 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 2535: }
2536: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2537: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2538: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2539: /* 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 2540: }
1.237 brouard 2541: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2542: if(Dummy[Tvar[Tage[k]]]){
2543: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2544: } else{
1.235 brouard 2545: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2546: }
1.235 brouard 2547: /* 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 2548: }
1.237 brouard 2549: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2550: /* 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 2551: if(Dummy[Tvard[k][1]==0]){
2552: if(Dummy[Tvard[k][2]==0]){
2553: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2554: }else{
2555: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2556: }
2557: }else{
2558: if(Dummy[Tvard[k][2]==0]){
2559: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2560: }else{
2561: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2562: }
2563: }
1.234 brouard 2564: }
1.138 brouard 2565: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2566: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2567: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2568: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2569: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2570: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2571: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2572:
1.126 brouard 2573: savm=oldm;
2574: oldm=newm;
1.209 brouard 2575:
2576: for(j=1; j<=nlstate; j++){
2577: max[j]=0.;
2578: min[j]=1.;
2579: }
2580: for(i=1;i<=nlstate;i++){
2581: sumnew=0;
2582: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2583: for(j=1; j<=nlstate; j++){
2584: prlim[i][j]= newm[i][j]/(1-sumnew);
2585: max[j]=FMAX(max[j],prlim[i][j]);
2586: min[j]=FMIN(min[j],prlim[i][j]);
2587: }
2588: }
2589:
1.126 brouard 2590: maxmax=0.;
1.209 brouard 2591: for(j=1; j<=nlstate; j++){
2592: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2593: maxmax=FMAX(maxmax,meandiff[j]);
2594: /* 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 2595: } /* j loop */
1.203 brouard 2596: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2597: /* 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 2598: if(maxmax < ftolpl){
1.209 brouard 2599: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2600: free_vector(min,1,nlstate);
2601: free_vector(max,1,nlstate);
2602: free_vector(meandiff,1,nlstate);
1.126 brouard 2603: return prlim;
2604: }
1.169 brouard 2605: } /* age loop */
1.208 brouard 2606: /* After some age loop it doesn't converge */
1.209 brouard 2607: 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 2608: 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 2609: /* 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); */
2610: free_vector(min,1,nlstate);
2611: free_vector(max,1,nlstate);
2612: free_vector(meandiff,1,nlstate);
1.208 brouard 2613:
1.169 brouard 2614: return prlim; /* should not reach here */
1.126 brouard 2615: }
2616:
1.217 brouard 2617:
2618: /**** Back Prevalence limit (stable or period prevalence) ****************/
2619:
1.218 brouard 2620: /* 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) */
2621: /* 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 2622: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2623: {
1.264 brouard 2624: /* 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 2625: matrix by transitions matrix until convergence is reached with precision ftolpl */
2626: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2627: /* Wx is row vector: population in state 1, population in state 2, population dead */
2628: /* or prevalence in state 1, prevalence in state 2, 0 */
2629: /* newm is the matrix after multiplications, its rows are identical at a factor */
2630: /* Initial matrix pimij */
2631: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2632: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2633: /* 0, 0 , 1} */
2634: /*
2635: * and after some iteration: */
2636: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2637: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2638: /* 0, 0 , 1} */
2639: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2640: /* {0.51571254859325999, 0.4842874514067399, */
2641: /* 0.51326036147820708, 0.48673963852179264} */
2642: /* If we start from prlim again, prlim tends to a constant matrix */
2643:
2644: int i, ii,j,k;
1.247 brouard 2645: int first=0;
1.217 brouard 2646: double *min, *max, *meandiff, maxmax,sumnew=0.;
2647: /* double **matprod2(); */ /* test */
2648: double **out, cov[NCOVMAX+1], **bmij();
2649: double **newm;
1.218 brouard 2650: double **dnewm, **doldm, **dsavm; /* for use */
2651: double **oldm, **savm; /* for use */
2652:
1.217 brouard 2653: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2654: int ncvloop=0;
2655:
2656: min=vector(1,nlstate);
2657: max=vector(1,nlstate);
2658: meandiff=vector(1,nlstate);
2659:
1.266 ! brouard 2660: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
! 2661: oldm=oldms; savm=savms;
! 2662:
! 2663: /* Starting with matrix unity */
! 2664: for (ii=1;ii<=nlstate+ndeath;ii++)
! 2665: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2666: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2667: }
2668:
2669: cov[1]=1.;
2670:
2671: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2672: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2673: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2674: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2675: ncvloop++;
1.218 brouard 2676: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2677: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2678: /* Covariates have to be included here again */
2679: cov[2]=agefin;
2680: if(nagesqr==1)
2681: cov[3]= agefin*agefin;;
1.242 brouard 2682: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2683: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2684: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2685: /* 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 2686: }
2687: /* for (k=1; k<=cptcovn;k++) { */
2688: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2689: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2690: /* /\* 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])]); *\/ */
2691: /* } */
2692: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2693: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2694: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2695: /* 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]); */
2696: }
2697: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2698: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2699: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2700: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2701: for (k=1; k<=cptcovage;k++){ /* For product with age */
2702: if(Dummy[Tvar[Tage[k]]]){
2703: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2704: } else{
2705: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2706: }
2707: /* 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]); */
2708: }
2709: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2710: /* 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]); */
2711: if(Dummy[Tvard[k][1]==0]){
2712: if(Dummy[Tvard[k][2]==0]){
2713: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2714: }else{
2715: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2716: }
2717: }else{
2718: if(Dummy[Tvard[k][2]==0]){
2719: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2720: }else{
2721: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2722: }
2723: }
1.217 brouard 2724: }
2725:
2726: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2727: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2728: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2729: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2730: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2731: /* ij should be linked to the correct index of cov */
2732: /* age and covariate values ij are in 'cov', but we need to pass
2733: * ij for the observed prevalence at age and status and covariate
2734: * number: prevacurrent[(int)agefin][ii][ij]
2735: */
2736: /* 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 *\/ */
2737: /* 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 *\/ */
2738: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.266 ! brouard 2739: /* if((int)age == 70){ */
! 2740: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
! 2741: /* for(i=1; i<=nlstate+ndeath; i++) { */
! 2742: /* printf("%d newm= ",i); */
! 2743: /* for(j=1;j<=nlstate+ndeath;j++) { */
! 2744: /* printf("%f ",newm[i][j]); */
! 2745: /* } */
! 2746: /* printf("oldm * "); */
! 2747: /* for(j=1;j<=nlstate+ndeath;j++) { */
! 2748: /* printf("%f ",oldm[i][j]); */
! 2749: /* } */
! 2750: /* printf(" pmmij "); */
! 2751: /* for(j=1;j<=nlstate+ndeath;j++) { */
! 2752: /* printf("%f ",pmmij[i][j]); */
! 2753: /* } */
! 2754: /* printf("\n"); */
! 2755: /* } */
! 2756: /* } */
1.217 brouard 2757: savm=oldm;
2758: oldm=newm;
1.266 ! brouard 2759:
1.217 brouard 2760: for(j=1; j<=nlstate; j++){
2761: max[j]=0.;
2762: min[j]=1.;
2763: }
2764: for(j=1; j<=nlstate; j++){
2765: for(i=1;i<=nlstate;i++){
1.234 brouard 2766: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2767: bprlim[i][j]= newm[i][j];
2768: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2769: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2770: }
2771: }
1.218 brouard 2772:
1.217 brouard 2773: maxmax=0.;
2774: for(i=1; i<=nlstate; i++){
2775: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2776: maxmax=FMAX(maxmax,meandiff[i]);
2777: /* 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); */
2778: } /* j loop */
2779: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2780: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2781: if(maxmax < ftolpl){
1.220 brouard 2782: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2783: free_vector(min,1,nlstate);
2784: free_vector(max,1,nlstate);
2785: free_vector(meandiff,1,nlstate);
2786: return bprlim;
2787: }
2788: } /* age loop */
2789: /* After some age loop it doesn't converge */
1.247 brouard 2790: if(first){
2791: first=1;
2792: 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\
2793: 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);
2794: }
2795: 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 2796: 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);
2797: /* 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); */
2798: free_vector(min,1,nlstate);
2799: free_vector(max,1,nlstate);
2800: free_vector(meandiff,1,nlstate);
2801:
2802: return bprlim; /* should not reach here */
2803: }
2804:
1.126 brouard 2805: /*************** transition probabilities ***************/
2806:
2807: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2808: {
1.138 brouard 2809: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 ! brouard 2810: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2811: model to the ncovmodel covariates (including constant and age).
2812: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2813: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2814: ncth covariate in the global vector x is given by the formula:
2815: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2816: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2817: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2818: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 ! brouard 2819: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2820: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 ! brouard 2821: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2822: */
2823: double s1, lnpijopii;
1.126 brouard 2824: /*double t34;*/
1.164 brouard 2825: int i,j, nc, ii, jj;
1.126 brouard 2826:
1.223 brouard 2827: for(i=1; i<= nlstate; i++){
2828: for(j=1; j<i;j++){
2829: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2830: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2831: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2832: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2833: }
2834: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2835: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2836: }
2837: for(j=i+1; j<=nlstate+ndeath;j++){
2838: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2839: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2840: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2841: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2842: }
2843: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2844: }
2845: }
1.218 brouard 2846:
1.223 brouard 2847: for(i=1; i<= nlstate; i++){
2848: s1=0;
2849: for(j=1; j<i; j++){
2850: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2851: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2852: }
2853: for(j=i+1; j<=nlstate+ndeath; j++){
2854: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2855: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2856: }
2857: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2858: ps[i][i]=1./(s1+1.);
2859: /* Computing other pijs */
2860: for(j=1; j<i; j++)
2861: ps[i][j]= exp(ps[i][j])*ps[i][i];
2862: for(j=i+1; j<=nlstate+ndeath; j++)
2863: ps[i][j]= exp(ps[i][j])*ps[i][i];
2864: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2865: } /* end i */
1.218 brouard 2866:
1.223 brouard 2867: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2868: for(jj=1; jj<= nlstate+ndeath; jj++){
2869: ps[ii][jj]=0;
2870: ps[ii][ii]=1;
2871: }
2872: }
1.218 brouard 2873:
2874:
1.223 brouard 2875: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2876: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2877: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2878: /* } */
2879: /* printf("\n "); */
2880: /* } */
2881: /* printf("\n ");printf("%lf ",cov[2]);*/
2882: /*
2883: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2884: goto end;*/
1.266 ! brouard 2885: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2886: }
2887:
1.218 brouard 2888: /*************** backward transition probabilities ***************/
2889:
2890: /* 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 ) */
2891: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2892: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2893: {
1.266 ! brouard 2894: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
! 2895: * 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 2896: */
1.218 brouard 2897: int i, ii, j,k;
1.222 brouard 2898:
2899: double **out, **pmij();
2900: double sumnew=0.;
1.218 brouard 2901: double agefin;
1.222 brouard 2902:
2903: double **dnewm, **dsavm, **doldm;
2904: double **bbmij;
2905:
1.218 brouard 2906: doldm=ddoldms; /* global pointers */
1.222 brouard 2907: dnewm=ddnewms;
2908: dsavm=ddsavms;
2909:
2910: agefin=cov[2];
2911: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 ! brouard 2912: the observed prevalence (with this covariate ij) at beginning of transition */
! 2913: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
! 2914: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
! 2915: /* outputs pmmij which is a stochastic matrix */
1.222 brouard 2916: /* We do have the matrix Px in savm and we need pij */
2917: for (j=1;j<=nlstate+ndeath;j++){
1.266 ! brouard 2918: sumnew=0.; /* 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 2919: for (ii=1;ii<=nlstate;ii++){
1.266 ! brouard 2920: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
! 2921: sumnew+=pmmij[ii][j]*prevacurrent[(int)agefin][ii][ij]; /* Yes prevalence at beginning of transition */
1.222 brouard 2922: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.266 ! brouard 2923: if(sumnew >= 1.e-10){
! 2924: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 2925: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2926: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2927: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2928: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2929: /* }else */
2930: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
1.266 ! brouard 2931: } /*End ii */
! 2932: }else{ /* We put the identity matrix */
! 2933: for (ii=1;ii<=nlstate+ndeath;ii++){
! 2934: doldm[ii][j]=(ii==j ? 1. : 0.0);
! 2935: } /*End ii */
! 2936: /* printf("Problem internal bmij A: sum_i w_i*p_ij=N.j/N.. <1.e-10 i=%d, j=%d, sumnew=%lf,agefin=%d\n",ii,j,sumnew, (int)agefin); */
! 2937: }
! 2938: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] or identity*/
! 2939: /* left Product of this diag matrix by dsavm=Px (dnewm=dsavm*doldm) */
! 2940: /* bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /\* Bug Valgrind *\/ */
! 2941: bbmij=matprod2(dnewm, pmmij,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
1.222 brouard 2942: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2943: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2944: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2945: /* left Product of this matrix by diag matrix of prevalences (savm) */
2946: for (j=1;j<=nlstate+ndeath;j++){
1.266 ! brouard 2947: sumnew=0.;
1.222 brouard 2948: for (ii=1;ii<=nlstate+ndeath;ii++){
1.266 ! brouard 2949: sumnew+=prevacurrent[(int)agefin][ii][ij];
1.222 brouard 2950: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2951: }
1.266 ! brouard 2952: /* if(sumnew <0.9){ */
! 2953: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
! 2954: /* } */
! 2955: } /* End j, At the end dsavm is diag[(w_i)] */
! 2956: /* What if dsavm doesn't sum ii to 1? */
! 2957: /* ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /\* Bug Valgrind *\/ */
! 2958: ps=matprod2(ps, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
1.222 brouard 2959: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2960: /* end bmij */
1.266 ! brouard 2961: return ps; /*pointer is unchanged */
1.218 brouard 2962: }
1.217 brouard 2963: /*************** transition probabilities ***************/
2964:
1.218 brouard 2965: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2966: {
2967: /* According to parameters values stored in x and the covariate's values stored in cov,
2968: computes the probability to be observed in state j being in state i by appying the
2969: model to the ncovmodel covariates (including constant and age).
2970: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2971: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2972: ncth covariate in the global vector x is given by the formula:
2973: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2974: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2975: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2976: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2977: Outputs ps[i][j] the probability to be observed in j being in j according to
2978: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2979: */
2980: double s1, lnpijopii;
2981: /*double t34;*/
2982: int i,j, nc, ii, jj;
2983:
1.234 brouard 2984: for(i=1; i<= nlstate; i++){
2985: for(j=1; j<i;j++){
2986: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2987: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2988: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2989: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2990: }
2991: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2992: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2993: }
2994: for(j=i+1; j<=nlstate+ndeath;j++){
2995: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2996: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2997: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2998: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2999: }
3000: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3001: }
3002: }
3003:
3004: for(i=1; i<= nlstate; i++){
3005: s1=0;
3006: for(j=1; j<i; j++){
3007: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3008: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3009: }
3010: for(j=i+1; j<=nlstate+ndeath; j++){
3011: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3012: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3013: }
3014: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3015: ps[i][i]=1./(s1+1.);
3016: /* Computing other pijs */
3017: for(j=1; j<i; j++)
3018: ps[i][j]= exp(ps[i][j])*ps[i][i];
3019: for(j=i+1; j<=nlstate+ndeath; j++)
3020: ps[i][j]= exp(ps[i][j])*ps[i][i];
3021: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3022: } /* end i */
3023:
3024: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3025: for(jj=1; jj<= nlstate+ndeath; jj++){
3026: ps[ii][jj]=0;
3027: ps[ii][ii]=1;
3028: }
3029: }
3030: /* Added for backcast */ /* Transposed matrix too */
3031: for(jj=1; jj<= nlstate+ndeath; jj++){
3032: s1=0.;
3033: for(ii=1; ii<= nlstate+ndeath; ii++){
3034: s1+=ps[ii][jj];
3035: }
3036: for(ii=1; ii<= nlstate; ii++){
3037: ps[ii][jj]=ps[ii][jj]/s1;
3038: }
3039: }
3040: /* Transposition */
3041: for(jj=1; jj<= nlstate+ndeath; jj++){
3042: for(ii=jj; ii<= nlstate+ndeath; ii++){
3043: s1=ps[ii][jj];
3044: ps[ii][jj]=ps[jj][ii];
3045: ps[jj][ii]=s1;
3046: }
3047: }
3048: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3049: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3050: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3051: /* } */
3052: /* printf("\n "); */
3053: /* } */
3054: /* printf("\n ");printf("%lf ",cov[2]);*/
3055: /*
3056: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3057: goto end;*/
3058: return ps;
1.217 brouard 3059: }
3060:
3061:
1.126 brouard 3062: /**************** Product of 2 matrices ******************/
3063:
1.145 brouard 3064: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3065: {
3066: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3067: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3068: /* in, b, out are matrice of pointers which should have been initialized
3069: before: only the contents of out is modified. The function returns
3070: a pointer to pointers identical to out */
1.145 brouard 3071: int i, j, k;
1.126 brouard 3072: for(i=nrl; i<= nrh; i++)
1.145 brouard 3073: for(k=ncolol; k<=ncoloh; k++){
3074: out[i][k]=0.;
3075: for(j=ncl; j<=nch; j++)
3076: out[i][k] +=in[i][j]*b[j][k];
3077: }
1.126 brouard 3078: return out;
3079: }
3080:
3081:
3082: /************* Higher Matrix Product ***************/
3083:
1.235 brouard 3084: 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 3085: {
1.218 brouard 3086: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3087: 'nhstepm*hstepm*stepm' months (i.e. until
3088: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3089: nhstepm*hstepm matrices.
3090: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3091: (typically every 2 years instead of every month which is too big
3092: for the memory).
3093: Model is determined by parameters x and covariates have to be
3094: included manually here.
3095:
3096: */
3097:
3098: int i, j, d, h, k;
1.131 brouard 3099: double **out, cov[NCOVMAX+1];
1.126 brouard 3100: double **newm;
1.187 brouard 3101: double agexact;
1.214 brouard 3102: double agebegin, ageend;
1.126 brouard 3103:
3104: /* Hstepm could be zero and should return the unit matrix */
3105: for (i=1;i<=nlstate+ndeath;i++)
3106: for (j=1;j<=nlstate+ndeath;j++){
3107: oldm[i][j]=(i==j ? 1.0 : 0.0);
3108: po[i][j][0]=(i==j ? 1.0 : 0.0);
3109: }
3110: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3111: for(h=1; h <=nhstepm; h++){
3112: for(d=1; d <=hstepm; d++){
3113: newm=savm;
3114: /* Covariates have to be included here again */
3115: cov[1]=1.;
1.214 brouard 3116: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3117: cov[2]=agexact;
3118: if(nagesqr==1)
1.227 brouard 3119: cov[3]= agexact*agexact;
1.235 brouard 3120: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3121: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3122: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3123: /* 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)); */
3124: }
3125: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3126: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3127: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3128: /* 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]); */
3129: }
3130: for (k=1; k<=cptcovage;k++){
3131: if(Dummy[Tvar[Tage[k]]]){
3132: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3133: } else{
3134: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3135: }
3136: /* 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]); */
3137: }
3138: for (k=1; k<=cptcovprod;k++){ /* */
3139: /* 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]); */
3140: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3141: }
3142: /* for (k=1; k<=cptcovn;k++) */
3143: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3144: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3145: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3146: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3147: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3148:
3149:
1.126 brouard 3150: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3151: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3152: /* right multiplication of oldm by the current matrix */
1.126 brouard 3153: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3154: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3155: /* if((int)age == 70){ */
3156: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3157: /* for(i=1; i<=nlstate+ndeath; i++) { */
3158: /* printf("%d pmmij ",i); */
3159: /* for(j=1;j<=nlstate+ndeath;j++) { */
3160: /* printf("%f ",pmmij[i][j]); */
3161: /* } */
3162: /* printf(" oldm "); */
3163: /* for(j=1;j<=nlstate+ndeath;j++) { */
3164: /* printf("%f ",oldm[i][j]); */
3165: /* } */
3166: /* printf("\n"); */
3167: /* } */
3168: /* } */
1.126 brouard 3169: savm=oldm;
3170: oldm=newm;
3171: }
3172: for(i=1; i<=nlstate+ndeath; i++)
3173: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3174: po[i][j][h]=newm[i][j];
3175: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3176: }
1.128 brouard 3177: /*printf("h=%d ",h);*/
1.126 brouard 3178: } /* end h */
1.218 brouard 3179: /* printf("\n H=%d \n",h); */
1.126 brouard 3180: return po;
3181: }
3182:
1.217 brouard 3183: /************* Higher Back Matrix Product ***************/
1.218 brouard 3184: /* 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 ) */
1.222 brouard 3185: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3186: {
1.266 ! brouard 3187: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3188: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3189: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3190: nhstepm*hstepm matrices.
3191: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3192: (typically every 2 years instead of every month which is too big
1.217 brouard 3193: for the memory).
1.218 brouard 3194: Model is determined by parameters x and covariates have to be
1.266 ! brouard 3195: included manually here. Then we use a call to bmij(x and cov)
! 3196: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3197: */
1.217 brouard 3198:
3199: int i, j, d, h, k;
1.266 ! brouard 3200: double **out, cov[NCOVMAX+1], **bmij();
! 3201: double **newm, ***newmm;
1.217 brouard 3202: double agexact;
3203: double agebegin, ageend;
1.222 brouard 3204: double **oldm, **savm;
1.217 brouard 3205:
1.266 ! brouard 3206: newmm=po; /* To be saved */
! 3207: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3208: /* Hstepm could be zero and should return the unit matrix */
3209: for (i=1;i<=nlstate+ndeath;i++)
3210: for (j=1;j<=nlstate+ndeath;j++){
3211: oldm[i][j]=(i==j ? 1.0 : 0.0);
3212: po[i][j][0]=(i==j ? 1.0 : 0.0);
3213: }
3214: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3215: for(h=1; h <=nhstepm; h++){
3216: for(d=1; d <=hstepm; d++){
3217: newm=savm;
3218: /* Covariates have to be included here again */
3219: cov[1]=1.;
1.266 ! brouard 3220: agexact=age-((h-1)*hstepm + (d))*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3221: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3222: cov[2]=agexact;
3223: if(nagesqr==1)
1.222 brouard 3224: cov[3]= agexact*agexact;
1.266 ! brouard 3225: for (k=1; k<=cptcovn;k++){
! 3226: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
! 3227: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
! 3228: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
! 3229: /* 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)); */
! 3230:
! 3231: }
1.217 brouard 3232: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3233: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3234: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3235: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3236: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3237: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3238: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
1.218 brouard 3239:
1.217 brouard 3240: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3241: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3242: /* Careful transposed matrix */
1.266 ! brouard 3243: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3244: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3245: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3246: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3247: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3248: /* if((int)age == 70){ */
3249: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3250: /* for(i=1; i<=nlstate+ndeath; i++) { */
3251: /* printf("%d pmmij ",i); */
3252: /* for(j=1;j<=nlstate+ndeath;j++) { */
3253: /* printf("%f ",pmmij[i][j]); */
3254: /* } */
3255: /* printf(" oldm "); */
3256: /* for(j=1;j<=nlstate+ndeath;j++) { */
3257: /* printf("%f ",oldm[i][j]); */
3258: /* } */
3259: /* printf("\n"); */
3260: /* } */
3261: /* } */
3262: savm=oldm;
3263: oldm=newm;
3264: }
3265: for(i=1; i<=nlstate+ndeath; i++)
3266: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3267: po[i][j][h]=newm[i][j];
3268: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3269: }
3270: /*printf("h=%d ",h);*/
3271: } /* end h */
1.222 brouard 3272: /* printf("\n H=%d \n",h); */
1.217 brouard 3273: return po;
3274: }
3275:
3276:
1.162 brouard 3277: #ifdef NLOPT
3278: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3279: double fret;
3280: double *xt;
3281: int j;
3282: myfunc_data *d2 = (myfunc_data *) pd;
3283: /* xt = (p1-1); */
3284: xt=vector(1,n);
3285: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3286:
3287: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3288: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3289: printf("Function = %.12lf ",fret);
3290: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3291: printf("\n");
3292: free_vector(xt,1,n);
3293: return fret;
3294: }
3295: #endif
1.126 brouard 3296:
3297: /*************** log-likelihood *************/
3298: double func( double *x)
3299: {
1.226 brouard 3300: int i, ii, j, k, mi, d, kk;
3301: int ioffset=0;
3302: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3303: double **out;
3304: double lli; /* Individual log likelihood */
3305: int s1, s2;
1.228 brouard 3306: 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 3307: double bbh, survp;
3308: long ipmx;
3309: double agexact;
3310: /*extern weight */
3311: /* We are differentiating ll according to initial status */
3312: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3313: /*for(i=1;i<imx;i++)
3314: printf(" %d\n",s[4][i]);
3315: */
1.162 brouard 3316:
1.226 brouard 3317: ++countcallfunc;
1.162 brouard 3318:
1.226 brouard 3319: cov[1]=1.;
1.126 brouard 3320:
1.226 brouard 3321: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3322: ioffset=0;
1.226 brouard 3323: if(mle==1){
3324: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3325: /* Computes the values of the ncovmodel covariates of the model
3326: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3327: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3328: to be observed in j being in i according to the model.
3329: */
1.243 brouard 3330: ioffset=2+nagesqr ;
1.233 brouard 3331: /* Fixed */
1.234 brouard 3332: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3333: 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)*/
3334: }
1.226 brouard 3335: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3336: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3337: has been calculated etc */
3338: /* For an individual i, wav[i] gives the number of effective waves */
3339: /* We compute the contribution to Likelihood of each effective transition
3340: mw[mi][i] is real wave of the mi th effectve wave */
3341: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3342: s2=s[mw[mi+1][i]][i];
3343: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3344: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3345: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3346: */
3347: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3348: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3349: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3350: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3351: }
3352: for (ii=1;ii<=nlstate+ndeath;ii++)
3353: for (j=1;j<=nlstate+ndeath;j++){
3354: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3355: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3356: }
3357: for(d=0; d<dh[mi][i]; d++){
3358: newm=savm;
3359: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3360: cov[2]=agexact;
3361: if(nagesqr==1)
3362: cov[3]= agexact*agexact; /* Should be changed here */
3363: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3364: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3365: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3366: else
3367: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3368: }
3369: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3370: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3371: savm=oldm;
3372: oldm=newm;
3373: } /* end mult */
3374:
3375: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3376: /* But now since version 0.9 we anticipate for bias at large stepm.
3377: * If stepm is larger than one month (smallest stepm) and if the exact delay
3378: * (in months) between two waves is not a multiple of stepm, we rounded to
3379: * the nearest (and in case of equal distance, to the lowest) interval but now
3380: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3381: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3382: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3383: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3384: * -stepm/2 to stepm/2 .
3385: * For stepm=1 the results are the same as for previous versions of Imach.
3386: * For stepm > 1 the results are less biased than in previous versions.
3387: */
1.234 brouard 3388: s1=s[mw[mi][i]][i];
3389: s2=s[mw[mi+1][i]][i];
3390: bbh=(double)bh[mi][i]/(double)stepm;
3391: /* bias bh is positive if real duration
3392: * is higher than the multiple of stepm and negative otherwise.
3393: */
3394: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3395: if( s2 > nlstate){
3396: /* i.e. if s2 is a death state and if the date of death is known
3397: then the contribution to the likelihood is the probability to
3398: die between last step unit time and current step unit time,
3399: which is also equal to probability to die before dh
3400: minus probability to die before dh-stepm .
3401: In version up to 0.92 likelihood was computed
3402: as if date of death was unknown. Death was treated as any other
3403: health state: the date of the interview describes the actual state
3404: and not the date of a change in health state. The former idea was
3405: to consider that at each interview the state was recorded
3406: (healthy, disable or death) and IMaCh was corrected; but when we
3407: introduced the exact date of death then we should have modified
3408: the contribution of an exact death to the likelihood. This new
3409: contribution is smaller and very dependent of the step unit
3410: stepm. It is no more the probability to die between last interview
3411: and month of death but the probability to survive from last
3412: interview up to one month before death multiplied by the
3413: probability to die within a month. Thanks to Chris
3414: Jackson for correcting this bug. Former versions increased
3415: mortality artificially. The bad side is that we add another loop
3416: which slows down the processing. The difference can be up to 10%
3417: lower mortality.
3418: */
3419: /* If, at the beginning of the maximization mostly, the
3420: cumulative probability or probability to be dead is
3421: constant (ie = 1) over time d, the difference is equal to
3422: 0. out[s1][3] = savm[s1][3]: probability, being at state
3423: s1 at precedent wave, to be dead a month before current
3424: wave is equal to probability, being at state s1 at
3425: precedent wave, to be dead at mont of the current
3426: wave. Then the observed probability (that this person died)
3427: is null according to current estimated parameter. In fact,
3428: it should be very low but not zero otherwise the log go to
3429: infinity.
3430: */
1.183 brouard 3431: /* #ifdef INFINITYORIGINAL */
3432: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3433: /* #else */
3434: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3435: /* lli=log(mytinydouble); */
3436: /* else */
3437: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3438: /* #endif */
1.226 brouard 3439: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3440:
1.226 brouard 3441: } else if ( s2==-1 ) { /* alive */
3442: for (j=1,survp=0. ; j<=nlstate; j++)
3443: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3444: /*survp += out[s1][j]; */
3445: lli= log(survp);
3446: }
3447: else if (s2==-4) {
3448: for (j=3,survp=0. ; j<=nlstate; j++)
3449: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3450: lli= log(survp);
3451: }
3452: else if (s2==-5) {
3453: for (j=1,survp=0. ; j<=2; j++)
3454: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3455: lli= log(survp);
3456: }
3457: else{
3458: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3459: /* 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 */
3460: }
3461: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3462: /*if(lli ==000.0)*/
3463: /*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); */
3464: ipmx +=1;
3465: sw += weight[i];
3466: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3467: /* if (lli < log(mytinydouble)){ */
3468: /* 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); */
3469: /* 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]); */
3470: /* } */
3471: } /* end of wave */
3472: } /* end of individual */
3473: } else if(mle==2){
3474: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3475: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3476: for(mi=1; mi<= wav[i]-1; mi++){
3477: for (ii=1;ii<=nlstate+ndeath;ii++)
3478: for (j=1;j<=nlstate+ndeath;j++){
3479: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3480: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3481: }
3482: for(d=0; d<=dh[mi][i]; d++){
3483: newm=savm;
3484: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3485: cov[2]=agexact;
3486: if(nagesqr==1)
3487: cov[3]= agexact*agexact;
3488: for (kk=1; kk<=cptcovage;kk++) {
3489: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3490: }
3491: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3492: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3493: savm=oldm;
3494: oldm=newm;
3495: } /* end mult */
3496:
3497: s1=s[mw[mi][i]][i];
3498: s2=s[mw[mi+1][i]][i];
3499: bbh=(double)bh[mi][i]/(double)stepm;
3500: 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 */
3501: ipmx +=1;
3502: sw += weight[i];
3503: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3504: } /* end of wave */
3505: } /* end of individual */
3506: } else if(mle==3){ /* exponential inter-extrapolation */
3507: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3508: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3509: for(mi=1; mi<= wav[i]-1; mi++){
3510: for (ii=1;ii<=nlstate+ndeath;ii++)
3511: for (j=1;j<=nlstate+ndeath;j++){
3512: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3513: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3514: }
3515: for(d=0; d<dh[mi][i]; d++){
3516: newm=savm;
3517: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3518: cov[2]=agexact;
3519: if(nagesqr==1)
3520: cov[3]= agexact*agexact;
3521: for (kk=1; kk<=cptcovage;kk++) {
3522: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3523: }
3524: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3525: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3526: savm=oldm;
3527: oldm=newm;
3528: } /* end mult */
3529:
3530: s1=s[mw[mi][i]][i];
3531: s2=s[mw[mi+1][i]][i];
3532: bbh=(double)bh[mi][i]/(double)stepm;
3533: 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 */
3534: ipmx +=1;
3535: sw += weight[i];
3536: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3537: } /* end of wave */
3538: } /* end of individual */
3539: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3540: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3541: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3542: for(mi=1; mi<= wav[i]-1; mi++){
3543: for (ii=1;ii<=nlstate+ndeath;ii++)
3544: for (j=1;j<=nlstate+ndeath;j++){
3545: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3546: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3547: }
3548: for(d=0; d<dh[mi][i]; d++){
3549: newm=savm;
3550: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3551: cov[2]=agexact;
3552: if(nagesqr==1)
3553: cov[3]= agexact*agexact;
3554: for (kk=1; kk<=cptcovage;kk++) {
3555: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3556: }
1.126 brouard 3557:
1.226 brouard 3558: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3559: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3560: savm=oldm;
3561: oldm=newm;
3562: } /* end mult */
3563:
3564: s1=s[mw[mi][i]][i];
3565: s2=s[mw[mi+1][i]][i];
3566: if( s2 > nlstate){
3567: lli=log(out[s1][s2] - savm[s1][s2]);
3568: } else if ( s2==-1 ) { /* alive */
3569: for (j=1,survp=0. ; j<=nlstate; j++)
3570: survp += out[s1][j];
3571: lli= log(survp);
3572: }else{
3573: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3574: }
3575: ipmx +=1;
3576: sw += weight[i];
3577: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3578: /* 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 3579: } /* end of wave */
3580: } /* end of individual */
3581: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3582: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3583: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3584: for(mi=1; mi<= wav[i]-1; mi++){
3585: for (ii=1;ii<=nlstate+ndeath;ii++)
3586: for (j=1;j<=nlstate+ndeath;j++){
3587: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3588: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3589: }
3590: for(d=0; d<dh[mi][i]; d++){
3591: newm=savm;
3592: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3593: cov[2]=agexact;
3594: if(nagesqr==1)
3595: cov[3]= agexact*agexact;
3596: for (kk=1; kk<=cptcovage;kk++) {
3597: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3598: }
1.126 brouard 3599:
1.226 brouard 3600: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3601: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3602: savm=oldm;
3603: oldm=newm;
3604: } /* end mult */
3605:
3606: s1=s[mw[mi][i]][i];
3607: s2=s[mw[mi+1][i]][i];
3608: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3609: ipmx +=1;
3610: sw += weight[i];
3611: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3612: /*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]);*/
3613: } /* end of wave */
3614: } /* end of individual */
3615: } /* End of if */
3616: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3617: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3618: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3619: return -l;
1.126 brouard 3620: }
3621:
3622: /*************** log-likelihood *************/
3623: double funcone( double *x)
3624: {
1.228 brouard 3625: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3626: int i, ii, j, k, mi, d, kk;
1.228 brouard 3627: int ioffset=0;
1.131 brouard 3628: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3629: double **out;
3630: double lli; /* Individual log likelihood */
3631: double llt;
3632: int s1, s2;
1.228 brouard 3633: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3634:
1.126 brouard 3635: double bbh, survp;
1.187 brouard 3636: double agexact;
1.214 brouard 3637: double agebegin, ageend;
1.126 brouard 3638: /*extern weight */
3639: /* We are differentiating ll according to initial status */
3640: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3641: /*for(i=1;i<imx;i++)
3642: printf(" %d\n",s[4][i]);
3643: */
3644: cov[1]=1.;
3645:
3646: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3647: ioffset=0;
3648: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3649: /* ioffset=2+nagesqr+cptcovage; */
3650: ioffset=2+nagesqr;
1.232 brouard 3651: /* Fixed */
1.224 brouard 3652: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3653: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3654: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3655: 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)*/
3656: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3657: /* cov[2+6]=covar[Tvar[6]][i]; */
3658: /* cov[2+6]=covar[2][i]; V2 */
3659: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3660: /* cov[2+7]=covar[Tvar[7]][i]; */
3661: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3662: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3663: /* cov[2+9]=covar[Tvar[9]][i]; */
3664: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3665: }
1.232 brouard 3666: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3667: /* 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?)*\/ */
3668: /* } */
1.231 brouard 3669: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3670: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3671: /* } */
1.225 brouard 3672:
1.233 brouard 3673:
3674: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3675: /* Wave varying (but not age varying) */
3676: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3677: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3678: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3679: }
1.232 brouard 3680: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3681: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3682: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3683: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3684: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3685: /* 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 3686: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3687: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3688: /* /\* 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]); *\/ */
3689: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3690: /* } */
1.126 brouard 3691: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3692: for (j=1;j<=nlstate+ndeath;j++){
3693: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3694: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3695: }
1.214 brouard 3696:
3697: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3698: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3699: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3700: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3701: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3702: and mw[mi+1][i]. dh depends on stepm.*/
3703: newm=savm;
1.247 brouard 3704: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3705: cov[2]=agexact;
3706: if(nagesqr==1)
3707: cov[3]= agexact*agexact;
3708: for (kk=1; kk<=cptcovage;kk++) {
3709: if(!FixedV[Tvar[Tage[kk]]])
3710: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3711: else
3712: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3713: }
3714: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3715: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3716: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3717: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3718: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3719: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3720: savm=oldm;
3721: oldm=newm;
1.126 brouard 3722: } /* end mult */
3723:
3724: s1=s[mw[mi][i]][i];
3725: s2=s[mw[mi+1][i]][i];
1.217 brouard 3726: /* if(s2==-1){ */
3727: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3728: /* /\* exit(1); *\/ */
3729: /* } */
1.126 brouard 3730: bbh=(double)bh[mi][i]/(double)stepm;
3731: /* bias is positive if real duration
3732: * is higher than the multiple of stepm and negative otherwise.
3733: */
3734: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3735: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3736: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3737: for (j=1,survp=0. ; j<=nlstate; j++)
3738: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3739: lli= log(survp);
1.126 brouard 3740: }else if (mle==1){
1.242 brouard 3741: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3742: } else if(mle==2){
1.242 brouard 3743: 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 3744: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3745: 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 3746: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3747: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3748: } else{ /* mle=0 back to 1 */
1.242 brouard 3749: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3750: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3751: } /* End of if */
3752: ipmx +=1;
3753: sw += weight[i];
3754: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3755: /*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 3756: if(globpr){
1.246 brouard 3757: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3758: %11.6f %11.6f %11.6f ", \
1.242 brouard 3759: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3760: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3761: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3762: llt +=ll[k]*gipmx/gsw;
3763: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3764: }
3765: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3766: }
1.232 brouard 3767: } /* end of wave */
3768: } /* end of individual */
3769: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3770: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3771: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3772: if(globpr==0){ /* First time we count the contributions and weights */
3773: gipmx=ipmx;
3774: gsw=sw;
3775: }
3776: return -l;
1.126 brouard 3777: }
3778:
3779:
3780: /*************** function likelione ***********/
3781: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3782: {
3783: /* This routine should help understanding what is done with
3784: the selection of individuals/waves and
3785: to check the exact contribution to the likelihood.
3786: Plotting could be done.
3787: */
3788: int k;
3789:
3790: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3791: strcpy(fileresilk,"ILK_");
1.202 brouard 3792: strcat(fileresilk,fileresu);
1.126 brouard 3793: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3794: printf("Problem with resultfile: %s\n", fileresilk);
3795: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3796: }
1.214 brouard 3797: 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");
3798: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3799: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3800: for(k=1; k<=nlstate; k++)
3801: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3802: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3803: }
3804:
3805: *fretone=(*funcone)(p);
3806: if(*globpri !=0){
3807: fclose(ficresilk);
1.205 brouard 3808: if (mle ==0)
3809: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3810: else if(mle >=1)
3811: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3812: 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));
1.207 brouard 3813:
1.208 brouard 3814:
3815: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3816: 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 3817: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3818: }
1.207 brouard 3819: 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 3820: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3821: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3822: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3823: fflush(fichtm);
1.205 brouard 3824: }
1.126 brouard 3825: return;
3826: }
3827:
3828:
3829: /*********** Maximum Likelihood Estimation ***************/
3830:
3831: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3832: {
1.165 brouard 3833: int i,j, iter=0;
1.126 brouard 3834: double **xi;
3835: double fret;
3836: double fretone; /* Only one call to likelihood */
3837: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3838:
3839: #ifdef NLOPT
3840: int creturn;
3841: nlopt_opt opt;
3842: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3843: double *lb;
3844: double minf; /* the minimum objective value, upon return */
3845: double * p1; /* Shifted parameters from 0 instead of 1 */
3846: myfunc_data dinst, *d = &dinst;
3847: #endif
3848:
3849:
1.126 brouard 3850: xi=matrix(1,npar,1,npar);
3851: for (i=1;i<=npar;i++)
3852: for (j=1;j<=npar;j++)
3853: xi[i][j]=(i==j ? 1.0 : 0.0);
3854: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3855: strcpy(filerespow,"POW_");
1.126 brouard 3856: strcat(filerespow,fileres);
3857: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3858: printf("Problem with resultfile: %s\n", filerespow);
3859: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3860: }
3861: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3862: for (i=1;i<=nlstate;i++)
3863: for(j=1;j<=nlstate+ndeath;j++)
3864: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3865: fprintf(ficrespow,"\n");
1.162 brouard 3866: #ifdef POWELL
1.126 brouard 3867: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3868: #endif
1.126 brouard 3869:
1.162 brouard 3870: #ifdef NLOPT
3871: #ifdef NEWUOA
3872: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3873: #else
3874: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3875: #endif
3876: lb=vector(0,npar-1);
3877: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3878: nlopt_set_lower_bounds(opt, lb);
3879: nlopt_set_initial_step1(opt, 0.1);
3880:
3881: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3882: d->function = func;
3883: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3884: nlopt_set_min_objective(opt, myfunc, d);
3885: nlopt_set_xtol_rel(opt, ftol);
3886: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3887: printf("nlopt failed! %d\n",creturn);
3888: }
3889: else {
3890: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3891: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3892: iter=1; /* not equal */
3893: }
3894: nlopt_destroy(opt);
3895: #endif
1.126 brouard 3896: free_matrix(xi,1,npar,1,npar);
3897: fclose(ficrespow);
1.203 brouard 3898: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3899: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3900: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3901:
3902: }
3903:
3904: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3905: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3906: {
3907: double **a,**y,*x,pd;
1.203 brouard 3908: /* double **hess; */
1.164 brouard 3909: int i, j;
1.126 brouard 3910: int *indx;
3911:
3912: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3913: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3914: void lubksb(double **a, int npar, int *indx, double b[]) ;
3915: void ludcmp(double **a, int npar, int *indx, double *d) ;
3916: double gompertz(double p[]);
1.203 brouard 3917: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3918:
3919: printf("\nCalculation of the hessian matrix. Wait...\n");
3920: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3921: for (i=1;i<=npar;i++){
1.203 brouard 3922: printf("%d-",i);fflush(stdout);
3923: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3924:
3925: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3926:
3927: /* printf(" %f ",p[i]);
3928: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3929: }
3930:
3931: for (i=1;i<=npar;i++) {
3932: for (j=1;j<=npar;j++) {
3933: if (j>i) {
1.203 brouard 3934: printf(".%d-%d",i,j);fflush(stdout);
3935: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3936: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3937:
3938: hess[j][i]=hess[i][j];
3939: /*printf(" %lf ",hess[i][j]);*/
3940: }
3941: }
3942: }
3943: printf("\n");
3944: fprintf(ficlog,"\n");
3945:
3946: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3947: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3948:
3949: a=matrix(1,npar,1,npar);
3950: y=matrix(1,npar,1,npar);
3951: x=vector(1,npar);
3952: indx=ivector(1,npar);
3953: for (i=1;i<=npar;i++)
3954: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3955: ludcmp(a,npar,indx,&pd);
3956:
3957: for (j=1;j<=npar;j++) {
3958: for (i=1;i<=npar;i++) x[i]=0;
3959: x[j]=1;
3960: lubksb(a,npar,indx,x);
3961: for (i=1;i<=npar;i++){
3962: matcov[i][j]=x[i];
3963: }
3964: }
3965:
3966: printf("\n#Hessian matrix#\n");
3967: fprintf(ficlog,"\n#Hessian matrix#\n");
3968: for (i=1;i<=npar;i++) {
3969: for (j=1;j<=npar;j++) {
1.203 brouard 3970: printf("%.6e ",hess[i][j]);
3971: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3972: }
3973: printf("\n");
3974: fprintf(ficlog,"\n");
3975: }
3976:
1.203 brouard 3977: /* printf("\n#Covariance matrix#\n"); */
3978: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3979: /* for (i=1;i<=npar;i++) { */
3980: /* for (j=1;j<=npar;j++) { */
3981: /* printf("%.6e ",matcov[i][j]); */
3982: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3983: /* } */
3984: /* printf("\n"); */
3985: /* fprintf(ficlog,"\n"); */
3986: /* } */
3987:
1.126 brouard 3988: /* Recompute Inverse */
1.203 brouard 3989: /* for (i=1;i<=npar;i++) */
3990: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3991: /* ludcmp(a,npar,indx,&pd); */
3992:
3993: /* printf("\n#Hessian matrix recomputed#\n"); */
3994:
3995: /* for (j=1;j<=npar;j++) { */
3996: /* for (i=1;i<=npar;i++) x[i]=0; */
3997: /* x[j]=1; */
3998: /* lubksb(a,npar,indx,x); */
3999: /* for (i=1;i<=npar;i++){ */
4000: /* y[i][j]=x[i]; */
4001: /* printf("%.3e ",y[i][j]); */
4002: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4003: /* } */
4004: /* printf("\n"); */
4005: /* fprintf(ficlog,"\n"); */
4006: /* } */
4007:
4008: /* Verifying the inverse matrix */
4009: #ifdef DEBUGHESS
4010: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4011:
1.203 brouard 4012: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4013: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4014:
4015: for (j=1;j<=npar;j++) {
4016: for (i=1;i<=npar;i++){
1.203 brouard 4017: printf("%.2f ",y[i][j]);
4018: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4019: }
4020: printf("\n");
4021: fprintf(ficlog,"\n");
4022: }
1.203 brouard 4023: #endif
1.126 brouard 4024:
4025: free_matrix(a,1,npar,1,npar);
4026: free_matrix(y,1,npar,1,npar);
4027: free_vector(x,1,npar);
4028: free_ivector(indx,1,npar);
1.203 brouard 4029: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4030:
4031:
4032: }
4033:
4034: /*************** hessian matrix ****************/
4035: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4036: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4037: int i;
4038: int l=1, lmax=20;
1.203 brouard 4039: double k1,k2, res, fx;
1.132 brouard 4040: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4041: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4042: int k=0,kmax=10;
4043: double l1;
4044:
4045: fx=func(x);
4046: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4047: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4048: l1=pow(10,l);
4049: delts=delt;
4050: for(k=1 ; k <kmax; k=k+1){
4051: delt = delta*(l1*k);
4052: p2[theta]=x[theta] +delt;
1.145 brouard 4053: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4054: p2[theta]=x[theta]-delt;
4055: k2=func(p2)-fx;
4056: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4057: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4058:
1.203 brouard 4059: #ifdef DEBUGHESSII
1.126 brouard 4060: 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);
4061: 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);
4062: #endif
4063: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4064: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4065: k=kmax;
4066: }
4067: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4068: k=kmax; l=lmax*10;
1.126 brouard 4069: }
4070: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4071: delts=delt;
4072: }
1.203 brouard 4073: } /* End loop k */
1.126 brouard 4074: }
4075: delti[theta]=delts;
4076: return res;
4077:
4078: }
4079:
1.203 brouard 4080: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4081: {
4082: int i;
1.164 brouard 4083: int l=1, lmax=20;
1.126 brouard 4084: double k1,k2,k3,k4,res,fx;
1.132 brouard 4085: double p2[MAXPARM+1];
1.203 brouard 4086: int k, kmax=1;
4087: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4088:
4089: int firstime=0;
1.203 brouard 4090:
1.126 brouard 4091: fx=func(x);
1.203 brouard 4092: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4093: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4094: p2[thetai]=x[thetai]+delti[thetai]*k;
4095: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4096: k1=func(p2)-fx;
4097:
1.203 brouard 4098: p2[thetai]=x[thetai]+delti[thetai]*k;
4099: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4100: k2=func(p2)-fx;
4101:
1.203 brouard 4102: p2[thetai]=x[thetai]-delti[thetai]*k;
4103: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4104: k3=func(p2)-fx;
4105:
1.203 brouard 4106: p2[thetai]=x[thetai]-delti[thetai]*k;
4107: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4108: k4=func(p2)-fx;
1.203 brouard 4109: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4110: if(k1*k2*k3*k4 <0.){
1.208 brouard 4111: firstime=1;
1.203 brouard 4112: kmax=kmax+10;
1.208 brouard 4113: }
4114: if(kmax >=10 || firstime ==1){
1.246 brouard 4115: 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);
4116: 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 4117: 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);
4118: 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);
4119: }
4120: #ifdef DEBUGHESSIJ
4121: v1=hess[thetai][thetai];
4122: v2=hess[thetaj][thetaj];
4123: cv12=res;
4124: /* Computing eigen value of Hessian matrix */
4125: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4126: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4127: if ((lc2 <0) || (lc1 <0) ){
4128: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4129: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4130: 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);
4131: 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);
4132: }
1.126 brouard 4133: #endif
4134: }
4135: return res;
4136: }
4137:
1.203 brouard 4138: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4139: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4140: /* { */
4141: /* int i; */
4142: /* int l=1, lmax=20; */
4143: /* double k1,k2,k3,k4,res,fx; */
4144: /* double p2[MAXPARM+1]; */
4145: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4146: /* int k=0,kmax=10; */
4147: /* double l1; */
4148:
4149: /* fx=func(x); */
4150: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4151: /* l1=pow(10,l); */
4152: /* delts=delt; */
4153: /* for(k=1 ; k <kmax; k=k+1){ */
4154: /* delt = delti*(l1*k); */
4155: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4156: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4157: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4158: /* k1=func(p2)-fx; */
4159:
4160: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4161: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4162: /* k2=func(p2)-fx; */
4163:
4164: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4165: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4166: /* k3=func(p2)-fx; */
4167:
4168: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4169: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4170: /* k4=func(p2)-fx; */
4171: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4172: /* #ifdef DEBUGHESSIJ */
4173: /* 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); */
4174: /* 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); */
4175: /* #endif */
4176: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4177: /* k=kmax; */
4178: /* } */
4179: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4180: /* k=kmax; l=lmax*10; */
4181: /* } */
4182: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4183: /* delts=delt; */
4184: /* } */
4185: /* } /\* End loop k *\/ */
4186: /* } */
4187: /* delti[theta]=delts; */
4188: /* return res; */
4189: /* } */
4190:
4191:
1.126 brouard 4192: /************** Inverse of matrix **************/
4193: void ludcmp(double **a, int n, int *indx, double *d)
4194: {
4195: int i,imax,j,k;
4196: double big,dum,sum,temp;
4197: double *vv;
4198:
4199: vv=vector(1,n);
4200: *d=1.0;
4201: for (i=1;i<=n;i++) {
4202: big=0.0;
4203: for (j=1;j<=n;j++)
4204: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4205: if (big == 0.0){
4206: printf(" Singular Hessian matrix at row %d:\n",i);
4207: for (j=1;j<=n;j++) {
4208: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4209: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4210: }
4211: fflush(ficlog);
4212: fclose(ficlog);
4213: nrerror("Singular matrix in routine ludcmp");
4214: }
1.126 brouard 4215: vv[i]=1.0/big;
4216: }
4217: for (j=1;j<=n;j++) {
4218: for (i=1;i<j;i++) {
4219: sum=a[i][j];
4220: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4221: a[i][j]=sum;
4222: }
4223: big=0.0;
4224: for (i=j;i<=n;i++) {
4225: sum=a[i][j];
4226: for (k=1;k<j;k++)
4227: sum -= a[i][k]*a[k][j];
4228: a[i][j]=sum;
4229: if ( (dum=vv[i]*fabs(sum)) >= big) {
4230: big=dum;
4231: imax=i;
4232: }
4233: }
4234: if (j != imax) {
4235: for (k=1;k<=n;k++) {
4236: dum=a[imax][k];
4237: a[imax][k]=a[j][k];
4238: a[j][k]=dum;
4239: }
4240: *d = -(*d);
4241: vv[imax]=vv[j];
4242: }
4243: indx[j]=imax;
4244: if (a[j][j] == 0.0) a[j][j]=TINY;
4245: if (j != n) {
4246: dum=1.0/(a[j][j]);
4247: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4248: }
4249: }
4250: free_vector(vv,1,n); /* Doesn't work */
4251: ;
4252: }
4253:
4254: void lubksb(double **a, int n, int *indx, double b[])
4255: {
4256: int i,ii=0,ip,j;
4257: double sum;
4258:
4259: for (i=1;i<=n;i++) {
4260: ip=indx[i];
4261: sum=b[ip];
4262: b[ip]=b[i];
4263: if (ii)
4264: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4265: else if (sum) ii=i;
4266: b[i]=sum;
4267: }
4268: for (i=n;i>=1;i--) {
4269: sum=b[i];
4270: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4271: b[i]=sum/a[i][i];
4272: }
4273: }
4274:
4275: void pstamp(FILE *fichier)
4276: {
1.196 brouard 4277: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4278: }
4279:
1.253 brouard 4280: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4281:
4282: /* y=a+bx regression */
4283: double sumx = 0.0; /* sum of x */
4284: double sumx2 = 0.0; /* sum of x**2 */
4285: double sumxy = 0.0; /* sum of x * y */
4286: double sumy = 0.0; /* sum of y */
4287: double sumy2 = 0.0; /* sum of y**2 */
4288: double sume2; /* sum of square or residuals */
4289: double yhat;
4290:
4291: double denom=0;
4292: int i;
4293: int ne=*no;
4294:
4295: for ( i=ifi, ne=0;i<=ila;i++) {
4296: if(!isfinite(x[i]) || !isfinite(y[i])){
4297: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4298: continue;
4299: }
4300: ne=ne+1;
4301: sumx += x[i];
4302: sumx2 += x[i]*x[i];
4303: sumxy += x[i] * y[i];
4304: sumy += y[i];
4305: sumy2 += y[i]*y[i];
4306: denom = (ne * sumx2 - sumx*sumx);
4307: /* 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); */
4308: }
4309:
4310: denom = (ne * sumx2 - sumx*sumx);
4311: if (denom == 0) {
4312: // vertical, slope m is infinity
4313: *b = INFINITY;
4314: *a = 0;
4315: if (r) *r = 0;
4316: return 1;
4317: }
4318:
4319: *b = (ne * sumxy - sumx * sumy) / denom;
4320: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4321: if (r!=NULL) {
4322: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4323: sqrt((sumx2 - sumx*sumx/ne) *
4324: (sumy2 - sumy*sumy/ne));
4325: }
4326: *no=ne;
4327: for ( i=ifi, ne=0;i<=ila;i++) {
4328: if(!isfinite(x[i]) || !isfinite(y[i])){
4329: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4330: continue;
4331: }
4332: ne=ne+1;
4333: yhat = y[i] - *a -*b* x[i];
4334: sume2 += yhat * yhat ;
4335:
4336: denom = (ne * sumx2 - sumx*sumx);
4337: /* 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); */
4338: }
4339: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4340: *sa= *sb * sqrt(sumx2/ne);
4341:
4342: return 0;
4343: }
4344:
1.126 brouard 4345: /************ Frequencies ********************/
1.251 brouard 4346: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4347: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4348: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4349: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4350:
1.265 brouard 4351: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4352: int iind=0, iage=0;
4353: int mi; /* Effective wave */
4354: int first;
4355: double ***freq; /* Frequencies */
1.253 brouard 4356: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4357: int no;
1.226 brouard 4358: double *meanq;
4359: double **meanqt;
4360: double *pp, **prop, *posprop, *pospropt;
4361: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4362: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4363: double agebegin, ageend;
4364:
4365: pp=vector(1,nlstate);
1.251 brouard 4366: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4367: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4368: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4369: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4370: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4371: meanqt=matrix(1,lastpass,1,nqtveff);
4372: strcpy(fileresp,"P_");
4373: strcat(fileresp,fileresu);
4374: /*strcat(fileresphtm,fileresu);*/
4375: if((ficresp=fopen(fileresp,"w"))==NULL) {
4376: printf("Problem with prevalence resultfile: %s\n", fileresp);
4377: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4378: exit(0);
4379: }
1.240 brouard 4380:
1.226 brouard 4381: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4382: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4383: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4384: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4385: fflush(ficlog);
4386: exit(70);
4387: }
4388: else{
4389: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4390: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4391: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4392: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4393: }
1.237 brouard 4394: 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 4395:
1.226 brouard 4396: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4397: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4398: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4399: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4400: fflush(ficlog);
4401: exit(70);
1.240 brouard 4402: } else{
1.226 brouard 4403: 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 4404: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4405: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4406: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4407: }
1.240 brouard 4408: 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);
4409:
1.253 brouard 4410: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4411: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4412: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4413: j1=0;
1.126 brouard 4414:
1.227 brouard 4415: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4416: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4417: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4418:
4419:
1.226 brouard 4420: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4421: reference=low_education V1=0,V2=0
4422: med_educ V1=1 V2=0,
4423: high_educ V1=0 V2=1
4424: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4425: */
1.249 brouard 4426: dateintsum=0;
4427: k2cpt=0;
4428:
1.253 brouard 4429: if(cptcoveff == 0 )
1.265 brouard 4430: nl=1; /* Constant and age model only */
1.253 brouard 4431: else
4432: nl=2;
1.265 brouard 4433:
4434: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4435: /* Loop on nj=1 or 2 if dummy covariates j!=0
4436: * Loop on j1(1 to 2**cptcoveff) covariate combination
4437: * freq[s1][s2][iage] =0.
4438: * Loop on iind
4439: * ++freq[s1][s2][iage] weighted
4440: * end iind
4441: * if covariate and j!0
4442: * headers Variable on one line
4443: * endif cov j!=0
4444: * header of frequency table by age
4445: * Loop on age
4446: * pp[s1]+=freq[s1][s2][iage] weighted
4447: * pos+=freq[s1][s2][iage] weighted
4448: * Loop on s1 initial state
4449: * fprintf(ficresp
4450: * end s1
4451: * end age
4452: * if j!=0 computes starting values
4453: * end compute starting values
4454: * end j1
4455: * end nl
4456: */
1.253 brouard 4457: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4458: if(nj==1)
4459: j=0; /* First pass for the constant */
1.265 brouard 4460: else{
1.253 brouard 4461: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4462: }
1.251 brouard 4463: first=1;
1.265 brouard 4464: 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 4465: posproptt=0.;
4466: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4467: scanf("%d", i);*/
4468: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4469: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4470: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4471: freq[i][s2][m]=0;
1.251 brouard 4472:
4473: for (i=1; i<=nlstate; i++) {
1.240 brouard 4474: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4475: prop[i][m]=0;
4476: posprop[i]=0;
4477: pospropt[i]=0;
4478: }
4479: /* for (z1=1; z1<= nqfveff; z1++) { */
4480: /* meanq[z1]+=0.; */
4481: /* for(m=1;m<=lastpass;m++){ */
4482: /* meanqt[m][z1]=0.; */
4483: /* } */
4484: /* } */
4485:
4486: /* dateintsum=0; */
4487: /* k2cpt=0; */
4488:
1.265 brouard 4489: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4490: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4491: bool=1;
4492: if(j !=0){
4493: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4494: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4495: /* for (z1=1; z1<= nqfveff; z1++) { */
4496: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4497: /* } */
4498: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4499: /* if(Tvaraff[z1] ==-20){ */
4500: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4501: /* }else if(Tvaraff[z1] ==-10){ */
4502: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4503: /* }else */
4504: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4505: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4506: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4507: /* 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",
4508: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4509: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4510: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4511: } /* Onlyf fixed */
4512: } /* end z1 */
4513: } /* cptcovn > 0 */
4514: } /* end any */
4515: }/* end j==0 */
1.265 brouard 4516: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4517: /* for(m=firstpass; m<=lastpass; m++){ */
4518: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4519: m=mw[mi][iind];
4520: if(j!=0){
4521: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4522: for (z1=1; z1<=cptcoveff; z1++) {
4523: if( Fixed[Tmodelind[z1]]==1){
4524: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4525: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4526: value is -1, we don't select. It differs from the
4527: constant and age model which counts them. */
4528: bool=0; /* not selected */
4529: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4530: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4531: bool=0;
4532: }
4533: }
4534: }
4535: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4536: } /* end j==0 */
4537: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4538: if(bool==1){
4539: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4540: and mw[mi+1][iind]. dh depends on stepm. */
4541: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4542: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4543: if(m >=firstpass && m <=lastpass){
4544: k2=anint[m][iind]+(mint[m][iind]/12.);
4545: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4546: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4547: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4548: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4549: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4550: if (m<lastpass) {
4551: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4552: /* 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]); */
4553: if(s[m][iind]==-1)
4554: 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.));
4555: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4556: /* if((int)agev[m][iind] == 55) */
4557: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4558: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4559: 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 4560: }
1.251 brouard 4561: } /* end if between passes */
4562: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4563: dateintsum=dateintsum+k2; /* on all covariates ?*/
4564: k2cpt++;
4565: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4566: }
1.251 brouard 4567: }else{
4568: bool=1;
4569: }/* end bool 2 */
4570: } /* end m */
4571: } /* end bool */
4572: } /* end iind = 1 to imx */
4573: /* prop[s][age] is feeded for any initial and valid live state as well as
4574: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4575:
4576:
4577: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4578: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4579: pstamp(ficresp);
1.251 brouard 4580: if (cptcoveff>0 && j!=0){
1.265 brouard 4581: pstamp(ficresp);
1.251 brouard 4582: printf( "\n#********** Variable ");
4583: fprintf(ficresp, "\n#********** Variable ");
4584: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4585: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4586: fprintf(ficlog, "\n#********** Variable ");
4587: for (z1=1; z1<=cptcoveff; z1++){
4588: if(!FixedV[Tvaraff[z1]]){
4589: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4590: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4591: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4592: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4593: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4594: }else{
1.251 brouard 4595: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4596: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4597: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4598: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4599: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4600: }
4601: }
4602: printf( "**********\n#");
4603: fprintf(ficresp, "**********\n#");
4604: fprintf(ficresphtm, "**********</h3>\n");
4605: fprintf(ficresphtmfr, "**********</h3>\n");
4606: fprintf(ficlog, "**********\n");
4607: }
4608: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4609: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4610: fprintf(ficresp, " Age");
4611: 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 4612: for(i=1; i<=nlstate;i++) {
1.265 brouard 4613: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4614: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4615: }
1.265 brouard 4616: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4617: fprintf(ficresphtm, "\n");
4618:
4619: /* Header of frequency table by age */
4620: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4621: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4622: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4623: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4624: if(s2!=0 && m!=0)
4625: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4626: }
1.226 brouard 4627: }
1.251 brouard 4628: fprintf(ficresphtmfr, "\n");
4629:
4630: /* For each age */
4631: for(iage=iagemin; iage <= iagemax+3; iage++){
4632: fprintf(ficresphtm,"<tr>");
4633: if(iage==iagemax+1){
4634: fprintf(ficlog,"1");
4635: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4636: }else if(iage==iagemax+2){
4637: fprintf(ficlog,"0");
4638: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4639: }else if(iage==iagemax+3){
4640: fprintf(ficlog,"Total");
4641: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4642: }else{
1.240 brouard 4643: if(first==1){
1.251 brouard 4644: first=0;
4645: printf("See log file for details...\n");
4646: }
4647: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4648: fprintf(ficlog,"Age %d", iage);
4649: }
1.265 brouard 4650: for(s1=1; s1 <=nlstate ; s1++){
4651: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4652: pp[s1] += freq[s1][m][iage];
1.251 brouard 4653: }
1.265 brouard 4654: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4655: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4656: pos += freq[s1][m][iage];
4657: if(pp[s1]>=1.e-10){
1.251 brouard 4658: if(first==1){
1.265 brouard 4659: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4660: }
1.265 brouard 4661: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4662: }else{
4663: if(first==1)
1.265 brouard 4664: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4665: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4666: }
4667: }
4668:
1.265 brouard 4669: for(s1=1; s1 <=nlstate ; s1++){
4670: /* posprop[s1]=0; */
4671: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4672: pp[s1] += freq[s1][m][iage];
4673: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4674:
4675: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4676: pos += pp[s1]; /* pos is the total number of transitions until this age */
4677: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4678: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4679: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4680: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4681: }
4682:
4683: /* Writing ficresp */
4684: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4685: if( iage <= iagemax){
4686: fprintf(ficresp," %d",iage);
4687: }
4688: }else if( nj==2){
4689: if( iage <= iagemax){
4690: fprintf(ficresp," %d",iage);
4691: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4692: }
1.240 brouard 4693: }
1.265 brouard 4694: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4695: if(pos>=1.e-5){
1.251 brouard 4696: if(first==1)
1.265 brouard 4697: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4698: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4699: }else{
4700: if(first==1)
1.265 brouard 4701: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4702: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4703: }
4704: if( iage <= iagemax){
4705: if(pos>=1.e-5){
1.265 brouard 4706: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4707: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4708: }else if( nj==2){
4709: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4710: }
4711: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4712: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4713: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4714: } else{
4715: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4716: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4717: }
1.240 brouard 4718: }
1.265 brouard 4719: pospropt[s1] +=posprop[s1];
4720: } /* end loop s1 */
1.251 brouard 4721: /* pospropt=0.; */
1.265 brouard 4722: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4723: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4724: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4725: if(first==1){
1.265 brouard 4726: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4727: }
1.265 brouard 4728: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4729: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4730: }
1.265 brouard 4731: if(s1!=0 && m!=0)
4732: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4733: }
1.265 brouard 4734: } /* end loop s1 */
1.251 brouard 4735: posproptt=0.;
1.265 brouard 4736: for(s1=1; s1 <=nlstate; s1++){
4737: posproptt += pospropt[s1];
1.251 brouard 4738: }
4739: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4740: fprintf(ficresphtm,"</tr>\n");
4741: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4742: if(iage <= iagemax)
4743: fprintf(ficresp,"\n");
1.240 brouard 4744: }
1.251 brouard 4745: if(first==1)
4746: printf("Others in log...\n");
4747: fprintf(ficlog,"\n");
4748: } /* end loop age iage */
1.265 brouard 4749:
1.251 brouard 4750: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4751: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4752: if(posproptt < 1.e-5){
1.265 brouard 4753: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4754: }else{
1.265 brouard 4755: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4756: }
1.226 brouard 4757: }
1.251 brouard 4758: fprintf(ficresphtm,"</tr>\n");
4759: fprintf(ficresphtm,"</table>\n");
4760: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4761: if(posproptt < 1.e-5){
1.251 brouard 4762: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4763: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4764: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4765: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4766: invalidvarcomb[j1]=1;
1.226 brouard 4767: }else{
1.251 brouard 4768: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4769: invalidvarcomb[j1]=0;
1.226 brouard 4770: }
1.251 brouard 4771: fprintf(ficresphtmfr,"</table>\n");
4772: fprintf(ficlog,"\n");
4773: if(j!=0){
4774: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4775: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4776: for(k=1; k <=(nlstate+ndeath); k++){
4777: if (k != i) {
1.265 brouard 4778: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4779: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4780: if(j1==1){ /* All dummy covariates to zero */
4781: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4782: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4783: printf("%d%d ",i,k);
4784: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4785: 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]));
4786: 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]));
4787: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4788: }
1.253 brouard 4789: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4790: for(iage=iagemin; iage <= iagemax+3; iage++){
4791: x[iage]= (double)iage;
4792: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4793: /* 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 4794: }
4795: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4796: pstart[s1]=b;
4797: pstart[s1-1]=a;
1.252 brouard 4798: }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 */
4799: 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]);
4800: 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 4801: 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 4802: printf("%d%d ",i,k);
4803: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4804: 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 4805: }else{ /* Other cases, like quantitative fixed or varying covariates */
4806: ;
4807: }
4808: /* printf("%12.7f )", param[i][jj][k]); */
4809: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4810: s1++;
1.251 brouard 4811: } /* end jj */
4812: } /* end k!= i */
4813: } /* end k */
1.265 brouard 4814: } /* end i, s1 */
1.251 brouard 4815: } /* end j !=0 */
4816: } /* end selected combination of covariate j1 */
4817: if(j==0){ /* We can estimate starting values from the occurences in each case */
4818: printf("#Freqsummary: Starting values for the constants:\n");
4819: fprintf(ficlog,"\n");
1.265 brouard 4820: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4821: for(k=1; k <=(nlstate+ndeath); k++){
4822: if (k != i) {
4823: printf("%d%d ",i,k);
4824: fprintf(ficlog,"%d%d ",i,k);
4825: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4826: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4827: if(jj==1){ /* Age has to be done */
1.265 brouard 4828: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4829: 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]));
4830: 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 4831: }
4832: /* printf("%12.7f )", param[i][jj][k]); */
4833: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4834: s1++;
1.250 brouard 4835: }
1.251 brouard 4836: printf("\n");
4837: fprintf(ficlog,"\n");
1.250 brouard 4838: }
4839: }
4840: }
1.251 brouard 4841: printf("#Freqsummary\n");
4842: fprintf(ficlog,"\n");
1.265 brouard 4843: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4844: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4845: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4846: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4847: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4848: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4849: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4850: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4851: /* } */
4852: }
1.265 brouard 4853: } /* end loop s1 */
1.251 brouard 4854:
4855: printf("\n");
4856: fprintf(ficlog,"\n");
4857: } /* end j=0 */
1.249 brouard 4858: } /* end j */
1.252 brouard 4859:
1.253 brouard 4860: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4861: for(i=1, jk=1; i <=nlstate; i++){
4862: for(j=1; j <=nlstate+ndeath; j++){
4863: if(j!=i){
4864: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4865: printf("%1d%1d",i,j);
4866: fprintf(ficparo,"%1d%1d",i,j);
4867: for(k=1; k<=ncovmodel;k++){
4868: /* printf(" %lf",param[i][j][k]); */
4869: /* fprintf(ficparo," %lf",param[i][j][k]); */
4870: p[jk]=pstart[jk];
4871: printf(" %f ",pstart[jk]);
4872: fprintf(ficparo," %f ",pstart[jk]);
4873: jk++;
4874: }
4875: printf("\n");
4876: fprintf(ficparo,"\n");
4877: }
4878: }
4879: }
4880: } /* end mle=-2 */
1.226 brouard 4881: dateintmean=dateintsum/k2cpt;
1.240 brouard 4882:
1.226 brouard 4883: fclose(ficresp);
4884: fclose(ficresphtm);
4885: fclose(ficresphtmfr);
4886: free_vector(meanq,1,nqfveff);
4887: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4888: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4889: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4890: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4891: free_vector(pospropt,1,nlstate);
4892: free_vector(posprop,1,nlstate);
1.251 brouard 4893: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4894: free_vector(pp,1,nlstate);
4895: /* End of freqsummary */
4896: }
1.126 brouard 4897:
4898: /************ Prevalence ********************/
1.227 brouard 4899: 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)
4900: {
4901: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4902: in each health status at the date of interview (if between dateprev1 and dateprev2).
4903: We still use firstpass and lastpass as another selection.
4904: */
1.126 brouard 4905:
1.227 brouard 4906: int i, m, jk, j1, bool, z1,j, iv;
4907: int mi; /* Effective wave */
4908: int iage;
4909: double agebegin, ageend;
4910:
4911: double **prop;
4912: double posprop;
4913: double y2; /* in fractional years */
4914: int iagemin, iagemax;
4915: int first; /** to stop verbosity which is redirected to log file */
4916:
4917: iagemin= (int) agemin;
4918: iagemax= (int) agemax;
4919: /*pp=vector(1,nlstate);*/
1.251 brouard 4920: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4921: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4922: j1=0;
1.222 brouard 4923:
1.227 brouard 4924: /*j=cptcoveff;*/
4925: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4926:
1.227 brouard 4927: first=1;
4928: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4929: for (i=1; i<=nlstate; i++)
1.251 brouard 4930: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4931: prop[i][iage]=0.0;
4932: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4933: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4934: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4935:
4936: for (i=1; i<=imx; i++) { /* Each individual */
4937: bool=1;
4938: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4939: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4940: m=mw[mi][i];
4941: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4942: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4943: for (z1=1; z1<=cptcoveff; z1++){
4944: if( Fixed[Tmodelind[z1]]==1){
4945: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4946: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4947: bool=0;
4948: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4949: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4950: bool=0;
4951: }
4952: }
4953: if(bool==1){ /* Otherwise we skip that wave/person */
4954: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4955: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4956: if(m >=firstpass && m <=lastpass){
4957: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4958: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4959: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4960: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4961: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4962: 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);
4963: exit(1);
4964: }
4965: if (s[m][i]>0 && s[m][i]<=nlstate) {
4966: /*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]]);*/
4967: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4968: prop[s[m][i]][iagemax+3] += weight[i];
4969: } /* end valid statuses */
4970: } /* end selection of dates */
4971: } /* end selection of waves */
4972: } /* end bool */
4973: } /* end wave */
4974: } /* end individual */
4975: for(i=iagemin; i <= iagemax+3; i++){
4976: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4977: posprop += prop[jk][i];
4978: }
4979:
4980: for(jk=1; jk <=nlstate ; jk++){
4981: if( i <= iagemax){
4982: if(posprop>=1.e-5){
4983: probs[i][jk][j1]= prop[jk][i]/posprop;
4984: } else{
4985: if(first==1){
4986: first=0;
1.266 ! brouard 4987: 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]);
! 4988: 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]);
! 4989: }else{
! 4990: 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 4991: }
4992: }
4993: }
4994: }/* end jk */
4995: }/* end i */
1.222 brouard 4996: /*} *//* end i1 */
1.227 brouard 4997: } /* end j1 */
1.222 brouard 4998:
1.227 brouard 4999: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5000: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5001: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5002: } /* End of prevalence */
1.126 brouard 5003:
5004: /************* Waves Concatenation ***************/
5005:
5006: 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)
5007: {
5008: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5009: Death is a valid wave (if date is known).
5010: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5011: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5012: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5013: */
1.126 brouard 5014:
1.224 brouard 5015: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5016: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5017: double sum=0., jmean=0.;*/
1.224 brouard 5018: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5019: int j, k=0,jk, ju, jl;
5020: double sum=0.;
5021: first=0;
1.214 brouard 5022: firstwo=0;
1.217 brouard 5023: firsthree=0;
1.218 brouard 5024: firstfour=0;
1.164 brouard 5025: jmin=100000;
1.126 brouard 5026: jmax=-1;
5027: jmean=0.;
1.224 brouard 5028:
5029: /* Treating live states */
1.214 brouard 5030: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5031: mi=0; /* First valid wave */
1.227 brouard 5032: mli=0; /* Last valid wave */
1.126 brouard 5033: m=firstpass;
1.214 brouard 5034: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5035: 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 */
5036: mli=m-1;/* mw[++mi][i]=m-1; */
5037: }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 */
5038: mw[++mi][i]=m;
5039: mli=m;
1.224 brouard 5040: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5041: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5042: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5043: }
1.227 brouard 5044: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5045: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5046: break;
1.224 brouard 5047: #else
1.227 brouard 5048: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5049: if(firsthree == 0){
1.262 brouard 5050: 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 5051: firsthree=1;
5052: }
1.262 brouard 5053: 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 5054: mw[++mi][i]=m;
5055: mli=m;
5056: }
5057: if(s[m][i]==-2){ /* Vital status is really unknown */
5058: nbwarn++;
5059: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5060: 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);
5061: 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);
5062: }
5063: break;
5064: }
5065: break;
1.224 brouard 5066: #endif
1.227 brouard 5067: }/* End m >= lastpass */
1.126 brouard 5068: }/* end while */
1.224 brouard 5069:
1.227 brouard 5070: /* 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 5071: /* After last pass */
1.224 brouard 5072: /* Treating death states */
1.214 brouard 5073: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5074: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5075: /* } */
1.126 brouard 5076: mi++; /* Death is another wave */
5077: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5078: /* Only death is a correct wave */
1.126 brouard 5079: mw[mi][i]=m;
1.257 brouard 5080: } /* else not in a death state */
1.224 brouard 5081: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5082: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5083: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5084: 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 */
5085: nbwarn++;
5086: if(firstfiv==0){
5087: 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 );
5088: firstfiv=1;
5089: }else{
5090: 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 );
5091: }
5092: }else{ /* Death occured afer last wave potential bias */
5093: nberr++;
5094: if(firstwo==0){
1.257 brouard 5095: 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 5096: firstwo=1;
5097: }
1.257 brouard 5098: 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 5099: }
1.257 brouard 5100: }else{ /* if date of interview is unknown */
1.227 brouard 5101: /* death is known but not confirmed by death status at any wave */
5102: if(firstfour==0){
5103: 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 );
5104: firstfour=1;
5105: }
5106: 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 5107: }
1.224 brouard 5108: } /* end if date of death is known */
5109: #endif
5110: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5111: /* wav[i]=mw[mi][i]; */
1.126 brouard 5112: if(mi==0){
5113: nbwarn++;
5114: if(first==0){
1.227 brouard 5115: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5116: first=1;
1.126 brouard 5117: }
5118: if(first==1){
1.227 brouard 5119: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5120: }
5121: } /* end mi==0 */
5122: } /* End individuals */
1.214 brouard 5123: /* wav and mw are no more changed */
1.223 brouard 5124:
1.214 brouard 5125:
1.126 brouard 5126: for(i=1; i<=imx; i++){
5127: for(mi=1; mi<wav[i];mi++){
5128: if (stepm <=0)
1.227 brouard 5129: dh[mi][i]=1;
1.126 brouard 5130: else{
1.260 brouard 5131: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5132: if (agedc[i] < 2*AGESUP) {
5133: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5134: if(j==0) j=1; /* Survives at least one month after exam */
5135: else if(j<0){
5136: nberr++;
5137: 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]);
5138: j=1; /* Temporary Dangerous patch */
5139: 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);
5140: 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]);
5141: 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);
5142: }
5143: k=k+1;
5144: if (j >= jmax){
5145: jmax=j;
5146: ijmax=i;
5147: }
5148: if (j <= jmin){
5149: jmin=j;
5150: ijmin=i;
5151: }
5152: sum=sum+j;
5153: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5154: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5155: }
5156: }
5157: else{
5158: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5159: /* 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 5160:
1.227 brouard 5161: k=k+1;
5162: if (j >= jmax) {
5163: jmax=j;
5164: ijmax=i;
5165: }
5166: else if (j <= jmin){
5167: jmin=j;
5168: ijmin=i;
5169: }
5170: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5171: /*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]);*/
5172: if(j<0){
5173: nberr++;
5174: 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]);
5175: 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]);
5176: }
5177: sum=sum+j;
5178: }
5179: jk= j/stepm;
5180: jl= j -jk*stepm;
5181: ju= j -(jk+1)*stepm;
5182: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5183: if(jl==0){
5184: dh[mi][i]=jk;
5185: bh[mi][i]=0;
5186: }else{ /* We want a negative bias in order to only have interpolation ie
5187: * to avoid the price of an extra matrix product in likelihood */
5188: dh[mi][i]=jk+1;
5189: bh[mi][i]=ju;
5190: }
5191: }else{
5192: if(jl <= -ju){
5193: dh[mi][i]=jk;
5194: bh[mi][i]=jl; /* bias is positive if real duration
5195: * is higher than the multiple of stepm and negative otherwise.
5196: */
5197: }
5198: else{
5199: dh[mi][i]=jk+1;
5200: bh[mi][i]=ju;
5201: }
5202: if(dh[mi][i]==0){
5203: dh[mi][i]=1; /* At least one step */
5204: bh[mi][i]=ju; /* At least one step */
5205: /* 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);*/
5206: }
5207: } /* end if mle */
1.126 brouard 5208: }
5209: } /* end wave */
5210: }
5211: jmean=sum/k;
5212: 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 5213: 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 5214: }
1.126 brouard 5215:
5216: /*********** Tricode ****************************/
1.220 brouard 5217: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5218: {
5219: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5220: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5221: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5222: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5223: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5224: */
1.130 brouard 5225:
1.242 brouard 5226: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5227: int modmaxcovj=0; /* Modality max of covariates j */
5228: int cptcode=0; /* Modality max of covariates j */
5229: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5230:
5231:
1.242 brouard 5232: /* cptcoveff=0; */
5233: /* *cptcov=0; */
1.126 brouard 5234:
1.242 brouard 5235: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5236:
1.242 brouard 5237: /* Loop on covariates without age and products and no quantitative variable */
5238: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5239: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5240: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5241: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5242: switch(Fixed[k]) {
5243: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5244: 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*/
5245: ij=(int)(covar[Tvar[k]][i]);
5246: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5247: * If product of Vn*Vm, still boolean *:
5248: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5249: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5250: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5251: modality of the nth covariate of individual i. */
5252: if (ij > modmaxcovj)
5253: modmaxcovj=ij;
5254: else if (ij < modmincovj)
5255: modmincovj=ij;
5256: if ((ij < -1) && (ij > NCOVMAX)){
5257: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5258: exit(1);
5259: }else
5260: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5261: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5262: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5263: /* getting the maximum value of the modality of the covariate
5264: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5265: female ies 1, then modmaxcovj=1.
5266: */
5267: } /* end for loop on individuals i */
5268: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5269: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5270: cptcode=modmaxcovj;
5271: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5272: /*for (i=0; i<=cptcode; i++) {*/
5273: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5274: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5275: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5276: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5277: if( j != -1){
5278: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5279: covariate for which somebody answered excluding
5280: undefined. Usually 2: 0 and 1. */
5281: }
5282: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5283: covariate for which somebody answered including
5284: undefined. Usually 3: -1, 0 and 1. */
5285: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5286: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5287: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5288:
1.242 brouard 5289: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5290: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5291: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5292: /* modmincovj=3; modmaxcovj = 7; */
5293: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5294: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5295: /* defining two dummy variables: variables V1_1 and V1_2.*/
5296: /* nbcode[Tvar[j]][ij]=k; */
5297: /* nbcode[Tvar[j]][1]=0; */
5298: /* nbcode[Tvar[j]][2]=1; */
5299: /* nbcode[Tvar[j]][3]=2; */
5300: /* To be continued (not working yet). */
5301: ij=0; /* ij is similar to i but can jump over null modalities */
5302: 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*/
5303: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5304: break;
5305: }
5306: ij++;
5307: 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*/
5308: cptcode = ij; /* New max modality for covar j */
5309: } /* end of loop on modality i=-1 to 1 or more */
5310: break;
5311: case 1: /* Testing on varying covariate, could be simple and
5312: * should look at waves or product of fixed *
5313: * varying. No time to test -1, assuming 0 and 1 only */
5314: ij=0;
5315: for(i=0; i<=1;i++){
5316: nbcode[Tvar[k]][++ij]=i;
5317: }
5318: break;
5319: default:
5320: break;
5321: } /* end switch */
5322: } /* end dummy test */
5323:
5324: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5325: /* /\*recode from 0 *\/ */
5326: /* k is a modality. If we have model=V1+V1*sex */
5327: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5328: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5329: /* } */
5330: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5331: /* if (ij > ncodemax[j]) { */
5332: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5333: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5334: /* break; */
5335: /* } */
5336: /* } /\* end of loop on modality k *\/ */
5337: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5338:
5339: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5340: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5341: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5342: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5343: 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 */
5344: 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 */
5345: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5346: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5347:
5348: ij=0;
5349: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5350: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5351: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5352: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5353: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5354: /* If product not in single variable we don't print results */
5355: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5356: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5357: 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*/
5358: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5359: 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 */
5360: if(Fixed[k]!=0)
5361: anyvaryingduminmodel=1;
5362: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5363: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5364: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5365: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5366: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5367: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5368: }
5369: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5370: /* ij--; */
5371: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5372: *cptcov=ij; /*Number of total real effective covariates: effective
5373: * because they can be excluded from the model and real
5374: * if in the model but excluded because missing values, but how to get k from ij?*/
5375: for(j=ij+1; j<= cptcovt; j++){
5376: Tvaraff[j]=0;
5377: Tmodelind[j]=0;
5378: }
5379: for(j=ntveff+1; j<= cptcovt; j++){
5380: TmodelInvind[j]=0;
5381: }
5382: /* To be sorted */
5383: ;
5384: }
1.126 brouard 5385:
1.145 brouard 5386:
1.126 brouard 5387: /*********** Health Expectancies ****************/
5388:
1.235 brouard 5389: 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 5390:
5391: {
5392: /* Health expectancies, no variances */
1.164 brouard 5393: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5394: int nhstepma, nstepma; /* Decreasing with age */
5395: double age, agelim, hf;
5396: double ***p3mat;
5397: double eip;
5398:
1.238 brouard 5399: /* pstamp(ficreseij); */
1.126 brouard 5400: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5401: fprintf(ficreseij,"# Age");
5402: for(i=1; i<=nlstate;i++){
5403: for(j=1; j<=nlstate;j++){
5404: fprintf(ficreseij," e%1d%1d ",i,j);
5405: }
5406: fprintf(ficreseij," e%1d. ",i);
5407: }
5408: fprintf(ficreseij,"\n");
5409:
5410:
5411: if(estepm < stepm){
5412: printf ("Problem %d lower than %d\n",estepm, stepm);
5413: }
5414: else hstepm=estepm;
5415: /* We compute the life expectancy from trapezoids spaced every estepm months
5416: * This is mainly to measure the difference between two models: for example
5417: * if stepm=24 months pijx are given only every 2 years and by summing them
5418: * we are calculating an estimate of the Life Expectancy assuming a linear
5419: * progression in between and thus overestimating or underestimating according
5420: * to the curvature of the survival function. If, for the same date, we
5421: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5422: * to compare the new estimate of Life expectancy with the same linear
5423: * hypothesis. A more precise result, taking into account a more precise
5424: * curvature will be obtained if estepm is as small as stepm. */
5425:
5426: /* For example we decided to compute the life expectancy with the smallest unit */
5427: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5428: nhstepm is the number of hstepm from age to agelim
5429: nstepm is the number of stepm from age to agelin.
5430: Look at hpijx to understand the reason of that which relies in memory size
5431: and note for a fixed period like estepm months */
5432: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5433: survival function given by stepm (the optimization length). Unfortunately it
5434: means that if the survival funtion is printed only each two years of age and if
5435: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5436: results. So we changed our mind and took the option of the best precision.
5437: */
5438: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5439:
5440: agelim=AGESUP;
5441: /* If stepm=6 months */
5442: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5443: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5444:
5445: /* nhstepm age range expressed in number of stepm */
5446: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5447: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5448: /* if (stepm >= YEARM) hstepm=1;*/
5449: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5450: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5451:
5452: for (age=bage; age<=fage; age ++){
5453: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5454: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5455: /* if (stepm >= YEARM) hstepm=1;*/
5456: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5457:
5458: /* If stepm=6 months */
5459: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5460: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5461:
1.235 brouard 5462: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5463:
5464: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5465:
5466: printf("%d|",(int)age);fflush(stdout);
5467: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5468:
5469: /* Computing expectancies */
5470: for(i=1; i<=nlstate;i++)
5471: for(j=1; j<=nlstate;j++)
5472: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5473: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5474:
5475: /* 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]);*/
5476:
5477: }
5478:
5479: fprintf(ficreseij,"%3.0f",age );
5480: for(i=1; i<=nlstate;i++){
5481: eip=0;
5482: for(j=1; j<=nlstate;j++){
5483: eip +=eij[i][j][(int)age];
5484: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5485: }
5486: fprintf(ficreseij,"%9.4f", eip );
5487: }
5488: fprintf(ficreseij,"\n");
5489:
5490: }
5491: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5492: printf("\n");
5493: fprintf(ficlog,"\n");
5494:
5495: }
5496:
1.235 brouard 5497: 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 5498:
5499: {
5500: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5501: to initial status i, ei. .
1.126 brouard 5502: */
5503: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5504: int nhstepma, nstepma; /* Decreasing with age */
5505: double age, agelim, hf;
5506: double ***p3matp, ***p3matm, ***varhe;
5507: double **dnewm,**doldm;
5508: double *xp, *xm;
5509: double **gp, **gm;
5510: double ***gradg, ***trgradg;
5511: int theta;
5512:
5513: double eip, vip;
5514:
5515: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5516: xp=vector(1,npar);
5517: xm=vector(1,npar);
5518: dnewm=matrix(1,nlstate*nlstate,1,npar);
5519: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5520:
5521: pstamp(ficresstdeij);
5522: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5523: fprintf(ficresstdeij,"# Age");
5524: for(i=1; i<=nlstate;i++){
5525: for(j=1; j<=nlstate;j++)
5526: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5527: fprintf(ficresstdeij," e%1d. ",i);
5528: }
5529: fprintf(ficresstdeij,"\n");
5530:
5531: pstamp(ficrescveij);
5532: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5533: fprintf(ficrescveij,"# Age");
5534: for(i=1; i<=nlstate;i++)
5535: for(j=1; j<=nlstate;j++){
5536: cptj= (j-1)*nlstate+i;
5537: for(i2=1; i2<=nlstate;i2++)
5538: for(j2=1; j2<=nlstate;j2++){
5539: cptj2= (j2-1)*nlstate+i2;
5540: if(cptj2 <= cptj)
5541: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5542: }
5543: }
5544: fprintf(ficrescveij,"\n");
5545:
5546: if(estepm < stepm){
5547: printf ("Problem %d lower than %d\n",estepm, stepm);
5548: }
5549: else hstepm=estepm;
5550: /* We compute the life expectancy from trapezoids spaced every estepm months
5551: * This is mainly to measure the difference between two models: for example
5552: * if stepm=24 months pijx are given only every 2 years and by summing them
5553: * we are calculating an estimate of the Life Expectancy assuming a linear
5554: * progression in between and thus overestimating or underestimating according
5555: * to the curvature of the survival function. If, for the same date, we
5556: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5557: * to compare the new estimate of Life expectancy with the same linear
5558: * hypothesis. A more precise result, taking into account a more precise
5559: * curvature will be obtained if estepm is as small as stepm. */
5560:
5561: /* For example we decided to compute the life expectancy with the smallest unit */
5562: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5563: nhstepm is the number of hstepm from age to agelim
5564: nstepm is the number of stepm from age to agelin.
5565: Look at hpijx to understand the reason of that which relies in memory size
5566: and note for a fixed period like estepm months */
5567: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5568: survival function given by stepm (the optimization length). Unfortunately it
5569: means that if the survival funtion is printed only each two years of age and if
5570: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5571: results. So we changed our mind and took the option of the best precision.
5572: */
5573: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5574:
5575: /* If stepm=6 months */
5576: /* nhstepm age range expressed in number of stepm */
5577: agelim=AGESUP;
5578: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5579: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5580: /* if (stepm >= YEARM) hstepm=1;*/
5581: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5582:
5583: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5584: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5585: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5586: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5587: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5588: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5589:
5590: for (age=bage; age<=fage; age ++){
5591: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5592: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5593: /* if (stepm >= YEARM) hstepm=1;*/
5594: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5595:
1.126 brouard 5596: /* If stepm=6 months */
5597: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5598: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5599:
5600: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5601:
1.126 brouard 5602: /* Computing Variances of health expectancies */
5603: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5604: decrease memory allocation */
5605: for(theta=1; theta <=npar; theta++){
5606: for(i=1; i<=npar; i++){
1.222 brouard 5607: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5608: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5609: }
1.235 brouard 5610: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5611: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5612:
1.126 brouard 5613: for(j=1; j<= nlstate; j++){
1.222 brouard 5614: for(i=1; i<=nlstate; i++){
5615: for(h=0; h<=nhstepm-1; h++){
5616: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5617: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5618: }
5619: }
1.126 brouard 5620: }
1.218 brouard 5621:
1.126 brouard 5622: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5623: for(h=0; h<=nhstepm-1; h++){
5624: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5625: }
1.126 brouard 5626: }/* End theta */
5627:
5628:
5629: for(h=0; h<=nhstepm-1; h++)
5630: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5631: for(theta=1; theta <=npar; theta++)
5632: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5633:
1.218 brouard 5634:
1.222 brouard 5635: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5636: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5637: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5638:
1.222 brouard 5639: printf("%d|",(int)age);fflush(stdout);
5640: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5641: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5642: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5643: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5644: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5645: for(ij=1;ij<=nlstate*nlstate;ij++)
5646: for(ji=1;ji<=nlstate*nlstate;ji++)
5647: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5648: }
5649: }
1.218 brouard 5650:
1.126 brouard 5651: /* Computing expectancies */
1.235 brouard 5652: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5653: for(i=1; i<=nlstate;i++)
5654: for(j=1; j<=nlstate;j++)
1.222 brouard 5655: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5656: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5657:
1.222 brouard 5658: /* 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 5659:
1.222 brouard 5660: }
1.218 brouard 5661:
1.126 brouard 5662: fprintf(ficresstdeij,"%3.0f",age );
5663: for(i=1; i<=nlstate;i++){
5664: eip=0.;
5665: vip=0.;
5666: for(j=1; j<=nlstate;j++){
1.222 brouard 5667: eip += eij[i][j][(int)age];
5668: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5669: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5670: 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 5671: }
5672: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5673: }
5674: fprintf(ficresstdeij,"\n");
1.218 brouard 5675:
1.126 brouard 5676: fprintf(ficrescveij,"%3.0f",age );
5677: for(i=1; i<=nlstate;i++)
5678: for(j=1; j<=nlstate;j++){
1.222 brouard 5679: cptj= (j-1)*nlstate+i;
5680: for(i2=1; i2<=nlstate;i2++)
5681: for(j2=1; j2<=nlstate;j2++){
5682: cptj2= (j2-1)*nlstate+i2;
5683: if(cptj2 <= cptj)
5684: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5685: }
1.126 brouard 5686: }
5687: fprintf(ficrescveij,"\n");
1.218 brouard 5688:
1.126 brouard 5689: }
5690: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5691: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5692: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5693: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5694: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5695: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5696: printf("\n");
5697: fprintf(ficlog,"\n");
1.218 brouard 5698:
1.126 brouard 5699: free_vector(xm,1,npar);
5700: free_vector(xp,1,npar);
5701: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5702: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5703: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5704: }
1.218 brouard 5705:
1.126 brouard 5706: /************ Variance ******************/
1.235 brouard 5707: 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 5708: {
5709: /* Variance of health expectancies */
5710: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5711: /* double **newm;*/
5712: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5713:
5714: /* int movingaverage(); */
5715: double **dnewm,**doldm;
5716: double **dnewmp,**doldmp;
5717: int i, j, nhstepm, hstepm, h, nstepm ;
5718: int k;
5719: double *xp;
5720: double **gp, **gm; /* for var eij */
5721: double ***gradg, ***trgradg; /*for var eij */
5722: double **gradgp, **trgradgp; /* for var p point j */
5723: double *gpp, *gmp; /* for var p point j */
5724: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5725: double ***p3mat;
5726: double age,agelim, hf;
5727: /* double ***mobaverage; */
5728: int theta;
5729: char digit[4];
5730: char digitp[25];
5731:
5732: char fileresprobmorprev[FILENAMELENGTH];
5733:
5734: if(popbased==1){
5735: if(mobilav!=0)
5736: strcpy(digitp,"-POPULBASED-MOBILAV_");
5737: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5738: }
5739: else
5740: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5741:
1.218 brouard 5742: /* if (mobilav!=0) { */
5743: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5744: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5745: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5746: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5747: /* } */
5748: /* } */
5749:
5750: strcpy(fileresprobmorprev,"PRMORPREV-");
5751: sprintf(digit,"%-d",ij);
5752: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5753: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5754: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5755: strcat(fileresprobmorprev,fileresu);
5756: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5757: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5758: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5759: }
5760: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5761: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5762: pstamp(ficresprobmorprev);
5763: 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 5764: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5765: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5766: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5767: }
5768: for(j=1;j<=cptcoveff;j++)
5769: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5770: fprintf(ficresprobmorprev,"\n");
5771:
1.218 brouard 5772: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5773: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5774: fprintf(ficresprobmorprev," p.%-d SE",j);
5775: for(i=1; i<=nlstate;i++)
5776: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5777: }
5778: fprintf(ficresprobmorprev,"\n");
5779:
5780: fprintf(ficgp,"\n# Routine varevsij");
5781: fprintf(ficgp,"\nunset title \n");
5782: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5783: 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");
5784: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5785: /* } */
5786: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5787: pstamp(ficresvij);
5788: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5789: if(popbased==1)
5790: 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);
5791: else
5792: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5793: fprintf(ficresvij,"# Age");
5794: for(i=1; i<=nlstate;i++)
5795: for(j=1; j<=nlstate;j++)
5796: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5797: fprintf(ficresvij,"\n");
5798:
5799: xp=vector(1,npar);
5800: dnewm=matrix(1,nlstate,1,npar);
5801: doldm=matrix(1,nlstate,1,nlstate);
5802: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5803: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5804:
5805: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5806: gpp=vector(nlstate+1,nlstate+ndeath);
5807: gmp=vector(nlstate+1,nlstate+ndeath);
5808: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5809:
1.218 brouard 5810: if(estepm < stepm){
5811: printf ("Problem %d lower than %d\n",estepm, stepm);
5812: }
5813: else hstepm=estepm;
5814: /* For example we decided to compute the life expectancy with the smallest unit */
5815: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5816: nhstepm is the number of hstepm from age to agelim
5817: nstepm is the number of stepm from age to agelim.
5818: Look at function hpijx to understand why because of memory size limitations,
5819: we decided (b) to get a life expectancy respecting the most precise curvature of the
5820: survival function given by stepm (the optimization length). Unfortunately it
5821: means that if the survival funtion is printed every two years of age and if
5822: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5823: results. So we changed our mind and took the option of the best precision.
5824: */
5825: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5826: agelim = AGESUP;
5827: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5828: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5829: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5830: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5831: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5832: gp=matrix(0,nhstepm,1,nlstate);
5833: gm=matrix(0,nhstepm,1,nlstate);
5834:
5835:
5836: for(theta=1; theta <=npar; theta++){
5837: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5838: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5839: }
5840:
1.242 brouard 5841: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5842:
5843: if (popbased==1) {
5844: if(mobilav ==0){
5845: for(i=1; i<=nlstate;i++)
5846: prlim[i][i]=probs[(int)age][i][ij];
5847: }else{ /* mobilav */
5848: for(i=1; i<=nlstate;i++)
5849: prlim[i][i]=mobaverage[(int)age][i][ij];
5850: }
5851: }
5852:
1.235 brouard 5853: 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 5854: for(j=1; j<= nlstate; j++){
5855: for(h=0; h<=nhstepm; h++){
5856: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5857: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5858: }
5859: }
5860: /* Next for computing probability of death (h=1 means
5861: computed over hstepm matrices product = hstepm*stepm months)
5862: as a weighted average of prlim.
5863: */
5864: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5865: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5866: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5867: }
5868: /* end probability of death */
5869:
5870: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5871: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5872:
1.242 brouard 5873: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5874:
5875: if (popbased==1) {
5876: if(mobilav ==0){
5877: for(i=1; i<=nlstate;i++)
5878: prlim[i][i]=probs[(int)age][i][ij];
5879: }else{ /* mobilav */
5880: for(i=1; i<=nlstate;i++)
5881: prlim[i][i]=mobaverage[(int)age][i][ij];
5882: }
5883: }
5884:
1.235 brouard 5885: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5886:
5887: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5888: for(h=0; h<=nhstepm; h++){
5889: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5890: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5891: }
5892: }
5893: /* This for computing probability of death (h=1 means
5894: computed over hstepm matrices product = hstepm*stepm months)
5895: as a weighted average of prlim.
5896: */
5897: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5898: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5899: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5900: }
5901: /* end probability of death */
5902:
5903: for(j=1; j<= nlstate; j++) /* vareij */
5904: for(h=0; h<=nhstepm; h++){
5905: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5906: }
5907:
5908: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5909: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5910: }
5911:
5912: } /* End theta */
5913:
5914: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5915:
5916: for(h=0; h<=nhstepm; h++) /* veij */
5917: for(j=1; j<=nlstate;j++)
5918: for(theta=1; theta <=npar; theta++)
5919: trgradg[h][j][theta]=gradg[h][theta][j];
5920:
5921: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5922: for(theta=1; theta <=npar; theta++)
5923: trgradgp[j][theta]=gradgp[theta][j];
5924:
5925:
5926: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5927: for(i=1;i<=nlstate;i++)
5928: for(j=1;j<=nlstate;j++)
5929: vareij[i][j][(int)age] =0.;
5930:
5931: for(h=0;h<=nhstepm;h++){
5932: for(k=0;k<=nhstepm;k++){
5933: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5934: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5935: for(i=1;i<=nlstate;i++)
5936: for(j=1;j<=nlstate;j++)
5937: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5938: }
5939: }
5940:
5941: /* pptj */
5942: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5943: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5944: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5945: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5946: varppt[j][i]=doldmp[j][i];
5947: /* end ppptj */
5948: /* x centered again */
5949:
1.242 brouard 5950: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5951:
5952: if (popbased==1) {
5953: if(mobilav ==0){
5954: for(i=1; i<=nlstate;i++)
5955: prlim[i][i]=probs[(int)age][i][ij];
5956: }else{ /* mobilav */
5957: for(i=1; i<=nlstate;i++)
5958: prlim[i][i]=mobaverage[(int)age][i][ij];
5959: }
5960: }
5961:
5962: /* This for computing probability of death (h=1 means
5963: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5964: as a weighted average of prlim.
5965: */
1.235 brouard 5966: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5967: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5968: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5969: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5970: }
5971: /* end probability of death */
5972:
5973: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5974: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5975: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5976: for(i=1; i<=nlstate;i++){
5977: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5978: }
5979: }
5980: fprintf(ficresprobmorprev,"\n");
5981:
5982: fprintf(ficresvij,"%.0f ",age );
5983: for(i=1; i<=nlstate;i++)
5984: for(j=1; j<=nlstate;j++){
5985: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5986: }
5987: fprintf(ficresvij,"\n");
5988: free_matrix(gp,0,nhstepm,1,nlstate);
5989: free_matrix(gm,0,nhstepm,1,nlstate);
5990: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5991: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5992: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5993: } /* End age */
5994: free_vector(gpp,nlstate+1,nlstate+ndeath);
5995: free_vector(gmp,nlstate+1,nlstate+ndeath);
5996: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5997: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5998: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5999: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6000: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6001: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6002: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6003: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6004: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6005: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6006: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6007: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6008: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6009: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6010: 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);
6011: /* 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 6012: */
1.218 brouard 6013: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6014: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6015:
1.218 brouard 6016: free_vector(xp,1,npar);
6017: free_matrix(doldm,1,nlstate,1,nlstate);
6018: free_matrix(dnewm,1,nlstate,1,npar);
6019: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6020: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6021: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6022: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6023: fclose(ficresprobmorprev);
6024: fflush(ficgp);
6025: fflush(fichtm);
6026: } /* end varevsij */
1.126 brouard 6027:
6028: /************ Variance of prevlim ******************/
1.235 brouard 6029: void varprevlim(char fileres[], 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 6030: {
1.205 brouard 6031: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6032: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6033:
1.126 brouard 6034: double **dnewm,**doldm;
6035: int i, j, nhstepm, hstepm;
6036: double *xp;
6037: double *gp, *gm;
6038: double **gradg, **trgradg;
1.208 brouard 6039: double **mgm, **mgp;
1.126 brouard 6040: double age,agelim;
6041: int theta;
6042:
6043: pstamp(ficresvpl);
6044: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6045: fprintf(ficresvpl,"# Age ");
6046: if(nresult >=1)
6047: fprintf(ficresvpl," Result# ");
1.126 brouard 6048: for(i=1; i<=nlstate;i++)
6049: fprintf(ficresvpl," %1d-%1d",i,i);
6050: fprintf(ficresvpl,"\n");
6051:
6052: xp=vector(1,npar);
6053: dnewm=matrix(1,nlstate,1,npar);
6054: doldm=matrix(1,nlstate,1,nlstate);
6055:
6056: hstepm=1*YEARM; /* Every year of age */
6057: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6058: agelim = AGESUP;
6059: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6060: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6061: if (stepm >= YEARM) hstepm=1;
6062: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6063: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6064: mgp=matrix(1,npar,1,nlstate);
6065: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6066: gp=vector(1,nlstate);
6067: gm=vector(1,nlstate);
6068:
6069: for(theta=1; theta <=npar; theta++){
6070: for(i=1; i<=npar; i++){ /* Computes gradient */
6071: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6072: }
1.209 brouard 6073: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6074: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6075: else
1.235 brouard 6076: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6077: for(i=1;i<=nlstate;i++){
1.126 brouard 6078: gp[i] = prlim[i][i];
1.208 brouard 6079: mgp[theta][i] = prlim[i][i];
6080: }
1.126 brouard 6081: for(i=1; i<=npar; i++) /* Computes gradient */
6082: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6083: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6084: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6085: else
1.235 brouard 6086: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6087: for(i=1;i<=nlstate;i++){
1.126 brouard 6088: gm[i] = prlim[i][i];
1.208 brouard 6089: mgm[theta][i] = prlim[i][i];
6090: }
1.126 brouard 6091: for(i=1;i<=nlstate;i++)
6092: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6093: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6094: } /* End theta */
6095:
6096: trgradg =matrix(1,nlstate,1,npar);
6097:
6098: for(j=1; j<=nlstate;j++)
6099: for(theta=1; theta <=npar; theta++)
6100: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6101: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6102: /* printf("\nmgm mgp %d ",(int)age); */
6103: /* for(j=1; j<=nlstate;j++){ */
6104: /* printf(" %d ",j); */
6105: /* for(theta=1; theta <=npar; theta++) */
6106: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6107: /* printf("\n "); */
6108: /* } */
6109: /* } */
6110: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6111: /* printf("\n gradg %d ",(int)age); */
6112: /* for(j=1; j<=nlstate;j++){ */
6113: /* printf("%d ",j); */
6114: /* for(theta=1; theta <=npar; theta++) */
6115: /* printf("%d %lf ",theta,gradg[theta][j]); */
6116: /* printf("\n "); */
6117: /* } */
6118: /* } */
1.126 brouard 6119:
6120: for(i=1;i<=nlstate;i++)
6121: varpl[i][(int)age] =0.;
1.209 brouard 6122: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 6123: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6124: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
6125: }else{
1.126 brouard 6126: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6127: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6128: }
1.126 brouard 6129: for(i=1;i<=nlstate;i++)
6130: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6131:
6132: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6133: if(nresult >=1)
6134: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6135: for(i=1; i<=nlstate;i++)
6136: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6137: fprintf(ficresvpl,"\n");
6138: free_vector(gp,1,nlstate);
6139: free_vector(gm,1,nlstate);
1.208 brouard 6140: free_matrix(mgm,1,npar,1,nlstate);
6141: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6142: free_matrix(gradg,1,npar,1,nlstate);
6143: free_matrix(trgradg,1,nlstate,1,npar);
6144: } /* End age */
6145:
6146: free_vector(xp,1,npar);
6147: free_matrix(doldm,1,nlstate,1,npar);
6148: free_matrix(dnewm,1,nlstate,1,nlstate);
6149:
6150: }
6151:
6152: /************ Variance of one-step probabilities ******************/
6153: 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 6154: {
6155: int i, j=0, k1, l1, tj;
6156: int k2, l2, j1, z1;
6157: int k=0, l;
6158: int first=1, first1, first2;
6159: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6160: double **dnewm,**doldm;
6161: double *xp;
6162: double *gp, *gm;
6163: double **gradg, **trgradg;
6164: double **mu;
6165: double age, cov[NCOVMAX+1];
6166: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6167: int theta;
6168: char fileresprob[FILENAMELENGTH];
6169: char fileresprobcov[FILENAMELENGTH];
6170: char fileresprobcor[FILENAMELENGTH];
6171: double ***varpij;
6172:
6173: strcpy(fileresprob,"PROB_");
6174: strcat(fileresprob,fileres);
6175: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6176: printf("Problem with resultfile: %s\n", fileresprob);
6177: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6178: }
6179: strcpy(fileresprobcov,"PROBCOV_");
6180: strcat(fileresprobcov,fileresu);
6181: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6182: printf("Problem with resultfile: %s\n", fileresprobcov);
6183: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6184: }
6185: strcpy(fileresprobcor,"PROBCOR_");
6186: strcat(fileresprobcor,fileresu);
6187: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6188: printf("Problem with resultfile: %s\n", fileresprobcor);
6189: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6190: }
6191: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6192: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6193: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6194: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6195: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6196: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6197: pstamp(ficresprob);
6198: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6199: fprintf(ficresprob,"# Age");
6200: pstamp(ficresprobcov);
6201: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6202: fprintf(ficresprobcov,"# Age");
6203: pstamp(ficresprobcor);
6204: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6205: fprintf(ficresprobcor,"# Age");
1.126 brouard 6206:
6207:
1.222 brouard 6208: for(i=1; i<=nlstate;i++)
6209: for(j=1; j<=(nlstate+ndeath);j++){
6210: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6211: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6212: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6213: }
6214: /* fprintf(ficresprob,"\n");
6215: fprintf(ficresprobcov,"\n");
6216: fprintf(ficresprobcor,"\n");
6217: */
6218: xp=vector(1,npar);
6219: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6220: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6221: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6222: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6223: first=1;
6224: fprintf(ficgp,"\n# Routine varprob");
6225: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6226: fprintf(fichtm,"\n");
6227:
1.266 ! brouard 6228: 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 6229: 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);
6230: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6231: and drawn. It helps understanding how is the covariance between two incidences.\
6232: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6233: 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 6234: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6235: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6236: standard deviations wide on each axis. <br>\
6237: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6238: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6239: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6240:
1.222 brouard 6241: cov[1]=1;
6242: /* tj=cptcoveff; */
1.225 brouard 6243: tj = (int) pow(2,cptcoveff);
1.222 brouard 6244: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6245: j1=0;
1.224 brouard 6246: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6247: if (cptcovn>0) {
6248: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6249: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6250: fprintf(ficresprob, "**********\n#\n");
6251: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6252: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6253: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6254:
1.222 brouard 6255: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6256: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6257: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6258:
6259:
1.222 brouard 6260: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6261: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6262: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6263:
1.222 brouard 6264: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6265: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6266: fprintf(ficresprobcor, "**********\n#");
6267: if(invalidvarcomb[j1]){
6268: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6269: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6270: continue;
6271: }
6272: }
6273: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6274: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6275: gp=vector(1,(nlstate)*(nlstate+ndeath));
6276: gm=vector(1,(nlstate)*(nlstate+ndeath));
6277: for (age=bage; age<=fage; age ++){
6278: cov[2]=age;
6279: if(nagesqr==1)
6280: cov[3]= age*age;
6281: for (k=1; k<=cptcovn;k++) {
6282: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6283: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6284: * 1 1 1 1 1
6285: * 2 2 1 1 1
6286: * 3 1 2 1 1
6287: */
6288: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6289: }
6290: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6291: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6292: for (k=1; k<=cptcovprod;k++)
6293: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6294:
6295:
1.222 brouard 6296: for(theta=1; theta <=npar; theta++){
6297: for(i=1; i<=npar; i++)
6298: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6299:
1.222 brouard 6300: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6301:
1.222 brouard 6302: k=0;
6303: for(i=1; i<= (nlstate); i++){
6304: for(j=1; j<=(nlstate+ndeath);j++){
6305: k=k+1;
6306: gp[k]=pmmij[i][j];
6307: }
6308: }
1.220 brouard 6309:
1.222 brouard 6310: for(i=1; i<=npar; i++)
6311: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6312:
1.222 brouard 6313: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6314: k=0;
6315: for(i=1; i<=(nlstate); i++){
6316: for(j=1; j<=(nlstate+ndeath);j++){
6317: k=k+1;
6318: gm[k]=pmmij[i][j];
6319: }
6320: }
1.220 brouard 6321:
1.222 brouard 6322: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6323: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6324: }
1.126 brouard 6325:
1.222 brouard 6326: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6327: for(theta=1; theta <=npar; theta++)
6328: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6329:
1.222 brouard 6330: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6331: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6332:
1.222 brouard 6333: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6334:
1.222 brouard 6335: k=0;
6336: for(i=1; i<=(nlstate); i++){
6337: for(j=1; j<=(nlstate+ndeath);j++){
6338: k=k+1;
6339: mu[k][(int) age]=pmmij[i][j];
6340: }
6341: }
6342: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6343: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6344: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6345:
1.222 brouard 6346: /*printf("\n%d ",(int)age);
6347: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6348: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6349: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6350: }*/
1.220 brouard 6351:
1.222 brouard 6352: fprintf(ficresprob,"\n%d ",(int)age);
6353: fprintf(ficresprobcov,"\n%d ",(int)age);
6354: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6355:
1.222 brouard 6356: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6357: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6358: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6359: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6360: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6361: }
6362: i=0;
6363: for (k=1; k<=(nlstate);k++){
6364: for (l=1; l<=(nlstate+ndeath);l++){
6365: i++;
6366: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6367: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6368: for (j=1; j<=i;j++){
6369: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6370: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6371: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6372: }
6373: }
6374: }/* end of loop for state */
6375: } /* end of loop for age */
6376: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6377: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6378: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6379: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6380:
6381: /* Confidence intervalle of pij */
6382: /*
6383: fprintf(ficgp,"\nunset parametric;unset label");
6384: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6385: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6386: 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);
6387: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6388: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6389: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6390: */
6391:
6392: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6393: first1=1;first2=2;
6394: for (k2=1; k2<=(nlstate);k2++){
6395: for (l2=1; l2<=(nlstate+ndeath);l2++){
6396: if(l2==k2) continue;
6397: j=(k2-1)*(nlstate+ndeath)+l2;
6398: for (k1=1; k1<=(nlstate);k1++){
6399: for (l1=1; l1<=(nlstate+ndeath);l1++){
6400: if(l1==k1) continue;
6401: i=(k1-1)*(nlstate+ndeath)+l1;
6402: if(i<=j) continue;
6403: for (age=bage; age<=fage; age ++){
6404: if ((int)age %5==0){
6405: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6406: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6407: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6408: mu1=mu[i][(int) age]/stepm*YEARM ;
6409: mu2=mu[j][(int) age]/stepm*YEARM;
6410: c12=cv12/sqrt(v1*v2);
6411: /* Computing eigen value of matrix of covariance */
6412: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6413: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6414: if ((lc2 <0) || (lc1 <0) ){
6415: if(first2==1){
6416: first1=0;
6417: 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);
6418: }
6419: 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);
6420: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6421: /* lc2=fabs(lc2); */
6422: }
1.220 brouard 6423:
1.222 brouard 6424: /* Eigen vectors */
6425: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6426: /*v21=sqrt(1.-v11*v11); *//* error */
6427: v21=(lc1-v1)/cv12*v11;
6428: v12=-v21;
6429: v22=v11;
6430: tnalp=v21/v11;
6431: if(first1==1){
6432: first1=0;
6433: 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);
6434: }
6435: 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);
6436: /*printf(fignu*/
6437: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6438: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6439: if(first==1){
6440: first=0;
6441: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6442: fprintf(ficgp,"\nset parametric;unset label");
6443: 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);
6444: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 ! brouard 6445: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6446: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6447: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6448: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6449: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6450: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6451: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6452: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6453: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6454: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6455: 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 6456: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
! 6457: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6458: }else{
6459: first=0;
6460: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6461: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6462: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6463: 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 6464: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
! 6465: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6466: }/* if first */
6467: } /* age mod 5 */
6468: } /* end loop age */
6469: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6470: first=1;
6471: } /*l12 */
6472: } /* k12 */
6473: } /*l1 */
6474: }/* k1 */
6475: } /* loop on combination of covariates j1 */
6476: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6477: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6478: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6479: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6480: free_vector(xp,1,npar);
6481: fclose(ficresprob);
6482: fclose(ficresprobcov);
6483: fclose(ficresprobcor);
6484: fflush(ficgp);
6485: fflush(fichtmcov);
6486: }
1.126 brouard 6487:
6488:
6489: /******************* Printing html file ***********/
1.201 brouard 6490: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6491: int lastpass, int stepm, int weightopt, char model[],\
6492: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6493: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.213 brouard 6494: double jprev1, double mprev1,double anprev1, double dateprev1, \
6495: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6496: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6497:
6498: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6499: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6500: </ul>");
1.237 brouard 6501: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6502: </ul>", model);
1.214 brouard 6503: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6504: 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",
6505: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6506: 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 6507: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6508: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6509: fprintf(fichtm,"\
6510: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6511: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6512: fprintf(fichtm,"\
1.217 brouard 6513: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6514: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6515: fprintf(fichtm,"\
1.126 brouard 6516: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6517: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6518: fprintf(fichtm,"\
1.217 brouard 6519: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6520: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6521: fprintf(fichtm,"\
1.211 brouard 6522: - (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 6523: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6524: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6525: if(prevfcast==1){
6526: fprintf(fichtm,"\
6527: - Prevalence projections by age and states: \
1.201 brouard 6528: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6529: }
1.126 brouard 6530:
6531:
1.225 brouard 6532: m=pow(2,cptcoveff);
1.222 brouard 6533: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6534:
1.264 brouard 6535: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6536:
6537: jj1=0;
6538:
6539: fprintf(fichtm," \n<ul>");
6540: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6541: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6542: if(m != 1 && TKresult[nres]!= k1)
6543: continue;
6544: jj1++;
6545: if (cptcovn > 0) {
6546: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6547: for (cpt=1; cpt<=cptcoveff;cpt++){
6548: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6549: }
6550: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6551: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6552: }
6553: fprintf(fichtm,"\">");
6554:
6555: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6556: fprintf(fichtm,"************ Results for covariates");
6557: for (cpt=1; cpt<=cptcoveff;cpt++){
6558: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6559: }
6560: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6561: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6562: }
6563: if(invalidvarcomb[k1]){
6564: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6565: continue;
6566: }
6567: fprintf(fichtm,"</a></li>");
6568: } /* cptcovn >0 */
6569: }
6570: fprintf(fichtm," \n</ul>");
6571:
1.222 brouard 6572: jj1=0;
1.237 brouard 6573:
6574: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6575: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6576: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6577: continue;
1.220 brouard 6578:
1.222 brouard 6579: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6580: jj1++;
6581: if (cptcovn > 0) {
1.264 brouard 6582: fprintf(fichtm,"\n<p><a name=\"rescov");
6583: for (cpt=1; cpt<=cptcoveff;cpt++){
6584: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6585: }
6586: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6587: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6588: }
6589: fprintf(fichtm,"\"</a>");
6590:
1.222 brouard 6591: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6592: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6593: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6594: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6595: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6596: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6597: }
1.237 brouard 6598: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6599: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6600: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6601: }
6602:
1.230 brouard 6603: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6604: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6605: if(invalidvarcomb[k1]){
6606: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6607: printf("\nCombination (%d) ignored because no cases \n",k1);
6608: continue;
6609: }
6610: }
6611: /* aij, bij */
1.259 brouard 6612: 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 6613: <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 6614: /* Pij */
1.241 brouard 6615: 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> \
6616: <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 6617: /* Quasi-incidences */
6618: 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 6619: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6620: 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 6621: 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> \
6622: <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 6623: /* Survival functions (period) in state j */
6624: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6625: 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> \
6626: <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 6627: }
6628: /* State specific survival functions (period) */
6629: for(cpt=1; cpt<=nlstate;cpt++){
6630: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6631: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6632: <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 6633: }
6634: /* Period (stable) prevalence in each health state */
6635: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6636: 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> \
6637: <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 6638: }
6639: if(backcast==1){
6640: /* Period (stable) back prevalence in each health state */
6641: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6642: 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 6643: <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 6644: }
1.217 brouard 6645: }
1.222 brouard 6646: if(prevfcast==1){
6647: /* Projection of prevalence up to period (stable) prevalence in each health state */
6648: for(cpt=1; cpt<=nlstate;cpt++){
1.258 brouard 6649: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d) up to period (stable) prevalence in state %d. Or probability to be 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> \
6650: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6651: }
6652: }
1.220 brouard 6653:
1.222 brouard 6654: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6655: 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> \
6656: <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 6657: }
6658: /* } /\* end i1 *\/ */
6659: }/* End k1 */
6660: fprintf(fichtm,"</ul>");
1.126 brouard 6661:
1.222 brouard 6662: fprintf(fichtm,"\
1.126 brouard 6663: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6664: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6665: - 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 6666: But because parameters are usually highly correlated (a higher incidence of disability \
6667: and a higher incidence of recovery can give very close observed transition) it might \
6668: be very useful to look not only at linear confidence intervals estimated from the \
6669: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6670: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6671: covariance matrix of the one-step probabilities. \
6672: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6673:
1.222 brouard 6674: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6675: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6676: fprintf(fichtm,"\
1.126 brouard 6677: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6678: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6679:
1.222 brouard 6680: fprintf(fichtm,"\
1.126 brouard 6681: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6682: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6683: fprintf(fichtm,"\
1.126 brouard 6684: - 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): \
6685: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6686: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6687: fprintf(fichtm,"\
1.126 brouard 6688: - (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): \
6689: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6690: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6691: fprintf(fichtm,"\
1.128 brouard 6692: - 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 6693: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6694: fprintf(fichtm,"\
1.128 brouard 6695: - 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 6696: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6697: fprintf(fichtm,"\
1.126 brouard 6698: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6699: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6700:
6701: /* if(popforecast==1) fprintf(fichtm,"\n */
6702: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6703: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6704: /* <br>",fileres,fileres,fileres,fileres); */
6705: /* else */
6706: /* 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 6707: fflush(fichtm);
6708: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6709:
1.225 brouard 6710: m=pow(2,cptcoveff);
1.222 brouard 6711: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6712:
1.222 brouard 6713: jj1=0;
1.237 brouard 6714:
1.241 brouard 6715: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6716: for(k1=1; k1<=m;k1++){
1.253 brouard 6717: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6718: continue;
1.222 brouard 6719: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6720: jj1++;
1.126 brouard 6721: if (cptcovn > 0) {
6722: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6723: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6724: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6725: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6726: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6727: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6728: }
6729:
1.126 brouard 6730: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6731:
1.222 brouard 6732: if(invalidvarcomb[k1]){
6733: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6734: continue;
6735: }
1.126 brouard 6736: }
6737: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6738: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6739: 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 6740: <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 6741: }
6742: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6743: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6744: true period expectancies (those weighted with period prevalences are also\
6745: drawn in addition to the population based expectancies computed using\
1.241 brouard 6746: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6747: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6748: /* } /\* end i1 *\/ */
6749: }/* End k1 */
1.241 brouard 6750: }/* End nres */
1.222 brouard 6751: fprintf(fichtm,"</ul>");
6752: fflush(fichtm);
1.126 brouard 6753: }
6754:
6755: /******************* Gnuplot file **************/
1.266 ! brouard 6756: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[], int offyear){
1.126 brouard 6757:
6758: char dirfileres[132],optfileres[132];
1.264 brouard 6759: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6760: 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 6761: int lv=0, vlv=0, kl=0;
1.130 brouard 6762: int ng=0;
1.201 brouard 6763: int vpopbased;
1.223 brouard 6764: int ioffset; /* variable offset for columns */
1.235 brouard 6765: int nres=0; /* Index of resultline */
1.266 ! brouard 6766: int istart=1; /* For starting graphs in projections */
1.219 brouard 6767:
1.126 brouard 6768: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6769: /* printf("Problem with file %s",optionfilegnuplot); */
6770: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6771: /* } */
6772:
6773: /*#ifdef windows */
6774: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6775: /*#endif */
1.225 brouard 6776: m=pow(2,cptcoveff);
1.126 brouard 6777:
1.202 brouard 6778: /* Contribution to likelihood */
6779: /* Plot the probability implied in the likelihood */
1.223 brouard 6780: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6781: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6782: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6783: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6784: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6785: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6786: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6787: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6788: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6789: 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));
6790: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6791: 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));
6792: for (i=1; i<= nlstate ; i ++) {
6793: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6794: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6795: 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);
6796: for (j=2; j<= nlstate+ndeath ; j ++) {
6797: 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);
6798: }
6799: fprintf(ficgp,";\nset out; unset ylabel;\n");
6800: }
6801: /* 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 */
6802: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6803: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6804: fprintf(ficgp,"\nset out;unset log\n");
6805: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6806:
1.126 brouard 6807: strcpy(dirfileres,optionfilefiname);
6808: strcpy(optfileres,"vpl");
1.223 brouard 6809: /* 1eme*/
1.238 brouard 6810: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6811: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6812: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6813: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6814: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6815: continue;
6816: /* We are interested in selected combination by the resultline */
1.246 brouard 6817: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6818: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 6819: strcpy(gplotlabel,"(");
1.238 brouard 6820: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6821: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6822: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6823: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6824: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6825: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6826: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6827: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6828: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 6829: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 6830: }
6831: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6832: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6833: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 6834: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6835: }
6836: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 6837: /* printf("\n#\n"); */
1.238 brouard 6838: fprintf(ficgp,"\n#\n");
6839: if(invalidvarcomb[k1]){
1.260 brouard 6840: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 6841: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6842: continue;
6843: }
1.235 brouard 6844:
1.241 brouard 6845: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6846: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.264 brouard 6847: 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 6848: 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);
6849: /* 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); */
6850: /* k1-1 error should be nres-1*/
1.238 brouard 6851: for (i=1; i<= nlstate ; i ++) {
6852: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6853: else fprintf(ficgp," %%*lf (%%*lf)");
6854: }
1.260 brouard 6855: 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 6856: for (i=1; i<= nlstate ; i ++) {
6857: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6858: else fprintf(ficgp," %%*lf (%%*lf)");
6859: }
1.260 brouard 6860: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 6861: for (i=1; i<= nlstate ; i ++) {
6862: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6863: else fprintf(ficgp," %%*lf (%%*lf)");
6864: }
1.265 brouard 6865: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
6866:
6867: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
6868: if(cptcoveff ==0){
6869: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
6870: }else{
6871: kl=0;
6872: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6873: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6874: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6875: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6876: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6877: vlv= nbcode[Tvaraff[k]][lv];
6878: kl++;
6879: /* 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 *\/ */
6880: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6881: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6882: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6883: if(k==cptcoveff){
6884: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
6885: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
6886: }else{
6887: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6888: kl++;
6889: }
6890: } /* end covariate */
6891: } /* end if no covariate */
6892:
1.238 brouard 6893: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6894: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6895: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6896: if(cptcoveff ==0){
1.245 brouard 6897: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6898: }else{
6899: kl=0;
6900: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6901: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6902: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6903: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6904: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6905: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6906: kl++;
1.238 brouard 6907: /* 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 *\/ */
6908: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6909: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6910: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6911: if(k==cptcoveff){
1.245 brouard 6912: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6913: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6914: }else{
6915: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6916: kl++;
6917: }
6918: } /* end covariate */
6919: } /* end if no covariate */
6920: } /* end if backcast */
1.264 brouard 6921: fprintf(ficgp,"\nset out ;unset label;\n");
1.238 brouard 6922: } /* nres */
1.201 brouard 6923: } /* k1 */
6924: } /* cpt */
1.235 brouard 6925:
6926:
1.126 brouard 6927: /*2 eme*/
1.238 brouard 6928: for (k1=1; k1<= m ; k1 ++){
6929: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6930: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6931: continue;
6932: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 6933: strcpy(gplotlabel,"(");
1.238 brouard 6934: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6935: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6936: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6937: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6938: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6939: vlv= nbcode[Tvaraff[k]][lv];
6940: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 6941: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6942: }
1.237 brouard 6943: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6944: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6945: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6946: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 6947: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6948: }
1.264 brouard 6949: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 6950: fprintf(ficgp,"\n#\n");
1.223 brouard 6951: if(invalidvarcomb[k1]){
6952: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6953: continue;
6954: }
1.219 brouard 6955:
1.241 brouard 6956: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6957: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 6958: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
6959: if(vpopbased==0){
1.238 brouard 6960: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 6961: }else
1.238 brouard 6962: fprintf(ficgp,"\nreplot ");
6963: for (i=1; i<= nlstate+1 ; i ++) {
6964: k=2*i;
1.261 brouard 6965: 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 6966: for (j=1; j<= nlstate+1 ; j ++) {
6967: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6968: else fprintf(ficgp," %%*lf (%%*lf)");
6969: }
6970: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6971: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 6972: 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 6973: for (j=1; j<= nlstate+1 ; j ++) {
6974: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6975: else fprintf(ficgp," %%*lf (%%*lf)");
6976: }
6977: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 6978: 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 6979: for (j=1; j<= nlstate+1 ; j ++) {
6980: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6981: else fprintf(ficgp," %%*lf (%%*lf)");
6982: }
6983: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6984: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6985: } /* state */
6986: } /* vpopbased */
1.264 brouard 6987: 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 6988: } /* end nres */
6989: } /* k1 end 2 eme*/
6990:
6991:
6992: /*3eme*/
6993: for (k1=1; k1<= m ; k1 ++){
6994: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6995: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6996: continue;
6997:
6998: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 6999: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7000: strcpy(gplotlabel,"(");
1.238 brouard 7001: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7002: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7003: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7004: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7005: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7006: vlv= nbcode[Tvaraff[k]][lv];
7007: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7008: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7009: }
7010: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7011: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7012: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7013: }
1.264 brouard 7014: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7015: fprintf(ficgp,"\n#\n");
7016: if(invalidvarcomb[k1]){
7017: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7018: continue;
7019: }
7020:
7021: /* k=2+nlstate*(2*cpt-2); */
7022: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7023: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7024: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7025: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7026: 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 7027: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7028: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7029: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7030: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7031: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7032: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7033:
1.238 brouard 7034: */
7035: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7036: 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 7037: /* 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 7038:
1.238 brouard 7039: }
1.261 brouard 7040: 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 7041: }
1.264 brouard 7042: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7043: } /* end nres */
7044: } /* end kl 3eme */
1.126 brouard 7045:
1.223 brouard 7046: /* 4eme */
1.201 brouard 7047: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7048: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7049: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7050: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7051: continue;
1.238 brouard 7052: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7053: strcpy(gplotlabel,"(");
1.238 brouard 7054: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7055: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7056: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7057: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7058: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7059: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7060: vlv= nbcode[Tvaraff[k]][lv];
7061: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7062: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7063: }
7064: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7065: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7066: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7067: }
1.264 brouard 7068: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7069: fprintf(ficgp,"\n#\n");
7070: if(invalidvarcomb[k1]){
7071: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7072: continue;
1.223 brouard 7073: }
1.238 brouard 7074:
1.241 brouard 7075: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7076: 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 7077: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7078: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7079: k=3;
7080: for (i=1; i<= nlstate ; i ++){
7081: if(i==1){
7082: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7083: }else{
7084: fprintf(ficgp,", '' ");
7085: }
7086: l=(nlstate+ndeath)*(i-1)+1;
7087: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7088: for (j=2; j<= nlstate+ndeath ; j ++)
7089: fprintf(ficgp,"+$%d",k+l+j-1);
7090: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7091: } /* nlstate */
1.264 brouard 7092: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7093: } /* end cpt state*/
7094: } /* end nres */
7095: } /* end covariate k1 */
7096:
1.220 brouard 7097: /* 5eme */
1.201 brouard 7098: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7099: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7100: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7101: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7102: continue;
1.238 brouard 7103: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7104: strcpy(gplotlabel,"(");
1.238 brouard 7105: 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);
7106: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7107: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7108: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7109: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7110: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7111: vlv= nbcode[Tvaraff[k]][lv];
7112: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7113: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7114: }
7115: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7116: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7117: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7118: }
1.264 brouard 7119: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7120: fprintf(ficgp,"\n#\n");
7121: if(invalidvarcomb[k1]){
7122: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7123: continue;
7124: }
1.227 brouard 7125:
1.241 brouard 7126: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7127: 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 7128: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7129: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7130: k=3;
7131: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7132: if(j==1)
7133: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7134: else
7135: fprintf(ficgp,", '' ");
7136: l=(nlstate+ndeath)*(cpt-1) +j;
7137: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7138: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7139: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7140: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7141: } /* nlstate */
7142: fprintf(ficgp,", '' ");
7143: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7144: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7145: l=(nlstate+ndeath)*(cpt-1) +j;
7146: if(j < nlstate)
7147: fprintf(ficgp,"$%d +",k+l);
7148: else
7149: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7150: }
1.264 brouard 7151: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7152: } /* end cpt state*/
7153: } /* end covariate */
7154: } /* end nres */
1.227 brouard 7155:
1.220 brouard 7156: /* 6eme */
1.202 brouard 7157: /* CV preval stable (period) for each covariate */
1.237 brouard 7158: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7159: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7160: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7161: continue;
1.255 brouard 7162: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7163: strcpy(gplotlabel,"(");
1.211 brouard 7164: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7165: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7166: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7167: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7168: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7169: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7170: vlv= nbcode[Tvaraff[k]][lv];
7171: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7172: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7173: }
1.237 brouard 7174: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7175: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7176: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7177: }
1.264 brouard 7178: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7179: fprintf(ficgp,"\n#\n");
1.223 brouard 7180: if(invalidvarcomb[k1]){
1.227 brouard 7181: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7182: continue;
1.223 brouard 7183: }
1.227 brouard 7184:
1.241 brouard 7185: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7186: 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 7187: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7188: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7189: k=3; /* Offset */
1.255 brouard 7190: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7191: if(i==1)
7192: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7193: else
7194: fprintf(ficgp,", '' ");
1.255 brouard 7195: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7196: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7197: for (j=2; j<= nlstate ; j ++)
7198: fprintf(ficgp,"+$%d",k+l+j-1);
7199: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7200: } /* nlstate */
1.264 brouard 7201: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7202: } /* end cpt state*/
7203: } /* end covariate */
1.227 brouard 7204:
7205:
1.220 brouard 7206: /* 7eme */
1.218 brouard 7207: if(backcast == 1){
1.217 brouard 7208: /* CV back preval stable (period) for each covariate */
1.237 brouard 7209: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7210: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7211: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7212: continue;
1.255 brouard 7213: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life ending state */
1.264 brouard 7214: strcpy(gplotlabel,"(");
7215: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7216: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7217: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7218: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7219: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7220: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7221: vlv= nbcode[Tvaraff[k]][lv];
7222: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7223: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7224: }
1.237 brouard 7225: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7226: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7227: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7228: }
1.264 brouard 7229: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7230: fprintf(ficgp,"\n#\n");
7231: if(invalidvarcomb[k1]){
7232: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7233: continue;
7234: }
7235:
1.241 brouard 7236: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.264 brouard 7237: fprintf(ficgp,"set label \"Ending alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 7238: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7239: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7240: k=3; /* Offset */
1.255 brouard 7241: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7242: if(i==1)
7243: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7244: else
7245: fprintf(ficgp,", '' ");
7246: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7247: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7248: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7249: /* 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 7250: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7251: /* for (j=2; j<= nlstate ; j ++) */
7252: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7253: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7254: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7255: } /* nlstate */
1.264 brouard 7256: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7257: } /* end cpt state*/
7258: } /* end covariate */
7259: } /* End if backcast */
7260:
1.223 brouard 7261: /* 8eme */
1.218 brouard 7262: if(prevfcast==1){
7263: /* Projection from cross-sectional to stable (period) for each covariate */
7264:
1.237 brouard 7265: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7266: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7267: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7268: continue;
1.211 brouard 7269: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7270: strcpy(gplotlabel,"(");
1.227 brouard 7271: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7272: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7273: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7274: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7275: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7276: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7277: vlv= nbcode[Tvaraff[k]][lv];
7278: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7279: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7280: }
1.237 brouard 7281: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7282: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7283: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7284: }
1.264 brouard 7285: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7286: fprintf(ficgp,"\n#\n");
7287: if(invalidvarcomb[k1]){
7288: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7289: continue;
7290: }
7291:
7292: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7293: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7294: 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 7295: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7296: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 ! brouard 7297:
! 7298: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
! 7299: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
! 7300: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
! 7301: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7302: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7303: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7304: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7305: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 ! brouard 7306: if(i==istart){
1.227 brouard 7307: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7308: }else{
7309: fprintf(ficgp,",\\\n '' ");
7310: }
7311: if(cptcoveff ==0){ /* No covariate */
7312: ioffset=2; /* Age is in 2 */
7313: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7314: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7315: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7316: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7317: fprintf(ficgp," u %d:(", ioffset);
1.266 ! brouard 7318: if(i==nlstate+1){
! 7319: fprintf(ficgp," $%d/(1.-$%d)):5 t 'pw.%d' with line lc variable ", \
! 7320: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
! 7321: fprintf(ficgp,",\\\n '' ");
! 7322: fprintf(ficgp," u %d:(",ioffset);
! 7323: fprintf(ficgp," (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", \
! 7324: offyear, \
1.227 brouard 7325: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
1.266 ! brouard 7326: }else
1.227 brouard 7327: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7328: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7329: }else{ /* more than 2 covariates */
7330: if(cptcoveff ==1){
7331: ioffset=4; /* Age is in 4 */
7332: }else{
7333: ioffset=6; /* Age is in 6 */
7334: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7335: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7336: }
7337: fprintf(ficgp," u %d:(",ioffset);
7338: kl=0;
7339: strcpy(gplotcondition,"(");
7340: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7341: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7342: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7343: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7344: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7345: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7346: kl++;
7347: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7348: kl++;
7349: if(k <cptcoveff && cptcoveff>1)
7350: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7351: }
7352: strcpy(gplotcondition+strlen(gplotcondition),")");
7353: /* 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 *\/ */
7354: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7355: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7356: /* '' 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*/
7357: if(i==nlstate+1){
1.266 ! brouard 7358: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):5 t 'p.%d' with line lc variable", gplotcondition, \
! 7359: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
! 7360: fprintf(ficgp,",\\\n '' ");
! 7361: fprintf(ficgp," u %d:(",ioffset);
! 7362: fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \
! 7363: offyear, \
1.227 brouard 7364: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
1.266 ! brouard 7365: /* '' 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 7366: }else{
7367: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7368: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7369: }
7370: } /* end if covariate */
7371: } /* nlstate */
1.264 brouard 7372: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7373: } /* end cpt state*/
7374: } /* end covariate */
7375: } /* End if prevfcast */
1.227 brouard 7376:
7377:
1.238 brouard 7378: /* 9eme writing MLE parameters */
7379: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7380: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7381: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7382: for(k=1; k <=(nlstate+ndeath); k++){
7383: if (k != i) {
1.227 brouard 7384: fprintf(ficgp,"# current state %d\n",k);
7385: for(j=1; j <=ncovmodel; j++){
7386: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7387: jk++;
7388: }
7389: fprintf(ficgp,"\n");
1.126 brouard 7390: }
7391: }
1.223 brouard 7392: }
1.187 brouard 7393: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7394:
1.145 brouard 7395: /*goto avoid;*/
1.238 brouard 7396: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7397: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7398: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7399: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7400: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7401: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7402: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7403: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7404: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7405: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7406: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7407: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7408: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7409: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7410: fprintf(ficgp,"#\n");
1.223 brouard 7411: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7412: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7413: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7414: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7415: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7416: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7417: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7418: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7419: continue;
1.264 brouard 7420: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7421: strcpy(gplotlabel,"(");
7422: sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);
7423: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7424: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7425: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7426: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7427: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7428: vlv= nbcode[Tvaraff[k]][lv];
7429: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7430: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7431: }
1.237 brouard 7432: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7433: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7434: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7435: }
1.264 brouard 7436: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7437: fprintf(ficgp,"\n#\n");
1.264 brouard 7438: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
7439: fprintf(ficgp,"\nset label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7440: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7441: if (ng==1){
7442: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7443: fprintf(ficgp,"\nunset log y");
7444: }else if (ng==2){
7445: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7446: fprintf(ficgp,"\nset log y");
7447: }else if (ng==3){
7448: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7449: fprintf(ficgp,"\nset log y");
7450: }else
7451: fprintf(ficgp,"\nunset title ");
7452: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7453: i=1;
7454: for(k2=1; k2<=nlstate; k2++) {
7455: k3=i;
7456: for(k=1; k<=(nlstate+ndeath); k++) {
7457: if (k != k2){
7458: switch( ng) {
7459: case 1:
7460: if(nagesqr==0)
7461: fprintf(ficgp," p%d+p%d*x",i,i+1);
7462: else /* nagesqr =1 */
7463: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7464: break;
7465: case 2: /* ng=2 */
7466: if(nagesqr==0)
7467: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7468: else /* nagesqr =1 */
7469: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7470: break;
7471: case 3:
7472: if(nagesqr==0)
7473: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7474: else /* nagesqr =1 */
7475: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7476: break;
7477: }
7478: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7479: ijp=1; /* product no age */
7480: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7481: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7482: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7483: if(j==Tage[ij]) { /* Product by age */
7484: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7485: if(DummyV[j]==0){
1.237 brouard 7486: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7487: }else{ /* quantitative */
7488: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
1.264 brouard 7489: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.237 brouard 7490: }
7491: ij++;
7492: }
7493: }else if(j==Tprod[ijp]) { /* */
7494: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7495: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7496: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7497: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.264 brouard 7498: /* 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)]); */
1.237 brouard 7499: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7500: }else{ /* Vn is dummy and Vm is quanti */
1.264 brouard 7501: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
1.237 brouard 7502: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7503: }
7504: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7505: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7506: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7507: }else{ /* Both quanti */
7508: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7509: }
7510: }
1.238 brouard 7511: ijp++;
1.237 brouard 7512: }
7513: } else{ /* simple covariate */
1.264 brouard 7514: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7515: if(Dummy[j]==0){
7516: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7517: }else{ /* quantitative */
7518: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7519: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7520: }
1.237 brouard 7521: } /* end simple */
7522: } /* end j */
1.223 brouard 7523: }else{
7524: i=i-ncovmodel;
7525: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7526: fprintf(ficgp," (1.");
7527: }
1.227 brouard 7528:
1.223 brouard 7529: if(ng != 1){
7530: fprintf(ficgp,")/(1");
1.227 brouard 7531:
1.264 brouard 7532: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7533: if(nagesqr==0)
1.264 brouard 7534: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7535: else /* nagesqr =1 */
1.264 brouard 7536: 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 7537:
1.223 brouard 7538: ij=1;
7539: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7540: if((j-2)==Tage[ij]) { /* Bug valgrind */
7541: if(ij <=cptcovage) { /* Bug valgrind */
1.264 brouard 7542: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7543: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7544: ij++;
7545: }
7546: }
7547: else
1.264 brouard 7548: 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 7549: }
7550: fprintf(ficgp,")");
7551: }
7552: fprintf(ficgp,")");
7553: if(ng ==2)
7554: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7555: else /* ng= 3 */
7556: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7557: }else{ /* end ng <> 1 */
7558: if( k !=k2) /* logit p11 is hard to draw */
7559: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7560: }
7561: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7562: fprintf(ficgp,",");
7563: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7564: fprintf(ficgp,",");
7565: i=i+ncovmodel;
7566: } /* end k */
7567: } /* end k2 */
1.264 brouard 7568: fprintf(ficgp,"\n set out; unset label;\n");
7569: } /* end k1 */
1.223 brouard 7570: } /* end ng */
7571: /* avoid: */
7572: fflush(ficgp);
1.126 brouard 7573: } /* end gnuplot */
7574:
7575:
7576: /*************** Moving average **************/
1.219 brouard 7577: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7578: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7579:
1.222 brouard 7580: int i, cpt, cptcod;
7581: int modcovmax =1;
7582: int mobilavrange, mob;
7583: int iage=0;
7584:
1.266 ! brouard 7585: double sum=0., sumr=0.;
1.222 brouard 7586: double age;
1.266 ! brouard 7587: double *sumnewp, *sumnewm, *sumnewmr;
! 7588: double *agemingood, *agemaxgood;
! 7589: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7590:
7591:
1.225 brouard 7592: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7593: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7594:
7595: sumnewp = vector(1,ncovcombmax);
7596: sumnewm = vector(1,ncovcombmax);
1.266 ! brouard 7597: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7598: agemingood = vector(1,ncovcombmax);
1.266 ! brouard 7599: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7600: agemaxgood = vector(1,ncovcombmax);
1.266 ! brouard 7601: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7602:
7603: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 ! brouard 7604: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7605: sumnewp[cptcod]=0.;
1.266 ! brouard 7606: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
! 7607: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7608: }
7609: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7610:
1.266 ! brouard 7611: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
! 7612: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7613: else mobilavrange=mobilav;
7614: for (age=bage; age<=fage; age++)
7615: for (i=1; i<=nlstate;i++)
7616: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7617: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7618: /* We keep the original values on the extreme ages bage, fage and for
7619: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7620: we use a 5 terms etc. until the borders are no more concerned.
7621: */
7622: for (mob=3;mob <=mobilavrange;mob=mob+2){
7623: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 ! brouard 7624: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
! 7625: sumnewm[cptcod]=0.;
! 7626: for (i=1; i<=nlstate;i++){
1.222 brouard 7627: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7628: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7629: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7630: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7631: }
7632: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 ! brouard 7633: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
! 7634: } /* end i */
! 7635: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
! 7636: } /* end cptcod */
1.222 brouard 7637: }/* end age */
7638: }/* end mob */
1.266 ! brouard 7639: }else{
! 7640: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 7641: return -1;
1.266 ! brouard 7642: }
! 7643:
! 7644: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 7645: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7646: if(invalidvarcomb[cptcod]){
7647: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7648: continue;
7649: }
1.219 brouard 7650:
1.266 ! brouard 7651: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
! 7652: sumnewm[cptcod]=0.;
! 7653: sumnewmr[cptcod]=0.;
! 7654: for (i=1; i<=nlstate;i++){
! 7655: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
! 7656: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
! 7657: }
! 7658: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
! 7659: agemingoodr[cptcod]=age;
! 7660: }
! 7661: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
! 7662: agemingood[cptcod]=age;
! 7663: }
! 7664: } /* age */
! 7665: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 7666: sumnewm[cptcod]=0.;
1.266 ! brouard 7667: sumnewmr[cptcod]=0.;
1.222 brouard 7668: for (i=1; i<=nlstate;i++){
7669: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 ! brouard 7670: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
! 7671: }
! 7672: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
! 7673: agemaxgoodr[cptcod]=age;
1.222 brouard 7674: }
7675: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 ! brouard 7676: agemaxgood[cptcod]=age;
! 7677: }
! 7678: } /* age */
! 7679: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
! 7680: /* but they will change */
! 7681: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
! 7682: sumnewm[cptcod]=0.;
! 7683: sumnewmr[cptcod]=0.;
! 7684: for (i=1; i<=nlstate;i++){
! 7685: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
! 7686: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
! 7687: }
! 7688: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
! 7689: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
! 7690: agemaxgoodr[cptcod]=age; /* age min */
! 7691: for (i=1; i<=nlstate;i++)
! 7692: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
! 7693: }else{ /* bad we change the value with the values of good ages */
! 7694: for (i=1; i<=nlstate;i++){
! 7695: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
! 7696: } /* i */
! 7697: } /* end bad */
! 7698: }else{
! 7699: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
! 7700: agemaxgood[cptcod]=age;
! 7701: }else{ /* bad we change the value with the values of good ages */
! 7702: for (i=1; i<=nlstate;i++){
! 7703: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
! 7704: } /* i */
! 7705: } /* end bad */
! 7706: }/* end else */
! 7707: sum=0.;sumr=0.;
! 7708: for (i=1; i<=nlstate;i++){
! 7709: sum+=mobaverage[(int)age][i][cptcod];
! 7710: sumr+=probs[(int)age][i][cptcod];
! 7711: }
! 7712: if(fabs(sum - 1.) > 1.e-3) { /* bad */
! 7713: 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, bage);
! 7714: } /* end bad */
! 7715: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
! 7716: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
! 7717: 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, bage);
1.222 brouard 7718: } /* end bad */
7719: }/* age */
1.266 ! brouard 7720:
! 7721: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 7722: sumnewm[cptcod]=0.;
1.266 ! brouard 7723: sumnewmr[cptcod]=0.;
1.222 brouard 7724: for (i=1; i<=nlstate;i++){
7725: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 ! brouard 7726: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
! 7727: }
! 7728: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
! 7729: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
! 7730: agemingoodr[cptcod]=age;
! 7731: for (i=1; i<=nlstate;i++)
! 7732: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
! 7733: }else{ /* bad we change the value with the values of good ages */
! 7734: for (i=1; i<=nlstate;i++){
! 7735: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
! 7736: } /* i */
! 7737: } /* end bad */
! 7738: }else{
! 7739: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
! 7740: agemingood[cptcod]=age;
! 7741: }else{ /* bad */
! 7742: for (i=1; i<=nlstate;i++){
! 7743: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
! 7744: } /* i */
! 7745: } /* end bad */
! 7746: }/* end else */
! 7747: sum=0.;sumr=0.;
! 7748: for (i=1; i<=nlstate;i++){
! 7749: sum+=mobaverage[(int)age][i][cptcod];
! 7750: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 7751: }
1.266 ! brouard 7752: if(fabs(sum - 1.) > 1.e-3) { /* bad */
! 7753: 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, fage);
! 7754: } /* end bad */
! 7755: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
! 7756: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
! 7757: 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, fage);
1.222 brouard 7758: } /* end bad */
7759: }/* age */
1.266 ! brouard 7760:
1.222 brouard 7761:
7762: for (age=bage; age<=fage; age++){
1.235 brouard 7763: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7764: sumnewp[cptcod]=0.;
7765: sumnewm[cptcod]=0.;
7766: for (i=1; i<=nlstate;i++){
7767: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7768: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7769: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7770: }
7771: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7772: }
7773: /* printf("\n"); */
7774: /* } */
1.266 ! brouard 7775:
1.222 brouard 7776: /* brutal averaging */
1.266 ! brouard 7777: /* for (i=1; i<=nlstate;i++){ */
! 7778: /* for (age=1; age<=bage; age++){ */
! 7779: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
! 7780: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
! 7781: /* } */
! 7782: /* for (age=fage; age<=AGESUP; age++){ */
! 7783: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
! 7784: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
! 7785: /* } */
! 7786: /* } /\* end i status *\/ */
! 7787: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
! 7788: /* for (age=1; age<=AGESUP; age++){ */
! 7789: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
! 7790: /* mobaverage[(int)age][i][cptcod]=0.; */
! 7791: /* } */
! 7792: /* } */
1.222 brouard 7793: }/* end cptcod */
1.266 ! brouard 7794: free_vector(agemaxgoodr,1, ncovcombmax);
! 7795: free_vector(agemaxgood,1, ncovcombmax);
! 7796: free_vector(agemingood,1, ncovcombmax);
! 7797: free_vector(agemingoodr,1, ncovcombmax);
! 7798: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 7799: free_vector(sumnewm,1, ncovcombmax);
7800: free_vector(sumnewp,1, ncovcombmax);
7801: return 0;
7802: }/* End movingaverage */
1.218 brouard 7803:
1.126 brouard 7804:
7805: /************** Forecasting ******************/
1.235 brouard 7806: void prevforecast(char fileres[], double anproj1, double mproj1, double jproj1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anproj2, double p[], int cptcoveff){
1.126 brouard 7807: /* proj1, year, month, day of starting projection
7808: agemin, agemax range of age
7809: dateprev1 dateprev2 range of dates during which prevalence is computed
7810: anproj2 year of en of projection (same day and month as proj1).
7811: */
1.235 brouard 7812: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7813: double agec; /* generic age */
7814: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7815: double *popeffectif,*popcount;
7816: double ***p3mat;
1.218 brouard 7817: /* double ***mobaverage; */
1.126 brouard 7818: char fileresf[FILENAMELENGTH];
7819:
7820: agelim=AGESUP;
1.211 brouard 7821: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7822: in each health status at the date of interview (if between dateprev1 and dateprev2).
7823: We still use firstpass and lastpass as another selection.
7824: */
1.214 brouard 7825: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7826: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7827:
1.201 brouard 7828: strcpy(fileresf,"F_");
7829: strcat(fileresf,fileresu);
1.126 brouard 7830: if((ficresf=fopen(fileresf,"w"))==NULL) {
7831: printf("Problem with forecast resultfile: %s\n", fileresf);
7832: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7833: }
1.235 brouard 7834: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7835: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7836:
1.225 brouard 7837: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7838:
7839:
7840: stepsize=(int) (stepm+YEARM-1)/YEARM;
7841: if (stepm<=12) stepsize=1;
7842: if(estepm < stepm){
7843: printf ("Problem %d lower than %d\n",estepm, stepm);
7844: }
7845: else hstepm=estepm;
7846:
7847: hstepm=hstepm/stepm;
7848: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7849: fractional in yp1 */
7850: anprojmean=yp;
7851: yp2=modf((yp1*12),&yp);
7852: mprojmean=yp;
7853: yp1=modf((yp2*30.5),&yp);
7854: jprojmean=yp;
7855: if(jprojmean==0) jprojmean=1;
7856: if(mprojmean==0) jprojmean=1;
7857:
1.227 brouard 7858: i1=pow(2,cptcoveff);
1.126 brouard 7859: if (cptcovn < 1){i1=1;}
7860:
7861: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7862:
7863: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7864:
1.126 brouard 7865: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7866: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7867: for(k=1; k<=i1;k++){
1.253 brouard 7868: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7869: continue;
1.227 brouard 7870: if(invalidvarcomb[k]){
7871: printf("\nCombination (%d) projection ignored because no cases \n",k);
7872: continue;
7873: }
7874: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7875: for(j=1;j<=cptcoveff;j++) {
7876: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7877: }
1.235 brouard 7878: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7879: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7880: }
1.227 brouard 7881: fprintf(ficresf," yearproj age");
7882: for(j=1; j<=nlstate+ndeath;j++){
7883: for(i=1; i<=nlstate;i++)
7884: fprintf(ficresf," p%d%d",i,j);
7885: fprintf(ficresf," wp.%d",j);
7886: }
7887: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7888: fprintf(ficresf,"\n");
7889: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7890: for (agec=fage; agec>=(ageminpar-1); agec--){
7891: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7892: nhstepm = nhstepm/hstepm;
7893: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7894: oldm=oldms;savm=savms;
1.235 brouard 7895: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7896:
7897: for (h=0; h<=nhstepm; h++){
7898: if (h*hstepm/YEARM*stepm ==yearp) {
7899: fprintf(ficresf,"\n");
7900: for(j=1;j<=cptcoveff;j++)
7901: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7902: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7903: }
7904: for(j=1; j<=nlstate+ndeath;j++) {
7905: ppij=0.;
7906: for(i=1; i<=nlstate;i++) {
1.266 ! brouard 7907: /* if (mobilav>=1) */
1.227 brouard 7908: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
1.266 ! brouard 7909: /* else { */ /* even if mobilav==-1 we use mobaverage */
! 7910: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
! 7911: /* } */
1.227 brouard 7912: if (h*hstepm/YEARM*stepm== yearp) {
7913: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7914: }
7915: } /* end i */
7916: if (h*hstepm/YEARM*stepm==yearp) {
7917: fprintf(ficresf," %.3f", ppij);
7918: }
7919: }/* end j */
7920: } /* end h */
7921: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7922: } /* end agec */
1.266 ! brouard 7923: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
! 7924: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 7925: } /* end yearp */
7926: } /* end k */
1.219 brouard 7927:
1.126 brouard 7928: fclose(ficresf);
1.215 brouard 7929: printf("End of Computing forecasting \n");
7930: fprintf(ficlog,"End of Computing forecasting\n");
7931:
1.126 brouard 7932: }
7933:
1.218 brouard 7934: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7935: /* void prevbackforecast(char fileres[], 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){ */
1.218 brouard 7936: /* /\* back1, year, month, day of starting backection */
7937: /* agemin, agemax range of age */
7938: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7939: /* anback2 year of en of backection (same day and month as back1). */
7940: /* *\/ */
7941: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7942: /* double agec; /\* generic age *\/ */
7943: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7944: /* double *popeffectif,*popcount; */
7945: /* double ***p3mat; */
7946: /* /\* double ***mobaverage; *\/ */
7947: /* char fileresfb[FILENAMELENGTH]; */
7948:
7949: /* agelim=AGESUP; */
7950: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7951: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7952: /* We still use firstpass and lastpass as another selection. */
7953: /* *\/ */
7954: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7955: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7956: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7957:
7958: /* strcpy(fileresfb,"FB_"); */
7959: /* strcat(fileresfb,fileresu); */
7960: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7961: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7962: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7963: /* } */
7964: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7965: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7966:
1.225 brouard 7967: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7968:
7969: /* /\* if (mobilav!=0) { *\/ */
7970: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7971: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7972: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7973: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7974: /* /\* } *\/ */
7975: /* /\* } *\/ */
7976:
7977: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7978: /* if (stepm<=12) stepsize=1; */
7979: /* if(estepm < stepm){ */
7980: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7981: /* } */
7982: /* else hstepm=estepm; */
7983:
7984: /* hstepm=hstepm/stepm; */
7985: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7986: /* fractional in yp1 *\/ */
7987: /* anprojmean=yp; */
7988: /* yp2=modf((yp1*12),&yp); */
7989: /* mprojmean=yp; */
7990: /* yp1=modf((yp2*30.5),&yp); */
7991: /* jprojmean=yp; */
7992: /* if(jprojmean==0) jprojmean=1; */
7993: /* if(mprojmean==0) jprojmean=1; */
7994:
1.225 brouard 7995: /* i1=cptcoveff; */
1.218 brouard 7996: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7997:
1.218 brouard 7998: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7999:
1.218 brouard 8000: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
8001:
8002: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
8003: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 8004: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 8005: /* k=k+1; */
8006: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 8007: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 8008: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8009: /* } */
8010: /* fprintf(ficresfb," yearbproj age"); */
8011: /* for(j=1; j<=nlstate+ndeath;j++){ */
8012: /* for(i=1; i<=nlstate;i++) */
8013: /* fprintf(ficresfb," p%d%d",i,j); */
8014: /* fprintf(ficresfb," p.%d",j); */
8015: /* } */
8016: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
8017: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
8018: /* fprintf(ficresfb,"\n"); */
8019: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
8020: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8021: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
8022: /* nhstepm = nhstepm/hstepm; */
8023: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8024: /* oldm=oldms;savm=savms; */
8025: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
8026: /* for (h=0; h<=nhstepm; h++){ */
8027: /* if (h*hstepm/YEARM*stepm ==yearp) { */
8028: /* fprintf(ficresfb,"\n"); */
1.225 brouard 8029: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 8030: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8031: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
8032: /* } */
8033: /* for(j=1; j<=nlstate+ndeath;j++) { */
8034: /* ppij=0.; */
8035: /* for(i=1; i<=nlstate;i++) { */
8036: /* if (mobilav==1) */
8037: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
8038: /* else { */
8039: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
8040: /* } */
8041: /* if (h*hstepm/YEARM*stepm== yearp) { */
8042: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
8043: /* } */
8044: /* } /\* end i *\/ */
8045: /* if (h*hstepm/YEARM*stepm==yearp) { */
8046: /* fprintf(ficresfb," %.3f", ppij); */
8047: /* } */
8048: /* }/\* end j *\/ */
8049: /* } /\* end h *\/ */
8050: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8051: /* } /\* end agec *\/ */
8052: /* } /\* end yearp *\/ */
8053: /* } /\* end cptcod *\/ */
8054: /* } /\* end cptcov *\/ */
8055:
8056: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8057:
8058: /* fclose(ficresfb); */
8059: /* printf("End of Computing Back forecasting \n"); */
8060: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 8061:
1.218 brouard 8062: /* } */
1.217 brouard 8063:
1.126 brouard 8064: /************** Forecasting *****not tested NB*************/
1.227 brouard 8065: /* 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 8066:
1.227 brouard 8067: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8068: /* int *popage; */
8069: /* double calagedatem, agelim, kk1, kk2; */
8070: /* double *popeffectif,*popcount; */
8071: /* double ***p3mat,***tabpop,***tabpopprev; */
8072: /* /\* double ***mobaverage; *\/ */
8073: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8074:
1.227 brouard 8075: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8076: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8077: /* agelim=AGESUP; */
8078: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8079:
1.227 brouard 8080: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8081:
8082:
1.227 brouard 8083: /* strcpy(filerespop,"POP_"); */
8084: /* strcat(filerespop,fileresu); */
8085: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8086: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8087: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8088: /* } */
8089: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8090: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8091:
1.227 brouard 8092: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8093:
1.227 brouard 8094: /* /\* if (mobilav!=0) { *\/ */
8095: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8096: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8097: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8098: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8099: /* /\* } *\/ */
8100: /* /\* } *\/ */
1.126 brouard 8101:
1.227 brouard 8102: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8103: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8104:
1.227 brouard 8105: /* agelim=AGESUP; */
1.126 brouard 8106:
1.227 brouard 8107: /* hstepm=1; */
8108: /* hstepm=hstepm/stepm; */
1.218 brouard 8109:
1.227 brouard 8110: /* if (popforecast==1) { */
8111: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8112: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8113: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8114: /* } */
8115: /* popage=ivector(0,AGESUP); */
8116: /* popeffectif=vector(0,AGESUP); */
8117: /* popcount=vector(0,AGESUP); */
1.126 brouard 8118:
1.227 brouard 8119: /* i=1; */
8120: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8121:
1.227 brouard 8122: /* imx=i; */
8123: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8124: /* } */
1.218 brouard 8125:
1.227 brouard 8126: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8127: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8128: /* k=k+1; */
8129: /* fprintf(ficrespop,"\n#******"); */
8130: /* for(j=1;j<=cptcoveff;j++) { */
8131: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8132: /* } */
8133: /* fprintf(ficrespop,"******\n"); */
8134: /* fprintf(ficrespop,"# Age"); */
8135: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8136: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8137:
1.227 brouard 8138: /* for (cpt=0; cpt<=0;cpt++) { */
8139: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8140:
1.227 brouard 8141: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8142: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8143: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8144:
1.227 brouard 8145: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8146: /* oldm=oldms;savm=savms; */
8147: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8148:
1.227 brouard 8149: /* for (h=0; h<=nhstepm; h++){ */
8150: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8151: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8152: /* } */
8153: /* for(j=1; j<=nlstate+ndeath;j++) { */
8154: /* kk1=0.;kk2=0; */
8155: /* for(i=1; i<=nlstate;i++) { */
8156: /* if (mobilav==1) */
8157: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8158: /* else { */
8159: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8160: /* } */
8161: /* } */
8162: /* if (h==(int)(calagedatem+12*cpt)){ */
8163: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8164: /* /\*fprintf(ficrespop," %.3f", kk1); */
8165: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8166: /* } */
8167: /* } */
8168: /* for(i=1; i<=nlstate;i++){ */
8169: /* kk1=0.; */
8170: /* for(j=1; j<=nlstate;j++){ */
8171: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8172: /* } */
8173: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8174: /* } */
1.218 brouard 8175:
1.227 brouard 8176: /* if (h==(int)(calagedatem+12*cpt)) */
8177: /* for(j=1; j<=nlstate;j++) */
8178: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8179: /* } */
8180: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8181: /* } */
8182: /* } */
1.218 brouard 8183:
1.227 brouard 8184: /* /\******\/ */
1.218 brouard 8185:
1.227 brouard 8186: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8187: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8188: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8189: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8190: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8191:
1.227 brouard 8192: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8193: /* oldm=oldms;savm=savms; */
8194: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8195: /* for (h=0; h<=nhstepm; h++){ */
8196: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8197: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8198: /* } */
8199: /* for(j=1; j<=nlstate+ndeath;j++) { */
8200: /* kk1=0.;kk2=0; */
8201: /* for(i=1; i<=nlstate;i++) { */
8202: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8203: /* } */
8204: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8205: /* } */
8206: /* } */
8207: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8208: /* } */
8209: /* } */
8210: /* } */
8211: /* } */
1.218 brouard 8212:
1.227 brouard 8213: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8214:
1.227 brouard 8215: /* if (popforecast==1) { */
8216: /* free_ivector(popage,0,AGESUP); */
8217: /* free_vector(popeffectif,0,AGESUP); */
8218: /* free_vector(popcount,0,AGESUP); */
8219: /* } */
8220: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8221: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8222: /* fclose(ficrespop); */
8223: /* } /\* End of popforecast *\/ */
1.218 brouard 8224:
1.126 brouard 8225: int fileappend(FILE *fichier, char *optionfich)
8226: {
8227: if((fichier=fopen(optionfich,"a"))==NULL) {
8228: printf("Problem with file: %s\n", optionfich);
8229: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8230: return (0);
8231: }
8232: fflush(fichier);
8233: return (1);
8234: }
8235:
8236:
8237: /**************** function prwizard **********************/
8238: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8239: {
8240:
8241: /* Wizard to print covariance matrix template */
8242:
1.164 brouard 8243: char ca[32], cb[32];
8244: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8245: int numlinepar;
8246:
8247: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8248: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8249: for(i=1; i <=nlstate; i++){
8250: jj=0;
8251: for(j=1; j <=nlstate+ndeath; j++){
8252: if(j==i) continue;
8253: jj++;
8254: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8255: printf("%1d%1d",i,j);
8256: fprintf(ficparo,"%1d%1d",i,j);
8257: for(k=1; k<=ncovmodel;k++){
8258: /* printf(" %lf",param[i][j][k]); */
8259: /* fprintf(ficparo," %lf",param[i][j][k]); */
8260: printf(" 0.");
8261: fprintf(ficparo," 0.");
8262: }
8263: printf("\n");
8264: fprintf(ficparo,"\n");
8265: }
8266: }
8267: printf("# Scales (for hessian or gradient estimation)\n");
8268: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8269: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8270: for(i=1; i <=nlstate; i++){
8271: jj=0;
8272: for(j=1; j <=nlstate+ndeath; j++){
8273: if(j==i) continue;
8274: jj++;
8275: fprintf(ficparo,"%1d%1d",i,j);
8276: printf("%1d%1d",i,j);
8277: fflush(stdout);
8278: for(k=1; k<=ncovmodel;k++){
8279: /* printf(" %le",delti3[i][j][k]); */
8280: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8281: printf(" 0.");
8282: fprintf(ficparo," 0.");
8283: }
8284: numlinepar++;
8285: printf("\n");
8286: fprintf(ficparo,"\n");
8287: }
8288: }
8289: printf("# Covariance matrix\n");
8290: /* # 121 Var(a12)\n\ */
8291: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8292: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8293: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8294: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8295: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8296: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8297: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8298: fflush(stdout);
8299: fprintf(ficparo,"# Covariance matrix\n");
8300: /* # 121 Var(a12)\n\ */
8301: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8302: /* # ...\n\ */
8303: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8304:
8305: for(itimes=1;itimes<=2;itimes++){
8306: jj=0;
8307: for(i=1; i <=nlstate; i++){
8308: for(j=1; j <=nlstate+ndeath; j++){
8309: if(j==i) continue;
8310: for(k=1; k<=ncovmodel;k++){
8311: jj++;
8312: ca[0]= k+'a'-1;ca[1]='\0';
8313: if(itimes==1){
8314: printf("#%1d%1d%d",i,j,k);
8315: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8316: }else{
8317: printf("%1d%1d%d",i,j,k);
8318: fprintf(ficparo,"%1d%1d%d",i,j,k);
8319: /* printf(" %.5le",matcov[i][j]); */
8320: }
8321: ll=0;
8322: for(li=1;li <=nlstate; li++){
8323: for(lj=1;lj <=nlstate+ndeath; lj++){
8324: if(lj==li) continue;
8325: for(lk=1;lk<=ncovmodel;lk++){
8326: ll++;
8327: if(ll<=jj){
8328: cb[0]= lk +'a'-1;cb[1]='\0';
8329: if(ll<jj){
8330: if(itimes==1){
8331: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8332: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8333: }else{
8334: printf(" 0.");
8335: fprintf(ficparo," 0.");
8336: }
8337: }else{
8338: if(itimes==1){
8339: printf(" Var(%s%1d%1d)",ca,i,j);
8340: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8341: }else{
8342: printf(" 0.");
8343: fprintf(ficparo," 0.");
8344: }
8345: }
8346: }
8347: } /* end lk */
8348: } /* end lj */
8349: } /* end li */
8350: printf("\n");
8351: fprintf(ficparo,"\n");
8352: numlinepar++;
8353: } /* end k*/
8354: } /*end j */
8355: } /* end i */
8356: } /* end itimes */
8357:
8358: } /* end of prwizard */
8359: /******************* Gompertz Likelihood ******************************/
8360: double gompertz(double x[])
8361: {
8362: double A,B,L=0.0,sump=0.,num=0.;
8363: int i,n=0; /* n is the size of the sample */
8364:
1.220 brouard 8365: for (i=1;i<=imx ; i++) {
1.126 brouard 8366: sump=sump+weight[i];
8367: /* sump=sump+1;*/
8368: num=num+1;
8369: }
8370:
8371:
8372: /* for (i=0; i<=imx; i++)
8373: 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]);*/
8374:
8375: for (i=1;i<=imx ; i++)
8376: {
8377: if (cens[i] == 1 && wav[i]>1)
8378: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8379:
8380: if (cens[i] == 0 && wav[i]>1)
8381: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8382: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8383:
8384: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8385: if (wav[i] > 1 ) { /* ??? */
8386: L=L+A*weight[i];
8387: /* 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]);*/
8388: }
8389: }
8390:
8391: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8392:
8393: return -2*L*num/sump;
8394: }
8395:
1.136 brouard 8396: #ifdef GSL
8397: /******************* Gompertz_f Likelihood ******************************/
8398: double gompertz_f(const gsl_vector *v, void *params)
8399: {
8400: double A,B,LL=0.0,sump=0.,num=0.;
8401: double *x= (double *) v->data;
8402: int i,n=0; /* n is the size of the sample */
8403:
8404: for (i=0;i<=imx-1 ; i++) {
8405: sump=sump+weight[i];
8406: /* sump=sump+1;*/
8407: num=num+1;
8408: }
8409:
8410:
8411: /* for (i=0; i<=imx; i++)
8412: 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]);*/
8413: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8414: for (i=1;i<=imx ; i++)
8415: {
8416: if (cens[i] == 1 && wav[i]>1)
8417: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8418:
8419: if (cens[i] == 0 && wav[i]>1)
8420: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8421: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8422:
8423: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8424: if (wav[i] > 1 ) { /* ??? */
8425: LL=LL+A*weight[i];
8426: /* 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]);*/
8427: }
8428: }
8429:
8430: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8431: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8432:
8433: return -2*LL*num/sump;
8434: }
8435: #endif
8436:
1.126 brouard 8437: /******************* Printing html file ***********/
1.201 brouard 8438: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8439: int lastpass, int stepm, int weightopt, char model[],\
8440: int imx, double p[],double **matcov,double agemortsup){
8441: int i,k;
8442:
8443: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8444: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8445: for (i=1;i<=2;i++)
8446: 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 8447: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8448: fprintf(fichtm,"</ul>");
8449:
8450: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8451:
8452: 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>");
8453:
8454: for (k=agegomp;k<(agemortsup-2);k++)
8455: 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]);
8456:
8457:
8458: fflush(fichtm);
8459: }
8460:
8461: /******************* Gnuplot file **************/
1.201 brouard 8462: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8463:
8464: char dirfileres[132],optfileres[132];
1.164 brouard 8465:
1.126 brouard 8466: int ng;
8467:
8468:
8469: /*#ifdef windows */
8470: fprintf(ficgp,"cd \"%s\" \n",pathc);
8471: /*#endif */
8472:
8473:
8474: strcpy(dirfileres,optionfilefiname);
8475: strcpy(optfileres,"vpl");
1.199 brouard 8476: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8477: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8478: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8479: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8480: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8481:
8482: }
8483:
1.136 brouard 8484: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8485: {
1.126 brouard 8486:
1.136 brouard 8487: /*-------- data file ----------*/
8488: FILE *fic;
8489: char dummy[]=" ";
1.240 brouard 8490: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8491: int lstra;
1.136 brouard 8492: int linei, month, year,iout;
8493: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8494: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8495: char *stratrunc;
1.223 brouard 8496:
1.240 brouard 8497: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8498: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8499:
1.240 brouard 8500: for(v=1; v <=ncovcol;v++){
8501: DummyV[v]=0;
8502: FixedV[v]=0;
8503: }
8504: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8505: DummyV[v]=1;
8506: FixedV[v]=0;
8507: }
8508: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8509: DummyV[v]=0;
8510: FixedV[v]=1;
8511: }
8512: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8513: DummyV[v]=1;
8514: FixedV[v]=1;
8515: }
8516: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8517: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8518: 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]);
8519: }
1.126 brouard 8520:
1.136 brouard 8521: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8522: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8523: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8524: }
1.126 brouard 8525:
1.136 brouard 8526: i=1;
8527: linei=0;
8528: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8529: linei=linei+1;
8530: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8531: if(line[j] == '\t')
8532: line[j] = ' ';
8533: }
8534: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8535: ;
8536: };
8537: line[j+1]=0; /* Trims blanks at end of line */
8538: if(line[0]=='#'){
8539: fprintf(ficlog,"Comment line\n%s\n",line);
8540: printf("Comment line\n%s\n",line);
8541: continue;
8542: }
8543: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8544: strcpy(line, linetmp);
1.223 brouard 8545:
8546: /* Loops on waves */
8547: for (j=maxwav;j>=1;j--){
8548: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8549: cutv(stra, strb, line, ' ');
8550: if(strb[0]=='.') { /* Missing value */
8551: lval=-1;
8552: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8553: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8554: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8555: 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);
8556: 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);
8557: return 1;
8558: }
8559: }else{
8560: errno=0;
8561: /* what_kind_of_number(strb); */
8562: dval=strtod(strb,&endptr);
8563: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8564: /* if(strb != endptr && *endptr == '\0') */
8565: /* dval=dlval; */
8566: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8567: if( strb[0]=='\0' || (*endptr != '\0')){
8568: 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);
8569: 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);
8570: return 1;
8571: }
8572: cotqvar[j][iv][i]=dval;
8573: cotvar[j][ntv+iv][i]=dval;
8574: }
8575: strcpy(line,stra);
1.223 brouard 8576: }/* end loop ntqv */
1.225 brouard 8577:
1.223 brouard 8578: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8579: cutv(stra, strb, line, ' ');
8580: if(strb[0]=='.') { /* Missing value */
8581: lval=-1;
8582: }else{
8583: errno=0;
8584: lval=strtol(strb,&endptr,10);
8585: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8586: if( strb[0]=='\0' || (*endptr != '\0')){
8587: 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);
8588: 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);
8589: return 1;
8590: }
8591: }
8592: if(lval <-1 || lval >1){
8593: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8594: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8595: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8596: For example, for multinomial values like 1, 2 and 3,\n \
8597: build V1=0 V2=0 for the reference value (1),\n \
8598: V1=1 V2=0 for (2) \n \
1.223 brouard 8599: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8600: output of IMaCh is often meaningless.\n \
1.223 brouard 8601: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8602: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8603: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8604: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8605: For example, for multinomial values like 1, 2 and 3,\n \
8606: build V1=0 V2=0 for the reference value (1),\n \
8607: V1=1 V2=0 for (2) \n \
1.223 brouard 8608: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8609: output of IMaCh is often meaningless.\n \
1.223 brouard 8610: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8611: return 1;
8612: }
8613: cotvar[j][iv][i]=(double)(lval);
8614: strcpy(line,stra);
1.223 brouard 8615: }/* end loop ntv */
1.225 brouard 8616:
1.223 brouard 8617: /* Statuses at wave */
1.137 brouard 8618: cutv(stra, strb, line, ' ');
1.223 brouard 8619: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8620: lval=-1;
1.136 brouard 8621: }else{
1.238 brouard 8622: errno=0;
8623: lval=strtol(strb,&endptr,10);
8624: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8625: if( strb[0]=='\0' || (*endptr != '\0')){
8626: 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);
8627: 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);
8628: return 1;
8629: }
1.136 brouard 8630: }
1.225 brouard 8631:
1.136 brouard 8632: s[j][i]=lval;
1.225 brouard 8633:
1.223 brouard 8634: /* Date of Interview */
1.136 brouard 8635: strcpy(line,stra);
8636: cutv(stra, strb,line,' ');
1.169 brouard 8637: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8638: }
1.169 brouard 8639: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8640: month=99;
8641: year=9999;
1.136 brouard 8642: }else{
1.225 brouard 8643: 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);
8644: 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);
8645: return 1;
1.136 brouard 8646: }
8647: anint[j][i]= (double) year;
8648: mint[j][i]= (double)month;
8649: strcpy(line,stra);
1.223 brouard 8650: } /* End loop on waves */
1.225 brouard 8651:
1.223 brouard 8652: /* Date of death */
1.136 brouard 8653: cutv(stra, strb,line,' ');
1.169 brouard 8654: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8655: }
1.169 brouard 8656: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8657: month=99;
8658: year=9999;
8659: }else{
1.141 brouard 8660: 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 8661: 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);
8662: return 1;
1.136 brouard 8663: }
8664: andc[i]=(double) year;
8665: moisdc[i]=(double) month;
8666: strcpy(line,stra);
8667:
1.223 brouard 8668: /* Date of birth */
1.136 brouard 8669: cutv(stra, strb,line,' ');
1.169 brouard 8670: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8671: }
1.169 brouard 8672: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8673: month=99;
8674: year=9999;
8675: }else{
1.141 brouard 8676: 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);
8677: 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 8678: return 1;
1.136 brouard 8679: }
8680: if (year==9999) {
1.141 brouard 8681: 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);
8682: 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 8683: return 1;
8684:
1.136 brouard 8685: }
8686: annais[i]=(double)(year);
8687: moisnais[i]=(double)(month);
8688: strcpy(line,stra);
1.225 brouard 8689:
1.223 brouard 8690: /* Sample weight */
1.136 brouard 8691: cutv(stra, strb,line,' ');
8692: errno=0;
8693: dval=strtod(strb,&endptr);
8694: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8695: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8696: 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 8697: fflush(ficlog);
8698: return 1;
8699: }
8700: weight[i]=dval;
8701: strcpy(line,stra);
1.225 brouard 8702:
1.223 brouard 8703: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8704: cutv(stra, strb, line, ' ');
8705: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8706: lval=-1;
1.223 brouard 8707: }else{
1.225 brouard 8708: errno=0;
8709: /* what_kind_of_number(strb); */
8710: dval=strtod(strb,&endptr);
8711: /* if(strb != endptr && *endptr == '\0') */
8712: /* dval=dlval; */
8713: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8714: if( strb[0]=='\0' || (*endptr != '\0')){
8715: 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);
8716: 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);
8717: return 1;
8718: }
8719: coqvar[iv][i]=dval;
1.226 brouard 8720: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8721: }
8722: strcpy(line,stra);
8723: }/* end loop nqv */
1.136 brouard 8724:
1.223 brouard 8725: /* Covariate values */
1.136 brouard 8726: for (j=ncovcol;j>=1;j--){
8727: cutv(stra, strb,line,' ');
1.223 brouard 8728: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8729: lval=-1;
1.136 brouard 8730: }else{
1.225 brouard 8731: errno=0;
8732: lval=strtol(strb,&endptr,10);
8733: if( strb[0]=='\0' || (*endptr != '\0')){
8734: 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);
8735: 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);
8736: return 1;
8737: }
1.136 brouard 8738: }
8739: if(lval <-1 || lval >1){
1.225 brouard 8740: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8741: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8742: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8743: For example, for multinomial values like 1, 2 and 3,\n \
8744: build V1=0 V2=0 for the reference value (1),\n \
8745: V1=1 V2=0 for (2) \n \
1.136 brouard 8746: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8747: output of IMaCh is often meaningless.\n \
1.136 brouard 8748: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8749: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8750: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8751: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8752: For example, for multinomial values like 1, 2 and 3,\n \
8753: build V1=0 V2=0 for the reference value (1),\n \
8754: V1=1 V2=0 for (2) \n \
1.136 brouard 8755: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8756: output of IMaCh is often meaningless.\n \
1.136 brouard 8757: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8758: return 1;
1.136 brouard 8759: }
8760: covar[j][i]=(double)(lval);
8761: strcpy(line,stra);
8762: }
8763: lstra=strlen(stra);
1.225 brouard 8764:
1.136 brouard 8765: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8766: stratrunc = &(stra[lstra-9]);
8767: num[i]=atol(stratrunc);
8768: }
8769: else
8770: num[i]=atol(stra);
8771: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8772: 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;}*/
8773:
8774: i=i+1;
8775: } /* End loop reading data */
1.225 brouard 8776:
1.136 brouard 8777: *imax=i-1; /* Number of individuals */
8778: fclose(fic);
1.225 brouard 8779:
1.136 brouard 8780: return (0);
1.164 brouard 8781: /* endread: */
1.225 brouard 8782: printf("Exiting readdata: ");
8783: fclose(fic);
8784: return (1);
1.223 brouard 8785: }
1.126 brouard 8786:
1.234 brouard 8787: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8788: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8789: while (*p2 == ' ')
1.234 brouard 8790: p2++;
8791: /* while ((*p1++ = *p2++) !=0) */
8792: /* ; */
8793: /* do */
8794: /* while (*p2 == ' ') */
8795: /* p2++; */
8796: /* while (*p1++ == *p2++); */
8797: *stri=p2;
1.145 brouard 8798: }
8799:
1.235 brouard 8800: int decoderesult ( char resultline[], int nres)
1.230 brouard 8801: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8802: {
1.235 brouard 8803: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8804: char resultsav[MAXLINE];
1.234 brouard 8805: int resultmodel[MAXLINE];
8806: int modelresult[MAXLINE];
1.230 brouard 8807: char stra[80], strb[80], strc[80], strd[80],stre[80];
8808:
1.234 brouard 8809: removefirstspace(&resultline);
1.233 brouard 8810: printf("decoderesult:%s\n",resultline);
1.230 brouard 8811:
8812: if (strstr(resultline,"v") !=0){
8813: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8814: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8815: return 1;
8816: }
8817: trimbb(resultsav, resultline);
8818: if (strlen(resultsav) >1){
8819: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8820: }
1.253 brouard 8821: if(j == 0){ /* Resultline but no = */
8822: TKresult[nres]=0; /* Combination for the nresult and the model */
8823: return (0);
8824: }
8825:
1.234 brouard 8826: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8827: 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);
8828: 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);
8829: }
8830: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8831: if(nbocc(resultsav,'=') >1){
8832: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8833: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8834: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8835: }else
8836: cutl(strc,strd,resultsav,'=');
1.230 brouard 8837: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8838:
1.230 brouard 8839: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8840: Tvarsel[k]=atoi(strc);
8841: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8842: /* cptcovsel++; */
8843: if (nbocc(stra,'=') >0)
8844: strcpy(resultsav,stra); /* and analyzes it */
8845: }
1.235 brouard 8846: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8847: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8848: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8849: match=0;
1.236 brouard 8850: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8851: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8852: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8853: match=1;
8854: break;
8855: }
8856: }
8857: if(match == 0){
8858: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8859: }
8860: }
8861: }
1.235 brouard 8862: /* Checking for missing or useless values in comparison of current model needs */
8863: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8864: match=0;
1.235 brouard 8865: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8866: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8867: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8868: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8869: ++match;
8870: }
8871: }
8872: }
8873: if(match == 0){
8874: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8875: }else if(match > 1){
8876: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8877: }
8878: }
1.235 brouard 8879:
1.234 brouard 8880: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8881: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8882: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8883: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8884: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8885: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8886: /* 1 0 0 0 */
8887: /* 2 1 0 0 */
8888: /* 3 0 1 0 */
8889: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8890: /* 5 0 0 1 */
8891: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8892: /* 7 0 1 1 */
8893: /* 8 1 1 1 */
1.237 brouard 8894: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8895: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8896: /* V5*age V5 known which value for nres? */
8897: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8898: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8899: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8900: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8901: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8902: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8903: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8904: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8905: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8906: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8907: k4++;;
8908: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8909: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8910: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8911: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8912: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8913: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8914: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8915: k4q++;;
8916: }
8917: }
1.234 brouard 8918:
1.235 brouard 8919: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8920: return (0);
8921: }
1.235 brouard 8922:
1.230 brouard 8923: int decodemodel( char model[], int lastobs)
8924: /**< This routine decodes the model and returns:
1.224 brouard 8925: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8926: * - nagesqr = 1 if age*age in the model, otherwise 0.
8927: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8928: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8929: * - cptcovage number of covariates with age*products =2
8930: * - cptcovs number of simple covariates
8931: * - 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
8932: * which is a new column after the 9 (ncovcol) variables.
8933: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8934: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8935: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8936: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8937: */
1.136 brouard 8938: {
1.238 brouard 8939: int i, j, k, ks, v;
1.227 brouard 8940: int j1, k1, k2, k3, k4;
1.136 brouard 8941: char modelsav[80];
1.145 brouard 8942: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8943: char *strpt;
1.136 brouard 8944:
1.145 brouard 8945: /*removespace(model);*/
1.136 brouard 8946: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8947: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8948: if (strstr(model,"AGE") !=0){
1.192 brouard 8949: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8950: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8951: return 1;
8952: }
1.141 brouard 8953: if (strstr(model,"v") !=0){
8954: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8955: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8956: return 1;
8957: }
1.187 brouard 8958: strcpy(modelsav,model);
8959: if ((strpt=strstr(model,"age*age")) !=0){
8960: printf(" strpt=%s, model=%s\n",strpt, model);
8961: if(strpt != model){
1.234 brouard 8962: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8963: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8964: corresponding column of parameters.\n",model);
1.234 brouard 8965: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8966: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8967: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8968: return 1;
1.225 brouard 8969: }
1.187 brouard 8970: nagesqr=1;
8971: if (strstr(model,"+age*age") !=0)
1.234 brouard 8972: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8973: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8974: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8975: else
1.234 brouard 8976: substrchaine(modelsav, model, "age*age");
1.187 brouard 8977: }else
8978: nagesqr=0;
8979: if (strlen(modelsav) >1){
8980: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8981: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8982: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8983: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8984: * cst, age and age*age
8985: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8986: /* including age products which are counted in cptcovage.
8987: * but the covariates which are products must be treated
8988: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8989: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8990: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8991:
8992:
1.187 brouard 8993: /* Design
8994: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8995: * < ncovcol=8 >
8996: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8997: * k= 1 2 3 4 5 6 7 8
8998: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8999: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9000: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9001: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9002: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9003: * Tage[++cptcovage]=k
9004: * if products, new covar are created after ncovcol with k1
9005: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9006: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9007: * 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
9008: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9009: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9010: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9011: * < ncovcol=8 >
9012: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9013: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9014: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9015: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9016: * p Tprod[1]@2={ 6, 5}
9017: *p Tvard[1][1]@4= {7, 8, 5, 6}
9018: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9019: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9020: *How to reorganize?
9021: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9022: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9023: * {2, 1, 4, 8, 5, 6, 3, 7}
9024: * Struct []
9025: */
1.225 brouard 9026:
1.187 brouard 9027: /* This loop fills the array Tvar from the string 'model'.*/
9028: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9029: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9030: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9031: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9032: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9033: /* k=1 Tvar[1]=2 (from V2) */
9034: /* k=5 Tvar[5] */
9035: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9036: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9037: /* } */
1.198 brouard 9038: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9039: /*
9040: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9041: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9042: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9043: }
1.187 brouard 9044: cptcovage=0;
9045: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9046: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9047: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9048: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9049: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9050: /*scanf("%d",i);*/
9051: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9052: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9053: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9054: /* covar is not filled and then is empty */
9055: cptcovprod--;
9056: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9057: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9058: Typevar[k]=1; /* 1 for age product */
9059: cptcovage++; /* Sums the number of covariates which include age as a product */
9060: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9061: /*printf("stre=%s ", stre);*/
9062: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9063: cptcovprod--;
9064: cutl(stre,strb,strc,'V');
9065: Tvar[k]=atoi(stre);
9066: Typevar[k]=1; /* 1 for age product */
9067: cptcovage++;
9068: Tage[cptcovage]=k;
9069: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9070: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9071: cptcovn++;
9072: cptcovprodnoage++;k1++;
9073: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9074: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9075: because this model-covariate is a construction we invent a new column
9076: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9077: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9078: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9079: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9080: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9081: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9082: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9083: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9084: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9085: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9086: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9087: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9088: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9089: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9090: for (i=1; i<=lastobs;i++){
9091: /* Computes the new covariate which is a product of
9092: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9093: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9094: }
9095: } /* End age is not in the model */
9096: } /* End if model includes a product */
9097: else { /* no more sum */
9098: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9099: /* scanf("%d",i);*/
9100: cutl(strd,strc,strb,'V');
9101: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9102: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9103: Tvar[k]=atoi(strd);
9104: Typevar[k]=0; /* 0 for simple covariates */
9105: }
9106: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9107: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9108: scanf("%d",i);*/
1.187 brouard 9109: } /* end of loop + on total covariates */
9110: } /* end if strlen(modelsave == 0) age*age might exist */
9111: } /* end if strlen(model == 0) */
1.136 brouard 9112:
9113: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9114: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9115:
1.136 brouard 9116: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9117: printf("cptcovprod=%d ", cptcovprod);
9118: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9119: scanf("%d ",i);*/
9120:
9121:
1.230 brouard 9122: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9123: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9124: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9125: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9126: k = 1 2 3 4 5 6 7 8 9
9127: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9128: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9129: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9130: Dummy[k] 1 0 0 0 3 1 1 2 3
9131: Tmodelind[combination of covar]=k;
1.225 brouard 9132: */
9133: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9134: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9135: /* 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 9136: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9137: printf("Model=%s\n\
9138: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9139: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9140: 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);
9141: fprintf(ficlog,"Model=%s\n\
9142: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9143: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9144: 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 9145: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9146: 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 */
9147: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9148: Fixed[k]= 0;
9149: Dummy[k]= 0;
1.225 brouard 9150: ncoveff++;
1.232 brouard 9151: ncovf++;
1.234 brouard 9152: nsd++;
9153: modell[k].maintype= FTYPE;
9154: TvarsD[nsd]=Tvar[k];
9155: TvarsDind[nsd]=k;
9156: TvarF[ncovf]=Tvar[k];
9157: TvarFind[ncovf]=k;
9158: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9159: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9160: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9161: Fixed[k]= 0;
9162: Dummy[k]= 0;
9163: ncoveff++;
9164: ncovf++;
9165: modell[k].maintype= FTYPE;
9166: TvarF[ncovf]=Tvar[k];
9167: TvarFind[ncovf]=k;
1.230 brouard 9168: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9169: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9170: }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 9171: Fixed[k]= 0;
9172: Dummy[k]= 1;
1.230 brouard 9173: nqfveff++;
1.234 brouard 9174: modell[k].maintype= FTYPE;
9175: modell[k].subtype= FQ;
9176: nsq++;
9177: TvarsQ[nsq]=Tvar[k];
9178: TvarsQind[nsq]=k;
1.232 brouard 9179: ncovf++;
1.234 brouard 9180: TvarF[ncovf]=Tvar[k];
9181: TvarFind[ncovf]=k;
1.231 brouard 9182: 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 9183: 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 9184: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9185: Fixed[k]= 1;
9186: Dummy[k]= 0;
1.225 brouard 9187: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9188: modell[k].maintype= VTYPE;
9189: modell[k].subtype= VD;
9190: nsd++;
9191: TvarsD[nsd]=Tvar[k];
9192: TvarsDind[nsd]=k;
9193: ncovv++; /* Only simple time varying variables */
9194: TvarV[ncovv]=Tvar[k];
1.242 brouard 9195: 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 9196: 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 */
9197: 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 9198: 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);
9199: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9200: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9201: Fixed[k]= 1;
9202: Dummy[k]= 1;
9203: nqtveff++;
9204: modell[k].maintype= VTYPE;
9205: modell[k].subtype= VQ;
9206: ncovv++; /* Only simple time varying variables */
9207: nsq++;
9208: TvarsQ[nsq]=Tvar[k];
9209: TvarsQind[nsq]=k;
9210: TvarV[ncovv]=Tvar[k];
1.242 brouard 9211: 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 9212: 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 */
9213: 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 9214: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9215: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9216: 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 9217: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9218: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9219: ncova++;
9220: TvarA[ncova]=Tvar[k];
9221: TvarAind[ncova]=k;
1.231 brouard 9222: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9223: Fixed[k]= 2;
9224: Dummy[k]= 2;
9225: modell[k].maintype= ATYPE;
9226: modell[k].subtype= APFD;
9227: /* ncoveff++; */
1.227 brouard 9228: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9229: Fixed[k]= 2;
9230: Dummy[k]= 3;
9231: modell[k].maintype= ATYPE;
9232: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9233: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9234: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9235: Fixed[k]= 3;
9236: Dummy[k]= 2;
9237: modell[k].maintype= ATYPE;
9238: modell[k].subtype= APVD; /* Product age * varying dummy */
9239: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9240: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9241: Fixed[k]= 3;
9242: Dummy[k]= 3;
9243: modell[k].maintype= ATYPE;
9244: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9245: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9246: }
9247: }else if (Typevar[k] == 2) { /* product without age */
9248: k1=Tposprod[k];
9249: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9250: if(Tvard[k1][2] <=ncovcol){
9251: Fixed[k]= 1;
9252: Dummy[k]= 0;
9253: modell[k].maintype= FTYPE;
9254: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9255: ncovf++; /* Fixed variables without age */
9256: TvarF[ncovf]=Tvar[k];
9257: TvarFind[ncovf]=k;
9258: }else if(Tvard[k1][2] <=ncovcol+nqv){
9259: Fixed[k]= 0; /* or 2 ?*/
9260: Dummy[k]= 1;
9261: modell[k].maintype= FTYPE;
9262: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9263: ncovf++; /* Varying variables without age */
9264: TvarF[ncovf]=Tvar[k];
9265: TvarFind[ncovf]=k;
9266: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9267: Fixed[k]= 1;
9268: Dummy[k]= 0;
9269: modell[k].maintype= VTYPE;
9270: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9271: ncovv++; /* Varying variables without age */
9272: TvarV[ncovv]=Tvar[k];
9273: TvarVind[ncovv]=k;
9274: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9275: Fixed[k]= 1;
9276: Dummy[k]= 1;
9277: modell[k].maintype= VTYPE;
9278: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9279: ncovv++; /* Varying variables without age */
9280: TvarV[ncovv]=Tvar[k];
9281: TvarVind[ncovv]=k;
9282: }
1.227 brouard 9283: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9284: if(Tvard[k1][2] <=ncovcol){
9285: Fixed[k]= 0; /* or 2 ?*/
9286: Dummy[k]= 1;
9287: modell[k].maintype= FTYPE;
9288: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9289: ncovf++; /* Fixed variables without age */
9290: TvarF[ncovf]=Tvar[k];
9291: TvarFind[ncovf]=k;
9292: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9293: Fixed[k]= 1;
9294: Dummy[k]= 1;
9295: modell[k].maintype= VTYPE;
9296: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9297: ncovv++; /* Varying variables without age */
9298: TvarV[ncovv]=Tvar[k];
9299: TvarVind[ncovv]=k;
9300: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9301: Fixed[k]= 1;
9302: Dummy[k]= 1;
9303: modell[k].maintype= VTYPE;
9304: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9305: ncovv++; /* Varying variables without age */
9306: TvarV[ncovv]=Tvar[k];
9307: TvarVind[ncovv]=k;
9308: ncovv++; /* Varying variables without age */
9309: TvarV[ncovv]=Tvar[k];
9310: TvarVind[ncovv]=k;
9311: }
1.227 brouard 9312: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9313: if(Tvard[k1][2] <=ncovcol){
9314: Fixed[k]= 1;
9315: Dummy[k]= 1;
9316: modell[k].maintype= VTYPE;
9317: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9318: ncovv++; /* Varying variables without age */
9319: TvarV[ncovv]=Tvar[k];
9320: TvarVind[ncovv]=k;
9321: }else if(Tvard[k1][2] <=ncovcol+nqv){
9322: Fixed[k]= 1;
9323: Dummy[k]= 1;
9324: modell[k].maintype= VTYPE;
9325: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9326: ncovv++; /* Varying variables without age */
9327: TvarV[ncovv]=Tvar[k];
9328: TvarVind[ncovv]=k;
9329: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9330: Fixed[k]= 1;
9331: Dummy[k]= 0;
9332: modell[k].maintype= VTYPE;
9333: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9334: ncovv++; /* Varying variables without age */
9335: TvarV[ncovv]=Tvar[k];
9336: TvarVind[ncovv]=k;
9337: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9338: Fixed[k]= 1;
9339: Dummy[k]= 1;
9340: modell[k].maintype= VTYPE;
9341: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9342: ncovv++; /* Varying variables without age */
9343: TvarV[ncovv]=Tvar[k];
9344: TvarVind[ncovv]=k;
9345: }
1.227 brouard 9346: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9347: if(Tvard[k1][2] <=ncovcol){
9348: Fixed[k]= 1;
9349: Dummy[k]= 1;
9350: modell[k].maintype= VTYPE;
9351: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9352: ncovv++; /* Varying variables without age */
9353: TvarV[ncovv]=Tvar[k];
9354: TvarVind[ncovv]=k;
9355: }else if(Tvard[k1][2] <=ncovcol+nqv){
9356: Fixed[k]= 1;
9357: Dummy[k]= 1;
9358: modell[k].maintype= VTYPE;
9359: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9360: ncovv++; /* Varying variables without age */
9361: TvarV[ncovv]=Tvar[k];
9362: TvarVind[ncovv]=k;
9363: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9364: Fixed[k]= 1;
9365: Dummy[k]= 1;
9366: modell[k].maintype= VTYPE;
9367: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9368: ncovv++; /* Varying variables without age */
9369: TvarV[ncovv]=Tvar[k];
9370: TvarVind[ncovv]=k;
9371: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9372: Fixed[k]= 1;
9373: Dummy[k]= 1;
9374: modell[k].maintype= VTYPE;
9375: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9376: ncovv++; /* Varying variables without age */
9377: TvarV[ncovv]=Tvar[k];
9378: TvarVind[ncovv]=k;
9379: }
1.227 brouard 9380: }else{
1.240 brouard 9381: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9382: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9383: } /*end k1*/
1.225 brouard 9384: }else{
1.226 brouard 9385: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9386: 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 9387: }
1.227 brouard 9388: 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 9389: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9390: 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]);
9391: }
9392: /* Searching for doublons in the model */
9393: for(k1=1; k1<= cptcovt;k1++){
9394: for(k2=1; k2 <k1;k2++){
9395: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9396: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9397: if(Tvar[k1]==Tvar[k2]){
9398: 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]]);
9399: 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);
9400: return(1);
9401: }
9402: }else if (Typevar[k1] ==2){
9403: k3=Tposprod[k1];
9404: k4=Tposprod[k2];
9405: 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])) ){
9406: 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]]);
9407: 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);
9408: return(1);
9409: }
9410: }
1.227 brouard 9411: }
9412: }
1.225 brouard 9413: }
9414: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9415: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9416: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9417: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9418: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9419: /*endread:*/
1.225 brouard 9420: printf("Exiting decodemodel: ");
9421: return (1);
1.136 brouard 9422: }
9423:
1.169 brouard 9424: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9425: {/* Check ages at death */
1.136 brouard 9426: int i, m;
1.218 brouard 9427: int firstone=0;
9428:
1.136 brouard 9429: for (i=1; i<=imx; i++) {
9430: for(m=2; (m<= maxwav); m++) {
9431: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9432: anint[m][i]=9999;
1.216 brouard 9433: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9434: s[m][i]=-1;
1.136 brouard 9435: }
9436: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9437: *nberr = *nberr + 1;
1.218 brouard 9438: if(firstone == 0){
9439: firstone=1;
1.260 brouard 9440: 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 9441: }
1.262 brouard 9442: 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 9443: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9444: }
9445: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9446: (*nberr)++;
1.259 brouard 9447: 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 9448: 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 9449: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9450: }
9451: }
9452: }
9453:
9454: for (i=1; i<=imx; i++) {
9455: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9456: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9457: 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 9458: if (s[m][i] >= nlstate+1) {
1.169 brouard 9459: if(agedc[i]>0){
9460: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9461: agev[m][i]=agedc[i];
1.214 brouard 9462: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9463: }else {
1.136 brouard 9464: if ((int)andc[i]!=9999){
9465: nbwarn++;
9466: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9467: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9468: agev[m][i]=-1;
9469: }
9470: }
1.169 brouard 9471: } /* agedc > 0 */
1.214 brouard 9472: } /* end if */
1.136 brouard 9473: else if(s[m][i] !=9){ /* Standard case, age in fractional
9474: years but with the precision of a month */
9475: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9476: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9477: agev[m][i]=1;
9478: else if(agev[m][i] < *agemin){
9479: *agemin=agev[m][i];
9480: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9481: }
9482: else if(agev[m][i] >*agemax){
9483: *agemax=agev[m][i];
1.156 brouard 9484: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9485: }
9486: /*agev[m][i]=anint[m][i]-annais[i];*/
9487: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9488: } /* en if 9*/
1.136 brouard 9489: else { /* =9 */
1.214 brouard 9490: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9491: agev[m][i]=1;
9492: s[m][i]=-1;
9493: }
9494: }
1.214 brouard 9495: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9496: agev[m][i]=1;
1.214 brouard 9497: else{
9498: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9499: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9500: agev[m][i]=0;
9501: }
9502: } /* End for lastpass */
9503: }
1.136 brouard 9504:
9505: for (i=1; i<=imx; i++) {
9506: for(m=firstpass; (m<=lastpass); m++){
9507: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9508: (*nberr)++;
1.136 brouard 9509: 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);
9510: 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);
9511: return 1;
9512: }
9513: }
9514: }
9515:
9516: /*for (i=1; i<=imx; i++){
9517: for (m=firstpass; (m<lastpass); m++){
9518: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9519: }
9520:
9521: }*/
9522:
9523:
1.139 brouard 9524: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9525: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9526:
9527: return (0);
1.164 brouard 9528: /* endread:*/
1.136 brouard 9529: printf("Exiting calandcheckages: ");
9530: return (1);
9531: }
9532:
1.172 brouard 9533: #if defined(_MSC_VER)
9534: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9535: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9536: //#include "stdafx.h"
9537: //#include <stdio.h>
9538: //#include <tchar.h>
9539: //#include <windows.h>
9540: //#include <iostream>
9541: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9542:
9543: LPFN_ISWOW64PROCESS fnIsWow64Process;
9544:
9545: BOOL IsWow64()
9546: {
9547: BOOL bIsWow64 = FALSE;
9548:
9549: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9550: // (HANDLE, PBOOL);
9551:
9552: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9553:
9554: HMODULE module = GetModuleHandle(_T("kernel32"));
9555: const char funcName[] = "IsWow64Process";
9556: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9557: GetProcAddress(module, funcName);
9558:
9559: if (NULL != fnIsWow64Process)
9560: {
9561: if (!fnIsWow64Process(GetCurrentProcess(),
9562: &bIsWow64))
9563: //throw std::exception("Unknown error");
9564: printf("Unknown error\n");
9565: }
9566: return bIsWow64 != FALSE;
9567: }
9568: #endif
1.177 brouard 9569:
1.191 brouard 9570: void syscompilerinfo(int logged)
1.167 brouard 9571: {
9572: /* #include "syscompilerinfo.h"*/
1.185 brouard 9573: /* command line Intel compiler 32bit windows, XP compatible:*/
9574: /* /GS /W3 /Gy
9575: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9576: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9577: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9578: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9579: */
9580: /* 64 bits */
1.185 brouard 9581: /*
9582: /GS /W3 /Gy
9583: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9584: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9585: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9586: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9587: /* Optimization are useless and O3 is slower than O2 */
9588: /*
9589: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9590: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9591: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9592: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9593: */
1.186 brouard 9594: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9595: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9596: /PDB:"visual studio
9597: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9598: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9599: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9600: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9601: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9602: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9603: uiAccess='false'"
9604: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9605: /NOLOGO /TLBID:1
9606: */
1.177 brouard 9607: #if defined __INTEL_COMPILER
1.178 brouard 9608: #if defined(__GNUC__)
9609: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9610: #endif
1.177 brouard 9611: #elif defined(__GNUC__)
1.179 brouard 9612: #ifndef __APPLE__
1.174 brouard 9613: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9614: #endif
1.177 brouard 9615: struct utsname sysInfo;
1.178 brouard 9616: int cross = CROSS;
9617: if (cross){
9618: printf("Cross-");
1.191 brouard 9619: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9620: }
1.174 brouard 9621: #endif
9622:
1.171 brouard 9623: #include <stdint.h>
1.178 brouard 9624:
1.191 brouard 9625: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9626: #if defined(__clang__)
1.191 brouard 9627: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9628: #endif
9629: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9630: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9631: #endif
9632: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9633: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9634: #endif
9635: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9636: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9637: #endif
9638: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9639: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9640: #endif
9641: #if defined(_MSC_VER)
1.191 brouard 9642: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9643: #endif
9644: #if defined(__PGI)
1.191 brouard 9645: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9646: #endif
9647: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9648: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9649: #endif
1.191 brouard 9650: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9651:
1.167 brouard 9652: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9653: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9654: // Windows (x64 and x86)
1.191 brouard 9655: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9656: #elif __unix__ // all unices, not all compilers
9657: // Unix
1.191 brouard 9658: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9659: #elif __linux__
9660: // linux
1.191 brouard 9661: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9662: #elif __APPLE__
1.174 brouard 9663: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9664: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9665: #endif
9666:
9667: /* __MINGW32__ */
9668: /* __CYGWIN__ */
9669: /* __MINGW64__ */
9670: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9671: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9672: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9673: /* _WIN64 // Defined for applications for Win64. */
9674: /* _M_X64 // Defined for compilations that target x64 processors. */
9675: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9676:
1.167 brouard 9677: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9678: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9679: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9680: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9681: #else
1.191 brouard 9682: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9683: #endif
9684:
1.169 brouard 9685: #if defined(__GNUC__)
9686: # if defined(__GNUC_PATCHLEVEL__)
9687: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9688: + __GNUC_MINOR__ * 100 \
9689: + __GNUC_PATCHLEVEL__)
9690: # else
9691: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9692: + __GNUC_MINOR__ * 100)
9693: # endif
1.174 brouard 9694: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9695: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9696:
9697: if (uname(&sysInfo) != -1) {
9698: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9699: 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 9700: }
9701: else
9702: perror("uname() error");
1.179 brouard 9703: //#ifndef __INTEL_COMPILER
9704: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9705: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9706: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9707: #endif
1.169 brouard 9708: #endif
1.172 brouard 9709:
9710: // void main()
9711: // {
1.169 brouard 9712: #if defined(_MSC_VER)
1.174 brouard 9713: if (IsWow64()){
1.191 brouard 9714: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9715: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9716: }
9717: else{
1.191 brouard 9718: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9719: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9720: }
1.172 brouard 9721: // printf("\nPress Enter to continue...");
9722: // getchar();
9723: // }
9724:
1.169 brouard 9725: #endif
9726:
1.167 brouard 9727:
1.219 brouard 9728: }
1.136 brouard 9729:
1.219 brouard 9730: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9731: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9732: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9733: /* double ftolpl = 1.e-10; */
1.180 brouard 9734: double age, agebase, agelim;
1.203 brouard 9735: double tot;
1.180 brouard 9736:
1.202 brouard 9737: strcpy(filerespl,"PL_");
9738: strcat(filerespl,fileresu);
9739: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9740: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9741: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9742: }
1.227 brouard 9743: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9744: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9745: pstamp(ficrespl);
1.203 brouard 9746: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9747: fprintf(ficrespl,"#Age ");
9748: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9749: fprintf(ficrespl,"\n");
1.180 brouard 9750:
1.219 brouard 9751: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9752:
1.219 brouard 9753: agebase=ageminpar;
9754: agelim=agemaxpar;
1.180 brouard 9755:
1.227 brouard 9756: /* i1=pow(2,ncoveff); */
1.234 brouard 9757: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9758: if (cptcovn < 1){i1=1;}
1.180 brouard 9759:
1.238 brouard 9760: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9761: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9762: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9763: continue;
1.235 brouard 9764:
1.238 brouard 9765: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9766: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9767: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9768: /* k=k+1; */
9769: /* to clean */
9770: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9771: fprintf(ficrespl,"#******");
9772: printf("#******");
9773: fprintf(ficlog,"#******");
9774: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9775: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9776: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9777: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9778: }
9779: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9780: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9781: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9782: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9783: }
9784: fprintf(ficrespl,"******\n");
9785: printf("******\n");
9786: fprintf(ficlog,"******\n");
9787: if(invalidvarcomb[k]){
9788: printf("\nCombination (%d) ignored because no case \n",k);
9789: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9790: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9791: continue;
9792: }
1.219 brouard 9793:
1.238 brouard 9794: fprintf(ficrespl,"#Age ");
9795: for(j=1;j<=cptcoveff;j++) {
9796: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9797: }
9798: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9799: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9800:
1.238 brouard 9801: for (age=agebase; age<=agelim; age++){
9802: /* for (age=agebase; age<=agebase; age++){ */
9803: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9804: fprintf(ficrespl,"%.0f ",age );
9805: for(j=1;j<=cptcoveff;j++)
9806: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9807: tot=0.;
9808: for(i=1; i<=nlstate;i++){
9809: tot += prlim[i][i];
9810: fprintf(ficrespl," %.5f", prlim[i][i]);
9811: }
9812: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9813: } /* Age */
9814: /* was end of cptcod */
9815: } /* cptcov */
9816: } /* nres */
1.219 brouard 9817: return 0;
1.180 brouard 9818: }
9819:
1.218 brouard 9820: 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){
9821: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9822:
9823: /* Computes the back prevalence limit for any combination of covariate values
9824: * at any age between ageminpar and agemaxpar
9825: */
1.235 brouard 9826: int i, j, k, i1, nres=0 ;
1.217 brouard 9827: /* double ftolpl = 1.e-10; */
9828: double age, agebase, agelim;
9829: double tot;
1.218 brouard 9830: /* double ***mobaverage; */
9831: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9832:
9833: strcpy(fileresplb,"PLB_");
9834: strcat(fileresplb,fileresu);
9835: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9836: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9837: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9838: }
9839: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9840: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9841: pstamp(ficresplb);
9842: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9843: fprintf(ficresplb,"#Age ");
9844: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9845: fprintf(ficresplb,"\n");
9846:
1.218 brouard 9847:
9848: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9849:
9850: agebase=ageminpar;
9851: agelim=agemaxpar;
9852:
9853:
1.227 brouard 9854: i1=pow(2,cptcoveff);
1.218 brouard 9855: if (cptcovn < 1){i1=1;}
1.227 brouard 9856:
1.238 brouard 9857: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9858: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9859: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9860: continue;
9861: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9862: fprintf(ficresplb,"#******");
9863: printf("#******");
9864: fprintf(ficlog,"#******");
9865: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9866: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9867: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9868: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9869: }
9870: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9871: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9872: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9873: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9874: }
9875: fprintf(ficresplb,"******\n");
9876: printf("******\n");
9877: fprintf(ficlog,"******\n");
9878: if(invalidvarcomb[k]){
9879: printf("\nCombination (%d) ignored because no cases \n",k);
9880: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9881: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9882: continue;
9883: }
1.218 brouard 9884:
1.238 brouard 9885: fprintf(ficresplb,"#Age ");
9886: for(j=1;j<=cptcoveff;j++) {
9887: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9888: }
9889: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9890: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9891:
9892:
1.238 brouard 9893: for (age=agebase; age<=agelim; age++){
9894: /* for (age=agebase; age<=agebase; age++){ */
9895: if(mobilavproj > 0){
9896: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9897: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9898: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9899: }else if (mobilavproj == 0){
9900: 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);
9901: 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);
9902: exit(1);
9903: }else{
9904: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9905: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 ! brouard 9906: /* printf("TOTOT\n"); */
! 9907: /* exit(1); */
1.238 brouard 9908: }
9909: fprintf(ficresplb,"%.0f ",age );
9910: for(j=1;j<=cptcoveff;j++)
9911: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9912: tot=0.;
9913: for(i=1; i<=nlstate;i++){
9914: tot += bprlim[i][i];
9915: fprintf(ficresplb," %.5f", bprlim[i][i]);
9916: }
9917: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9918: } /* Age */
9919: /* was end of cptcod */
1.255 brouard 9920: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 9921: } /* end of any combination */
9922: } /* end of nres */
1.218 brouard 9923: /* hBijx(p, bage, fage); */
9924: /* fclose(ficrespijb); */
9925:
9926: return 0;
1.217 brouard 9927: }
1.218 brouard 9928:
1.180 brouard 9929: int hPijx(double *p, int bage, int fage){
9930: /*------------- h Pij x at various ages ------------*/
9931:
9932: int stepsize;
9933: int agelim;
9934: int hstepm;
9935: int nhstepm;
1.235 brouard 9936: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9937:
9938: double agedeb;
9939: double ***p3mat;
9940:
1.201 brouard 9941: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9942: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9943: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9944: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9945: }
9946: printf("Computing pij: result on file '%s' \n", filerespij);
9947: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9948:
9949: stepsize=(int) (stepm+YEARM-1)/YEARM;
9950: /*if (stepm<=24) stepsize=2;*/
9951:
9952: agelim=AGESUP;
9953: hstepm=stepsize*YEARM; /* Every year of age */
9954: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9955:
1.180 brouard 9956: /* hstepm=1; aff par mois*/
9957: pstamp(ficrespij);
9958: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9959: i1= pow(2,cptcoveff);
1.218 brouard 9960: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9961: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9962: /* k=k+1; */
1.235 brouard 9963: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9964: for(k=1; k<=i1;k++){
1.253 brouard 9965: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9966: continue;
1.183 brouard 9967: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9968: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9969: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9970: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9971: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9972: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9973: }
1.183 brouard 9974: fprintf(ficrespij,"******\n");
9975:
9976: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9977: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9978: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9979:
9980: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9981:
1.183 brouard 9982: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9983: oldm=oldms;savm=savms;
1.235 brouard 9984: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9985: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9986: for(i=1; i<=nlstate;i++)
9987: for(j=1; j<=nlstate+ndeath;j++)
9988: fprintf(ficrespij," %1d-%1d",i,j);
9989: fprintf(ficrespij,"\n");
9990: for (h=0; h<=nhstepm; h++){
9991: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9992: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9993: for(i=1; i<=nlstate;i++)
9994: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9995: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9996: fprintf(ficrespij,"\n");
9997: }
1.183 brouard 9998: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9999: fprintf(ficrespij,"\n");
10000: }
1.180 brouard 10001: /*}*/
10002: }
1.218 brouard 10003: return 0;
1.180 brouard 10004: }
1.218 brouard 10005:
10006: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10007: /*------------- h Bij x at various ages ------------*/
10008:
10009: int stepsize;
1.218 brouard 10010: /* int agelim; */
10011: int ageminl;
1.217 brouard 10012: int hstepm;
10013: int nhstepm;
1.238 brouard 10014: int h, i, i1, j, k, nres;
1.218 brouard 10015:
1.217 brouard 10016: double agedeb;
10017: double ***p3mat;
1.218 brouard 10018:
10019: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10020: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10021: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10022: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10023: }
10024: printf("Computing pij back: result on file '%s' \n", filerespijb);
10025: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10026:
10027: stepsize=(int) (stepm+YEARM-1)/YEARM;
10028: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10029:
1.218 brouard 10030: /* agelim=AGESUP; */
10031: ageminl=30;
10032: hstepm=stepsize*YEARM; /* Every year of age */
10033: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10034:
10035: /* hstepm=1; aff par mois*/
10036: pstamp(ficrespijb);
1.255 brouard 10037: 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 10038: i1= pow(2,cptcoveff);
1.218 brouard 10039: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10040: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10041: /* k=k+1; */
1.238 brouard 10042: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10043: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10044: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10045: continue;
10046: fprintf(ficrespijb,"\n#****** ");
10047: for(j=1;j<=cptcoveff;j++)
10048: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10049: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10050: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10051: }
10052: fprintf(ficrespijb,"******\n");
1.264 brouard 10053: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10054: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10055: continue;
10056: }
10057:
10058: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10059: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10060: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10061: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10062: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10063:
10064: /* nhstepm=nhstepm*YEARM; aff par mois*/
10065:
1.266 ! brouard 10066: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
! 10067: /* and memory limitations if stepm is small */
! 10068:
1.238 brouard 10069: /* oldm=oldms;savm=savms; */
10070: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
10071: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
10072: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10073: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10074: for(i=1; i<=nlstate;i++)
10075: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10076: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10077: fprintf(ficrespijb,"\n");
1.238 brouard 10078: for (h=0; h<=nhstepm; h++){
10079: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10080: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10081: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10082: for(i=1; i<=nlstate;i++)
10083: for(j=1; j<=nlstate+ndeath;j++)
10084: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10085: fprintf(ficrespijb,"\n");
10086: }
10087: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10088: fprintf(ficrespijb,"\n");
10089: } /* end age deb */
10090: } /* end combination */
10091: } /* end nres */
1.218 brouard 10092: return 0;
10093: } /* hBijx */
1.217 brouard 10094:
1.180 brouard 10095:
1.136 brouard 10096: /***********************************************/
10097: /**************** Main Program *****************/
10098: /***********************************************/
10099:
10100: int main(int argc, char *argv[])
10101: {
10102: #ifdef GSL
10103: const gsl_multimin_fminimizer_type *T;
10104: size_t iteri = 0, it;
10105: int rval = GSL_CONTINUE;
10106: int status = GSL_SUCCESS;
10107: double ssval;
10108: #endif
10109: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10110: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10111: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10112: int jj, ll, li, lj, lk;
1.136 brouard 10113: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10114: int num_filled;
1.136 brouard 10115: int itimes;
10116: int NDIM=2;
10117: int vpopbased=0;
1.235 brouard 10118: int nres=0;
1.258 brouard 10119: int endishere=0;
1.136 brouard 10120:
1.164 brouard 10121: char ca[32], cb[32];
1.136 brouard 10122: /* FILE *fichtm; *//* Html File */
10123: /* FILE *ficgp;*/ /*Gnuplot File */
10124: struct stat info;
1.191 brouard 10125: double agedeb=0.;
1.194 brouard 10126:
10127: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10128: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10129:
1.165 brouard 10130: double fret;
1.191 brouard 10131: double dum=0.; /* Dummy variable */
1.136 brouard 10132: double ***p3mat;
1.218 brouard 10133: /* double ***mobaverage; */
1.164 brouard 10134:
10135: char line[MAXLINE];
1.197 brouard 10136: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10137:
1.234 brouard 10138: char modeltemp[MAXLINE];
1.230 brouard 10139: char resultline[MAXLINE];
10140:
1.136 brouard 10141: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10142: char *tok, *val; /* pathtot */
1.136 brouard 10143: int firstobs=1, lastobs=10;
1.195 brouard 10144: int c, h , cpt, c2;
1.191 brouard 10145: int jl=0;
10146: int i1, j1, jk, stepsize=0;
1.194 brouard 10147: int count=0;
10148:
1.164 brouard 10149: int *tab;
1.136 brouard 10150: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10151: int backcast=0;
1.136 brouard 10152: int mobilav=0,popforecast=0;
1.191 brouard 10153: int hstepm=0, nhstepm=0;
1.136 brouard 10154: int agemortsup;
10155: float sumlpop=0.;
10156: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10157: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10158:
1.191 brouard 10159: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10160: double ftolpl=FTOL;
10161: double **prlim;
1.217 brouard 10162: double **bprlim;
1.136 brouard 10163: double ***param; /* Matrix of parameters */
1.251 brouard 10164: double ***paramstart; /* Matrix of starting parameter values */
10165: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10166: double **matcov; /* Matrix of covariance */
1.203 brouard 10167: double **hess; /* Hessian matrix */
1.136 brouard 10168: double ***delti3; /* Scale */
10169: double *delti; /* Scale */
10170: double ***eij, ***vareij;
10171: double **varpl; /* Variances of prevalence limits by age */
10172: double *epj, vepp;
1.164 brouard 10173:
1.136 brouard 10174: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 10175: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
10176:
1.136 brouard 10177: double **ximort;
1.145 brouard 10178: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10179: int *dcwave;
10180:
1.164 brouard 10181: char z[1]="c";
1.136 brouard 10182:
10183: /*char *strt;*/
10184: char strtend[80];
1.126 brouard 10185:
1.164 brouard 10186:
1.126 brouard 10187: /* setlocale (LC_ALL, ""); */
10188: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10189: /* textdomain (PACKAGE); */
10190: /* setlocale (LC_CTYPE, ""); */
10191: /* setlocale (LC_MESSAGES, ""); */
10192:
10193: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10194: rstart_time = time(NULL);
10195: /* (void) gettimeofday(&start_time,&tzp);*/
10196: start_time = *localtime(&rstart_time);
1.126 brouard 10197: curr_time=start_time;
1.157 brouard 10198: /*tml = *localtime(&start_time.tm_sec);*/
10199: /* strcpy(strstart,asctime(&tml)); */
10200: strcpy(strstart,asctime(&start_time));
1.126 brouard 10201:
10202: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10203: /* tp.tm_sec = tp.tm_sec +86400; */
10204: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10205: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10206: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10207: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10208: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10209: /* strt=asctime(&tmg); */
10210: /* printf("Time(after) =%s",strstart); */
10211: /* (void) time (&time_value);
10212: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10213: * tm = *localtime(&time_value);
10214: * strstart=asctime(&tm);
10215: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10216: */
10217:
10218: nberr=0; /* Number of errors and warnings */
10219: nbwarn=0;
1.184 brouard 10220: #ifdef WIN32
10221: _getcwd(pathcd, size);
10222: #else
1.126 brouard 10223: getcwd(pathcd, size);
1.184 brouard 10224: #endif
1.191 brouard 10225: syscompilerinfo(0);
1.196 brouard 10226: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10227: if(argc <=1){
10228: printf("\nEnter the parameter file name: ");
1.205 brouard 10229: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10230: printf("ERROR Empty parameter file name\n");
10231: goto end;
10232: }
1.126 brouard 10233: i=strlen(pathr);
10234: if(pathr[i-1]=='\n')
10235: pathr[i-1]='\0';
1.156 brouard 10236: i=strlen(pathr);
1.205 brouard 10237: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10238: pathr[i-1]='\0';
1.205 brouard 10239: }
10240: i=strlen(pathr);
10241: if( i==0 ){
10242: printf("ERROR Empty parameter file name\n");
10243: goto end;
10244: }
10245: for (tok = pathr; tok != NULL; ){
1.126 brouard 10246: printf("Pathr |%s|\n",pathr);
10247: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10248: printf("val= |%s| pathr=%s\n",val,pathr);
10249: strcpy (pathtot, val);
10250: if(pathr[0] == '\0') break; /* Dirty */
10251: }
10252: }
10253: else{
10254: strcpy(pathtot,argv[1]);
10255: }
10256: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10257: /*cygwin_split_path(pathtot,path,optionfile);
10258: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10259: /* cutv(path,optionfile,pathtot,'\\');*/
10260:
10261: /* Split argv[0], imach program to get pathimach */
10262: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10263: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10264: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10265: /* strcpy(pathimach,argv[0]); */
10266: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10267: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10268: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10269: #ifdef WIN32
10270: _chdir(path); /* Can be a relative path */
10271: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10272: #else
1.126 brouard 10273: chdir(path); /* Can be a relative path */
1.184 brouard 10274: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10275: #endif
10276: printf("Current directory %s!\n",pathcd);
1.126 brouard 10277: strcpy(command,"mkdir ");
10278: strcat(command,optionfilefiname);
10279: if((outcmd=system(command)) != 0){
1.169 brouard 10280: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10281: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10282: /* fclose(ficlog); */
10283: /* exit(1); */
10284: }
10285: /* if((imk=mkdir(optionfilefiname))<0){ */
10286: /* perror("mkdir"); */
10287: /* } */
10288:
10289: /*-------- arguments in the command line --------*/
10290:
1.186 brouard 10291: /* Main Log file */
1.126 brouard 10292: strcat(filelog, optionfilefiname);
10293: strcat(filelog,".log"); /* */
10294: if((ficlog=fopen(filelog,"w"))==NULL) {
10295: printf("Problem with logfile %s\n",filelog);
10296: goto end;
10297: }
10298: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10299: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10300: fprintf(ficlog,"\nEnter the parameter file name: \n");
10301: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10302: path=%s \n\
10303: optionfile=%s\n\
10304: optionfilext=%s\n\
1.156 brouard 10305: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10306:
1.197 brouard 10307: syscompilerinfo(1);
1.167 brouard 10308:
1.126 brouard 10309: printf("Local time (at start):%s",strstart);
10310: fprintf(ficlog,"Local time (at start): %s",strstart);
10311: fflush(ficlog);
10312: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10313: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10314:
10315: /* */
10316: strcpy(fileres,"r");
10317: strcat(fileres, optionfilefiname);
1.201 brouard 10318: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10319: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10320: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10321:
1.186 brouard 10322: /* Main ---------arguments file --------*/
1.126 brouard 10323:
10324: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10325: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10326: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10327: fflush(ficlog);
1.149 brouard 10328: /* goto end; */
10329: exit(70);
1.126 brouard 10330: }
10331:
10332:
10333:
10334: strcpy(filereso,"o");
1.201 brouard 10335: strcat(filereso,fileresu);
1.126 brouard 10336: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10337: printf("Problem with Output resultfile: %s\n", filereso);
10338: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10339: fflush(ficlog);
10340: goto end;
10341: }
10342:
10343: /* Reads comments: lines beginning with '#' */
10344: numlinepar=0;
1.197 brouard 10345:
10346: /* First parameter line */
10347: while(fgets(line, MAXLINE, ficpar)) {
10348: /* If line starts with a # it is a comment */
10349: if (line[0] == '#') {
10350: numlinepar++;
10351: fputs(line,stdout);
10352: fputs(line,ficparo);
10353: fputs(line,ficlog);
10354: continue;
10355: }else
10356: break;
10357: }
10358: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10359: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10360: if (num_filled != 5) {
10361: printf("Should be 5 parameters\n");
10362: }
1.126 brouard 10363: numlinepar++;
1.197 brouard 10364: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10365: }
10366: /* Second parameter line */
10367: while(fgets(line, MAXLINE, ficpar)) {
10368: /* If line starts with a # it is a comment */
10369: if (line[0] == '#') {
10370: numlinepar++;
10371: fputs(line,stdout);
10372: fputs(line,ficparo);
10373: fputs(line,ficlog);
10374: continue;
10375: }else
10376: break;
10377: }
1.223 brouard 10378: 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", \
10379: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10380: if (num_filled != 11) {
10381: 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 10382: printf("but line=%s\n",line);
1.197 brouard 10383: }
1.223 brouard 10384: 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 10385: }
1.203 brouard 10386: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10387: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10388: /* Third parameter line */
10389: while(fgets(line, MAXLINE, ficpar)) {
10390: /* If line starts with a # it is a comment */
10391: if (line[0] == '#') {
10392: numlinepar++;
10393: fputs(line,stdout);
10394: fputs(line,ficparo);
10395: fputs(line,ficlog);
10396: continue;
10397: }else
10398: break;
10399: }
1.201 brouard 10400: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 brouard 10401: if (num_filled == 0){
10402: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10403: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10404: model[0]='\0';
10405: goto end;
10406: } else if (num_filled != 1){
1.197 brouard 10407: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10408: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10409: model[0]='\0';
10410: goto end;
10411: }
10412: else{
10413: if (model[0]=='+'){
10414: for(i=1; i<=strlen(model);i++)
10415: modeltemp[i-1]=model[i];
1.201 brouard 10416: strcpy(model,modeltemp);
1.197 brouard 10417: }
10418: }
1.199 brouard 10419: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10420: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10421: }
10422: /* 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); */
10423: /* numlinepar=numlinepar+3; /\* In general *\/ */
10424: /* 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 10425: 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);
10426: 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 10427: fflush(ficlog);
1.190 brouard 10428: /* if(model[0]=='#'|| model[0]== '\0'){ */
10429: if(model[0]=='#'){
1.187 brouard 10430: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10431: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10432: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10433: if(mle != -1){
10434: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10435: exit(1);
10436: }
10437: }
1.126 brouard 10438: while((c=getc(ficpar))=='#' && c!= EOF){
10439: ungetc(c,ficpar);
10440: fgets(line, MAXLINE, ficpar);
10441: numlinepar++;
1.195 brouard 10442: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10443: z[0]=line[1];
10444: }
10445: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10446: fputs(line, stdout);
10447: //puts(line);
1.126 brouard 10448: fputs(line,ficparo);
10449: fputs(line,ficlog);
10450: }
10451: ungetc(c,ficpar);
10452:
10453:
1.145 brouard 10454: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10455: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10456: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10457: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10458: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10459: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10460: v1+v2*age+v2*v3 makes cptcovn = 3
10461: */
10462: if (strlen(model)>1)
1.187 brouard 10463: 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 10464: else
1.187 brouard 10465: ncovmodel=2; /* Constant and age */
1.133 brouard 10466: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10467: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10468: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10469: 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);
10470: 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);
10471: fflush(stdout);
10472: fclose (ficlog);
10473: goto end;
10474: }
1.126 brouard 10475: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10476: delti=delti3[1][1];
10477: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10478: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10479: /* We could also provide initial parameters values giving by simple logistic regression
10480: * only one way, that is without matrix product. We will have nlstate maximizations */
10481: /* for(i=1;i<nlstate;i++){ */
10482: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10483: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10484: /* } */
1.126 brouard 10485: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10486: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10487: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10488: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10489: fclose (ficparo);
10490: fclose (ficlog);
10491: goto end;
10492: exit(0);
1.220 brouard 10493: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10494: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10495: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10496: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10497: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10498: matcov=matrix(1,npar,1,npar);
1.203 brouard 10499: hess=matrix(1,npar,1,npar);
1.220 brouard 10500: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10501: /* Read guessed parameters */
1.126 brouard 10502: /* Reads comments: lines beginning with '#' */
10503: while((c=getc(ficpar))=='#' && c!= EOF){
10504: ungetc(c,ficpar);
10505: fgets(line, MAXLINE, ficpar);
10506: numlinepar++;
1.141 brouard 10507: fputs(line,stdout);
1.126 brouard 10508: fputs(line,ficparo);
10509: fputs(line,ficlog);
10510: }
10511: ungetc(c,ficpar);
10512:
10513: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10514: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10515: for(i=1; i <=nlstate; i++){
1.234 brouard 10516: j=0;
1.126 brouard 10517: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10518: if(jj==i) continue;
10519: j++;
10520: fscanf(ficpar,"%1d%1d",&i1,&j1);
10521: if ((i1 != i) || (j1 != jj)){
10522: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10523: It might be a problem of design; if ncovcol and the model are correct\n \
10524: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10525: exit(1);
10526: }
10527: fprintf(ficparo,"%1d%1d",i1,j1);
10528: if(mle==1)
10529: printf("%1d%1d",i,jj);
10530: fprintf(ficlog,"%1d%1d",i,jj);
10531: for(k=1; k<=ncovmodel;k++){
10532: fscanf(ficpar," %lf",¶m[i][j][k]);
10533: if(mle==1){
10534: printf(" %lf",param[i][j][k]);
10535: fprintf(ficlog," %lf",param[i][j][k]);
10536: }
10537: else
10538: fprintf(ficlog," %lf",param[i][j][k]);
10539: fprintf(ficparo," %lf",param[i][j][k]);
10540: }
10541: fscanf(ficpar,"\n");
10542: numlinepar++;
10543: if(mle==1)
10544: printf("\n");
10545: fprintf(ficlog,"\n");
10546: fprintf(ficparo,"\n");
1.126 brouard 10547: }
10548: }
10549: fflush(ficlog);
1.234 brouard 10550:
1.251 brouard 10551: /* Reads parameters values */
1.126 brouard 10552: p=param[1][1];
1.251 brouard 10553: pstart=paramstart[1][1];
1.126 brouard 10554:
10555: /* Reads comments: lines beginning with '#' */
10556: while((c=getc(ficpar))=='#' && c!= EOF){
10557: ungetc(c,ficpar);
10558: fgets(line, MAXLINE, ficpar);
10559: numlinepar++;
1.141 brouard 10560: fputs(line,stdout);
1.126 brouard 10561: fputs(line,ficparo);
10562: fputs(line,ficlog);
10563: }
10564: ungetc(c,ficpar);
10565:
10566: for(i=1; i <=nlstate; i++){
10567: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10568: fscanf(ficpar,"%1d%1d",&i1,&j1);
10569: if ( (i1-i) * (j1-j) != 0){
10570: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10571: exit(1);
10572: }
10573: printf("%1d%1d",i,j);
10574: fprintf(ficparo,"%1d%1d",i1,j1);
10575: fprintf(ficlog,"%1d%1d",i1,j1);
10576: for(k=1; k<=ncovmodel;k++){
10577: fscanf(ficpar,"%le",&delti3[i][j][k]);
10578: printf(" %le",delti3[i][j][k]);
10579: fprintf(ficparo," %le",delti3[i][j][k]);
10580: fprintf(ficlog," %le",delti3[i][j][k]);
10581: }
10582: fscanf(ficpar,"\n");
10583: numlinepar++;
10584: printf("\n");
10585: fprintf(ficparo,"\n");
10586: fprintf(ficlog,"\n");
1.126 brouard 10587: }
10588: }
10589: fflush(ficlog);
1.234 brouard 10590:
1.145 brouard 10591: /* Reads covariance matrix */
1.126 brouard 10592: delti=delti3[1][1];
1.220 brouard 10593:
10594:
1.126 brouard 10595: /* 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 10596:
1.126 brouard 10597: /* Reads comments: lines beginning with '#' */
10598: while((c=getc(ficpar))=='#' && c!= EOF){
10599: ungetc(c,ficpar);
10600: fgets(line, MAXLINE, ficpar);
10601: numlinepar++;
1.141 brouard 10602: fputs(line,stdout);
1.126 brouard 10603: fputs(line,ficparo);
10604: fputs(line,ficlog);
10605: }
10606: ungetc(c,ficpar);
1.220 brouard 10607:
1.126 brouard 10608: matcov=matrix(1,npar,1,npar);
1.203 brouard 10609: hess=matrix(1,npar,1,npar);
1.131 brouard 10610: for(i=1; i <=npar; i++)
10611: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10612:
1.194 brouard 10613: /* Scans npar lines */
1.126 brouard 10614: for(i=1; i <=npar; i++){
1.226 brouard 10615: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10616: if(count != 3){
1.226 brouard 10617: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10618: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10619: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10620: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10621: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10622: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10623: exit(1);
1.220 brouard 10624: }else{
1.226 brouard 10625: if(mle==1)
10626: printf("%1d%1d%d",i1,j1,jk);
10627: }
10628: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10629: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10630: for(j=1; j <=i; j++){
1.226 brouard 10631: fscanf(ficpar," %le",&matcov[i][j]);
10632: if(mle==1){
10633: printf(" %.5le",matcov[i][j]);
10634: }
10635: fprintf(ficlog," %.5le",matcov[i][j]);
10636: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10637: }
10638: fscanf(ficpar,"\n");
10639: numlinepar++;
10640: if(mle==1)
1.220 brouard 10641: printf("\n");
1.126 brouard 10642: fprintf(ficlog,"\n");
10643: fprintf(ficparo,"\n");
10644: }
1.194 brouard 10645: /* End of read covariance matrix npar lines */
1.126 brouard 10646: for(i=1; i <=npar; i++)
10647: for(j=i+1;j<=npar;j++)
1.226 brouard 10648: matcov[i][j]=matcov[j][i];
1.126 brouard 10649:
10650: if(mle==1)
10651: printf("\n");
10652: fprintf(ficlog,"\n");
10653:
10654: fflush(ficlog);
10655:
10656: /*-------- Rewriting parameter file ----------*/
10657: strcpy(rfileres,"r"); /* "Rparameterfile */
10658: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10659: strcat(rfileres,"."); /* */
10660: strcat(rfileres,optionfilext); /* Other files have txt extension */
10661: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10662: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10663: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10664: }
10665: fprintf(ficres,"#%s\n",version);
10666: } /* End of mle != -3 */
1.218 brouard 10667:
1.186 brouard 10668: /* Main data
10669: */
1.126 brouard 10670: n= lastobs;
10671: num=lvector(1,n);
10672: moisnais=vector(1,n);
10673: annais=vector(1,n);
10674: moisdc=vector(1,n);
10675: andc=vector(1,n);
1.220 brouard 10676: weight=vector(1,n);
1.126 brouard 10677: agedc=vector(1,n);
10678: cod=ivector(1,n);
1.220 brouard 10679: for(i=1;i<=n;i++){
1.234 brouard 10680: num[i]=0;
10681: moisnais[i]=0;
10682: annais[i]=0;
10683: moisdc[i]=0;
10684: andc[i]=0;
10685: agedc[i]=0;
10686: cod[i]=0;
10687: weight[i]=1.0; /* Equal weights, 1 by default */
10688: }
1.126 brouard 10689: mint=matrix(1,maxwav,1,n);
10690: anint=matrix(1,maxwav,1,n);
1.131 brouard 10691: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10692: tab=ivector(1,NCOVMAX);
1.144 brouard 10693: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10694: 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 10695:
1.136 brouard 10696: /* Reads data from file datafile */
10697: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10698: goto end;
10699:
10700: /* Calculation of the number of parameters from char model */
1.234 brouard 10701: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10702: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10703: k=3 V4 Tvar[k=3]= 4 (from V4)
10704: k=2 V1 Tvar[k=2]= 1 (from V1)
10705: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10706: */
10707:
10708: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10709: TvarsDind=ivector(1,NCOVMAX); /* */
10710: TvarsD=ivector(1,NCOVMAX); /* */
10711: TvarsQind=ivector(1,NCOVMAX); /* */
10712: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10713: TvarF=ivector(1,NCOVMAX); /* */
10714: TvarFind=ivector(1,NCOVMAX); /* */
10715: TvarV=ivector(1,NCOVMAX); /* */
10716: TvarVind=ivector(1,NCOVMAX); /* */
10717: TvarA=ivector(1,NCOVMAX); /* */
10718: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10719: TvarFD=ivector(1,NCOVMAX); /* */
10720: TvarFDind=ivector(1,NCOVMAX); /* */
10721: TvarFQ=ivector(1,NCOVMAX); /* */
10722: TvarFQind=ivector(1,NCOVMAX); /* */
10723: TvarVD=ivector(1,NCOVMAX); /* */
10724: TvarVDind=ivector(1,NCOVMAX); /* */
10725: TvarVQ=ivector(1,NCOVMAX); /* */
10726: TvarVQind=ivector(1,NCOVMAX); /* */
10727:
1.230 brouard 10728: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10729: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10730: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10731: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10732: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10733: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10734: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10735: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10736: */
10737: /* For model-covariate k tells which data-covariate to use but
10738: because this model-covariate is a construction we invent a new column
10739: ncovcol + k1
10740: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10741: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10742: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10743: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10744: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10745: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10746: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10747: */
1.145 brouard 10748: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10749: 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 10750: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10751: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10752: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10753: 4 covariates (3 plus signs)
10754: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10755: */
1.230 brouard 10756: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10757: * individual dummy, fixed or varying:
10758: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10759: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10760: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10761: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10762: * Tmodelind[1]@9={9,0,3,2,}*/
10763: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10764: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10765: * individual quantitative, fixed or varying:
10766: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10767: * 3, 1, 0, 0, 0, 0, 0, 0},
10768: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10769: /* Main decodemodel */
10770:
1.187 brouard 10771:
1.223 brouard 10772: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10773: goto end;
10774:
1.137 brouard 10775: if((double)(lastobs-imx)/(double)imx > 1.10){
10776: nbwarn++;
10777: 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);
10778: 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);
10779: }
1.136 brouard 10780: /* if(mle==1){*/
1.137 brouard 10781: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10782: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10783: }
10784:
10785: /*-calculation of age at interview from date of interview and age at death -*/
10786: agev=matrix(1,maxwav,1,imx);
10787:
10788: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10789: goto end;
10790:
1.126 brouard 10791:
1.136 brouard 10792: agegomp=(int)agemin;
10793: free_vector(moisnais,1,n);
10794: free_vector(annais,1,n);
1.126 brouard 10795: /* free_matrix(mint,1,maxwav,1,n);
10796: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10797: /* free_vector(moisdc,1,n); */
10798: /* free_vector(andc,1,n); */
1.145 brouard 10799: /* */
10800:
1.126 brouard 10801: wav=ivector(1,imx);
1.214 brouard 10802: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10803: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10804: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10805: 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.*/
10806: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10807: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10808:
10809: /* Concatenates waves */
1.214 brouard 10810: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10811: Death is a valid wave (if date is known).
10812: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10813: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10814: and mw[mi+1][i]. dh depends on stepm.
10815: */
10816:
1.126 brouard 10817: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10818: /* Concatenates waves */
1.145 brouard 10819:
1.215 brouard 10820: free_vector(moisdc,1,n);
10821: free_vector(andc,1,n);
10822:
1.126 brouard 10823: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10824: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10825: ncodemax[1]=1;
1.145 brouard 10826: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10827: cptcoveff=0;
1.220 brouard 10828: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10829: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10830: }
10831:
10832: ncovcombmax=pow(2,cptcoveff);
10833: invalidvarcomb=ivector(1, ncovcombmax);
10834: for(i=1;i<ncovcombmax;i++)
10835: invalidvarcomb[i]=0;
10836:
1.211 brouard 10837: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10838: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10839: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10840:
1.200 brouard 10841: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10842: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10843: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10844: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10845: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10846: * (currently 0 or 1) in the data.
10847: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10848: * corresponding modality (h,j).
10849: */
10850:
1.145 brouard 10851: h=0;
10852: /*if (cptcovn > 0) */
1.126 brouard 10853: m=pow(2,cptcoveff);
10854:
1.144 brouard 10855: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10856: * For k=4 covariates, h goes from 1 to m=2**k
10857: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10858: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10859: * h\k 1 2 3 4
1.143 brouard 10860: *______________________________
10861: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10862: * 2 2 1 1 1
10863: * 3 i=2 1 2 1 1
10864: * 4 2 2 1 1
10865: * 5 i=3 1 i=2 1 2 1
10866: * 6 2 1 2 1
10867: * 7 i=4 1 2 2 1
10868: * 8 2 2 2 1
1.197 brouard 10869: * 9 i=5 1 i=3 1 i=2 1 2
10870: * 10 2 1 1 2
10871: * 11 i=6 1 2 1 2
10872: * 12 2 2 1 2
10873: * 13 i=7 1 i=4 1 2 2
10874: * 14 2 1 2 2
10875: * 15 i=8 1 2 2 2
10876: * 16 2 2 2 2
1.143 brouard 10877: */
1.212 brouard 10878: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10879: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10880: * and the value of each covariate?
10881: * V1=1, V2=1, V3=2, V4=1 ?
10882: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10883: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10884: * In order to get the real value in the data, we use nbcode
10885: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10886: * We are keeping this crazy system in order to be able (in the future?)
10887: * to have more than 2 values (0 or 1) for a covariate.
10888: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10889: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10890: * bbbbbbbb
10891: * 76543210
10892: * h-1 00000101 (6-1=5)
1.219 brouard 10893: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10894: * &
10895: * 1 00000001 (1)
1.219 brouard 10896: * 00000000 = 1 & ((h-1) >> (k-1))
10897: * +1= 00000001 =1
1.211 brouard 10898: *
10899: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10900: * h' 1101 =2^3+2^2+0x2^1+2^0
10901: * >>k' 11
10902: * & 00000001
10903: * = 00000001
10904: * +1 = 00000010=2 = codtabm(14,3)
10905: * Reverse h=6 and m=16?
10906: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10907: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10908: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10909: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10910: * V3=decodtabm(14,3,2**4)=2
10911: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10912: *(h-1) >> (j-1) 0011 =13 >> 2
10913: * &1 000000001
10914: * = 000000001
10915: * +1= 000000010 =2
10916: * 2211
10917: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10918: * V3=2
1.220 brouard 10919: * codtabm and decodtabm are identical
1.211 brouard 10920: */
10921:
1.145 brouard 10922:
10923: free_ivector(Ndum,-1,NCOVMAX);
10924:
10925:
1.126 brouard 10926:
1.186 brouard 10927: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10928: strcpy(optionfilegnuplot,optionfilefiname);
10929: if(mle==-3)
1.201 brouard 10930: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10931: strcat(optionfilegnuplot,".gp");
10932:
10933: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10934: printf("Problem with file %s",optionfilegnuplot);
10935: }
10936: else{
1.204 brouard 10937: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10938: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10939: //fprintf(ficgp,"set missing 'NaNq'\n");
10940: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10941: }
10942: /* fclose(ficgp);*/
1.186 brouard 10943:
10944:
10945: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10946:
10947: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10948: if(mle==-3)
1.201 brouard 10949: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10950: strcat(optionfilehtm,".htm");
10951: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10952: printf("Problem with %s \n",optionfilehtm);
10953: exit(0);
1.126 brouard 10954: }
10955:
10956: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10957: strcat(optionfilehtmcov,"-cov.htm");
10958: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10959: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10960: }
10961: else{
10962: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10963: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10964: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10965: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10966: }
10967:
1.213 brouard 10968: 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 10969: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10970: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10971: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10972: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10973: \n\
10974: <hr size=\"2\" color=\"#EC5E5E\">\
10975: <ul><li><h4>Parameter files</h4>\n\
10976: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10977: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10978: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10979: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10980: - Date and time at start: %s</ul>\n",\
10981: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10982: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10983: fileres,fileres,\
10984: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10985: fflush(fichtm);
10986:
10987: strcpy(pathr,path);
10988: strcat(pathr,optionfilefiname);
1.184 brouard 10989: #ifdef WIN32
10990: _chdir(optionfilefiname); /* Move to directory named optionfile */
10991: #else
1.126 brouard 10992: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10993: #endif
10994:
1.126 brouard 10995:
1.220 brouard 10996: /* Calculates basic frequencies. Computes observed prevalence at single age
10997: and for any valid combination of covariates
1.126 brouard 10998: and prints on file fileres'p'. */
1.251 brouard 10999: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11000: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11001:
11002: fprintf(fichtm,"\n");
11003: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
11004: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11005: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
11006: imx,agemin,agemax,jmin,jmax,jmean);
11007: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 11008: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11009: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11010: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11011: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11012:
1.126 brouard 11013: /* For Powell, parameters are in a vector p[] starting at p[1]
11014: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11015: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11016:
11017: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11018: /* For mortality only */
1.126 brouard 11019: if (mle==-3){
1.136 brouard 11020: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11021: for(i=1;i<=NDIM;i++)
11022: for(j=1;j<=NDIM;j++)
11023: ximort[i][j]=0.;
1.186 brouard 11024: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11025: cens=ivector(1,n);
11026: ageexmed=vector(1,n);
11027: agecens=vector(1,n);
11028: dcwave=ivector(1,n);
1.223 brouard 11029:
1.126 brouard 11030: for (i=1; i<=imx; i++){
11031: dcwave[i]=-1;
11032: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11033: if (s[m][i]>nlstate) {
11034: dcwave[i]=m;
11035: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11036: break;
11037: }
1.126 brouard 11038: }
1.226 brouard 11039:
1.126 brouard 11040: for (i=1; i<=imx; i++) {
11041: if (wav[i]>0){
1.226 brouard 11042: ageexmed[i]=agev[mw[1][i]][i];
11043: j=wav[i];
11044: agecens[i]=1.;
11045:
11046: if (ageexmed[i]> 1 && wav[i] > 0){
11047: agecens[i]=agev[mw[j][i]][i];
11048: cens[i]= 1;
11049: }else if (ageexmed[i]< 1)
11050: cens[i]= -1;
11051: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11052: cens[i]=0 ;
1.126 brouard 11053: }
11054: else cens[i]=-1;
11055: }
11056:
11057: for (i=1;i<=NDIM;i++) {
11058: for (j=1;j<=NDIM;j++)
1.226 brouard 11059: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11060: }
11061:
1.145 brouard 11062: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11063: /*printf("%lf %lf", p[1], p[2]);*/
11064:
11065:
1.136 brouard 11066: #ifdef GSL
11067: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11068: #else
1.126 brouard 11069: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11070: #endif
1.201 brouard 11071: strcpy(filerespow,"POW-MORT_");
11072: strcat(filerespow,fileresu);
1.126 brouard 11073: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11074: printf("Problem with resultfile: %s\n", filerespow);
11075: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11076: }
1.136 brouard 11077: #ifdef GSL
11078: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11079: #else
1.126 brouard 11080: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11081: #endif
1.126 brouard 11082: /* for (i=1;i<=nlstate;i++)
11083: for(j=1;j<=nlstate+ndeath;j++)
11084: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11085: */
11086: fprintf(ficrespow,"\n");
1.136 brouard 11087: #ifdef GSL
11088: /* gsl starts here */
11089: T = gsl_multimin_fminimizer_nmsimplex;
11090: gsl_multimin_fminimizer *sfm = NULL;
11091: gsl_vector *ss, *x;
11092: gsl_multimin_function minex_func;
11093:
11094: /* Initial vertex size vector */
11095: ss = gsl_vector_alloc (NDIM);
11096:
11097: if (ss == NULL){
11098: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11099: }
11100: /* Set all step sizes to 1 */
11101: gsl_vector_set_all (ss, 0.001);
11102:
11103: /* Starting point */
1.126 brouard 11104:
1.136 brouard 11105: x = gsl_vector_alloc (NDIM);
11106:
11107: if (x == NULL){
11108: gsl_vector_free(ss);
11109: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11110: }
11111:
11112: /* Initialize method and iterate */
11113: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11114: /* gsl_vector_set(x, 0, 0.0268); */
11115: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11116: gsl_vector_set(x, 0, p[1]);
11117: gsl_vector_set(x, 1, p[2]);
11118:
11119: minex_func.f = &gompertz_f;
11120: minex_func.n = NDIM;
11121: minex_func.params = (void *)&p; /* ??? */
11122:
11123: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11124: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11125:
11126: printf("Iterations beginning .....\n\n");
11127: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11128:
11129: iteri=0;
11130: while (rval == GSL_CONTINUE){
11131: iteri++;
11132: status = gsl_multimin_fminimizer_iterate(sfm);
11133:
11134: if (status) printf("error: %s\n", gsl_strerror (status));
11135: fflush(0);
11136:
11137: if (status)
11138: break;
11139:
11140: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11141: ssval = gsl_multimin_fminimizer_size (sfm);
11142:
11143: if (rval == GSL_SUCCESS)
11144: printf ("converged to a local maximum at\n");
11145:
11146: printf("%5d ", iteri);
11147: for (it = 0; it < NDIM; it++){
11148: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11149: }
11150: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11151: }
11152:
11153: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11154:
11155: gsl_vector_free(x); /* initial values */
11156: gsl_vector_free(ss); /* inital step size */
11157: for (it=0; it<NDIM; it++){
11158: p[it+1]=gsl_vector_get(sfm->x,it);
11159: fprintf(ficrespow," %.12lf", p[it]);
11160: }
11161: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11162: #endif
11163: #ifdef POWELL
11164: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11165: #endif
1.126 brouard 11166: fclose(ficrespow);
11167:
1.203 brouard 11168: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11169:
11170: for(i=1; i <=NDIM; i++)
11171: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11172: matcov[i][j]=matcov[j][i];
1.126 brouard 11173:
11174: printf("\nCovariance matrix\n ");
1.203 brouard 11175: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11176: for(i=1; i <=NDIM; i++) {
11177: for(j=1;j<=NDIM;j++){
1.220 brouard 11178: printf("%f ",matcov[i][j]);
11179: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11180: }
1.203 brouard 11181: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11182: }
11183:
11184: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11185: for (i=1;i<=NDIM;i++) {
1.126 brouard 11186: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11187: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11188: }
1.126 brouard 11189: lsurv=vector(1,AGESUP);
11190: lpop=vector(1,AGESUP);
11191: tpop=vector(1,AGESUP);
11192: lsurv[agegomp]=100000;
11193:
11194: for (k=agegomp;k<=AGESUP;k++) {
11195: agemortsup=k;
11196: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11197: }
11198:
11199: for (k=agegomp;k<agemortsup;k++)
11200: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11201:
11202: for (k=agegomp;k<agemortsup;k++){
11203: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11204: sumlpop=sumlpop+lpop[k];
11205: }
11206:
11207: tpop[agegomp]=sumlpop;
11208: for (k=agegomp;k<(agemortsup-3);k++){
11209: /* tpop[k+1]=2;*/
11210: tpop[k+1]=tpop[k]-lpop[k];
11211: }
11212:
11213:
11214: printf("\nAge lx qx dx Lx Tx e(x)\n");
11215: for (k=agegomp;k<(agemortsup-2);k++)
11216: 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]);
11217:
11218:
11219: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11220: ageminpar=50;
11221: agemaxpar=100;
1.194 brouard 11222: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11223: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11224: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11225: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11226: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11227: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11228: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11229: }else{
11230: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11231: 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 11232: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11233: }
1.201 brouard 11234: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11235: stepm, weightopt,\
11236: model,imx,p,matcov,agemortsup);
11237:
11238: free_vector(lsurv,1,AGESUP);
11239: free_vector(lpop,1,AGESUP);
11240: free_vector(tpop,1,AGESUP);
1.220 brouard 11241: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11242: free_ivector(cens,1,n);
11243: free_vector(agecens,1,n);
11244: free_ivector(dcwave,1,n);
1.220 brouard 11245: #ifdef GSL
1.136 brouard 11246: #endif
1.186 brouard 11247: } /* Endof if mle==-3 mortality only */
1.205 brouard 11248: /* Standard */
11249: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11250: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11251: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11252: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11253: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11254: for (k=1; k<=npar;k++)
11255: printf(" %d %8.5f",k,p[k]);
11256: printf("\n");
1.205 brouard 11257: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11258: /* mlikeli uses func not funcone */
1.247 brouard 11259: /* for(i=1;i<nlstate;i++){ */
11260: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11261: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11262: /* } */
1.205 brouard 11263: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11264: }
11265: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11266: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11267: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11268: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11269: }
11270: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11271: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11272: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11273: for (k=1; k<=npar;k++)
11274: printf(" %d %8.5f",k,p[k]);
11275: printf("\n");
11276:
11277: /*--------- results files --------------*/
1.224 brouard 11278: 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 11279:
11280:
11281: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11282: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11283: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11284: for(i=1,jk=1; i <=nlstate; i++){
11285: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11286: if (k != i) {
11287: printf("%d%d ",i,k);
11288: fprintf(ficlog,"%d%d ",i,k);
11289: fprintf(ficres,"%1d%1d ",i,k);
11290: for(j=1; j <=ncovmodel; j++){
11291: printf("%12.7f ",p[jk]);
11292: fprintf(ficlog,"%12.7f ",p[jk]);
11293: fprintf(ficres,"%12.7f ",p[jk]);
11294: jk++;
11295: }
11296: printf("\n");
11297: fprintf(ficlog,"\n");
11298: fprintf(ficres,"\n");
11299: }
1.126 brouard 11300: }
11301: }
1.203 brouard 11302: if(mle != 0){
11303: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11304: ftolhess=ftol; /* Usually correct */
1.203 brouard 11305: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11306: 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");
11307: 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");
11308: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11309: for(k=1; k <=(nlstate+ndeath); k++){
11310: if (k != i) {
11311: printf("%d%d ",i,k);
11312: fprintf(ficlog,"%d%d ",i,k);
11313: for(j=1; j <=ncovmodel; j++){
11314: 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]));
11315: 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]));
11316: jk++;
11317: }
11318: printf("\n");
11319: fprintf(ficlog,"\n");
11320: }
11321: }
1.193 brouard 11322: }
1.203 brouard 11323: } /* end of hesscov and Wald tests */
1.225 brouard 11324:
1.203 brouard 11325: /* */
1.126 brouard 11326: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11327: printf("# Scales (for hessian or gradient estimation)\n");
11328: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11329: for(i=1,jk=1; i <=nlstate; i++){
11330: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11331: if (j!=i) {
11332: fprintf(ficres,"%1d%1d",i,j);
11333: printf("%1d%1d",i,j);
11334: fprintf(ficlog,"%1d%1d",i,j);
11335: for(k=1; k<=ncovmodel;k++){
11336: printf(" %.5e",delti[jk]);
11337: fprintf(ficlog," %.5e",delti[jk]);
11338: fprintf(ficres," %.5e",delti[jk]);
11339: jk++;
11340: }
11341: printf("\n");
11342: fprintf(ficlog,"\n");
11343: fprintf(ficres,"\n");
11344: }
1.126 brouard 11345: }
11346: }
11347:
11348: 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 11349: if(mle >= 1) /* To big for the screen */
1.126 brouard 11350: 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");
11351: 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");
11352: /* # 121 Var(a12)\n\ */
11353: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11354: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11355: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11356: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11357: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11358: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11359: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11360:
11361:
11362: /* Just to have a covariance matrix which will be more understandable
11363: even is we still don't want to manage dictionary of variables
11364: */
11365: for(itimes=1;itimes<=2;itimes++){
11366: jj=0;
11367: for(i=1; i <=nlstate; i++){
1.225 brouard 11368: for(j=1; j <=nlstate+ndeath; j++){
11369: if(j==i) continue;
11370: for(k=1; k<=ncovmodel;k++){
11371: jj++;
11372: ca[0]= k+'a'-1;ca[1]='\0';
11373: if(itimes==1){
11374: if(mle>=1)
11375: printf("#%1d%1d%d",i,j,k);
11376: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11377: fprintf(ficres,"#%1d%1d%d",i,j,k);
11378: }else{
11379: if(mle>=1)
11380: printf("%1d%1d%d",i,j,k);
11381: fprintf(ficlog,"%1d%1d%d",i,j,k);
11382: fprintf(ficres,"%1d%1d%d",i,j,k);
11383: }
11384: ll=0;
11385: for(li=1;li <=nlstate; li++){
11386: for(lj=1;lj <=nlstate+ndeath; lj++){
11387: if(lj==li) continue;
11388: for(lk=1;lk<=ncovmodel;lk++){
11389: ll++;
11390: if(ll<=jj){
11391: cb[0]= lk +'a'-1;cb[1]='\0';
11392: if(ll<jj){
11393: if(itimes==1){
11394: if(mle>=1)
11395: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11396: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11397: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11398: }else{
11399: if(mle>=1)
11400: printf(" %.5e",matcov[jj][ll]);
11401: fprintf(ficlog," %.5e",matcov[jj][ll]);
11402: fprintf(ficres," %.5e",matcov[jj][ll]);
11403: }
11404: }else{
11405: if(itimes==1){
11406: if(mle>=1)
11407: printf(" Var(%s%1d%1d)",ca,i,j);
11408: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11409: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11410: }else{
11411: if(mle>=1)
11412: printf(" %.7e",matcov[jj][ll]);
11413: fprintf(ficlog," %.7e",matcov[jj][ll]);
11414: fprintf(ficres," %.7e",matcov[jj][ll]);
11415: }
11416: }
11417: }
11418: } /* end lk */
11419: } /* end lj */
11420: } /* end li */
11421: if(mle>=1)
11422: printf("\n");
11423: fprintf(ficlog,"\n");
11424: fprintf(ficres,"\n");
11425: numlinepar++;
11426: } /* end k*/
11427: } /*end j */
1.126 brouard 11428: } /* end i */
11429: } /* end itimes */
11430:
11431: fflush(ficlog);
11432: fflush(ficres);
1.225 brouard 11433: while(fgets(line, MAXLINE, ficpar)) {
11434: /* If line starts with a # it is a comment */
11435: if (line[0] == '#') {
11436: numlinepar++;
11437: fputs(line,stdout);
11438: fputs(line,ficparo);
11439: fputs(line,ficlog);
11440: continue;
11441: }else
11442: break;
11443: }
11444:
1.209 brouard 11445: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11446: /* ungetc(c,ficpar); */
11447: /* fgets(line, MAXLINE, ficpar); */
11448: /* fputs(line,stdout); */
11449: /* fputs(line,ficparo); */
11450: /* } */
11451: /* ungetc(c,ficpar); */
1.126 brouard 11452:
11453: estepm=0;
1.209 brouard 11454: 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 11455:
11456: if (num_filled != 6) {
11457: 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);
11458: 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);
11459: goto end;
11460: }
11461: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11462: }
11463: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11464: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11465:
1.209 brouard 11466: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11467: if (estepm==0 || estepm < stepm) estepm=stepm;
11468: if (fage <= 2) {
11469: bage = ageminpar;
11470: fage = agemaxpar;
11471: }
11472:
11473: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11474: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11475: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11476:
1.186 brouard 11477: /* Other stuffs, more or less useful */
1.254 brouard 11478: while(fgets(line, MAXLINE, ficpar)) {
11479: /* If line starts with a # it is a comment */
11480: if (line[0] == '#') {
11481: numlinepar++;
11482: fputs(line,stdout);
11483: fputs(line,ficparo);
11484: fputs(line,ficlog);
11485: continue;
11486: }else
11487: break;
11488: }
11489:
11490: 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){
11491:
11492: if (num_filled != 7) {
11493: 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);
11494: 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);
11495: goto end;
11496: }
11497: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11498: 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);
11499: 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);
11500: 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 11501: }
1.254 brouard 11502:
11503: while(fgets(line, MAXLINE, ficpar)) {
11504: /* If line starts with a # it is a comment */
11505: if (line[0] == '#') {
11506: numlinepar++;
11507: fputs(line,stdout);
11508: fputs(line,ficparo);
11509: fputs(line,ficlog);
11510: continue;
11511: }else
11512: break;
1.126 brouard 11513: }
11514:
11515:
11516: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11517: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11518:
1.254 brouard 11519: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11520: if (num_filled != 1) {
11521: 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);
11522: 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);
11523: goto end;
11524: }
11525: printf("pop_based=%d\n",popbased);
11526: fprintf(ficlog,"pop_based=%d\n",popbased);
11527: fprintf(ficparo,"pop_based=%d\n",popbased);
11528: fprintf(ficres,"pop_based=%d\n",popbased);
11529: }
11530:
1.258 brouard 11531: /* Results */
11532: nresult=0;
11533: do{
11534: if(!fgets(line, MAXLINE, ficpar)){
11535: endishere=1;
11536: parameterline=14;
11537: }else if (line[0] == '#') {
11538: /* If line starts with a # it is a comment */
1.254 brouard 11539: numlinepar++;
11540: fputs(line,stdout);
11541: fputs(line,ficparo);
11542: fputs(line,ficlog);
11543: continue;
1.258 brouard 11544: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
11545: parameterline=11;
11546: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
11547: parameterline=12;
11548: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
11549: parameterline=13;
11550: else{
11551: parameterline=14;
1.254 brouard 11552: }
1.258 brouard 11553: switch (parameterline){
11554: case 11:
11555: 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){
11556: if (num_filled != 8) {
11557: 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);
11558: 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);
11559: goto end;
11560: }
11561: 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);
11562: 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);
11563: 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);
11564: 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);
11565: /* day and month of proj2 are not used but only year anproj2.*/
11566: }
1.254 brouard 11567: break;
1.258 brouard 11568: case 12:
11569: /*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);*/
11570: 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){
11571: if (num_filled != 8) {
1.262 brouard 11572: 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);
11573: 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 11574: goto end;
11575: }
11576: 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);
11577: 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);
11578: 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);
11579: 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);
11580: /* day and month of proj2 are not used but only year anproj2.*/
11581: }
1.230 brouard 11582: break;
1.258 brouard 11583: case 13:
11584: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
11585: if (num_filled == 0){
11586: resultline[0]='\0';
11587: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
11588: 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);
11589: break;
11590: } else if (num_filled != 1){
11591: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11592: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11593: }
11594: nresult++; /* Sum of resultlines */
11595: printf("Result %d: result=%s\n",nresult, resultline);
11596: if(nresult > MAXRESULTLINES){
11597: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11598: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11599: goto end;
11600: }
11601: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
11602: fprintf(ficparo,"result: %s\n",resultline);
11603: fprintf(ficres,"result: %s\n",resultline);
11604: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11605: break;
1.258 brouard 11606: case 14:
1.259 brouard 11607: if(ncovmodel >2 && nresult==0 ){
11608: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 11609: goto end;
11610: }
1.259 brouard 11611: break;
1.258 brouard 11612: default:
11613: nresult=1;
11614: decoderesult(".",nresult ); /* No covariate */
11615: }
11616: } /* End switch parameterline */
11617: }while(endishere==0); /* End do */
1.126 brouard 11618:
1.230 brouard 11619: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11620: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11621:
11622: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11623: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11624: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11625: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11626: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11627: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11628: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11629: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11630: }else{
1.266 ! brouard 11631: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)ageminpar);
1.220 brouard 11632: }
11633: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 11634: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.225 brouard 11635: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11636:
1.225 brouard 11637: /*------------ free_vector -------------*/
11638: /* chdir(path); */
1.220 brouard 11639:
1.215 brouard 11640: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11641: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11642: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11643: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11644: free_lvector(num,1,n);
11645: free_vector(agedc,1,n);
11646: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11647: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11648: fclose(ficparo);
11649: fclose(ficres);
1.220 brouard 11650:
11651:
1.186 brouard 11652: /* Other results (useful)*/
1.220 brouard 11653:
11654:
1.126 brouard 11655: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11656: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11657: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11658: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11659: fclose(ficrespl);
11660:
11661: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11662: /*#include "hpijx.h"*/
11663: hPijx(p, bage, fage);
1.145 brouard 11664: fclose(ficrespij);
1.227 brouard 11665:
1.220 brouard 11666: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11667: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11668: k=1;
1.126 brouard 11669: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11670:
1.219 brouard 11671: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11672: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11673: for(i=1;i<=AGESUP;i++)
1.219 brouard 11674: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11675: for(k=1;k<=ncovcombmax;k++)
11676: probs[i][j][k]=0.;
1.219 brouard 11677: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11678: if (mobilav!=0 ||mobilavproj !=0 ) {
11679: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11680: for(i=1;i<=AGESUP;i++)
11681: for(j=1;j<=nlstate;j++)
11682: for(k=1;k<=ncovcombmax;k++)
11683: mobaverages[i][j][k]=0.;
1.219 brouard 11684: mobaverage=mobaverages;
11685: if (mobilav!=0) {
1.235 brouard 11686: printf("Movingaveraging observed prevalence\n");
1.258 brouard 11687: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 11688: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11689: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11690: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11691: }
1.219 brouard 11692: }
1.266 ! brouard 11693: /* else if(mobilavproj==-1){ /\* Forcing raw observed prevalences *\/ */
! 11694: /* for(i=1;i<=AGESUP;i++) */
! 11695: /* for(j=1;j<=nlstate;j++) */
! 11696: /* for(k=1;k<=ncovcombmax;k++) */
! 11697: /* mobaverages[i][j][k]=probs[i][j][k]; */
! 11698: /* /\* /\\* Prevalence for each covariates in probs[age][status][cov] *\\/ *\/ */
! 11699: /* /\* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); *\/ */
! 11700: /* } */
1.219 brouard 11701: else if (mobilavproj !=0) {
1.235 brouard 11702: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 11703: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 11704: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11705: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11706: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11707: }
1.219 brouard 11708: }
11709: }/* end if moving average */
1.227 brouard 11710:
1.126 brouard 11711: /*---------- Forecasting ------------------*/
11712: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11713: if(prevfcast==1){
11714: /* if(stepm ==1){*/
1.225 brouard 11715: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11716: }
1.217 brouard 11717: if(backcast==1){
1.219 brouard 11718: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11719: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11720: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11721:
11722: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11723:
11724: bprlim=matrix(1,nlstate,1,nlstate);
11725: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11726: fclose(ficresplb);
11727:
1.222 brouard 11728: hBijx(p, bage, fage, mobaverage);
11729: fclose(ficrespijb);
1.219 brouard 11730: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11731:
11732: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11733: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11734: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11735: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11736: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11737: }
1.217 brouard 11738:
1.186 brouard 11739:
11740: /* ------ Other prevalence ratios------------ */
1.126 brouard 11741:
1.215 brouard 11742: free_ivector(wav,1,imx);
11743: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11744: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11745: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11746:
11747:
1.127 brouard 11748: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11749:
1.201 brouard 11750: strcpy(filerese,"E_");
11751: strcat(filerese,fileresu);
1.126 brouard 11752: if((ficreseij=fopen(filerese,"w"))==NULL) {
11753: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11754: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11755: }
1.208 brouard 11756: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11757: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11758:
11759: pstamp(ficreseij);
1.219 brouard 11760:
1.235 brouard 11761: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11762: if (cptcovn < 1){i1=1;}
11763:
11764: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11765: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11766: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11767: continue;
1.219 brouard 11768: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11769: printf("\n#****** ");
1.225 brouard 11770: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11771: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11772: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11773: }
11774: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11775: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11776: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11777: }
11778: fprintf(ficreseij,"******\n");
1.235 brouard 11779: printf("******\n");
1.219 brouard 11780:
11781: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11782: oldm=oldms;savm=savms;
1.235 brouard 11783: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11784:
1.219 brouard 11785: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11786: }
11787: fclose(ficreseij);
1.208 brouard 11788: printf("done evsij\n");fflush(stdout);
11789: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11790:
1.227 brouard 11791: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11792:
11793:
1.201 brouard 11794: strcpy(filerest,"T_");
11795: strcat(filerest,fileresu);
1.127 brouard 11796: if((ficrest=fopen(filerest,"w"))==NULL) {
11797: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11798: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11799: }
1.208 brouard 11800: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11801: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11802:
1.126 brouard 11803:
1.201 brouard 11804: strcpy(fileresstde,"STDE_");
11805: strcat(fileresstde,fileresu);
1.126 brouard 11806: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11807: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11808: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11809: }
1.227 brouard 11810: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11811: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11812:
1.201 brouard 11813: strcpy(filerescve,"CVE_");
11814: strcat(filerescve,fileresu);
1.126 brouard 11815: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11816: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11817: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11818: }
1.227 brouard 11819: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11820: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11821:
1.201 brouard 11822: strcpy(fileresv,"V_");
11823: strcat(fileresv,fileresu);
1.126 brouard 11824: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11825: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11826: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11827: }
1.227 brouard 11828: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11829: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11830:
1.145 brouard 11831: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11832: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11833:
1.235 brouard 11834: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11835: if (cptcovn < 1){i1=1;}
11836:
11837: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11838: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11839: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11840: continue;
1.242 brouard 11841: printf("\n#****** Result for:");
11842: fprintf(ficrest,"\n#****** Result for:");
11843: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11844: for(j=1;j<=cptcoveff;j++){
11845: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11846: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11847: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11848: }
1.235 brouard 11849: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11850: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11851: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11852: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11853: }
1.208 brouard 11854: fprintf(ficrest,"******\n");
1.227 brouard 11855: fprintf(ficlog,"******\n");
11856: printf("******\n");
1.208 brouard 11857:
11858: fprintf(ficresstdeij,"\n#****** ");
11859: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11860: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11861: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11862: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11863: }
1.235 brouard 11864: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11865: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11866: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11867: }
1.208 brouard 11868: fprintf(ficresstdeij,"******\n");
11869: fprintf(ficrescveij,"******\n");
11870:
11871: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11872: /* pstamp(ficresvij); */
1.225 brouard 11873: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11874: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11875: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11876: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11877: }
1.208 brouard 11878: fprintf(ficresvij,"******\n");
11879:
11880: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11881: oldm=oldms;savm=savms;
1.235 brouard 11882: printf(" cvevsij ");
11883: fprintf(ficlog, " cvevsij ");
11884: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11885: printf(" end cvevsij \n ");
11886: fprintf(ficlog, " end cvevsij \n ");
11887:
11888: /*
11889: */
11890: /* goto endfree; */
11891:
11892: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11893: pstamp(ficrest);
11894:
11895:
11896: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11897: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11898: cptcod= 0; /* To be deleted */
11899: printf("varevsij vpopbased=%d \n",vpopbased);
11900: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11901: 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 11902: 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 ");
11903: if(vpopbased==1)
11904: 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);
11905: else
11906: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11907: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11908: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11909: fprintf(ficrest,"\n");
11910: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11911: epj=vector(1,nlstate+1);
11912: printf("Computing age specific period (stable) prevalences in each health state \n");
11913: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11914: for(age=bage; age <=fage ;age++){
1.235 brouard 11915: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11916: if (vpopbased==1) {
11917: if(mobilav ==0){
11918: for(i=1; i<=nlstate;i++)
11919: prlim[i][i]=probs[(int)age][i][k];
11920: }else{ /* mobilav */
11921: for(i=1; i<=nlstate;i++)
11922: prlim[i][i]=mobaverage[(int)age][i][k];
11923: }
11924: }
1.219 brouard 11925:
1.227 brouard 11926: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11927: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11928: /* printf(" age %4.0f ",age); */
11929: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11930: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11931: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11932: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11933: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11934: }
11935: epj[nlstate+1] +=epj[j];
11936: }
11937: /* printf(" age %4.0f \n",age); */
1.219 brouard 11938:
1.227 brouard 11939: for(i=1, vepp=0.;i <=nlstate;i++)
11940: for(j=1;j <=nlstate;j++)
11941: vepp += vareij[i][j][(int)age];
11942: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11943: for(j=1;j <=nlstate;j++){
11944: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11945: }
11946: fprintf(ficrest,"\n");
11947: }
1.208 brouard 11948: } /* End vpopbased */
11949: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11950: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11951: free_vector(epj,1,nlstate+1);
1.235 brouard 11952: printf("done selection\n");fflush(stdout);
11953: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11954:
1.145 brouard 11955: /*}*/
1.235 brouard 11956: } /* End k selection */
1.227 brouard 11957:
11958: printf("done State-specific expectancies\n");fflush(stdout);
11959: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11960:
1.126 brouard 11961: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11962:
1.201 brouard 11963: strcpy(fileresvpl,"VPL_");
11964: strcat(fileresvpl,fileresu);
1.126 brouard 11965: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11966: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11967: exit(0);
11968: }
1.208 brouard 11969: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11970: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11971:
1.145 brouard 11972: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11973: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11974:
1.235 brouard 11975: i1=pow(2,cptcoveff);
11976: if (cptcovn < 1){i1=1;}
11977:
11978: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11979: for(k=1; k<=i1;k++){
1.253 brouard 11980: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11981: continue;
1.227 brouard 11982: fprintf(ficresvpl,"\n#****** ");
11983: printf("\n#****** ");
11984: fprintf(ficlog,"\n#****** ");
11985: for(j=1;j<=cptcoveff;j++) {
11986: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11987: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11988: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11989: }
1.235 brouard 11990: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11991: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11992: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11993: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11994: }
1.227 brouard 11995: fprintf(ficresvpl,"******\n");
11996: printf("******\n");
11997: fprintf(ficlog,"******\n");
11998:
11999: varpl=matrix(1,nlstate,(int) bage, (int) fage);
12000: oldm=oldms;savm=savms;
1.235 brouard 12001: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 12002: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 12003: /*}*/
1.126 brouard 12004: }
1.227 brouard 12005:
1.126 brouard 12006: fclose(ficresvpl);
1.208 brouard 12007: printf("done variance-covariance of period prevalence\n");fflush(stdout);
12008: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 12009:
12010: free_vector(weight,1,n);
12011: free_imatrix(Tvard,1,NCOVMAX,1,2);
12012: free_imatrix(s,1,maxwav+1,1,n);
12013: free_matrix(anint,1,maxwav,1,n);
12014: free_matrix(mint,1,maxwav,1,n);
12015: free_ivector(cod,1,n);
12016: free_ivector(tab,1,NCOVMAX);
12017: fclose(ficresstdeij);
12018: fclose(ficrescveij);
12019: fclose(ficresvij);
12020: fclose(ficrest);
12021: fclose(ficpar);
12022:
12023:
1.126 brouard 12024: /*---------- End : free ----------------*/
1.219 brouard 12025: if (mobilav!=0 ||mobilavproj !=0)
12026: free_ma3x(mobaverages,1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
1.218 brouard 12027: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12028: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12029: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12030: } /* mle==-3 arrives here for freeing */
1.227 brouard 12031: /* endfree:*/
12032: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12033: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12034: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
12035: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 12036: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 12037: free_matrix(coqvar,1,maxwav,1,n);
12038: free_matrix(covar,0,NCOVMAX,1,n);
12039: free_matrix(matcov,1,npar,1,npar);
12040: free_matrix(hess,1,npar,1,npar);
12041: /*free_vector(delti,1,npar);*/
12042: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12043: free_matrix(agev,1,maxwav,1,imx);
12044: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12045:
12046: free_ivector(ncodemax,1,NCOVMAX);
12047: free_ivector(ncodemaxwundef,1,NCOVMAX);
12048: free_ivector(Dummy,-1,NCOVMAX);
12049: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12050: free_ivector(DummyV,1,NCOVMAX);
12051: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12052: free_ivector(Typevar,-1,NCOVMAX);
12053: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12054: free_ivector(TvarsQ,1,NCOVMAX);
12055: free_ivector(TvarsQind,1,NCOVMAX);
12056: free_ivector(TvarsD,1,NCOVMAX);
12057: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12058: free_ivector(TvarFD,1,NCOVMAX);
12059: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12060: free_ivector(TvarF,1,NCOVMAX);
12061: free_ivector(TvarFind,1,NCOVMAX);
12062: free_ivector(TvarV,1,NCOVMAX);
12063: free_ivector(TvarVind,1,NCOVMAX);
12064: free_ivector(TvarA,1,NCOVMAX);
12065: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12066: free_ivector(TvarFQ,1,NCOVMAX);
12067: free_ivector(TvarFQind,1,NCOVMAX);
12068: free_ivector(TvarVD,1,NCOVMAX);
12069: free_ivector(TvarVDind,1,NCOVMAX);
12070: free_ivector(TvarVQ,1,NCOVMAX);
12071: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12072: free_ivector(Tvarsel,1,NCOVMAX);
12073: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12074: free_ivector(Tposprod,1,NCOVMAX);
12075: free_ivector(Tprod,1,NCOVMAX);
12076: free_ivector(Tvaraff,1,NCOVMAX);
12077: free_ivector(invalidvarcomb,1,ncovcombmax);
12078: free_ivector(Tage,1,NCOVMAX);
12079: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12080: free_ivector(TmodelInvind,1,NCOVMAX);
12081: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12082:
12083: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12084: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12085: fflush(fichtm);
12086: fflush(ficgp);
12087:
1.227 brouard 12088:
1.126 brouard 12089: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12090: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12091: 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 12092: }else{
12093: printf("End of Imach\n");
12094: fprintf(ficlog,"End of Imach\n");
12095: }
12096: printf("See log file on %s\n",filelog);
12097: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12098: /*(void) gettimeofday(&end_time,&tzp);*/
12099: rend_time = time(NULL);
12100: end_time = *localtime(&rend_time);
12101: /* tml = *localtime(&end_time.tm_sec); */
12102: strcpy(strtend,asctime(&end_time));
1.126 brouard 12103: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12104: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12105: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12106:
1.157 brouard 12107: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12108: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12109: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12110: /* printf("Total time was %d uSec.\n", total_usecs);*/
12111: /* if(fileappend(fichtm,optionfilehtm)){ */
12112: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12113: fclose(fichtm);
12114: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12115: fclose(fichtmcov);
12116: fclose(ficgp);
12117: fclose(ficlog);
12118: /*------ End -----------*/
1.227 brouard 12119:
12120:
12121: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12122: #ifdef WIN32
1.227 brouard 12123: if (_chdir(pathcd) != 0)
12124: printf("Can't move to directory %s!\n",path);
12125: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12126: #else
1.227 brouard 12127: if(chdir(pathcd) != 0)
12128: printf("Can't move to directory %s!\n", path);
12129: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12130: #endif
1.126 brouard 12131: printf("Current directory %s!\n",pathcd);
12132: /*strcat(plotcmd,CHARSEPARATOR);*/
12133: sprintf(plotcmd,"gnuplot");
1.157 brouard 12134: #ifdef _WIN32
1.126 brouard 12135: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12136: #endif
12137: if(!stat(plotcmd,&info)){
1.158 brouard 12138: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12139: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12140: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12141: }else
12142: strcpy(pplotcmd,plotcmd);
1.157 brouard 12143: #ifdef __unix
1.126 brouard 12144: strcpy(plotcmd,GNUPLOTPROGRAM);
12145: if(!stat(plotcmd,&info)){
1.158 brouard 12146: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12147: }else
12148: strcpy(pplotcmd,plotcmd);
12149: #endif
12150: }else
12151: strcpy(pplotcmd,plotcmd);
12152:
12153: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12154: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12155:
1.126 brouard 12156: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12157: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12158: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12159: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12160: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12161: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12162: }
1.158 brouard 12163: printf(" Successful, please wait...");
1.126 brouard 12164: while (z[0] != 'q') {
12165: /* chdir(path); */
1.154 brouard 12166: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12167: scanf("%s",z);
12168: /* if (z[0] == 'c') system("./imach"); */
12169: if (z[0] == 'e') {
1.158 brouard 12170: #ifdef __APPLE__
1.152 brouard 12171: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12172: #elif __linux
12173: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12174: #else
1.152 brouard 12175: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12176: #endif
12177: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12178: system(pplotcmd);
1.126 brouard 12179: }
12180: else if (z[0] == 'g') system(plotcmd);
12181: else if (z[0] == 'q') exit(0);
12182: }
1.227 brouard 12183: end:
1.126 brouard 12184: while (z[0] != 'q') {
1.195 brouard 12185: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12186: scanf("%s",z);
12187: }
12188: }
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