Annotation of imach/src/imach.c, revision 1.278
1.278 ! brouard 1: /* $Id: imach.c,v 1.277 2017/07/17 08:53:49 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.278 ! brouard 4: Revision 1.277 2017/07/17 08:53:49 brouard
! 5: Summary: BOM files can be read now
! 6:
1.277 brouard 7: Revision 1.276 2017/06/30 15:48:31 brouard
8: Summary: Graphs improvements
9:
1.276 brouard 10: Revision 1.275 2017/06/30 13:39:33 brouard
11: Summary: Saito's color
12:
1.275 brouard 13: Revision 1.274 2017/06/29 09:47:08 brouard
14: Summary: Version 0.99r14
15:
1.274 brouard 16: Revision 1.273 2017/06/27 11:06:02 brouard
17: Summary: More documentation on projections
18:
1.273 brouard 19: Revision 1.272 2017/06/27 10:22:40 brouard
20: Summary: Color of backprojection changed from 6 to 5(yellow)
21:
1.272 brouard 22: Revision 1.271 2017/06/27 10:17:50 brouard
23: Summary: Some bug with rint
24:
1.271 brouard 25: Revision 1.270 2017/05/24 05:45:29 brouard
26: *** empty log message ***
27:
1.270 brouard 28: Revision 1.269 2017/05/23 08:39:25 brouard
29: Summary: Code into subroutine, cleanings
30:
1.269 brouard 31: Revision 1.268 2017/05/18 20:09:32 brouard
32: Summary: backprojection and confidence intervals of backprevalence
33:
1.268 brouard 34: Revision 1.267 2017/05/13 10:25:05 brouard
35: Summary: temporary save for backprojection
36:
1.267 brouard 37: Revision 1.266 2017/05/13 07:26:12 brouard
38: Summary: Version 0.99r13 (improvements and bugs fixed)
39:
1.266 brouard 40: Revision 1.265 2017/04/26 16:22:11 brouard
41: Summary: imach 0.99r13 Some bugs fixed
42:
1.265 brouard 43: Revision 1.264 2017/04/26 06:01:29 brouard
44: Summary: Labels in graphs
45:
1.264 brouard 46: Revision 1.263 2017/04/24 15:23:15 brouard
47: Summary: to save
48:
1.263 brouard 49: Revision 1.262 2017/04/18 16:48:12 brouard
50: *** empty log message ***
51:
1.262 brouard 52: Revision 1.261 2017/04/05 10:14:09 brouard
53: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
54:
1.261 brouard 55: Revision 1.260 2017/04/04 17:46:59 brouard
56: Summary: Gnuplot indexations fixed (humm)
57:
1.260 brouard 58: Revision 1.259 2017/04/04 13:01:16 brouard
59: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
60:
1.259 brouard 61: Revision 1.258 2017/04/03 10:17:47 brouard
62: Summary: Version 0.99r12
63:
64: Some cleanings, conformed with updated documentation.
65:
1.258 brouard 66: Revision 1.257 2017/03/29 16:53:30 brouard
67: Summary: Temp
68:
1.257 brouard 69: Revision 1.256 2017/03/27 05:50:23 brouard
70: Summary: Temporary
71:
1.256 brouard 72: Revision 1.255 2017/03/08 16:02:28 brouard
73: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
74:
1.255 brouard 75: Revision 1.254 2017/03/08 07:13:00 brouard
76: Summary: Fixing data parameter line
77:
1.254 brouard 78: Revision 1.253 2016/12/15 11:59:41 brouard
79: Summary: 0.99 in progress
80:
1.253 brouard 81: Revision 1.252 2016/09/15 21:15:37 brouard
82: *** empty log message ***
83:
1.252 brouard 84: Revision 1.251 2016/09/15 15:01:13 brouard
85: Summary: not working
86:
1.251 brouard 87: Revision 1.250 2016/09/08 16:07:27 brouard
88: Summary: continue
89:
1.250 brouard 90: Revision 1.249 2016/09/07 17:14:18 brouard
91: Summary: Starting values from frequencies
92:
1.249 brouard 93: Revision 1.248 2016/09/07 14:10:18 brouard
94: *** empty log message ***
95:
1.248 brouard 96: Revision 1.247 2016/09/02 11:11:21 brouard
97: *** empty log message ***
98:
1.247 brouard 99: Revision 1.246 2016/09/02 08:49:22 brouard
100: *** empty log message ***
101:
1.246 brouard 102: Revision 1.245 2016/09/02 07:25:01 brouard
103: *** empty log message ***
104:
1.245 brouard 105: Revision 1.244 2016/09/02 07:17:34 brouard
106: *** empty log message ***
107:
1.244 brouard 108: Revision 1.243 2016/09/02 06:45:35 brouard
109: *** empty log message ***
110:
1.243 brouard 111: Revision 1.242 2016/08/30 15:01:20 brouard
112: Summary: Fixing a lots
113:
1.242 brouard 114: Revision 1.241 2016/08/29 17:17:25 brouard
115: Summary: gnuplot problem in Back projection to fix
116:
1.241 brouard 117: Revision 1.240 2016/08/29 07:53:18 brouard
118: Summary: Better
119:
1.240 brouard 120: Revision 1.239 2016/08/26 15:51:03 brouard
121: Summary: Improvement in Powell output in order to copy and paste
122:
123: Author:
124:
1.239 brouard 125: Revision 1.238 2016/08/26 14:23:35 brouard
126: Summary: Starting tests of 0.99
127:
1.238 brouard 128: Revision 1.237 2016/08/26 09:20:19 brouard
129: Summary: to valgrind
130:
1.237 brouard 131: Revision 1.236 2016/08/25 10:50:18 brouard
132: *** empty log message ***
133:
1.236 brouard 134: Revision 1.235 2016/08/25 06:59:23 brouard
135: *** empty log message ***
136:
1.235 brouard 137: Revision 1.234 2016/08/23 16:51:20 brouard
138: *** empty log message ***
139:
1.234 brouard 140: Revision 1.233 2016/08/23 07:40:50 brouard
141: Summary: not working
142:
1.233 brouard 143: Revision 1.232 2016/08/22 14:20:21 brouard
144: Summary: not working
145:
1.232 brouard 146: Revision 1.231 2016/08/22 07:17:15 brouard
147: Summary: not working
148:
1.231 brouard 149: Revision 1.230 2016/08/22 06:55:53 brouard
150: Summary: Not working
151:
1.230 brouard 152: Revision 1.229 2016/07/23 09:45:53 brouard
153: Summary: Completing for func too
154:
1.229 brouard 155: Revision 1.228 2016/07/22 17:45:30 brouard
156: Summary: Fixing some arrays, still debugging
157:
1.227 brouard 158: Revision 1.226 2016/07/12 18:42:34 brouard
159: Summary: temp
160:
1.226 brouard 161: Revision 1.225 2016/07/12 08:40:03 brouard
162: Summary: saving but not running
163:
1.225 brouard 164: Revision 1.224 2016/07/01 13:16:01 brouard
165: Summary: Fixes
166:
1.224 brouard 167: Revision 1.223 2016/02/19 09:23:35 brouard
168: Summary: temporary
169:
1.223 brouard 170: Revision 1.222 2016/02/17 08:14:50 brouard
171: Summary: Probably last 0.98 stable version 0.98r6
172:
1.222 brouard 173: Revision 1.221 2016/02/15 23:35:36 brouard
174: Summary: minor bug
175:
1.220 brouard 176: Revision 1.219 2016/02/15 00:48:12 brouard
177: *** empty log message ***
178:
1.219 brouard 179: Revision 1.218 2016/02/12 11:29:23 brouard
180: Summary: 0.99 Back projections
181:
1.218 brouard 182: Revision 1.217 2015/12/23 17:18:31 brouard
183: Summary: Experimental backcast
184:
1.217 brouard 185: Revision 1.216 2015/12/18 17:32:11 brouard
186: Summary: 0.98r4 Warning and status=-2
187:
188: Version 0.98r4 is now:
189: - displaying an error when status is -1, date of interview unknown and date of death known;
190: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
191: Older changes concerning s=-2, dating from 2005 have been supersed.
192:
1.216 brouard 193: Revision 1.215 2015/12/16 08:52:24 brouard
194: Summary: 0.98r4 working
195:
1.215 brouard 196: Revision 1.214 2015/12/16 06:57:54 brouard
197: Summary: temporary not working
198:
1.214 brouard 199: Revision 1.213 2015/12/11 18:22:17 brouard
200: Summary: 0.98r4
201:
1.213 brouard 202: Revision 1.212 2015/11/21 12:47:24 brouard
203: Summary: minor typo
204:
1.212 brouard 205: Revision 1.211 2015/11/21 12:41:11 brouard
206: Summary: 0.98r3 with some graph of projected cross-sectional
207:
208: Author: Nicolas Brouard
209:
1.211 brouard 210: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 211: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 212: Summary: Adding ftolpl parameter
213: Author: N Brouard
214:
215: We had difficulties to get smoothed confidence intervals. It was due
216: to the period prevalence which wasn't computed accurately. The inner
217: parameter ftolpl is now an outer parameter of the .imach parameter
218: file after estepm. If ftolpl is small 1.e-4 and estepm too,
219: computation are long.
220:
1.209 brouard 221: Revision 1.208 2015/11/17 14:31:57 brouard
222: Summary: temporary
223:
1.208 brouard 224: Revision 1.207 2015/10/27 17:36:57 brouard
225: *** empty log message ***
226:
1.207 brouard 227: Revision 1.206 2015/10/24 07:14:11 brouard
228: *** empty log message ***
229:
1.206 brouard 230: Revision 1.205 2015/10/23 15:50:53 brouard
231: Summary: 0.98r3 some clarification for graphs on likelihood contributions
232:
1.205 brouard 233: Revision 1.204 2015/10/01 16:20:26 brouard
234: Summary: Some new graphs of contribution to likelihood
235:
1.204 brouard 236: Revision 1.203 2015/09/30 17:45:14 brouard
237: Summary: looking at better estimation of the hessian
238:
239: Also a better criteria for convergence to the period prevalence And
240: therefore adding the number of years needed to converge. (The
241: prevalence in any alive state shold sum to one
242:
1.203 brouard 243: Revision 1.202 2015/09/22 19:45:16 brouard
244: Summary: Adding some overall graph on contribution to likelihood. Might change
245:
1.202 brouard 246: Revision 1.201 2015/09/15 17:34:58 brouard
247: Summary: 0.98r0
248:
249: - Some new graphs like suvival functions
250: - Some bugs fixed like model=1+age+V2.
251:
1.201 brouard 252: Revision 1.200 2015/09/09 16:53:55 brouard
253: Summary: Big bug thanks to Flavia
254:
255: Even model=1+age+V2. did not work anymore
256:
1.200 brouard 257: Revision 1.199 2015/09/07 14:09:23 brouard
258: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
259:
1.199 brouard 260: Revision 1.198 2015/09/03 07:14:39 brouard
261: Summary: 0.98q5 Flavia
262:
1.198 brouard 263: Revision 1.197 2015/09/01 18:24:39 brouard
264: *** empty log message ***
265:
1.197 brouard 266: Revision 1.196 2015/08/18 23:17:52 brouard
267: Summary: 0.98q5
268:
1.196 brouard 269: Revision 1.195 2015/08/18 16:28:39 brouard
270: Summary: Adding a hack for testing purpose
271:
272: After reading the title, ftol and model lines, if the comment line has
273: a q, starting with #q, the answer at the end of the run is quit. It
274: permits to run test files in batch with ctest. The former workaround was
275: $ echo q | imach foo.imach
276:
1.195 brouard 277: Revision 1.194 2015/08/18 13:32:00 brouard
278: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
279:
1.194 brouard 280: Revision 1.193 2015/08/04 07:17:42 brouard
281: Summary: 0.98q4
282:
1.193 brouard 283: Revision 1.192 2015/07/16 16:49:02 brouard
284: Summary: Fixing some outputs
285:
1.192 brouard 286: Revision 1.191 2015/07/14 10:00:33 brouard
287: Summary: Some fixes
288:
1.191 brouard 289: Revision 1.190 2015/05/05 08:51:13 brouard
290: Summary: Adding digits in output parameters (7 digits instead of 6)
291:
292: Fix 1+age+.
293:
1.190 brouard 294: Revision 1.189 2015/04/30 14:45:16 brouard
295: Summary: 0.98q2
296:
1.189 brouard 297: Revision 1.188 2015/04/30 08:27:53 brouard
298: *** empty log message ***
299:
1.188 brouard 300: Revision 1.187 2015/04/29 09:11:15 brouard
301: *** empty log message ***
302:
1.187 brouard 303: Revision 1.186 2015/04/23 12:01:52 brouard
304: Summary: V1*age is working now, version 0.98q1
305:
306: Some codes had been disabled in order to simplify and Vn*age was
307: working in the optimization phase, ie, giving correct MLE parameters,
308: but, as usual, outputs were not correct and program core dumped.
309:
1.186 brouard 310: Revision 1.185 2015/03/11 13:26:42 brouard
311: Summary: Inclusion of compile and links command line for Intel Compiler
312:
1.185 brouard 313: Revision 1.184 2015/03/11 11:52:39 brouard
314: Summary: Back from Windows 8. Intel Compiler
315:
1.184 brouard 316: Revision 1.183 2015/03/10 20:34:32 brouard
317: Summary: 0.98q0, trying with directest, mnbrak fixed
318:
319: We use directest instead of original Powell test; probably no
320: incidence on the results, but better justifications;
321: We fixed Numerical Recipes mnbrak routine which was wrong and gave
322: wrong results.
323:
1.183 brouard 324: Revision 1.182 2015/02/12 08:19:57 brouard
325: Summary: Trying to keep directest which seems simpler and more general
326: Author: Nicolas Brouard
327:
1.182 brouard 328: Revision 1.181 2015/02/11 23:22:24 brouard
329: Summary: Comments on Powell added
330:
331: Author:
332:
1.181 brouard 333: Revision 1.180 2015/02/11 17:33:45 brouard
334: Summary: Finishing move from main to function (hpijx and prevalence_limit)
335:
1.180 brouard 336: Revision 1.179 2015/01/04 09:57:06 brouard
337: Summary: back to OS/X
338:
1.179 brouard 339: Revision 1.178 2015/01/04 09:35:48 brouard
340: *** empty log message ***
341:
1.178 brouard 342: Revision 1.177 2015/01/03 18:40:56 brouard
343: Summary: Still testing ilc32 on OSX
344:
1.177 brouard 345: Revision 1.176 2015/01/03 16:45:04 brouard
346: *** empty log message ***
347:
1.176 brouard 348: Revision 1.175 2015/01/03 16:33:42 brouard
349: *** empty log message ***
350:
1.175 brouard 351: Revision 1.174 2015/01/03 16:15:49 brouard
352: Summary: Still in cross-compilation
353:
1.174 brouard 354: Revision 1.173 2015/01/03 12:06:26 brouard
355: Summary: trying to detect cross-compilation
356:
1.173 brouard 357: Revision 1.172 2014/12/27 12:07:47 brouard
358: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
359:
1.172 brouard 360: Revision 1.171 2014/12/23 13:26:59 brouard
361: Summary: Back from Visual C
362:
363: Still problem with utsname.h on Windows
364:
1.171 brouard 365: Revision 1.170 2014/12/23 11:17:12 brouard
366: Summary: Cleaning some \%% back to %%
367:
368: The escape was mandatory for a specific compiler (which one?), but too many warnings.
369:
1.170 brouard 370: Revision 1.169 2014/12/22 23:08:31 brouard
371: Summary: 0.98p
372:
373: Outputs some informations on compiler used, OS etc. Testing on different platforms.
374:
1.169 brouard 375: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 376: Summary: update
1.169 brouard 377:
1.168 brouard 378: Revision 1.167 2014/12/22 13:50:56 brouard
379: Summary: Testing uname and compiler version and if compiled 32 or 64
380:
381: Testing on Linux 64
382:
1.167 brouard 383: Revision 1.166 2014/12/22 11:40:47 brouard
384: *** empty log message ***
385:
1.166 brouard 386: Revision 1.165 2014/12/16 11:20:36 brouard
387: Summary: After compiling on Visual C
388:
389: * imach.c (Module): Merging 1.61 to 1.162
390:
1.165 brouard 391: Revision 1.164 2014/12/16 10:52:11 brouard
392: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
393:
394: * imach.c (Module): Merging 1.61 to 1.162
395:
1.164 brouard 396: Revision 1.163 2014/12/16 10:30:11 brouard
397: * imach.c (Module): Merging 1.61 to 1.162
398:
1.163 brouard 399: Revision 1.162 2014/09/25 11:43:39 brouard
400: Summary: temporary backup 0.99!
401:
1.162 brouard 402: Revision 1.1 2014/09/16 11:06:58 brouard
403: Summary: With some code (wrong) for nlopt
404:
405: Author:
406:
407: Revision 1.161 2014/09/15 20:41:41 brouard
408: Summary: Problem with macro SQR on Intel compiler
409:
1.161 brouard 410: Revision 1.160 2014/09/02 09:24:05 brouard
411: *** empty log message ***
412:
1.160 brouard 413: Revision 1.159 2014/09/01 10:34:10 brouard
414: Summary: WIN32
415: Author: Brouard
416:
1.159 brouard 417: Revision 1.158 2014/08/27 17:11:51 brouard
418: *** empty log message ***
419:
1.158 brouard 420: Revision 1.157 2014/08/27 16:26:55 brouard
421: Summary: Preparing windows Visual studio version
422: Author: Brouard
423:
424: In order to compile on Visual studio, time.h is now correct and time_t
425: and tm struct should be used. difftime should be used but sometimes I
426: just make the differences in raw time format (time(&now).
427: Trying to suppress #ifdef LINUX
428: Add xdg-open for __linux in order to open default browser.
429:
1.157 brouard 430: Revision 1.156 2014/08/25 20:10:10 brouard
431: *** empty log message ***
432:
1.156 brouard 433: Revision 1.155 2014/08/25 18:32:34 brouard
434: Summary: New compile, minor changes
435: Author: Brouard
436:
1.155 brouard 437: Revision 1.154 2014/06/20 17:32:08 brouard
438: Summary: Outputs now all graphs of convergence to period prevalence
439:
1.154 brouard 440: Revision 1.153 2014/06/20 16:45:46 brouard
441: Summary: If 3 live state, convergence to period prevalence on same graph
442: Author: Brouard
443:
1.153 brouard 444: Revision 1.152 2014/06/18 17:54:09 brouard
445: Summary: open browser, use gnuplot on same dir than imach if not found in the path
446:
1.152 brouard 447: Revision 1.151 2014/06/18 16:43:30 brouard
448: *** empty log message ***
449:
1.151 brouard 450: Revision 1.150 2014/06/18 16:42:35 brouard
451: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
452: Author: brouard
453:
1.150 brouard 454: Revision 1.149 2014/06/18 15:51:14 brouard
455: Summary: Some fixes in parameter files errors
456: Author: Nicolas Brouard
457:
1.149 brouard 458: Revision 1.148 2014/06/17 17:38:48 brouard
459: Summary: Nothing new
460: Author: Brouard
461:
462: Just a new packaging for OS/X version 0.98nS
463:
1.148 brouard 464: Revision 1.147 2014/06/16 10:33:11 brouard
465: *** empty log message ***
466:
1.147 brouard 467: Revision 1.146 2014/06/16 10:20:28 brouard
468: Summary: Merge
469: Author: Brouard
470:
471: Merge, before building revised version.
472:
1.146 brouard 473: Revision 1.145 2014/06/10 21:23:15 brouard
474: Summary: Debugging with valgrind
475: Author: Nicolas Brouard
476:
477: Lot of changes in order to output the results with some covariates
478: After the Edimburgh REVES conference 2014, it seems mandatory to
479: improve the code.
480: No more memory valgrind error but a lot has to be done in order to
481: continue the work of splitting the code into subroutines.
482: Also, decodemodel has been improved. Tricode is still not
483: optimal. nbcode should be improved. Documentation has been added in
484: the source code.
485:
1.144 brouard 486: Revision 1.143 2014/01/26 09:45:38 brouard
487: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
488:
489: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
490: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
491:
1.143 brouard 492: Revision 1.142 2014/01/26 03:57:36 brouard
493: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
494:
495: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
496:
1.142 brouard 497: Revision 1.141 2014/01/26 02:42:01 brouard
498: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
499:
1.141 brouard 500: Revision 1.140 2011/09/02 10:37:54 brouard
501: Summary: times.h is ok with mingw32 now.
502:
1.140 brouard 503: Revision 1.139 2010/06/14 07:50:17 brouard
504: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
505: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
506:
1.139 brouard 507: Revision 1.138 2010/04/30 18:19:40 brouard
508: *** empty log message ***
509:
1.138 brouard 510: Revision 1.137 2010/04/29 18:11:38 brouard
511: (Module): Checking covariates for more complex models
512: than V1+V2. A lot of change to be done. Unstable.
513:
1.137 brouard 514: Revision 1.136 2010/04/26 20:30:53 brouard
515: (Module): merging some libgsl code. Fixing computation
516: of likelione (using inter/intrapolation if mle = 0) in order to
517: get same likelihood as if mle=1.
518: Some cleaning of code and comments added.
519:
1.136 brouard 520: Revision 1.135 2009/10/29 15:33:14 brouard
521: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
522:
1.135 brouard 523: Revision 1.134 2009/10/29 13:18:53 brouard
524: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
525:
1.134 brouard 526: Revision 1.133 2009/07/06 10:21:25 brouard
527: just nforces
528:
1.133 brouard 529: Revision 1.132 2009/07/06 08:22:05 brouard
530: Many tings
531:
1.132 brouard 532: Revision 1.131 2009/06/20 16:22:47 brouard
533: Some dimensions resccaled
534:
1.131 brouard 535: Revision 1.130 2009/05/26 06:44:34 brouard
536: (Module): Max Covariate is now set to 20 instead of 8. A
537: lot of cleaning with variables initialized to 0. Trying to make
538: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
539:
1.130 brouard 540: Revision 1.129 2007/08/31 13:49:27 lievre
541: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
542:
1.129 lievre 543: Revision 1.128 2006/06/30 13:02:05 brouard
544: (Module): Clarifications on computing e.j
545:
1.128 brouard 546: Revision 1.127 2006/04/28 18:11:50 brouard
547: (Module): Yes the sum of survivors was wrong since
548: imach-114 because nhstepm was no more computed in the age
549: loop. Now we define nhstepma in the age loop.
550: (Module): In order to speed up (in case of numerous covariates) we
551: compute health expectancies (without variances) in a first step
552: and then all the health expectancies with variances or standard
553: deviation (needs data from the Hessian matrices) which slows the
554: computation.
555: In the future we should be able to stop the program is only health
556: expectancies and graph are needed without standard deviations.
557:
1.127 brouard 558: Revision 1.126 2006/04/28 17:23:28 brouard
559: (Module): Yes the sum of survivors was wrong since
560: imach-114 because nhstepm was no more computed in the age
561: loop. Now we define nhstepma in the age loop.
562: Version 0.98h
563:
1.126 brouard 564: Revision 1.125 2006/04/04 15:20:31 lievre
565: Errors in calculation of health expectancies. Age was not initialized.
566: Forecasting file added.
567:
568: Revision 1.124 2006/03/22 17:13:53 lievre
569: Parameters are printed with %lf instead of %f (more numbers after the comma).
570: The log-likelihood is printed in the log file
571:
572: Revision 1.123 2006/03/20 10:52:43 brouard
573: * imach.c (Module): <title> changed, corresponds to .htm file
574: name. <head> headers where missing.
575:
576: * imach.c (Module): Weights can have a decimal point as for
577: English (a comma might work with a correct LC_NUMERIC environment,
578: otherwise the weight is truncated).
579: Modification of warning when the covariates values are not 0 or
580: 1.
581: Version 0.98g
582:
583: Revision 1.122 2006/03/20 09:45:41 brouard
584: (Module): Weights can have a decimal point as for
585: English (a comma might work with a correct LC_NUMERIC environment,
586: otherwise the weight is truncated).
587: Modification of warning when the covariates values are not 0 or
588: 1.
589: Version 0.98g
590:
591: Revision 1.121 2006/03/16 17:45:01 lievre
592: * imach.c (Module): Comments concerning covariates added
593:
594: * imach.c (Module): refinements in the computation of lli if
595: status=-2 in order to have more reliable computation if stepm is
596: not 1 month. Version 0.98f
597:
598: Revision 1.120 2006/03/16 15:10:38 lievre
599: (Module): refinements in the computation of lli if
600: status=-2 in order to have more reliable computation if stepm is
601: not 1 month. Version 0.98f
602:
603: Revision 1.119 2006/03/15 17:42:26 brouard
604: (Module): Bug if status = -2, the loglikelihood was
605: computed as likelihood omitting the logarithm. Version O.98e
606:
607: Revision 1.118 2006/03/14 18:20:07 brouard
608: (Module): varevsij Comments added explaining the second
609: table of variances if popbased=1 .
610: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
611: (Module): Function pstamp added
612: (Module): Version 0.98d
613:
614: Revision 1.117 2006/03/14 17:16:22 brouard
615: (Module): varevsij Comments added explaining the second
616: table of variances if popbased=1 .
617: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
618: (Module): Function pstamp added
619: (Module): Version 0.98d
620:
621: Revision 1.116 2006/03/06 10:29:27 brouard
622: (Module): Variance-covariance wrong links and
623: varian-covariance of ej. is needed (Saito).
624:
625: Revision 1.115 2006/02/27 12:17:45 brouard
626: (Module): One freematrix added in mlikeli! 0.98c
627:
628: Revision 1.114 2006/02/26 12:57:58 brouard
629: (Module): Some improvements in processing parameter
630: filename with strsep.
631:
632: Revision 1.113 2006/02/24 14:20:24 brouard
633: (Module): Memory leaks checks with valgrind and:
634: datafile was not closed, some imatrix were not freed and on matrix
635: allocation too.
636:
637: Revision 1.112 2006/01/30 09:55:26 brouard
638: (Module): Back to gnuplot.exe instead of wgnuplot.exe
639:
640: Revision 1.111 2006/01/25 20:38:18 brouard
641: (Module): Lots of cleaning and bugs added (Gompertz)
642: (Module): Comments can be added in data file. Missing date values
643: can be a simple dot '.'.
644:
645: Revision 1.110 2006/01/25 00:51:50 brouard
646: (Module): Lots of cleaning and bugs added (Gompertz)
647:
648: Revision 1.109 2006/01/24 19:37:15 brouard
649: (Module): Comments (lines starting with a #) are allowed in data.
650:
651: Revision 1.108 2006/01/19 18:05:42 lievre
652: Gnuplot problem appeared...
653: To be fixed
654:
655: Revision 1.107 2006/01/19 16:20:37 brouard
656: Test existence of gnuplot in imach path
657:
658: Revision 1.106 2006/01/19 13:24:36 brouard
659: Some cleaning and links added in html output
660:
661: Revision 1.105 2006/01/05 20:23:19 lievre
662: *** empty log message ***
663:
664: Revision 1.104 2005/09/30 16:11:43 lievre
665: (Module): sump fixed, loop imx fixed, and simplifications.
666: (Module): If the status is missing at the last wave but we know
667: that the person is alive, then we can code his/her status as -2
668: (instead of missing=-1 in earlier versions) and his/her
669: contributions to the likelihood is 1 - Prob of dying from last
670: health status (= 1-p13= p11+p12 in the easiest case of somebody in
671: the healthy state at last known wave). Version is 0.98
672:
673: Revision 1.103 2005/09/30 15:54:49 lievre
674: (Module): sump fixed, loop imx fixed, and simplifications.
675:
676: Revision 1.102 2004/09/15 17:31:30 brouard
677: Add the possibility to read data file including tab characters.
678:
679: Revision 1.101 2004/09/15 10:38:38 brouard
680: Fix on curr_time
681:
682: Revision 1.100 2004/07/12 18:29:06 brouard
683: Add version for Mac OS X. Just define UNIX in Makefile
684:
685: Revision 1.99 2004/06/05 08:57:40 brouard
686: *** empty log message ***
687:
688: Revision 1.98 2004/05/16 15:05:56 brouard
689: New version 0.97 . First attempt to estimate force of mortality
690: directly from the data i.e. without the need of knowing the health
691: state at each age, but using a Gompertz model: log u =a + b*age .
692: This is the basic analysis of mortality and should be done before any
693: other analysis, in order to test if the mortality estimated from the
694: cross-longitudinal survey is different from the mortality estimated
695: from other sources like vital statistic data.
696:
697: The same imach parameter file can be used but the option for mle should be -3.
698:
1.133 brouard 699: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 700: former routines in order to include the new code within the former code.
701:
702: The output is very simple: only an estimate of the intercept and of
703: the slope with 95% confident intervals.
704:
705: Current limitations:
706: A) Even if you enter covariates, i.e. with the
707: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
708: B) There is no computation of Life Expectancy nor Life Table.
709:
710: Revision 1.97 2004/02/20 13:25:42 lievre
711: Version 0.96d. Population forecasting command line is (temporarily)
712: suppressed.
713:
714: Revision 1.96 2003/07/15 15:38:55 brouard
715: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
716: rewritten within the same printf. Workaround: many printfs.
717:
718: Revision 1.95 2003/07/08 07:54:34 brouard
719: * imach.c (Repository):
720: (Repository): Using imachwizard code to output a more meaningful covariance
721: matrix (cov(a12,c31) instead of numbers.
722:
723: Revision 1.94 2003/06/27 13:00:02 brouard
724: Just cleaning
725:
726: Revision 1.93 2003/06/25 16:33:55 brouard
727: (Module): On windows (cygwin) function asctime_r doesn't
728: exist so I changed back to asctime which exists.
729: (Module): Version 0.96b
730:
731: Revision 1.92 2003/06/25 16:30:45 brouard
732: (Module): On windows (cygwin) function asctime_r doesn't
733: exist so I changed back to asctime which exists.
734:
735: Revision 1.91 2003/06/25 15:30:29 brouard
736: * imach.c (Repository): Duplicated warning errors corrected.
737: (Repository): Elapsed time after each iteration is now output. It
738: helps to forecast when convergence will be reached. Elapsed time
739: is stamped in powell. We created a new html file for the graphs
740: concerning matrix of covariance. It has extension -cov.htm.
741:
742: Revision 1.90 2003/06/24 12:34:15 brouard
743: (Module): Some bugs corrected for windows. Also, when
744: mle=-1 a template is output in file "or"mypar.txt with the design
745: of the covariance matrix to be input.
746:
747: Revision 1.89 2003/06/24 12:30:52 brouard
748: (Module): Some bugs corrected for windows. Also, when
749: mle=-1 a template is output in file "or"mypar.txt with the design
750: of the covariance matrix to be input.
751:
752: Revision 1.88 2003/06/23 17:54:56 brouard
753: * 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.
754:
755: Revision 1.87 2003/06/18 12:26:01 brouard
756: Version 0.96
757:
758: Revision 1.86 2003/06/17 20:04:08 brouard
759: (Module): Change position of html and gnuplot routines and added
760: routine fileappend.
761:
762: Revision 1.85 2003/06/17 13:12:43 brouard
763: * imach.c (Repository): Check when date of death was earlier that
764: current date of interview. It may happen when the death was just
765: prior to the death. In this case, dh was negative and likelihood
766: was wrong (infinity). We still send an "Error" but patch by
767: assuming that the date of death was just one stepm after the
768: interview.
769: (Repository): Because some people have very long ID (first column)
770: we changed int to long in num[] and we added a new lvector for
771: memory allocation. But we also truncated to 8 characters (left
772: truncation)
773: (Repository): No more line truncation errors.
774:
775: Revision 1.84 2003/06/13 21:44:43 brouard
776: * imach.c (Repository): Replace "freqsummary" at a correct
777: place. It differs from routine "prevalence" which may be called
778: many times. Probs is memory consuming and must be used with
779: parcimony.
780: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
781:
782: Revision 1.83 2003/06/10 13:39:11 lievre
783: *** empty log message ***
784:
785: Revision 1.82 2003/06/05 15:57:20 brouard
786: Add log in imach.c and fullversion number is now printed.
787:
788: */
789: /*
790: Interpolated Markov Chain
791:
792: Short summary of the programme:
793:
1.227 brouard 794: This program computes Healthy Life Expectancies or State-specific
795: (if states aren't health statuses) Expectancies from
796: cross-longitudinal data. Cross-longitudinal data consist in:
797:
798: -1- a first survey ("cross") where individuals from different ages
799: are interviewed on their health status or degree of disability (in
800: the case of a health survey which is our main interest)
801:
802: -2- at least a second wave of interviews ("longitudinal") which
803: measure each change (if any) in individual health status. Health
804: expectancies are computed from the time spent in each health state
805: according to a model. More health states you consider, more time is
806: necessary to reach the Maximum Likelihood of the parameters involved
807: in the model. The simplest model is the multinomial logistic model
808: where pij is the probability to be observed in state j at the second
809: wave conditional to be observed in state i at the first
810: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
811: etc , where 'age' is age and 'sex' is a covariate. If you want to
812: have a more complex model than "constant and age", you should modify
813: the program where the markup *Covariates have to be included here
814: again* invites you to do it. More covariates you add, slower the
1.126 brouard 815: convergence.
816:
817: The advantage of this computer programme, compared to a simple
818: multinomial logistic model, is clear when the delay between waves is not
819: identical for each individual. Also, if a individual missed an
820: intermediate interview, the information is lost, but taken into
821: account using an interpolation or extrapolation.
822:
823: hPijx is the probability to be observed in state i at age x+h
824: conditional to the observed state i at age x. The delay 'h' can be
825: split into an exact number (nh*stepm) of unobserved intermediate
826: states. This elementary transition (by month, quarter,
827: semester or year) is modelled as a multinomial logistic. The hPx
828: matrix is simply the matrix product of nh*stepm elementary matrices
829: and the contribution of each individual to the likelihood is simply
830: hPijx.
831:
832: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 833: of the life expectancies. It also computes the period (stable) prevalence.
834:
835: Back prevalence and projections:
1.227 brouard 836:
837: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
838: double agemaxpar, double ftolpl, int *ncvyearp, double
839: dateprev1,double dateprev2, int firstpass, int lastpass, int
840: mobilavproj)
841:
842: Computes the back prevalence limit for any combination of
843: covariate values k at any age between ageminpar and agemaxpar and
844: returns it in **bprlim. In the loops,
845:
846: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
847: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
848:
849: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 850: Computes for any combination of covariates k and any age between bage and fage
851: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
852: oldm=oldms;savm=savms;
1.227 brouard 853:
1.267 brouard 854: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 855: Computes the transition matrix starting at age 'age' over
856: 'nhstepm*hstepm*stepm' months (i.e. until
857: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 858: nhstepm*hstepm matrices.
859:
860: Returns p3mat[i][j][h] after calling
861: p3mat[i][j][h]=matprod2(newm,
862: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
863: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
864: oldm);
1.226 brouard 865:
866: Important routines
867:
868: - func (or funcone), computes logit (pij) distinguishing
869: o fixed variables (single or product dummies or quantitative);
870: o varying variables by:
871: (1) wave (single, product dummies, quantitative),
872: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
873: % fixed dummy (treated) or quantitative (not done because time-consuming);
874: % varying dummy (not done) or quantitative (not done);
875: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
876: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
877: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
878: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
879: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 880:
1.226 brouard 881:
882:
1.133 brouard 883: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
884: Institut national d'études démographiques, Paris.
1.126 brouard 885: This software have been partly granted by Euro-REVES, a concerted action
886: from the European Union.
887: It is copyrighted identically to a GNU software product, ie programme and
888: software can be distributed freely for non commercial use. Latest version
889: can be accessed at http://euroreves.ined.fr/imach .
890:
891: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
892: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
893:
894: **********************************************************************/
895: /*
896: main
897: read parameterfile
898: read datafile
899: concatwav
900: freqsummary
901: if (mle >= 1)
902: mlikeli
903: print results files
904: if mle==1
905: computes hessian
906: read end of parameter file: agemin, agemax, bage, fage, estepm
907: begin-prev-date,...
908: open gnuplot file
909: open html file
1.145 brouard 910: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
911: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
912: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
913: freexexit2 possible for memory heap.
914:
915: h Pij x | pij_nom ficrestpij
916: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
917: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
918: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
919:
920: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
921: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
922: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
923: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
924: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
925:
1.126 brouard 926: forecasting if prevfcast==1 prevforecast call prevalence()
927: health expectancies
928: Variance-covariance of DFLE
929: prevalence()
930: movingaverage()
931: varevsij()
932: if popbased==1 varevsij(,popbased)
933: total life expectancies
934: Variance of period (stable) prevalence
935: end
936: */
937:
1.187 brouard 938: /* #define DEBUG */
939: /* #define DEBUGBRENT */
1.203 brouard 940: /* #define DEBUGLINMIN */
941: /* #define DEBUGHESS */
942: #define DEBUGHESSIJ
1.224 brouard 943: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 944: #define POWELL /* Instead of NLOPT */
1.224 brouard 945: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 946: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
947: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 948:
949: #include <math.h>
950: #include <stdio.h>
951: #include <stdlib.h>
952: #include <string.h>
1.226 brouard 953: #include <ctype.h>
1.159 brouard 954:
955: #ifdef _WIN32
956: #include <io.h>
1.172 brouard 957: #include <windows.h>
958: #include <tchar.h>
1.159 brouard 959: #else
1.126 brouard 960: #include <unistd.h>
1.159 brouard 961: #endif
1.126 brouard 962:
963: #include <limits.h>
964: #include <sys/types.h>
1.171 brouard 965:
966: #if defined(__GNUC__)
967: #include <sys/utsname.h> /* Doesn't work on Windows */
968: #endif
969:
1.126 brouard 970: #include <sys/stat.h>
971: #include <errno.h>
1.159 brouard 972: /* extern int errno; */
1.126 brouard 973:
1.157 brouard 974: /* #ifdef LINUX */
975: /* #include <time.h> */
976: /* #include "timeval.h" */
977: /* #else */
978: /* #include <sys/time.h> */
979: /* #endif */
980:
1.126 brouard 981: #include <time.h>
982:
1.136 brouard 983: #ifdef GSL
984: #include <gsl/gsl_errno.h>
985: #include <gsl/gsl_multimin.h>
986: #endif
987:
1.167 brouard 988:
1.162 brouard 989: #ifdef NLOPT
990: #include <nlopt.h>
991: typedef struct {
992: double (* function)(double [] );
993: } myfunc_data ;
994: #endif
995:
1.126 brouard 996: /* #include <libintl.h> */
997: /* #define _(String) gettext (String) */
998:
1.251 brouard 999: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1000:
1001: #define GNUPLOTPROGRAM "gnuplot"
1002: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1003: #define FILENAMELENGTH 132
1004:
1005: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1006: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1007:
1.144 brouard 1008: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1009: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1010:
1011: #define NINTERVMAX 8
1.144 brouard 1012: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1013: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1014: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1015: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1016: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1017: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1018: #define MAXN 20000
1.144 brouard 1019: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1020: /* #define AGESUP 130 */
1021: #define AGESUP 150
1.268 brouard 1022: #define AGEINF 0
1.218 brouard 1023: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1024: #define AGEBASE 40
1.194 brouard 1025: #define AGEOVERFLOW 1.e20
1.164 brouard 1026: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1027: #ifdef _WIN32
1028: #define DIRSEPARATOR '\\'
1029: #define CHARSEPARATOR "\\"
1030: #define ODIRSEPARATOR '/'
1031: #else
1.126 brouard 1032: #define DIRSEPARATOR '/'
1033: #define CHARSEPARATOR "/"
1034: #define ODIRSEPARATOR '\\'
1035: #endif
1036:
1.278 ! brouard 1037: /* $Id: imach.c,v 1.277 2017/07/17 08:53:49 brouard Exp $ */
1.126 brouard 1038: /* $State: Exp $ */
1.196 brouard 1039: #include "version.h"
1040: char version[]=__IMACH_VERSION__;
1.224 brouard 1041: 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";
1.278 ! brouard 1042: char fullversion[]="$Revision: 1.277 $ $Date: 2017/07/17 08:53:49 $";
1.126 brouard 1043: char strstart[80];
1044: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1045: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1046: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1047: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1048: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1049: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1050: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1051: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1052: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1053: int cptcovprodnoage=0; /**< Number of covariate products without age */
1054: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1055: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1056: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1057: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1058: int nsd=0; /**< Total number of single dummy variables (output) */
1059: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1060: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1061: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1062: int ntveff=0; /**< ntveff number of effective time varying variables */
1063: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1064: int cptcov=0; /* Working variable */
1.218 brouard 1065: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1066: int npar=NPARMAX;
1067: int nlstate=2; /* Number of live states */
1068: int ndeath=1; /* Number of dead states */
1.130 brouard 1069: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1070: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1071: int popbased=0;
1072:
1073: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1074: int maxwav=0; /* Maxim number of waves */
1075: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1076: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1077: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1078: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1079: int mle=1, weightopt=0;
1.126 brouard 1080: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1081: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1082: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1083: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1084: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1085: int selected(int kvar); /* Is covariate kvar selected for printing results */
1086:
1.130 brouard 1087: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1088: double **matprod2(); /* test */
1.126 brouard 1089: double **oldm, **newm, **savm; /* Working pointers to matrices */
1090: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1091: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1092:
1.136 brouard 1093: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1094: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1095: FILE *ficlog, *ficrespow;
1.130 brouard 1096: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1097: double fretone; /* Only one call to likelihood */
1.130 brouard 1098: long ipmx=0; /* Number of contributions */
1.126 brouard 1099: double sw; /* Sum of weights */
1100: char filerespow[FILENAMELENGTH];
1101: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1102: FILE *ficresilk;
1103: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1104: FILE *ficresprobmorprev;
1105: FILE *fichtm, *fichtmcov; /* Html File */
1106: FILE *ficreseij;
1107: char filerese[FILENAMELENGTH];
1108: FILE *ficresstdeij;
1109: char fileresstde[FILENAMELENGTH];
1110: FILE *ficrescveij;
1111: char filerescve[FILENAMELENGTH];
1112: FILE *ficresvij;
1113: char fileresv[FILENAMELENGTH];
1.269 brouard 1114:
1.126 brouard 1115: char title[MAXLINE];
1.234 brouard 1116: char model[MAXLINE]; /**< The model line */
1.217 brouard 1117: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1118: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1119: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1120: char command[FILENAMELENGTH];
1121: int outcmd=0;
1122:
1.217 brouard 1123: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1124: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1125: char filelog[FILENAMELENGTH]; /* Log file */
1126: char filerest[FILENAMELENGTH];
1127: char fileregp[FILENAMELENGTH];
1128: char popfile[FILENAMELENGTH];
1129:
1130: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1131:
1.157 brouard 1132: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1133: /* struct timezone tzp; */
1134: /* extern int gettimeofday(); */
1135: struct tm tml, *gmtime(), *localtime();
1136:
1137: extern time_t time();
1138:
1139: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1140: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1141: struct tm tm;
1142:
1.126 brouard 1143: char strcurr[80], strfor[80];
1144:
1145: char *endptr;
1146: long lval;
1147: double dval;
1148:
1149: #define NR_END 1
1150: #define FREE_ARG char*
1151: #define FTOL 1.0e-10
1152:
1153: #define NRANSI
1.240 brouard 1154: #define ITMAX 200
1155: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1156:
1157: #define TOL 2.0e-4
1158:
1159: #define CGOLD 0.3819660
1160: #define ZEPS 1.0e-10
1161: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1162:
1163: #define GOLD 1.618034
1164: #define GLIMIT 100.0
1165: #define TINY 1.0e-20
1166:
1167: static double maxarg1,maxarg2;
1168: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1169: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1170:
1171: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1172: #define rint(a) floor(a+0.5)
1.166 brouard 1173: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1174: #define mytinydouble 1.0e-16
1.166 brouard 1175: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1176: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1177: /* static double dsqrarg; */
1178: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1179: static double sqrarg;
1180: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1181: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1182: int agegomp= AGEGOMP;
1183:
1184: int imx;
1185: int stepm=1;
1186: /* Stepm, step in month: minimum step interpolation*/
1187:
1188: int estepm;
1189: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1190:
1191: int m,nb;
1192: long *num;
1.197 brouard 1193: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1194: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1195: covariate for which somebody answered excluding
1196: undefined. Usually 2: 0 and 1. */
1197: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1198: covariate for which somebody answered including
1199: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1200: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1201: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1202: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1203: double *ageexmed,*agecens;
1204: double dateintmean=0;
1205:
1206: double *weight;
1207: int **s; /* Status */
1.141 brouard 1208: double *agedc;
1.145 brouard 1209: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1210: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1211: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1212: double **coqvar; /* Fixed quantitative covariate nqv */
1213: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1214: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1215: double idx;
1216: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1217: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1218: /*k 1 2 3 4 5 6 7 8 9 */
1219: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1220: /* Tndvar[k] 1 2 3 4 5 */
1221: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1222: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1223: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1224: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1225: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1226: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1227: /* Tprod[i]=k 4 7 */
1228: /* Tage[i]=k 5 8 */
1229: /* */
1230: /* Type */
1231: /* V 1 2 3 4 5 */
1232: /* F F V V V */
1233: /* D Q D D Q */
1234: /* */
1235: int *TvarsD;
1236: int *TvarsDind;
1237: int *TvarsQ;
1238: int *TvarsQind;
1239:
1.235 brouard 1240: #define MAXRESULTLINES 10
1241: int nresult=0;
1.258 brouard 1242: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1243: int TKresult[MAXRESULTLINES];
1.237 brouard 1244: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1245: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1246: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1247: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1248: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1249: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1250:
1.234 brouard 1251: /* 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 1252: 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 */
1253: 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 */
1254: 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 */
1255: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1256: 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 */
1257: 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 1258: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1259: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1260: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1261: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1262: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1263: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1264: 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 */
1265: 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 */
1266:
1.230 brouard 1267: int *Tvarsel; /**< Selected covariates for output */
1268: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1269: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1270: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1271: 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 1272: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1273: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1274: int *Tage;
1.227 brouard 1275: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1276: 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 1277: 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*/
1278: 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 1279: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1280: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1281: int **Tvard;
1282: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1283: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1284: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1285: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1286: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1287: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1288: double *lsurv, *lpop, *tpop;
1289:
1.231 brouard 1290: #define FD 1; /* Fixed dummy covariate */
1291: #define FQ 2; /* Fixed quantitative covariate */
1292: #define FP 3; /* Fixed product covariate */
1293: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1294: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1295: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1296: #define VD 10; /* Varying dummy covariate */
1297: #define VQ 11; /* Varying quantitative covariate */
1298: #define VP 12; /* Varying product covariate */
1299: #define VPDD 13; /* Varying product dummy*dummy covariate */
1300: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1301: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1302: #define APFD 16; /* Age product * fixed dummy covariate */
1303: #define APFQ 17; /* Age product * fixed quantitative covariate */
1304: #define APVD 18; /* Age product * varying dummy covariate */
1305: #define APVQ 19; /* Age product * varying quantitative covariate */
1306:
1307: #define FTYPE 1; /* Fixed covariate */
1308: #define VTYPE 2; /* Varying covariate (loop in wave) */
1309: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1310:
1311: struct kmodel{
1312: int maintype; /* main type */
1313: int subtype; /* subtype */
1314: };
1315: struct kmodel modell[NCOVMAX];
1316:
1.143 brouard 1317: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1318: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1319:
1320: /**************** split *************************/
1321: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1322: {
1323: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1324: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1325: */
1326: char *ss; /* pointer */
1.186 brouard 1327: int l1=0, l2=0; /* length counters */
1.126 brouard 1328:
1329: l1 = strlen(path ); /* length of path */
1330: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1331: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1332: if ( ss == NULL ) { /* no directory, so determine current directory */
1333: strcpy( name, path ); /* we got the fullname name because no directory */
1334: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1335: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1336: /* get current working directory */
1337: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1338: #ifdef WIN32
1339: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1340: #else
1341: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1342: #endif
1.126 brouard 1343: return( GLOCK_ERROR_GETCWD );
1344: }
1345: /* got dirc from getcwd*/
1346: printf(" DIRC = %s \n",dirc);
1.205 brouard 1347: } else { /* strip directory from path */
1.126 brouard 1348: ss++; /* after this, the filename */
1349: l2 = strlen( ss ); /* length of filename */
1350: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1351: strcpy( name, ss ); /* save file name */
1352: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1353: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1354: printf(" DIRC2 = %s \n",dirc);
1355: }
1356: /* We add a separator at the end of dirc if not exists */
1357: l1 = strlen( dirc ); /* length of directory */
1358: if( dirc[l1-1] != DIRSEPARATOR ){
1359: dirc[l1] = DIRSEPARATOR;
1360: dirc[l1+1] = 0;
1361: printf(" DIRC3 = %s \n",dirc);
1362: }
1363: ss = strrchr( name, '.' ); /* find last / */
1364: if (ss >0){
1365: ss++;
1366: strcpy(ext,ss); /* save extension */
1367: l1= strlen( name);
1368: l2= strlen(ss)+1;
1369: strncpy( finame, name, l1-l2);
1370: finame[l1-l2]= 0;
1371: }
1372:
1373: return( 0 ); /* we're done */
1374: }
1375:
1376:
1377: /******************************************/
1378:
1379: void replace_back_to_slash(char *s, char*t)
1380: {
1381: int i;
1382: int lg=0;
1383: i=0;
1384: lg=strlen(t);
1385: for(i=0; i<= lg; i++) {
1386: (s[i] = t[i]);
1387: if (t[i]== '\\') s[i]='/';
1388: }
1389: }
1390:
1.132 brouard 1391: char *trimbb(char *out, char *in)
1.137 brouard 1392: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1393: char *s;
1394: s=out;
1395: while (*in != '\0'){
1.137 brouard 1396: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1397: in++;
1398: }
1399: *out++ = *in++;
1400: }
1401: *out='\0';
1402: return s;
1403: }
1404:
1.187 brouard 1405: /* char *substrchaine(char *out, char *in, char *chain) */
1406: /* { */
1407: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1408: /* char *s, *t; */
1409: /* t=in;s=out; */
1410: /* while ((*in != *chain) && (*in != '\0')){ */
1411: /* *out++ = *in++; */
1412: /* } */
1413:
1414: /* /\* *in matches *chain *\/ */
1415: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1416: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1417: /* } */
1418: /* in--; chain--; */
1419: /* while ( (*in != '\0')){ */
1420: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1421: /* *out++ = *in++; */
1422: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1423: /* } */
1424: /* *out='\0'; */
1425: /* out=s; */
1426: /* return out; */
1427: /* } */
1428: char *substrchaine(char *out, char *in, char *chain)
1429: {
1430: /* Substract chain 'chain' from 'in', return and output 'out' */
1431: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1432:
1433: char *strloc;
1434:
1435: strcpy (out, in);
1436: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1437: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1438: if(strloc != NULL){
1439: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1440: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1441: /* strcpy (strloc, strloc +strlen(chain));*/
1442: }
1443: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1444: return out;
1445: }
1446:
1447:
1.145 brouard 1448: char *cutl(char *blocc, char *alocc, char *in, char occ)
1449: {
1.187 brouard 1450: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1451: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1452: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1453: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1454: */
1.160 brouard 1455: char *s, *t;
1.145 brouard 1456: t=in;s=in;
1457: while ((*in != occ) && (*in != '\0')){
1458: *alocc++ = *in++;
1459: }
1460: if( *in == occ){
1461: *(alocc)='\0';
1462: s=++in;
1463: }
1464:
1465: if (s == t) {/* occ not found */
1466: *(alocc-(in-s))='\0';
1467: in=s;
1468: }
1469: while ( *in != '\0'){
1470: *blocc++ = *in++;
1471: }
1472:
1473: *blocc='\0';
1474: return t;
1475: }
1.137 brouard 1476: char *cutv(char *blocc, char *alocc, char *in, char occ)
1477: {
1.187 brouard 1478: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1479: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1480: gives blocc="abcdef2ghi" and alocc="j".
1481: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1482: */
1483: char *s, *t;
1484: t=in;s=in;
1485: while (*in != '\0'){
1486: while( *in == occ){
1487: *blocc++ = *in++;
1488: s=in;
1489: }
1490: *blocc++ = *in++;
1491: }
1492: if (s == t) /* occ not found */
1493: *(blocc-(in-s))='\0';
1494: else
1495: *(blocc-(in-s)-1)='\0';
1496: in=s;
1497: while ( *in != '\0'){
1498: *alocc++ = *in++;
1499: }
1500:
1501: *alocc='\0';
1502: return s;
1503: }
1504:
1.126 brouard 1505: int nbocc(char *s, char occ)
1506: {
1507: int i,j=0;
1508: int lg=20;
1509: i=0;
1510: lg=strlen(s);
1511: for(i=0; i<= lg; i++) {
1.234 brouard 1512: if (s[i] == occ ) j++;
1.126 brouard 1513: }
1514: return j;
1515: }
1516:
1.137 brouard 1517: /* void cutv(char *u,char *v, char*t, char occ) */
1518: /* { */
1519: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1520: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1521: /* gives u="abcdef2ghi" and v="j" *\/ */
1522: /* int i,lg,j,p=0; */
1523: /* i=0; */
1524: /* lg=strlen(t); */
1525: /* for(j=0; j<=lg-1; j++) { */
1526: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1527: /* } */
1.126 brouard 1528:
1.137 brouard 1529: /* for(j=0; j<p; j++) { */
1530: /* (u[j] = t[j]); */
1531: /* } */
1532: /* u[p]='\0'; */
1.126 brouard 1533:
1.137 brouard 1534: /* for(j=0; j<= lg; j++) { */
1535: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1536: /* } */
1537: /* } */
1.126 brouard 1538:
1.160 brouard 1539: #ifdef _WIN32
1540: char * strsep(char **pp, const char *delim)
1541: {
1542: char *p, *q;
1543:
1544: if ((p = *pp) == NULL)
1545: return 0;
1546: if ((q = strpbrk (p, delim)) != NULL)
1547: {
1548: *pp = q + 1;
1549: *q = '\0';
1550: }
1551: else
1552: *pp = 0;
1553: return p;
1554: }
1555: #endif
1556:
1.126 brouard 1557: /********************** nrerror ********************/
1558:
1559: void nrerror(char error_text[])
1560: {
1561: fprintf(stderr,"ERREUR ...\n");
1562: fprintf(stderr,"%s\n",error_text);
1563: exit(EXIT_FAILURE);
1564: }
1565: /*********************** vector *******************/
1566: double *vector(int nl, int nh)
1567: {
1568: double *v;
1569: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1570: if (!v) nrerror("allocation failure in vector");
1571: return v-nl+NR_END;
1572: }
1573:
1574: /************************ free vector ******************/
1575: void free_vector(double*v, int nl, int nh)
1576: {
1577: free((FREE_ARG)(v+nl-NR_END));
1578: }
1579:
1580: /************************ivector *******************************/
1581: int *ivector(long nl,long nh)
1582: {
1583: int *v;
1584: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1585: if (!v) nrerror("allocation failure in ivector");
1586: return v-nl+NR_END;
1587: }
1588:
1589: /******************free ivector **************************/
1590: void free_ivector(int *v, long nl, long nh)
1591: {
1592: free((FREE_ARG)(v+nl-NR_END));
1593: }
1594:
1595: /************************lvector *******************************/
1596: long *lvector(long nl,long nh)
1597: {
1598: long *v;
1599: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1600: if (!v) nrerror("allocation failure in ivector");
1601: return v-nl+NR_END;
1602: }
1603:
1604: /******************free lvector **************************/
1605: void free_lvector(long *v, long nl, long nh)
1606: {
1607: free((FREE_ARG)(v+nl-NR_END));
1608: }
1609:
1610: /******************* imatrix *******************************/
1611: int **imatrix(long nrl, long nrh, long ncl, long nch)
1612: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1613: {
1614: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1615: int **m;
1616:
1617: /* allocate pointers to rows */
1618: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1619: if (!m) nrerror("allocation failure 1 in matrix()");
1620: m += NR_END;
1621: m -= nrl;
1622:
1623:
1624: /* allocate rows and set pointers to them */
1625: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1626: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1627: m[nrl] += NR_END;
1628: m[nrl] -= ncl;
1629:
1630: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1631:
1632: /* return pointer to array of pointers to rows */
1633: return m;
1634: }
1635:
1636: /****************** free_imatrix *************************/
1637: void free_imatrix(m,nrl,nrh,ncl,nch)
1638: int **m;
1639: long nch,ncl,nrh,nrl;
1640: /* free an int matrix allocated by imatrix() */
1641: {
1642: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1643: free((FREE_ARG) (m+nrl-NR_END));
1644: }
1645:
1646: /******************* matrix *******************************/
1647: double **matrix(long nrl, long nrh, long ncl, long nch)
1648: {
1649: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1650: double **m;
1651:
1652: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1653: if (!m) nrerror("allocation failure 1 in matrix()");
1654: m += NR_END;
1655: m -= nrl;
1656:
1657: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1658: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1659: m[nrl] += NR_END;
1660: m[nrl] -= ncl;
1661:
1662: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1663: return m;
1.145 brouard 1664: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1665: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1666: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1667: */
1668: }
1669:
1670: /*************************free matrix ************************/
1671: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1672: {
1673: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1674: free((FREE_ARG)(m+nrl-NR_END));
1675: }
1676:
1677: /******************* ma3x *******************************/
1678: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1679: {
1680: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1681: double ***m;
1682:
1683: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1684: if (!m) nrerror("allocation failure 1 in matrix()");
1685: m += NR_END;
1686: m -= nrl;
1687:
1688: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1689: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1690: m[nrl] += NR_END;
1691: m[nrl] -= ncl;
1692:
1693: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1694:
1695: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1696: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1697: m[nrl][ncl] += NR_END;
1698: m[nrl][ncl] -= nll;
1699: for (j=ncl+1; j<=nch; j++)
1700: m[nrl][j]=m[nrl][j-1]+nlay;
1701:
1702: for (i=nrl+1; i<=nrh; i++) {
1703: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1704: for (j=ncl+1; j<=nch; j++)
1705: m[i][j]=m[i][j-1]+nlay;
1706: }
1707: return m;
1708: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1709: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1710: */
1711: }
1712:
1713: /*************************free ma3x ************************/
1714: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1715: {
1716: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1717: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1718: free((FREE_ARG)(m+nrl-NR_END));
1719: }
1720:
1721: /*************** function subdirf ***********/
1722: char *subdirf(char fileres[])
1723: {
1724: /* Caution optionfilefiname is hidden */
1725: strcpy(tmpout,optionfilefiname);
1726: strcat(tmpout,"/"); /* Add to the right */
1727: strcat(tmpout,fileres);
1728: return tmpout;
1729: }
1730:
1731: /*************** function subdirf2 ***********/
1732: char *subdirf2(char fileres[], char *preop)
1733: {
1734:
1735: /* Caution optionfilefiname is hidden */
1736: strcpy(tmpout,optionfilefiname);
1737: strcat(tmpout,"/");
1738: strcat(tmpout,preop);
1739: strcat(tmpout,fileres);
1740: return tmpout;
1741: }
1742:
1743: /*************** function subdirf3 ***********/
1744: char *subdirf3(char fileres[], char *preop, char *preop2)
1745: {
1746:
1747: /* Caution optionfilefiname is hidden */
1748: strcpy(tmpout,optionfilefiname);
1749: strcat(tmpout,"/");
1750: strcat(tmpout,preop);
1751: strcat(tmpout,preop2);
1752: strcat(tmpout,fileres);
1753: return tmpout;
1754: }
1.213 brouard 1755:
1756: /*************** function subdirfext ***********/
1757: char *subdirfext(char fileres[], char *preop, char *postop)
1758: {
1759:
1760: strcpy(tmpout,preop);
1761: strcat(tmpout,fileres);
1762: strcat(tmpout,postop);
1763: return tmpout;
1764: }
1.126 brouard 1765:
1.213 brouard 1766: /*************** function subdirfext3 ***********/
1767: char *subdirfext3(char fileres[], char *preop, char *postop)
1768: {
1769:
1770: /* Caution optionfilefiname is hidden */
1771: strcpy(tmpout,optionfilefiname);
1772: strcat(tmpout,"/");
1773: strcat(tmpout,preop);
1774: strcat(tmpout,fileres);
1775: strcat(tmpout,postop);
1776: return tmpout;
1777: }
1778:
1.162 brouard 1779: char *asc_diff_time(long time_sec, char ascdiff[])
1780: {
1781: long sec_left, days, hours, minutes;
1782: days = (time_sec) / (60*60*24);
1783: sec_left = (time_sec) % (60*60*24);
1784: hours = (sec_left) / (60*60) ;
1785: sec_left = (sec_left) %(60*60);
1786: minutes = (sec_left) /60;
1787: sec_left = (sec_left) % (60);
1788: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1789: return ascdiff;
1790: }
1791:
1.126 brouard 1792: /***************** f1dim *************************/
1793: extern int ncom;
1794: extern double *pcom,*xicom;
1795: extern double (*nrfunc)(double []);
1796:
1797: double f1dim(double x)
1798: {
1799: int j;
1800: double f;
1801: double *xt;
1802:
1803: xt=vector(1,ncom);
1804: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1805: f=(*nrfunc)(xt);
1806: free_vector(xt,1,ncom);
1807: return f;
1808: }
1809:
1810: /*****************brent *************************/
1811: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1812: {
1813: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1814: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1815: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1816: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1817: * returned function value.
1818: */
1.126 brouard 1819: int iter;
1820: double a,b,d,etemp;
1.159 brouard 1821: double fu=0,fv,fw,fx;
1.164 brouard 1822: double ftemp=0.;
1.126 brouard 1823: double p,q,r,tol1,tol2,u,v,w,x,xm;
1824: double e=0.0;
1825:
1826: a=(ax < cx ? ax : cx);
1827: b=(ax > cx ? ax : cx);
1828: x=w=v=bx;
1829: fw=fv=fx=(*f)(x);
1830: for (iter=1;iter<=ITMAX;iter++) {
1831: xm=0.5*(a+b);
1832: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1833: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1834: printf(".");fflush(stdout);
1835: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1836: #ifdef DEBUGBRENT
1.126 brouard 1837: 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);
1838: 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);
1839: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1840: #endif
1841: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1842: *xmin=x;
1843: return fx;
1844: }
1845: ftemp=fu;
1846: if (fabs(e) > tol1) {
1847: r=(x-w)*(fx-fv);
1848: q=(x-v)*(fx-fw);
1849: p=(x-v)*q-(x-w)*r;
1850: q=2.0*(q-r);
1851: if (q > 0.0) p = -p;
1852: q=fabs(q);
1853: etemp=e;
1854: e=d;
1855: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1856: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1857: else {
1.224 brouard 1858: d=p/q;
1859: u=x+d;
1860: if (u-a < tol2 || b-u < tol2)
1861: d=SIGN(tol1,xm-x);
1.126 brouard 1862: }
1863: } else {
1864: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1865: }
1866: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1867: fu=(*f)(u);
1868: if (fu <= fx) {
1869: if (u >= x) a=x; else b=x;
1870: SHFT(v,w,x,u)
1.183 brouard 1871: SHFT(fv,fw,fx,fu)
1872: } else {
1873: if (u < x) a=u; else b=u;
1874: if (fu <= fw || w == x) {
1.224 brouard 1875: v=w;
1876: w=u;
1877: fv=fw;
1878: fw=fu;
1.183 brouard 1879: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1880: v=u;
1881: fv=fu;
1.183 brouard 1882: }
1883: }
1.126 brouard 1884: }
1885: nrerror("Too many iterations in brent");
1886: *xmin=x;
1887: return fx;
1888: }
1889:
1890: /****************** mnbrak ***********************/
1891:
1892: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1893: double (*func)(double))
1.183 brouard 1894: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1895: the downhill direction (defined by the function as evaluated at the initial points) and returns
1896: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1897: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1898: */
1.126 brouard 1899: double ulim,u,r,q, dum;
1900: double fu;
1.187 brouard 1901:
1902: double scale=10.;
1903: int iterscale=0;
1904:
1905: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1906: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1907:
1908:
1909: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1910: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1911: /* *bx = *ax - (*ax - *bx)/scale; */
1912: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1913: /* } */
1914:
1.126 brouard 1915: if (*fb > *fa) {
1916: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1917: SHFT(dum,*fb,*fa,dum)
1918: }
1.126 brouard 1919: *cx=(*bx)+GOLD*(*bx-*ax);
1920: *fc=(*func)(*cx);
1.183 brouard 1921: #ifdef DEBUG
1.224 brouard 1922: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1923: 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 1924: #endif
1.224 brouard 1925: 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 1926: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1927: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1928: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1929: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1930: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1931: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1932: fu=(*func)(u);
1.163 brouard 1933: #ifdef DEBUG
1934: /* f(x)=A(x-u)**2+f(u) */
1935: double A, fparabu;
1936: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1937: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1938: 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);
1939: 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 1940: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1941: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1942: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1943: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1944: #endif
1.184 brouard 1945: #ifdef MNBRAKORIGINAL
1.183 brouard 1946: #else
1.191 brouard 1947: /* if (fu > *fc) { */
1948: /* #ifdef DEBUG */
1949: /* printf("mnbrak4 fu > fc \n"); */
1950: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1951: /* #endif */
1952: /* /\* 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 *\\/ *\/ */
1953: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1954: /* dum=u; /\* Shifting c and u *\/ */
1955: /* u = *cx; */
1956: /* *cx = dum; */
1957: /* dum = fu; */
1958: /* fu = *fc; */
1959: /* *fc =dum; */
1960: /* } else { /\* end *\/ */
1961: /* #ifdef DEBUG */
1962: /* printf("mnbrak3 fu < fc \n"); */
1963: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1964: /* #endif */
1965: /* dum=u; /\* Shifting c and u *\/ */
1966: /* u = *cx; */
1967: /* *cx = dum; */
1968: /* dum = fu; */
1969: /* fu = *fc; */
1970: /* *fc =dum; */
1971: /* } */
1.224 brouard 1972: #ifdef DEBUGMNBRAK
1973: double A, fparabu;
1974: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1975: fparabu= *fa - A*(*ax-u)*(*ax-u);
1976: 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);
1977: 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 1978: #endif
1.191 brouard 1979: dum=u; /* Shifting c and u */
1980: u = *cx;
1981: *cx = dum;
1982: dum = fu;
1983: fu = *fc;
1984: *fc =dum;
1.183 brouard 1985: #endif
1.162 brouard 1986: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1987: #ifdef DEBUG
1.224 brouard 1988: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1989: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1990: #endif
1.126 brouard 1991: fu=(*func)(u);
1992: if (fu < *fc) {
1.183 brouard 1993: #ifdef DEBUG
1.224 brouard 1994: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1995: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1996: #endif
1997: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1998: SHFT(*fb,*fc,fu,(*func)(u))
1999: #ifdef DEBUG
2000: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2001: #endif
2002: }
1.162 brouard 2003: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2004: #ifdef DEBUG
1.224 brouard 2005: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2006: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2007: #endif
1.126 brouard 2008: u=ulim;
2009: fu=(*func)(u);
1.183 brouard 2010: } else { /* u could be left to b (if r > q parabola has a maximum) */
2011: #ifdef DEBUG
1.224 brouard 2012: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2013: 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 2014: #endif
1.126 brouard 2015: u=(*cx)+GOLD*(*cx-*bx);
2016: fu=(*func)(u);
1.224 brouard 2017: #ifdef DEBUG
2018: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2019: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2020: #endif
1.183 brouard 2021: } /* end tests */
1.126 brouard 2022: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2023: SHFT(*fa,*fb,*fc,fu)
2024: #ifdef DEBUG
1.224 brouard 2025: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2026: 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 2027: #endif
2028: } /* 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 2029: }
2030:
2031: /*************** linmin ************************/
1.162 brouard 2032: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2033: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2034: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2035: the value of func at the returned location p . This is actually all accomplished by calling the
2036: routines mnbrak and brent .*/
1.126 brouard 2037: int ncom;
2038: double *pcom,*xicom;
2039: double (*nrfunc)(double []);
2040:
1.224 brouard 2041: #ifdef LINMINORIGINAL
1.126 brouard 2042: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2043: #else
2044: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2045: #endif
1.126 brouard 2046: {
2047: double brent(double ax, double bx, double cx,
2048: double (*f)(double), double tol, double *xmin);
2049: double f1dim(double x);
2050: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2051: double *fc, double (*func)(double));
2052: int j;
2053: double xx,xmin,bx,ax;
2054: double fx,fb,fa;
1.187 brouard 2055:
1.203 brouard 2056: #ifdef LINMINORIGINAL
2057: #else
2058: double scale=10., axs, xxs; /* Scale added for infinity */
2059: #endif
2060:
1.126 brouard 2061: ncom=n;
2062: pcom=vector(1,n);
2063: xicom=vector(1,n);
2064: nrfunc=func;
2065: for (j=1;j<=n;j++) {
2066: pcom[j]=p[j];
1.202 brouard 2067: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2068: }
1.187 brouard 2069:
1.203 brouard 2070: #ifdef LINMINORIGINAL
2071: xx=1.;
2072: #else
2073: axs=0.0;
2074: xxs=1.;
2075: do{
2076: xx= xxs;
2077: #endif
1.187 brouard 2078: ax=0.;
2079: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2080: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2081: /* 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)) */
2082: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2083: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2084: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2085: /* 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 2086: #ifdef LINMINORIGINAL
2087: #else
2088: if (fx != fx){
1.224 brouard 2089: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2090: printf("|");
2091: fprintf(ficlog,"|");
1.203 brouard 2092: #ifdef DEBUGLINMIN
1.224 brouard 2093: 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 2094: #endif
2095: }
1.224 brouard 2096: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2097: #endif
2098:
1.191 brouard 2099: #ifdef DEBUGLINMIN
2100: 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 2101: 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 2102: #endif
1.224 brouard 2103: #ifdef LINMINORIGINAL
2104: #else
2105: if(fb == fx){ /* Flat function in the direction */
2106: xmin=xx;
2107: *flat=1;
2108: }else{
2109: *flat=0;
2110: #endif
2111: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2112: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2113: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2114: /* fmin = f(p[j] + xmin * xi[j]) */
2115: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2116: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2117: #ifdef DEBUG
1.224 brouard 2118: 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);
2119: 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);
2120: #endif
2121: #ifdef LINMINORIGINAL
2122: #else
2123: }
1.126 brouard 2124: #endif
1.191 brouard 2125: #ifdef DEBUGLINMIN
2126: printf("linmin end ");
1.202 brouard 2127: fprintf(ficlog,"linmin end ");
1.191 brouard 2128: #endif
1.126 brouard 2129: for (j=1;j<=n;j++) {
1.203 brouard 2130: #ifdef LINMINORIGINAL
2131: xi[j] *= xmin;
2132: #else
2133: #ifdef DEBUGLINMIN
2134: if(xxs <1.0)
2135: printf(" before xi[%d]=%12.8f", j,xi[j]);
2136: #endif
2137: 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) */
2138: #ifdef DEBUGLINMIN
2139: if(xxs <1.0)
2140: 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 );
2141: #endif
2142: #endif
1.187 brouard 2143: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2144: }
1.191 brouard 2145: #ifdef DEBUGLINMIN
1.203 brouard 2146: printf("\n");
1.191 brouard 2147: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2148: 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 2149: for (j=1;j<=n;j++) {
1.202 brouard 2150: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2151: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2152: if(j % ncovmodel == 0){
1.191 brouard 2153: printf("\n");
1.202 brouard 2154: fprintf(ficlog,"\n");
2155: }
1.191 brouard 2156: }
1.203 brouard 2157: #else
1.191 brouard 2158: #endif
1.126 brouard 2159: free_vector(xicom,1,n);
2160: free_vector(pcom,1,n);
2161: }
2162:
2163:
2164: /*************** powell ************************/
1.162 brouard 2165: /*
2166: Minimization of a function func of n variables. Input consists of an initial starting point
2167: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2168: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2169: such that failure to decrease by more than this amount on one iteration signals doneness. On
2170: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2171: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2172: */
1.224 brouard 2173: #ifdef LINMINORIGINAL
2174: #else
2175: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2176: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2177: #endif
1.126 brouard 2178: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2179: double (*func)(double []))
2180: {
1.224 brouard 2181: #ifdef LINMINORIGINAL
2182: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2183: double (*func)(double []));
1.224 brouard 2184: #else
1.241 brouard 2185: void linmin(double p[], double xi[], int n, double *fret,
2186: double (*func)(double []),int *flat);
1.224 brouard 2187: #endif
1.239 brouard 2188: int i,ibig,j,jk,k;
1.126 brouard 2189: double del,t,*pt,*ptt,*xit;
1.181 brouard 2190: double directest;
1.126 brouard 2191: double fp,fptt;
2192: double *xits;
2193: int niterf, itmp;
1.224 brouard 2194: #ifdef LINMINORIGINAL
2195: #else
2196:
2197: flatdir=ivector(1,n);
2198: for (j=1;j<=n;j++) flatdir[j]=0;
2199: #endif
1.126 brouard 2200:
2201: pt=vector(1,n);
2202: ptt=vector(1,n);
2203: xit=vector(1,n);
2204: xits=vector(1,n);
2205: *fret=(*func)(p);
2206: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2207: rcurr_time = time(NULL);
1.126 brouard 2208: for (*iter=1;;++(*iter)) {
1.187 brouard 2209: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2210: ibig=0;
2211: del=0.0;
1.157 brouard 2212: rlast_time=rcurr_time;
2213: /* (void) gettimeofday(&curr_time,&tzp); */
2214: rcurr_time = time(NULL);
2215: curr_time = *localtime(&rcurr_time);
2216: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2217: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2218: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2219: for (i=1;i<=n;i++) {
1.126 brouard 2220: fprintf(ficrespow," %.12lf", p[i]);
2221: }
1.239 brouard 2222: fprintf(ficrespow,"\n");fflush(ficrespow);
2223: printf("\n#model= 1 + age ");
2224: fprintf(ficlog,"\n#model= 1 + age ");
2225: if(nagesqr==1){
1.241 brouard 2226: printf(" + age*age ");
2227: fprintf(ficlog," + age*age ");
1.239 brouard 2228: }
2229: for(j=1;j <=ncovmodel-2;j++){
2230: if(Typevar[j]==0) {
2231: printf(" + V%d ",Tvar[j]);
2232: fprintf(ficlog," + V%d ",Tvar[j]);
2233: }else if(Typevar[j]==1) {
2234: printf(" + V%d*age ",Tvar[j]);
2235: fprintf(ficlog," + V%d*age ",Tvar[j]);
2236: }else if(Typevar[j]==2) {
2237: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2238: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2239: }
2240: }
1.126 brouard 2241: printf("\n");
1.239 brouard 2242: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2243: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2244: fprintf(ficlog,"\n");
1.239 brouard 2245: for(i=1,jk=1; i <=nlstate; i++){
2246: for(k=1; k <=(nlstate+ndeath); k++){
2247: if (k != i) {
2248: printf("%d%d ",i,k);
2249: fprintf(ficlog,"%d%d ",i,k);
2250: for(j=1; j <=ncovmodel; j++){
2251: printf("%12.7f ",p[jk]);
2252: fprintf(ficlog,"%12.7f ",p[jk]);
2253: jk++;
2254: }
2255: printf("\n");
2256: fprintf(ficlog,"\n");
2257: }
2258: }
2259: }
1.241 brouard 2260: if(*iter <=3 && *iter >1){
1.157 brouard 2261: tml = *localtime(&rcurr_time);
2262: strcpy(strcurr,asctime(&tml));
2263: rforecast_time=rcurr_time;
1.126 brouard 2264: itmp = strlen(strcurr);
2265: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2266: strcurr[itmp-1]='\0';
1.162 brouard 2267: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2268: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2269: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2270: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2271: forecast_time = *localtime(&rforecast_time);
2272: strcpy(strfor,asctime(&forecast_time));
2273: itmp = strlen(strfor);
2274: if(strfor[itmp-1]=='\n')
2275: strfor[itmp-1]='\0';
2276: 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);
2277: 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 2278: }
2279: }
1.187 brouard 2280: for (i=1;i<=n;i++) { /* For each direction i */
2281: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2282: fptt=(*fret);
2283: #ifdef DEBUG
1.203 brouard 2284: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2285: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2286: #endif
1.203 brouard 2287: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2288: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2289: #ifdef LINMINORIGINAL
1.188 brouard 2290: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2291: #else
2292: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2293: flatdir[i]=flat; /* Function is vanishing in that direction i */
2294: #endif
2295: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2296: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2297: /* because that direction will be replaced unless the gain del is small */
2298: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2299: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2300: /* with the new direction. */
2301: del=fabs(fptt-(*fret));
2302: ibig=i;
1.126 brouard 2303: }
2304: #ifdef DEBUG
2305: printf("%d %.12e",i,(*fret));
2306: fprintf(ficlog,"%d %.12e",i,(*fret));
2307: for (j=1;j<=n;j++) {
1.224 brouard 2308: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2309: printf(" x(%d)=%.12e",j,xit[j]);
2310: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2311: }
2312: for(j=1;j<=n;j++) {
1.225 brouard 2313: printf(" p(%d)=%.12e",j,p[j]);
2314: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2315: }
2316: printf("\n");
2317: fprintf(ficlog,"\n");
2318: #endif
1.187 brouard 2319: } /* end loop on each direction i */
2320: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2321: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2322: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2323: for(j=1;j<=n;j++) {
1.225 brouard 2324: if(flatdir[j] >0){
2325: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2326: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2327: }
2328: /* printf("\n"); */
2329: /* fprintf(ficlog,"\n"); */
2330: }
1.243 brouard 2331: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2332: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2333: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2334: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2335: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2336: /* decreased of more than 3.84 */
2337: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2338: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2339: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2340:
1.188 brouard 2341: /* Starting the program with initial values given by a former maximization will simply change */
2342: /* the scales of the directions and the directions, because the are reset to canonical directions */
2343: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2344: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2345: #ifdef DEBUG
2346: int k[2],l;
2347: k[0]=1;
2348: k[1]=-1;
2349: printf("Max: %.12e",(*func)(p));
2350: fprintf(ficlog,"Max: %.12e",(*func)(p));
2351: for (j=1;j<=n;j++) {
2352: printf(" %.12e",p[j]);
2353: fprintf(ficlog," %.12e",p[j]);
2354: }
2355: printf("\n");
2356: fprintf(ficlog,"\n");
2357: for(l=0;l<=1;l++) {
2358: for (j=1;j<=n;j++) {
2359: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2360: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2361: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2362: }
2363: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2364: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2365: }
2366: #endif
2367:
1.224 brouard 2368: #ifdef LINMINORIGINAL
2369: #else
2370: free_ivector(flatdir,1,n);
2371: #endif
1.126 brouard 2372: free_vector(xit,1,n);
2373: free_vector(xits,1,n);
2374: free_vector(ptt,1,n);
2375: free_vector(pt,1,n);
2376: return;
1.192 brouard 2377: } /* enough precision */
1.240 brouard 2378: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2379: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2380: ptt[j]=2.0*p[j]-pt[j];
2381: xit[j]=p[j]-pt[j];
2382: pt[j]=p[j];
2383: }
1.181 brouard 2384: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2385: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2386: if (*iter <=4) {
1.225 brouard 2387: #else
2388: #endif
1.224 brouard 2389: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2390: #else
1.161 brouard 2391: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2392: #endif
1.162 brouard 2393: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2394: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2395: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2396: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2397: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2398: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2399: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2400: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2401: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2402: /* Even if f3 <f1, directest can be negative and t >0 */
2403: /* mu² and del² are equal when f3=f1 */
2404: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2405: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2406: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2407: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2408: #ifdef NRCORIGINAL
2409: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2410: #else
2411: 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 2412: t= t- del*SQR(fp-fptt);
1.183 brouard 2413: #endif
1.202 brouard 2414: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2415: #ifdef DEBUG
1.181 brouard 2416: 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);
2417: 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 2418: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2419: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2420: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2421: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2422: 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);
2423: 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);
2424: #endif
1.183 brouard 2425: #ifdef POWELLORIGINAL
2426: if (t < 0.0) { /* Then we use it for new direction */
2427: #else
1.182 brouard 2428: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2429: 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 2430: 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 2431: 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 2432: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2433: }
1.181 brouard 2434: if (directest < 0.0) { /* Then we use it for new direction */
2435: #endif
1.191 brouard 2436: #ifdef DEBUGLINMIN
1.234 brouard 2437: printf("Before linmin in direction P%d-P0\n",n);
2438: for (j=1;j<=n;j++) {
2439: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2440: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2441: if(j % ncovmodel == 0){
2442: printf("\n");
2443: fprintf(ficlog,"\n");
2444: }
2445: }
1.224 brouard 2446: #endif
2447: #ifdef LINMINORIGINAL
1.234 brouard 2448: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2449: #else
1.234 brouard 2450: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2451: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2452: #endif
1.234 brouard 2453:
1.191 brouard 2454: #ifdef DEBUGLINMIN
1.234 brouard 2455: for (j=1;j<=n;j++) {
2456: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2457: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2458: if(j % ncovmodel == 0){
2459: printf("\n");
2460: fprintf(ficlog,"\n");
2461: }
2462: }
1.224 brouard 2463: #endif
1.234 brouard 2464: for (j=1;j<=n;j++) {
2465: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2466: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2467: }
1.224 brouard 2468: #ifdef LINMINORIGINAL
2469: #else
1.234 brouard 2470: for (j=1, flatd=0;j<=n;j++) {
2471: if(flatdir[j]>0)
2472: flatd++;
2473: }
2474: if(flatd >0){
1.255 brouard 2475: printf("%d flat directions: ",flatd);
2476: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2477: for (j=1;j<=n;j++) {
2478: if(flatdir[j]>0){
2479: printf("%d ",j);
2480: fprintf(ficlog,"%d ",j);
2481: }
2482: }
2483: printf("\n");
2484: fprintf(ficlog,"\n");
2485: }
1.191 brouard 2486: #endif
1.234 brouard 2487: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2488: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2489:
1.126 brouard 2490: #ifdef DEBUG
1.234 brouard 2491: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2492: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2493: for(j=1;j<=n;j++){
2494: printf(" %lf",xit[j]);
2495: fprintf(ficlog," %lf",xit[j]);
2496: }
2497: printf("\n");
2498: fprintf(ficlog,"\n");
1.126 brouard 2499: #endif
1.192 brouard 2500: } /* end of t or directest negative */
1.224 brouard 2501: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2502: #else
1.234 brouard 2503: } /* end if (fptt < fp) */
1.192 brouard 2504: #endif
1.225 brouard 2505: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2506: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2507: #else
1.224 brouard 2508: #endif
1.234 brouard 2509: } /* loop iteration */
1.126 brouard 2510: }
1.234 brouard 2511:
1.126 brouard 2512: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2513:
1.235 brouard 2514: 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 2515: {
1.235 brouard 2516: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2517: (and selected quantitative values in nres)
2518: by left multiplying the unit
1.234 brouard 2519: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2520: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2521: /* Wx is row vector: population in state 1, population in state 2, population dead */
2522: /* or prevalence in state 1, prevalence in state 2, 0 */
2523: /* newm is the matrix after multiplications, its rows are identical at a factor */
2524: /* Initial matrix pimij */
2525: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2526: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2527: /* 0, 0 , 1} */
2528: /*
2529: * and after some iteration: */
2530: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2531: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2532: /* 0, 0 , 1} */
2533: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2534: /* {0.51571254859325999, 0.4842874514067399, */
2535: /* 0.51326036147820708, 0.48673963852179264} */
2536: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2537:
1.126 brouard 2538: int i, ii,j,k;
1.209 brouard 2539: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2540: /* double **matprod2(); */ /* test */
1.218 brouard 2541: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2542: double **newm;
1.209 brouard 2543: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2544: int ncvloop=0;
1.169 brouard 2545:
1.209 brouard 2546: min=vector(1,nlstate);
2547: max=vector(1,nlstate);
2548: meandiff=vector(1,nlstate);
2549:
1.218 brouard 2550: /* Starting with matrix unity */
1.126 brouard 2551: for (ii=1;ii<=nlstate+ndeath;ii++)
2552: for (j=1;j<=nlstate+ndeath;j++){
2553: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2554: }
1.169 brouard 2555:
2556: cov[1]=1.;
2557:
2558: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2559: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2560: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2561: ncvloop++;
1.126 brouard 2562: newm=savm;
2563: /* Covariates have to be included here again */
1.138 brouard 2564: cov[2]=agefin;
1.187 brouard 2565: if(nagesqr==1)
2566: cov[3]= agefin*agefin;;
1.234 brouard 2567: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2568: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2569: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2570: /* 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 2571: }
2572: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2573: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2574: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2575: /* 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 2576: }
1.237 brouard 2577: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2578: if(Dummy[Tvar[Tage[k]]]){
2579: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2580: } else{
1.235 brouard 2581: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2582: }
1.235 brouard 2583: /* 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 2584: }
1.237 brouard 2585: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2586: /* 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 2587: if(Dummy[Tvard[k][1]==0]){
2588: if(Dummy[Tvard[k][2]==0]){
2589: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2590: }else{
2591: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2592: }
2593: }else{
2594: if(Dummy[Tvard[k][2]==0]){
2595: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2596: }else{
2597: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2598: }
2599: }
1.234 brouard 2600: }
1.138 brouard 2601: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2602: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2603: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2604: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2605: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2606: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2607: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2608:
1.126 brouard 2609: savm=oldm;
2610: oldm=newm;
1.209 brouard 2611:
2612: for(j=1; j<=nlstate; j++){
2613: max[j]=0.;
2614: min[j]=1.;
2615: }
2616: for(i=1;i<=nlstate;i++){
2617: sumnew=0;
2618: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2619: for(j=1; j<=nlstate; j++){
2620: prlim[i][j]= newm[i][j]/(1-sumnew);
2621: max[j]=FMAX(max[j],prlim[i][j]);
2622: min[j]=FMIN(min[j],prlim[i][j]);
2623: }
2624: }
2625:
1.126 brouard 2626: maxmax=0.;
1.209 brouard 2627: for(j=1; j<=nlstate; j++){
2628: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2629: maxmax=FMAX(maxmax,meandiff[j]);
2630: /* 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 2631: } /* j loop */
1.203 brouard 2632: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2633: /* 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 2634: if(maxmax < ftolpl){
1.209 brouard 2635: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2636: free_vector(min,1,nlstate);
2637: free_vector(max,1,nlstate);
2638: free_vector(meandiff,1,nlstate);
1.126 brouard 2639: return prlim;
2640: }
1.169 brouard 2641: } /* age loop */
1.208 brouard 2642: /* After some age loop it doesn't converge */
1.209 brouard 2643: 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 2644: 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 2645: /* 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); */
2646: free_vector(min,1,nlstate);
2647: free_vector(max,1,nlstate);
2648: free_vector(meandiff,1,nlstate);
1.208 brouard 2649:
1.169 brouard 2650: return prlim; /* should not reach here */
1.126 brouard 2651: }
2652:
1.217 brouard 2653:
2654: /**** Back Prevalence limit (stable or period prevalence) ****************/
2655:
1.218 brouard 2656: /* 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) */
2657: /* 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 2658: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2659: {
1.264 brouard 2660: /* 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 2661: matrix by transitions matrix until convergence is reached with precision ftolpl */
2662: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2663: /* Wx is row vector: population in state 1, population in state 2, population dead */
2664: /* or prevalence in state 1, prevalence in state 2, 0 */
2665: /* newm is the matrix after multiplications, its rows are identical at a factor */
2666: /* Initial matrix pimij */
2667: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2668: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2669: /* 0, 0 , 1} */
2670: /*
2671: * and after some iteration: */
2672: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2673: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2674: /* 0, 0 , 1} */
2675: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2676: /* {0.51571254859325999, 0.4842874514067399, */
2677: /* 0.51326036147820708, 0.48673963852179264} */
2678: /* If we start from prlim again, prlim tends to a constant matrix */
2679:
2680: int i, ii,j,k;
1.247 brouard 2681: int first=0;
1.217 brouard 2682: double *min, *max, *meandiff, maxmax,sumnew=0.;
2683: /* double **matprod2(); */ /* test */
2684: double **out, cov[NCOVMAX+1], **bmij();
2685: double **newm;
1.218 brouard 2686: double **dnewm, **doldm, **dsavm; /* for use */
2687: double **oldm, **savm; /* for use */
2688:
1.217 brouard 2689: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2690: int ncvloop=0;
2691:
2692: min=vector(1,nlstate);
2693: max=vector(1,nlstate);
2694: meandiff=vector(1,nlstate);
2695:
1.266 brouard 2696: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2697: oldm=oldms; savm=savms;
2698:
2699: /* Starting with matrix unity */
2700: for (ii=1;ii<=nlstate+ndeath;ii++)
2701: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2702: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2703: }
2704:
2705: cov[1]=1.;
2706:
2707: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2708: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2709: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2710: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2711: ncvloop++;
1.218 brouard 2712: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2713: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2714: /* Covariates have to be included here again */
2715: cov[2]=agefin;
2716: if(nagesqr==1)
2717: cov[3]= agefin*agefin;;
1.242 brouard 2718: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2719: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2720: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2721: /* 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 2722: }
2723: /* for (k=1; k<=cptcovn;k++) { */
2724: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2725: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2726: /* /\* 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])]); *\/ */
2727: /* } */
2728: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2729: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2730: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2731: /* 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]); */
2732: }
2733: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2734: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2735: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2736: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2737: for (k=1; k<=cptcovage;k++){ /* For product with age */
2738: if(Dummy[Tvar[Tage[k]]]){
2739: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2740: } else{
2741: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2742: }
2743: /* 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]); */
2744: }
2745: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2746: /* 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]); */
2747: if(Dummy[Tvard[k][1]==0]){
2748: if(Dummy[Tvard[k][2]==0]){
2749: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2750: }else{
2751: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2752: }
2753: }else{
2754: if(Dummy[Tvard[k][2]==0]){
2755: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2756: }else{
2757: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2758: }
2759: }
1.217 brouard 2760: }
2761:
2762: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2763: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2764: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2765: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2766: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2767: /* ij should be linked to the correct index of cov */
2768: /* age and covariate values ij are in 'cov', but we need to pass
2769: * ij for the observed prevalence at age and status and covariate
2770: * number: prevacurrent[(int)agefin][ii][ij]
2771: */
2772: /* 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 *\/ */
2773: /* 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 *\/ */
2774: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.268 brouard 2775: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2776: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2777: /* for(i=1; i<=nlstate+ndeath; i++) { */
2778: /* printf("%d newm= ",i); */
2779: /* for(j=1;j<=nlstate+ndeath;j++) { */
2780: /* printf("%f ",newm[i][j]); */
2781: /* } */
2782: /* printf("oldm * "); */
2783: /* for(j=1;j<=nlstate+ndeath;j++) { */
2784: /* printf("%f ",oldm[i][j]); */
2785: /* } */
1.268 brouard 2786: /* printf(" bmmij "); */
1.266 brouard 2787: /* for(j=1;j<=nlstate+ndeath;j++) { */
2788: /* printf("%f ",pmmij[i][j]); */
2789: /* } */
2790: /* printf("\n"); */
2791: /* } */
2792: /* } */
1.217 brouard 2793: savm=oldm;
2794: oldm=newm;
1.266 brouard 2795:
1.217 brouard 2796: for(j=1; j<=nlstate; j++){
2797: max[j]=0.;
2798: min[j]=1.;
2799: }
2800: for(j=1; j<=nlstate; j++){
2801: for(i=1;i<=nlstate;i++){
1.234 brouard 2802: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2803: bprlim[i][j]= newm[i][j];
2804: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2805: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2806: }
2807: }
1.218 brouard 2808:
1.217 brouard 2809: maxmax=0.;
2810: for(i=1; i<=nlstate; i++){
2811: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2812: maxmax=FMAX(maxmax,meandiff[i]);
2813: /* printf("Back age= %d meandiff[%d]=%f, agefin=%d max[%d]=%f min[%d]=%f maxmax=%f\n", (int)age, i, meandiff[i],(int)agefin, i, max[i], i, min[i],maxmax); */
1.268 brouard 2814: } /* i loop */
1.217 brouard 2815: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2816: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2817: if(maxmax < ftolpl){
1.220 brouard 2818: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2819: free_vector(min,1,nlstate);
2820: free_vector(max,1,nlstate);
2821: free_vector(meandiff,1,nlstate);
2822: return bprlim;
2823: }
2824: } /* age loop */
2825: /* After some age loop it doesn't converge */
1.247 brouard 2826: if(first){
2827: first=1;
2828: 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\
2829: 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);
2830: }
2831: 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 2832: 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);
2833: /* 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); */
2834: free_vector(min,1,nlstate);
2835: free_vector(max,1,nlstate);
2836: free_vector(meandiff,1,nlstate);
2837:
2838: return bprlim; /* should not reach here */
2839: }
2840:
1.126 brouard 2841: /*************** transition probabilities ***************/
2842:
2843: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2844: {
1.138 brouard 2845: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2846: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2847: model to the ncovmodel covariates (including constant and age).
2848: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2849: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2850: ncth covariate in the global vector x is given by the formula:
2851: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2852: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2853: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2854: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2855: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2856: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2857: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2858: */
2859: double s1, lnpijopii;
1.126 brouard 2860: /*double t34;*/
1.164 brouard 2861: int i,j, nc, ii, jj;
1.126 brouard 2862:
1.223 brouard 2863: for(i=1; i<= nlstate; i++){
2864: for(j=1; j<i;j++){
2865: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2866: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2867: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2868: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2869: }
2870: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2871: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2872: }
2873: for(j=i+1; j<=nlstate+ndeath;j++){
2874: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2875: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2876: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2877: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2878: }
2879: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2880: }
2881: }
1.218 brouard 2882:
1.223 brouard 2883: for(i=1; i<= nlstate; i++){
2884: s1=0;
2885: for(j=1; j<i; j++){
2886: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2887: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2888: }
2889: for(j=i+1; j<=nlstate+ndeath; j++){
2890: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2891: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2892: }
2893: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2894: ps[i][i]=1./(s1+1.);
2895: /* Computing other pijs */
2896: for(j=1; j<i; j++)
2897: ps[i][j]= exp(ps[i][j])*ps[i][i];
2898: for(j=i+1; j<=nlstate+ndeath; j++)
2899: ps[i][j]= exp(ps[i][j])*ps[i][i];
2900: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2901: } /* end i */
1.218 brouard 2902:
1.223 brouard 2903: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2904: for(jj=1; jj<= nlstate+ndeath; jj++){
2905: ps[ii][jj]=0;
2906: ps[ii][ii]=1;
2907: }
2908: }
1.218 brouard 2909:
2910:
1.223 brouard 2911: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2912: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2913: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2914: /* } */
2915: /* printf("\n "); */
2916: /* } */
2917: /* printf("\n ");printf("%lf ",cov[2]);*/
2918: /*
2919: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2920: goto end;*/
1.266 brouard 2921: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2922: }
2923:
1.218 brouard 2924: /*************** backward transition probabilities ***************/
2925:
2926: /* 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 ) */
2927: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2928: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2929: {
1.266 brouard 2930: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2931: * 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 2932: */
1.218 brouard 2933: int i, ii, j,k;
1.222 brouard 2934:
2935: double **out, **pmij();
2936: double sumnew=0.;
1.218 brouard 2937: double agefin;
1.268 brouard 2938: double k3=0.; /* constant of the w_x diagonal matrixe (in order for B to sum to 1 even for death state) */
1.222 brouard 2939: double **dnewm, **dsavm, **doldm;
2940: double **bbmij;
2941:
1.218 brouard 2942: doldm=ddoldms; /* global pointers */
1.222 brouard 2943: dnewm=ddnewms;
2944: dsavm=ddsavms;
2945:
2946: agefin=cov[2];
1.268 brouard 2947: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2948: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2949: the observed prevalence (with this covariate ij) at beginning of transition */
2950: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2951:
2952: /* P_x */
1.266 brouard 2953: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2954: /* outputs pmmij which is a stochastic matrix in row */
2955:
2956: /* Diag(w_x) */
2957: /* Problem with prevacurrent which can be zero */
2958: sumnew=0.;
1.269 brouard 2959: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2960: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2961: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2962: sumnew+=prevacurrent[(int)agefin][ii][ij];
2963: }
2964: if(sumnew >0.01){ /* At least some value in the prevalence */
2965: for (ii=1;ii<=nlstate+ndeath;ii++){
2966: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2967: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2968: }
2969: }else{
2970: for (ii=1;ii<=nlstate+ndeath;ii++){
2971: for (j=1;j<=nlstate+ndeath;j++)
2972: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
2973: }
2974: /* if(sumnew <0.9){ */
2975: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2976: /* } */
2977: }
2978: k3=0.0; /* We put the last diagonal to 0 */
2979: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
2980: doldm[ii][ii]= k3;
2981: }
2982: /* End doldm, At the end doldm is diag[(w_i)] */
2983:
2984: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
2985: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
2986:
2987: /* Diag(Sum_i w^i_x p^ij_x */
2988: /* 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 2989: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 2990: sumnew=0.;
1.222 brouard 2991: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 2992: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 2993: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 2994: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 2995: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 2996: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 2997: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2998: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 2999: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3000: /* }else */
1.268 brouard 3001: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3002: } /*End ii */
3003: } /* End j, At the end dsavm is diag[1/(w_1p1i+w_2 p2i)] for ALL states even if the sum is only for live states */
3004:
3005: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
3006: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3007: /* end bmij */
1.266 brouard 3008: return ps; /*pointer is unchanged */
1.218 brouard 3009: }
1.217 brouard 3010: /*************** transition probabilities ***************/
3011:
1.218 brouard 3012: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3013: {
3014: /* According to parameters values stored in x and the covariate's values stored in cov,
3015: computes the probability to be observed in state j being in state i by appying the
3016: model to the ncovmodel covariates (including constant and age).
3017: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3018: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3019: ncth covariate in the global vector x is given by the formula:
3020: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3021: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3022: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3023: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3024: Outputs ps[i][j] the probability to be observed in j being in j according to
3025: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3026: */
3027: double s1, lnpijopii;
3028: /*double t34;*/
3029: int i,j, nc, ii, jj;
3030:
1.234 brouard 3031: for(i=1; i<= nlstate; i++){
3032: for(j=1; j<i;j++){
3033: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3034: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3035: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3036: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3037: }
3038: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3039: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3040: }
3041: for(j=i+1; j<=nlstate+ndeath;j++){
3042: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3043: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3044: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3045: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3046: }
3047: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3048: }
3049: }
3050:
3051: for(i=1; i<= nlstate; i++){
3052: s1=0;
3053: for(j=1; j<i; j++){
3054: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3055: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3056: }
3057: for(j=i+1; j<=nlstate+ndeath; j++){
3058: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3059: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3060: }
3061: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3062: ps[i][i]=1./(s1+1.);
3063: /* Computing other pijs */
3064: for(j=1; j<i; j++)
3065: ps[i][j]= exp(ps[i][j])*ps[i][i];
3066: for(j=i+1; j<=nlstate+ndeath; j++)
3067: ps[i][j]= exp(ps[i][j])*ps[i][i];
3068: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3069: } /* end i */
3070:
3071: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3072: for(jj=1; jj<= nlstate+ndeath; jj++){
3073: ps[ii][jj]=0;
3074: ps[ii][ii]=1;
3075: }
3076: }
3077: /* Added for backcast */ /* Transposed matrix too */
3078: for(jj=1; jj<= nlstate+ndeath; jj++){
3079: s1=0.;
3080: for(ii=1; ii<= nlstate+ndeath; ii++){
3081: s1+=ps[ii][jj];
3082: }
3083: for(ii=1; ii<= nlstate; ii++){
3084: ps[ii][jj]=ps[ii][jj]/s1;
3085: }
3086: }
3087: /* Transposition */
3088: for(jj=1; jj<= nlstate+ndeath; jj++){
3089: for(ii=jj; ii<= nlstate+ndeath; ii++){
3090: s1=ps[ii][jj];
3091: ps[ii][jj]=ps[jj][ii];
3092: ps[jj][ii]=s1;
3093: }
3094: }
3095: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3096: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3097: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3098: /* } */
3099: /* printf("\n "); */
3100: /* } */
3101: /* printf("\n ");printf("%lf ",cov[2]);*/
3102: /*
3103: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3104: goto end;*/
3105: return ps;
1.217 brouard 3106: }
3107:
3108:
1.126 brouard 3109: /**************** Product of 2 matrices ******************/
3110:
1.145 brouard 3111: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3112: {
3113: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3114: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3115: /* in, b, out are matrice of pointers which should have been initialized
3116: before: only the contents of out is modified. The function returns
3117: a pointer to pointers identical to out */
1.145 brouard 3118: int i, j, k;
1.126 brouard 3119: for(i=nrl; i<= nrh; i++)
1.145 brouard 3120: for(k=ncolol; k<=ncoloh; k++){
3121: out[i][k]=0.;
3122: for(j=ncl; j<=nch; j++)
3123: out[i][k] +=in[i][j]*b[j][k];
3124: }
1.126 brouard 3125: return out;
3126: }
3127:
3128:
3129: /************* Higher Matrix Product ***************/
3130:
1.235 brouard 3131: 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 3132: {
1.218 brouard 3133: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3134: 'nhstepm*hstepm*stepm' months (i.e. until
3135: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3136: nhstepm*hstepm matrices.
3137: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3138: (typically every 2 years instead of every month which is too big
3139: for the memory).
3140: Model is determined by parameters x and covariates have to be
3141: included manually here.
3142:
3143: */
3144:
3145: int i, j, d, h, k;
1.131 brouard 3146: double **out, cov[NCOVMAX+1];
1.126 brouard 3147: double **newm;
1.187 brouard 3148: double agexact;
1.214 brouard 3149: double agebegin, ageend;
1.126 brouard 3150:
3151: /* Hstepm could be zero and should return the unit matrix */
3152: for (i=1;i<=nlstate+ndeath;i++)
3153: for (j=1;j<=nlstate+ndeath;j++){
3154: oldm[i][j]=(i==j ? 1.0 : 0.0);
3155: po[i][j][0]=(i==j ? 1.0 : 0.0);
3156: }
3157: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3158: for(h=1; h <=nhstepm; h++){
3159: for(d=1; d <=hstepm; d++){
3160: newm=savm;
3161: /* Covariates have to be included here again */
3162: cov[1]=1.;
1.214 brouard 3163: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3164: cov[2]=agexact;
3165: if(nagesqr==1)
1.227 brouard 3166: cov[3]= agexact*agexact;
1.235 brouard 3167: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3168: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3169: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3170: /* 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)); */
3171: }
3172: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3173: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3174: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3175: /* 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]); */
3176: }
3177: for (k=1; k<=cptcovage;k++){
3178: if(Dummy[Tvar[Tage[k]]]){
3179: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3180: } else{
3181: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3182: }
3183: /* 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]); */
3184: }
3185: for (k=1; k<=cptcovprod;k++){ /* */
3186: /* 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]); */
3187: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3188: }
3189: /* for (k=1; k<=cptcovn;k++) */
3190: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3191: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3192: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3193: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3194: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3195:
3196:
1.126 brouard 3197: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3198: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3199: /* right multiplication of oldm by the current matrix */
1.126 brouard 3200: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3201: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3202: /* if((int)age == 70){ */
3203: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3204: /* for(i=1; i<=nlstate+ndeath; i++) { */
3205: /* printf("%d pmmij ",i); */
3206: /* for(j=1;j<=nlstate+ndeath;j++) { */
3207: /* printf("%f ",pmmij[i][j]); */
3208: /* } */
3209: /* printf(" oldm "); */
3210: /* for(j=1;j<=nlstate+ndeath;j++) { */
3211: /* printf("%f ",oldm[i][j]); */
3212: /* } */
3213: /* printf("\n"); */
3214: /* } */
3215: /* } */
1.126 brouard 3216: savm=oldm;
3217: oldm=newm;
3218: }
3219: for(i=1; i<=nlstate+ndeath; i++)
3220: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3221: po[i][j][h]=newm[i][j];
3222: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3223: }
1.128 brouard 3224: /*printf("h=%d ",h);*/
1.126 brouard 3225: } /* end h */
1.267 brouard 3226: /* printf("\n H=%d \n",h); */
1.126 brouard 3227: return po;
3228: }
3229:
1.217 brouard 3230: /************* Higher Back Matrix Product ***************/
1.218 brouard 3231: /* double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, int ij ) */
1.267 brouard 3232: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij, int nres )
1.217 brouard 3233: {
1.266 brouard 3234: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3235: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3236: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3237: nhstepm*hstepm matrices.
3238: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3239: (typically every 2 years instead of every month which is too big
1.217 brouard 3240: for the memory).
1.218 brouard 3241: Model is determined by parameters x and covariates have to be
1.266 brouard 3242: included manually here. Then we use a call to bmij(x and cov)
3243: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3244: */
1.217 brouard 3245:
3246: int i, j, d, h, k;
1.266 brouard 3247: double **out, cov[NCOVMAX+1], **bmij();
3248: double **newm, ***newmm;
1.217 brouard 3249: double agexact;
3250: double agebegin, ageend;
1.222 brouard 3251: double **oldm, **savm;
1.217 brouard 3252:
1.266 brouard 3253: newmm=po; /* To be saved */
3254: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3255: /* Hstepm could be zero and should return the unit matrix */
3256: for (i=1;i<=nlstate+ndeath;i++)
3257: for (j=1;j<=nlstate+ndeath;j++){
3258: oldm[i][j]=(i==j ? 1.0 : 0.0);
3259: po[i][j][0]=(i==j ? 1.0 : 0.0);
3260: }
3261: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3262: for(h=1; h <=nhstepm; h++){
3263: for(d=1; d <=hstepm; d++){
3264: newm=savm;
3265: /* Covariates have to be included here again */
3266: cov[1]=1.;
1.271 brouard 3267: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3268: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3269: cov[2]=agexact;
3270: if(nagesqr==1)
1.222 brouard 3271: cov[3]= agexact*agexact;
1.266 brouard 3272: for (k=1; k<=cptcovn;k++){
3273: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3274: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3275: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3276: /* 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)); */
3277: }
1.267 brouard 3278: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3279: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3280: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3281: /* 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]); */
3282: }
3283: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3284: if(Dummy[Tvar[Tage[k]]]){
3285: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3286: } else{
3287: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3288: }
3289: /* printf("hBxij Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); */
3290: }
3291: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3292: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3293: }
1.217 brouard 3294: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3295: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3296:
1.218 brouard 3297: /* Careful transposed matrix */
1.266 brouard 3298: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3299: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3300: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3301: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3302: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3303: /* if((int)age == 70){ */
3304: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3305: /* for(i=1; i<=nlstate+ndeath; i++) { */
3306: /* printf("%d pmmij ",i); */
3307: /* for(j=1;j<=nlstate+ndeath;j++) { */
3308: /* printf("%f ",pmmij[i][j]); */
3309: /* } */
3310: /* printf(" oldm "); */
3311: /* for(j=1;j<=nlstate+ndeath;j++) { */
3312: /* printf("%f ",oldm[i][j]); */
3313: /* } */
3314: /* printf("\n"); */
3315: /* } */
3316: /* } */
3317: savm=oldm;
3318: oldm=newm;
3319: }
3320: for(i=1; i<=nlstate+ndeath; i++)
3321: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3322: po[i][j][h]=newm[i][j];
1.268 brouard 3323: /* if(h==nhstepm) */
3324: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3325: }
1.268 brouard 3326: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3327: } /* end h */
1.268 brouard 3328: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3329: return po;
3330: }
3331:
3332:
1.162 brouard 3333: #ifdef NLOPT
3334: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3335: double fret;
3336: double *xt;
3337: int j;
3338: myfunc_data *d2 = (myfunc_data *) pd;
3339: /* xt = (p1-1); */
3340: xt=vector(1,n);
3341: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3342:
3343: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3344: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3345: printf("Function = %.12lf ",fret);
3346: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3347: printf("\n");
3348: free_vector(xt,1,n);
3349: return fret;
3350: }
3351: #endif
1.126 brouard 3352:
3353: /*************** log-likelihood *************/
3354: double func( double *x)
3355: {
1.226 brouard 3356: int i, ii, j, k, mi, d, kk;
3357: int ioffset=0;
3358: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3359: double **out;
3360: double lli; /* Individual log likelihood */
3361: int s1, s2;
1.228 brouard 3362: 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 3363: double bbh, survp;
3364: long ipmx;
3365: double agexact;
3366: /*extern weight */
3367: /* We are differentiating ll according to initial status */
3368: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3369: /*for(i=1;i<imx;i++)
3370: printf(" %d\n",s[4][i]);
3371: */
1.162 brouard 3372:
1.226 brouard 3373: ++countcallfunc;
1.162 brouard 3374:
1.226 brouard 3375: cov[1]=1.;
1.126 brouard 3376:
1.226 brouard 3377: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3378: ioffset=0;
1.226 brouard 3379: if(mle==1){
3380: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3381: /* Computes the values of the ncovmodel covariates of the model
3382: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3383: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3384: to be observed in j being in i according to the model.
3385: */
1.243 brouard 3386: ioffset=2+nagesqr ;
1.233 brouard 3387: /* Fixed */
1.234 brouard 3388: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3389: 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)*/
3390: }
1.226 brouard 3391: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3392: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3393: has been calculated etc */
3394: /* For an individual i, wav[i] gives the number of effective waves */
3395: /* We compute the contribution to Likelihood of each effective transition
3396: mw[mi][i] is real wave of the mi th effectve wave */
3397: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3398: s2=s[mw[mi+1][i]][i];
3399: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3400: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3401: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3402: */
3403: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3404: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3405: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3406: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3407: }
3408: for (ii=1;ii<=nlstate+ndeath;ii++)
3409: for (j=1;j<=nlstate+ndeath;j++){
3410: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3411: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3412: }
3413: for(d=0; d<dh[mi][i]; d++){
3414: newm=savm;
3415: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3416: cov[2]=agexact;
3417: if(nagesqr==1)
3418: cov[3]= agexact*agexact; /* Should be changed here */
3419: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3420: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3421: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3422: else
3423: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3424: }
3425: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3426: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3427: savm=oldm;
3428: oldm=newm;
3429: } /* end mult */
3430:
3431: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3432: /* But now since version 0.9 we anticipate for bias at large stepm.
3433: * If stepm is larger than one month (smallest stepm) and if the exact delay
3434: * (in months) between two waves is not a multiple of stepm, we rounded to
3435: * the nearest (and in case of equal distance, to the lowest) interval but now
3436: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3437: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3438: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3439: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3440: * -stepm/2 to stepm/2 .
3441: * For stepm=1 the results are the same as for previous versions of Imach.
3442: * For stepm > 1 the results are less biased than in previous versions.
3443: */
1.234 brouard 3444: s1=s[mw[mi][i]][i];
3445: s2=s[mw[mi+1][i]][i];
3446: bbh=(double)bh[mi][i]/(double)stepm;
3447: /* bias bh is positive if real duration
3448: * is higher than the multiple of stepm and negative otherwise.
3449: */
3450: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3451: if( s2 > nlstate){
3452: /* i.e. if s2 is a death state and if the date of death is known
3453: then the contribution to the likelihood is the probability to
3454: die between last step unit time and current step unit time,
3455: which is also equal to probability to die before dh
3456: minus probability to die before dh-stepm .
3457: In version up to 0.92 likelihood was computed
3458: as if date of death was unknown. Death was treated as any other
3459: health state: the date of the interview describes the actual state
3460: and not the date of a change in health state. The former idea was
3461: to consider that at each interview the state was recorded
3462: (healthy, disable or death) and IMaCh was corrected; but when we
3463: introduced the exact date of death then we should have modified
3464: the contribution of an exact death to the likelihood. This new
3465: contribution is smaller and very dependent of the step unit
3466: stepm. It is no more the probability to die between last interview
3467: and month of death but the probability to survive from last
3468: interview up to one month before death multiplied by the
3469: probability to die within a month. Thanks to Chris
3470: Jackson for correcting this bug. Former versions increased
3471: mortality artificially. The bad side is that we add another loop
3472: which slows down the processing. The difference can be up to 10%
3473: lower mortality.
3474: */
3475: /* If, at the beginning of the maximization mostly, the
3476: cumulative probability or probability to be dead is
3477: constant (ie = 1) over time d, the difference is equal to
3478: 0. out[s1][3] = savm[s1][3]: probability, being at state
3479: s1 at precedent wave, to be dead a month before current
3480: wave is equal to probability, being at state s1 at
3481: precedent wave, to be dead at mont of the current
3482: wave. Then the observed probability (that this person died)
3483: is null according to current estimated parameter. In fact,
3484: it should be very low but not zero otherwise the log go to
3485: infinity.
3486: */
1.183 brouard 3487: /* #ifdef INFINITYORIGINAL */
3488: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3489: /* #else */
3490: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3491: /* lli=log(mytinydouble); */
3492: /* else */
3493: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3494: /* #endif */
1.226 brouard 3495: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3496:
1.226 brouard 3497: } else if ( s2==-1 ) { /* alive */
3498: for (j=1,survp=0. ; j<=nlstate; j++)
3499: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3500: /*survp += out[s1][j]; */
3501: lli= log(survp);
3502: }
3503: else if (s2==-4) {
3504: for (j=3,survp=0. ; j<=nlstate; j++)
3505: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3506: lli= log(survp);
3507: }
3508: else if (s2==-5) {
3509: for (j=1,survp=0. ; j<=2; j++)
3510: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3511: lli= log(survp);
3512: }
3513: else{
3514: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3515: /* 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 */
3516: }
3517: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3518: /*if(lli ==000.0)*/
3519: /*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); */
3520: ipmx +=1;
3521: sw += weight[i];
3522: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3523: /* if (lli < log(mytinydouble)){ */
3524: /* 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); */
3525: /* 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]); */
3526: /* } */
3527: } /* end of wave */
3528: } /* end of individual */
3529: } else if(mle==2){
3530: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3531: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3532: for(mi=1; mi<= wav[i]-1; mi++){
3533: for (ii=1;ii<=nlstate+ndeath;ii++)
3534: for (j=1;j<=nlstate+ndeath;j++){
3535: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3536: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3537: }
3538: for(d=0; d<=dh[mi][i]; d++){
3539: newm=savm;
3540: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3541: cov[2]=agexact;
3542: if(nagesqr==1)
3543: cov[3]= agexact*agexact;
3544: for (kk=1; kk<=cptcovage;kk++) {
3545: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3546: }
3547: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3548: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3549: savm=oldm;
3550: oldm=newm;
3551: } /* end mult */
3552:
3553: s1=s[mw[mi][i]][i];
3554: s2=s[mw[mi+1][i]][i];
3555: bbh=(double)bh[mi][i]/(double)stepm;
3556: 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 */
3557: ipmx +=1;
3558: sw += weight[i];
3559: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3560: } /* end of wave */
3561: } /* end of individual */
3562: } else if(mle==3){ /* exponential inter-extrapolation */
3563: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3564: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3565: for(mi=1; mi<= wav[i]-1; mi++){
3566: for (ii=1;ii<=nlstate+ndeath;ii++)
3567: for (j=1;j<=nlstate+ndeath;j++){
3568: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3569: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3570: }
3571: for(d=0; d<dh[mi][i]; d++){
3572: newm=savm;
3573: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3574: cov[2]=agexact;
3575: if(nagesqr==1)
3576: cov[3]= agexact*agexact;
3577: for (kk=1; kk<=cptcovage;kk++) {
3578: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3579: }
3580: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3581: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3582: savm=oldm;
3583: oldm=newm;
3584: } /* end mult */
3585:
3586: s1=s[mw[mi][i]][i];
3587: s2=s[mw[mi+1][i]][i];
3588: bbh=(double)bh[mi][i]/(double)stepm;
3589: 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 */
3590: ipmx +=1;
3591: sw += weight[i];
3592: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3593: } /* end of wave */
3594: } /* end of individual */
3595: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3596: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3597: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3598: for(mi=1; mi<= wav[i]-1; mi++){
3599: for (ii=1;ii<=nlstate+ndeath;ii++)
3600: for (j=1;j<=nlstate+ndeath;j++){
3601: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3602: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3603: }
3604: for(d=0; d<dh[mi][i]; d++){
3605: newm=savm;
3606: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3607: cov[2]=agexact;
3608: if(nagesqr==1)
3609: cov[3]= agexact*agexact;
3610: for (kk=1; kk<=cptcovage;kk++) {
3611: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3612: }
1.126 brouard 3613:
1.226 brouard 3614: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3615: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3616: savm=oldm;
3617: oldm=newm;
3618: } /* end mult */
3619:
3620: s1=s[mw[mi][i]][i];
3621: s2=s[mw[mi+1][i]][i];
3622: if( s2 > nlstate){
3623: lli=log(out[s1][s2] - savm[s1][s2]);
3624: } else if ( s2==-1 ) { /* alive */
3625: for (j=1,survp=0. ; j<=nlstate; j++)
3626: survp += out[s1][j];
3627: lli= log(survp);
3628: }else{
3629: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3630: }
3631: ipmx +=1;
3632: sw += weight[i];
3633: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3634: /* 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 3635: } /* end of wave */
3636: } /* end of individual */
3637: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3638: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3639: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3640: for(mi=1; mi<= wav[i]-1; mi++){
3641: for (ii=1;ii<=nlstate+ndeath;ii++)
3642: for (j=1;j<=nlstate+ndeath;j++){
3643: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3644: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3645: }
3646: for(d=0; d<dh[mi][i]; d++){
3647: newm=savm;
3648: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3649: cov[2]=agexact;
3650: if(nagesqr==1)
3651: cov[3]= agexact*agexact;
3652: for (kk=1; kk<=cptcovage;kk++) {
3653: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3654: }
1.126 brouard 3655:
1.226 brouard 3656: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3657: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3658: savm=oldm;
3659: oldm=newm;
3660: } /* end mult */
3661:
3662: s1=s[mw[mi][i]][i];
3663: s2=s[mw[mi+1][i]][i];
3664: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3665: ipmx +=1;
3666: sw += weight[i];
3667: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3668: /*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]);*/
3669: } /* end of wave */
3670: } /* end of individual */
3671: } /* End of if */
3672: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3673: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3674: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3675: return -l;
1.126 brouard 3676: }
3677:
3678: /*************** log-likelihood *************/
3679: double funcone( double *x)
3680: {
1.228 brouard 3681: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3682: int i, ii, j, k, mi, d, kk;
1.228 brouard 3683: int ioffset=0;
1.131 brouard 3684: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3685: double **out;
3686: double lli; /* Individual log likelihood */
3687: double llt;
3688: int s1, s2;
1.228 brouard 3689: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3690:
1.126 brouard 3691: double bbh, survp;
1.187 brouard 3692: double agexact;
1.214 brouard 3693: double agebegin, ageend;
1.126 brouard 3694: /*extern weight */
3695: /* We are differentiating ll according to initial status */
3696: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3697: /*for(i=1;i<imx;i++)
3698: printf(" %d\n",s[4][i]);
3699: */
3700: cov[1]=1.;
3701:
3702: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3703: ioffset=0;
3704: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3705: /* ioffset=2+nagesqr+cptcovage; */
3706: ioffset=2+nagesqr;
1.232 brouard 3707: /* Fixed */
1.224 brouard 3708: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3709: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3710: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3711: 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)*/
3712: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3713: /* cov[2+6]=covar[Tvar[6]][i]; */
3714: /* cov[2+6]=covar[2][i]; V2 */
3715: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3716: /* cov[2+7]=covar[Tvar[7]][i]; */
3717: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3718: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3719: /* cov[2+9]=covar[Tvar[9]][i]; */
3720: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3721: }
1.232 brouard 3722: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3723: /* 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?)*\/ */
3724: /* } */
1.231 brouard 3725: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3726: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3727: /* } */
1.225 brouard 3728:
1.233 brouard 3729:
3730: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3731: /* Wave varying (but not age varying) */
3732: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3733: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3734: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3735: }
1.232 brouard 3736: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3737: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3738: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3739: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3740: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3741: /* 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 3742: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3743: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3744: /* /\* 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]); *\/ */
3745: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3746: /* } */
1.126 brouard 3747: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3748: for (j=1;j<=nlstate+ndeath;j++){
3749: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3750: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3751: }
1.214 brouard 3752:
3753: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3754: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3755: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3756: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3757: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3758: and mw[mi+1][i]. dh depends on stepm.*/
3759: newm=savm;
1.247 brouard 3760: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3761: cov[2]=agexact;
3762: if(nagesqr==1)
3763: cov[3]= agexact*agexact;
3764: for (kk=1; kk<=cptcovage;kk++) {
3765: if(!FixedV[Tvar[Tage[kk]]])
3766: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3767: else
3768: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3769: }
3770: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3771: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3772: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3773: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3774: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3775: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3776: savm=oldm;
3777: oldm=newm;
1.126 brouard 3778: } /* end mult */
3779:
3780: s1=s[mw[mi][i]][i];
3781: s2=s[mw[mi+1][i]][i];
1.217 brouard 3782: /* if(s2==-1){ */
1.268 brouard 3783: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3784: /* /\* exit(1); *\/ */
3785: /* } */
1.126 brouard 3786: bbh=(double)bh[mi][i]/(double)stepm;
3787: /* bias is positive if real duration
3788: * is higher than the multiple of stepm and negative otherwise.
3789: */
3790: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3791: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3792: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3793: for (j=1,survp=0. ; j<=nlstate; j++)
3794: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3795: lli= log(survp);
1.126 brouard 3796: }else if (mle==1){
1.242 brouard 3797: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3798: } else if(mle==2){
1.242 brouard 3799: 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 3800: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3801: 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 3802: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3803: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3804: } else{ /* mle=0 back to 1 */
1.242 brouard 3805: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3806: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3807: } /* End of if */
3808: ipmx +=1;
3809: sw += weight[i];
3810: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3811: /*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 3812: if(globpr){
1.246 brouard 3813: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3814: %11.6f %11.6f %11.6f ", \
1.242 brouard 3815: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
1.268 brouard 3816: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3817: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3818: llt +=ll[k]*gipmx/gsw;
3819: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3820: }
3821: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3822: }
1.232 brouard 3823: } /* end of wave */
3824: } /* end of individual */
3825: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3826: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3827: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3828: if(globpr==0){ /* First time we count the contributions and weights */
3829: gipmx=ipmx;
3830: gsw=sw;
3831: }
3832: return -l;
1.126 brouard 3833: }
3834:
3835:
3836: /*************** function likelione ***********/
3837: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3838: {
3839: /* This routine should help understanding what is done with
3840: the selection of individuals/waves and
3841: to check the exact contribution to the likelihood.
3842: Plotting could be done.
3843: */
3844: int k;
3845:
3846: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3847: strcpy(fileresilk,"ILK_");
1.202 brouard 3848: strcat(fileresilk,fileresu);
1.126 brouard 3849: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3850: printf("Problem with resultfile: %s\n", fileresilk);
3851: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3852: }
1.214 brouard 3853: 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");
3854: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3855: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3856: for(k=1; k<=nlstate; k++)
3857: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3858: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3859: }
3860:
3861: *fretone=(*funcone)(p);
3862: if(*globpri !=0){
3863: fclose(ficresilk);
1.205 brouard 3864: if (mle ==0)
3865: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3866: else if(mle >=1)
3867: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3868: fprintf(fichtm," You should at least run with mle >= 1 to get starting values corresponding to the optimized parameters in order to visualize the real contribution of each individual/wave: <a href=\"%s\">%s</a><br>\n",subdirf(fileresilk),subdirf(fileresilk));
1.274 brouard 3869: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3870:
3871: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3872: 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 3873: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3874: }
1.207 brouard 3875: 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 3876: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3877: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3878: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3879: fflush(fichtm);
1.205 brouard 3880: }
1.126 brouard 3881: return;
3882: }
3883:
3884:
3885: /*********** Maximum Likelihood Estimation ***************/
3886:
3887: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3888: {
1.165 brouard 3889: int i,j, iter=0;
1.126 brouard 3890: double **xi;
3891: double fret;
3892: double fretone; /* Only one call to likelihood */
3893: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3894:
3895: #ifdef NLOPT
3896: int creturn;
3897: nlopt_opt opt;
3898: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3899: double *lb;
3900: double minf; /* the minimum objective value, upon return */
3901: double * p1; /* Shifted parameters from 0 instead of 1 */
3902: myfunc_data dinst, *d = &dinst;
3903: #endif
3904:
3905:
1.126 brouard 3906: xi=matrix(1,npar,1,npar);
3907: for (i=1;i<=npar;i++)
3908: for (j=1;j<=npar;j++)
3909: xi[i][j]=(i==j ? 1.0 : 0.0);
3910: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3911: strcpy(filerespow,"POW_");
1.126 brouard 3912: strcat(filerespow,fileres);
3913: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3914: printf("Problem with resultfile: %s\n", filerespow);
3915: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3916: }
3917: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3918: for (i=1;i<=nlstate;i++)
3919: for(j=1;j<=nlstate+ndeath;j++)
3920: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3921: fprintf(ficrespow,"\n");
1.162 brouard 3922: #ifdef POWELL
1.126 brouard 3923: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3924: #endif
1.126 brouard 3925:
1.162 brouard 3926: #ifdef NLOPT
3927: #ifdef NEWUOA
3928: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3929: #else
3930: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3931: #endif
3932: lb=vector(0,npar-1);
3933: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3934: nlopt_set_lower_bounds(opt, lb);
3935: nlopt_set_initial_step1(opt, 0.1);
3936:
3937: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3938: d->function = func;
3939: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3940: nlopt_set_min_objective(opt, myfunc, d);
3941: nlopt_set_xtol_rel(opt, ftol);
3942: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3943: printf("nlopt failed! %d\n",creturn);
3944: }
3945: else {
3946: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3947: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3948: iter=1; /* not equal */
3949: }
3950: nlopt_destroy(opt);
3951: #endif
1.126 brouard 3952: free_matrix(xi,1,npar,1,npar);
3953: fclose(ficrespow);
1.203 brouard 3954: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3955: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3956: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3957:
3958: }
3959:
3960: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3961: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3962: {
3963: double **a,**y,*x,pd;
1.203 brouard 3964: /* double **hess; */
1.164 brouard 3965: int i, j;
1.126 brouard 3966: int *indx;
3967:
3968: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3969: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3970: void lubksb(double **a, int npar, int *indx, double b[]) ;
3971: void ludcmp(double **a, int npar, int *indx, double *d) ;
3972: double gompertz(double p[]);
1.203 brouard 3973: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3974:
3975: printf("\nCalculation of the hessian matrix. Wait...\n");
3976: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3977: for (i=1;i<=npar;i++){
1.203 brouard 3978: printf("%d-",i);fflush(stdout);
3979: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3980:
3981: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3982:
3983: /* printf(" %f ",p[i]);
3984: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3985: }
3986:
3987: for (i=1;i<=npar;i++) {
3988: for (j=1;j<=npar;j++) {
3989: if (j>i) {
1.203 brouard 3990: printf(".%d-%d",i,j);fflush(stdout);
3991: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3992: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3993:
3994: hess[j][i]=hess[i][j];
3995: /*printf(" %lf ",hess[i][j]);*/
3996: }
3997: }
3998: }
3999: printf("\n");
4000: fprintf(ficlog,"\n");
4001:
4002: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4003: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4004:
4005: a=matrix(1,npar,1,npar);
4006: y=matrix(1,npar,1,npar);
4007: x=vector(1,npar);
4008: indx=ivector(1,npar);
4009: for (i=1;i<=npar;i++)
4010: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4011: ludcmp(a,npar,indx,&pd);
4012:
4013: for (j=1;j<=npar;j++) {
4014: for (i=1;i<=npar;i++) x[i]=0;
4015: x[j]=1;
4016: lubksb(a,npar,indx,x);
4017: for (i=1;i<=npar;i++){
4018: matcov[i][j]=x[i];
4019: }
4020: }
4021:
4022: printf("\n#Hessian matrix#\n");
4023: fprintf(ficlog,"\n#Hessian matrix#\n");
4024: for (i=1;i<=npar;i++) {
4025: for (j=1;j<=npar;j++) {
1.203 brouard 4026: printf("%.6e ",hess[i][j]);
4027: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4028: }
4029: printf("\n");
4030: fprintf(ficlog,"\n");
4031: }
4032:
1.203 brouard 4033: /* printf("\n#Covariance matrix#\n"); */
4034: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4035: /* for (i=1;i<=npar;i++) { */
4036: /* for (j=1;j<=npar;j++) { */
4037: /* printf("%.6e ",matcov[i][j]); */
4038: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4039: /* } */
4040: /* printf("\n"); */
4041: /* fprintf(ficlog,"\n"); */
4042: /* } */
4043:
1.126 brouard 4044: /* Recompute Inverse */
1.203 brouard 4045: /* for (i=1;i<=npar;i++) */
4046: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4047: /* ludcmp(a,npar,indx,&pd); */
4048:
4049: /* printf("\n#Hessian matrix recomputed#\n"); */
4050:
4051: /* for (j=1;j<=npar;j++) { */
4052: /* for (i=1;i<=npar;i++) x[i]=0; */
4053: /* x[j]=1; */
4054: /* lubksb(a,npar,indx,x); */
4055: /* for (i=1;i<=npar;i++){ */
4056: /* y[i][j]=x[i]; */
4057: /* printf("%.3e ",y[i][j]); */
4058: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4059: /* } */
4060: /* printf("\n"); */
4061: /* fprintf(ficlog,"\n"); */
4062: /* } */
4063:
4064: /* Verifying the inverse matrix */
4065: #ifdef DEBUGHESS
4066: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4067:
1.203 brouard 4068: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4069: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4070:
4071: for (j=1;j<=npar;j++) {
4072: for (i=1;i<=npar;i++){
1.203 brouard 4073: printf("%.2f ",y[i][j]);
4074: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4075: }
4076: printf("\n");
4077: fprintf(ficlog,"\n");
4078: }
1.203 brouard 4079: #endif
1.126 brouard 4080:
4081: free_matrix(a,1,npar,1,npar);
4082: free_matrix(y,1,npar,1,npar);
4083: free_vector(x,1,npar);
4084: free_ivector(indx,1,npar);
1.203 brouard 4085: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4086:
4087:
4088: }
4089:
4090: /*************** hessian matrix ****************/
4091: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4092: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4093: int i;
4094: int l=1, lmax=20;
1.203 brouard 4095: double k1,k2, res, fx;
1.132 brouard 4096: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4097: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4098: int k=0,kmax=10;
4099: double l1;
4100:
4101: fx=func(x);
4102: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4103: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4104: l1=pow(10,l);
4105: delts=delt;
4106: for(k=1 ; k <kmax; k=k+1){
4107: delt = delta*(l1*k);
4108: p2[theta]=x[theta] +delt;
1.145 brouard 4109: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4110: p2[theta]=x[theta]-delt;
4111: k2=func(p2)-fx;
4112: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4113: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4114:
1.203 brouard 4115: #ifdef DEBUGHESSII
1.126 brouard 4116: 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);
4117: 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);
4118: #endif
4119: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4120: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4121: k=kmax;
4122: }
4123: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4124: k=kmax; l=lmax*10;
1.126 brouard 4125: }
4126: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4127: delts=delt;
4128: }
1.203 brouard 4129: } /* End loop k */
1.126 brouard 4130: }
4131: delti[theta]=delts;
4132: return res;
4133:
4134: }
4135:
1.203 brouard 4136: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4137: {
4138: int i;
1.164 brouard 4139: int l=1, lmax=20;
1.126 brouard 4140: double k1,k2,k3,k4,res,fx;
1.132 brouard 4141: double p2[MAXPARM+1];
1.203 brouard 4142: int k, kmax=1;
4143: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4144:
4145: int firstime=0;
1.203 brouard 4146:
1.126 brouard 4147: fx=func(x);
1.203 brouard 4148: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4149: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4150: p2[thetai]=x[thetai]+delti[thetai]*k;
4151: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4152: k1=func(p2)-fx;
4153:
1.203 brouard 4154: p2[thetai]=x[thetai]+delti[thetai]*k;
4155: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4156: k2=func(p2)-fx;
4157:
1.203 brouard 4158: p2[thetai]=x[thetai]-delti[thetai]*k;
4159: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4160: k3=func(p2)-fx;
4161:
1.203 brouard 4162: p2[thetai]=x[thetai]-delti[thetai]*k;
4163: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4164: k4=func(p2)-fx;
1.203 brouard 4165: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4166: if(k1*k2*k3*k4 <0.){
1.208 brouard 4167: firstime=1;
1.203 brouard 4168: kmax=kmax+10;
1.208 brouard 4169: }
4170: if(kmax >=10 || firstime ==1){
1.246 brouard 4171: 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);
4172: 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 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: }
4176: #ifdef DEBUGHESSIJ
4177: v1=hess[thetai][thetai];
4178: v2=hess[thetaj][thetaj];
4179: cv12=res;
4180: /* Computing eigen value of Hessian matrix */
4181: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4182: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4183: if ((lc2 <0) || (lc1 <0) ){
4184: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4185: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4186: 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);
4187: 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);
4188: }
1.126 brouard 4189: #endif
4190: }
4191: return res;
4192: }
4193:
1.203 brouard 4194: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4195: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4196: /* { */
4197: /* int i; */
4198: /* int l=1, lmax=20; */
4199: /* double k1,k2,k3,k4,res,fx; */
4200: /* double p2[MAXPARM+1]; */
4201: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4202: /* int k=0,kmax=10; */
4203: /* double l1; */
4204:
4205: /* fx=func(x); */
4206: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4207: /* l1=pow(10,l); */
4208: /* delts=delt; */
4209: /* for(k=1 ; k <kmax; k=k+1){ */
4210: /* delt = delti*(l1*k); */
4211: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4212: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4213: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4214: /* k1=func(p2)-fx; */
4215:
4216: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4217: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4218: /* k2=func(p2)-fx; */
4219:
4220: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4221: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4222: /* k3=func(p2)-fx; */
4223:
4224: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4225: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4226: /* k4=func(p2)-fx; */
4227: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4228: /* #ifdef DEBUGHESSIJ */
4229: /* 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); */
4230: /* 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); */
4231: /* #endif */
4232: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4233: /* k=kmax; */
4234: /* } */
4235: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4236: /* k=kmax; l=lmax*10; */
4237: /* } */
4238: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4239: /* delts=delt; */
4240: /* } */
4241: /* } /\* End loop k *\/ */
4242: /* } */
4243: /* delti[theta]=delts; */
4244: /* return res; */
4245: /* } */
4246:
4247:
1.126 brouard 4248: /************** Inverse of matrix **************/
4249: void ludcmp(double **a, int n, int *indx, double *d)
4250: {
4251: int i,imax,j,k;
4252: double big,dum,sum,temp;
4253: double *vv;
4254:
4255: vv=vector(1,n);
4256: *d=1.0;
4257: for (i=1;i<=n;i++) {
4258: big=0.0;
4259: for (j=1;j<=n;j++)
4260: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4261: if (big == 0.0){
4262: printf(" Singular Hessian matrix at row %d:\n",i);
4263: for (j=1;j<=n;j++) {
4264: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4265: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4266: }
4267: fflush(ficlog);
4268: fclose(ficlog);
4269: nrerror("Singular matrix in routine ludcmp");
4270: }
1.126 brouard 4271: vv[i]=1.0/big;
4272: }
4273: for (j=1;j<=n;j++) {
4274: for (i=1;i<j;i++) {
4275: sum=a[i][j];
4276: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4277: a[i][j]=sum;
4278: }
4279: big=0.0;
4280: for (i=j;i<=n;i++) {
4281: sum=a[i][j];
4282: for (k=1;k<j;k++)
4283: sum -= a[i][k]*a[k][j];
4284: a[i][j]=sum;
4285: if ( (dum=vv[i]*fabs(sum)) >= big) {
4286: big=dum;
4287: imax=i;
4288: }
4289: }
4290: if (j != imax) {
4291: for (k=1;k<=n;k++) {
4292: dum=a[imax][k];
4293: a[imax][k]=a[j][k];
4294: a[j][k]=dum;
4295: }
4296: *d = -(*d);
4297: vv[imax]=vv[j];
4298: }
4299: indx[j]=imax;
4300: if (a[j][j] == 0.0) a[j][j]=TINY;
4301: if (j != n) {
4302: dum=1.0/(a[j][j]);
4303: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4304: }
4305: }
4306: free_vector(vv,1,n); /* Doesn't work */
4307: ;
4308: }
4309:
4310: void lubksb(double **a, int n, int *indx, double b[])
4311: {
4312: int i,ii=0,ip,j;
4313: double sum;
4314:
4315: for (i=1;i<=n;i++) {
4316: ip=indx[i];
4317: sum=b[ip];
4318: b[ip]=b[i];
4319: if (ii)
4320: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4321: else if (sum) ii=i;
4322: b[i]=sum;
4323: }
4324: for (i=n;i>=1;i--) {
4325: sum=b[i];
4326: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4327: b[i]=sum/a[i][i];
4328: }
4329: }
4330:
4331: void pstamp(FILE *fichier)
4332: {
1.196 brouard 4333: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4334: }
4335:
1.253 brouard 4336:
4337:
1.126 brouard 4338: /************ Frequencies ********************/
1.251 brouard 4339: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4340: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4341: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4342: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4343:
1.265 brouard 4344: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4345: int iind=0, iage=0;
4346: int mi; /* Effective wave */
4347: int first;
4348: double ***freq; /* Frequencies */
1.268 brouard 4349: double *x, *y, a=0.,b=0.,r=1., sa=0., sb=0.; /* for regression, y=b+m*x and r is the correlation coefficient */
4350: int no=0, linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb);
1.226 brouard 4351: double *meanq;
4352: double **meanqt;
4353: double *pp, **prop, *posprop, *pospropt;
4354: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4355: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4356: double agebegin, ageend;
4357:
4358: pp=vector(1,nlstate);
1.251 brouard 4359: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4360: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4361: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4362: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4363: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4364: meanqt=matrix(1,lastpass,1,nqtveff);
4365: strcpy(fileresp,"P_");
4366: strcat(fileresp,fileresu);
4367: /*strcat(fileresphtm,fileresu);*/
4368: if((ficresp=fopen(fileresp,"w"))==NULL) {
4369: printf("Problem with prevalence resultfile: %s\n", fileresp);
4370: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4371: exit(0);
4372: }
1.240 brouard 4373:
1.226 brouard 4374: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4375: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4376: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4377: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4378: fflush(ficlog);
4379: exit(70);
4380: }
4381: else{
4382: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4383: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4384: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4385: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4386: }
1.237 brouard 4387: 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 4388:
1.226 brouard 4389: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4390: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4391: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4392: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4393: fflush(ficlog);
4394: exit(70);
1.240 brouard 4395: } else{
1.226 brouard 4396: 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 4397: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4398: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4399: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4400: }
1.240 brouard 4401: 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);
4402:
1.253 brouard 4403: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4404: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4405: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4406: j1=0;
1.126 brouard 4407:
1.227 brouard 4408: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4409: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4410: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4411:
4412:
1.226 brouard 4413: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4414: reference=low_education V1=0,V2=0
4415: med_educ V1=1 V2=0,
4416: high_educ V1=0 V2=1
4417: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4418: */
1.249 brouard 4419: dateintsum=0;
4420: k2cpt=0;
4421:
1.253 brouard 4422: if(cptcoveff == 0 )
1.265 brouard 4423: nl=1; /* Constant and age model only */
1.253 brouard 4424: else
4425: nl=2;
1.265 brouard 4426:
4427: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4428: /* Loop on nj=1 or 2 if dummy covariates j!=0
4429: * Loop on j1(1 to 2**cptcoveff) covariate combination
4430: * freq[s1][s2][iage] =0.
4431: * Loop on iind
4432: * ++freq[s1][s2][iage] weighted
4433: * end iind
4434: * if covariate and j!0
4435: * headers Variable on one line
4436: * endif cov j!=0
4437: * header of frequency table by age
4438: * Loop on age
4439: * pp[s1]+=freq[s1][s2][iage] weighted
4440: * pos+=freq[s1][s2][iage] weighted
4441: * Loop on s1 initial state
4442: * fprintf(ficresp
4443: * end s1
4444: * end age
4445: * if j!=0 computes starting values
4446: * end compute starting values
4447: * end j1
4448: * end nl
4449: */
1.253 brouard 4450: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4451: if(nj==1)
4452: j=0; /* First pass for the constant */
1.265 brouard 4453: else{
1.253 brouard 4454: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4455: }
1.251 brouard 4456: first=1;
1.265 brouard 4457: 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 4458: posproptt=0.;
4459: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4460: scanf("%d", i);*/
4461: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4462: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4463: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4464: freq[i][s2][m]=0;
1.251 brouard 4465:
4466: for (i=1; i<=nlstate; i++) {
1.240 brouard 4467: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4468: prop[i][m]=0;
4469: posprop[i]=0;
4470: pospropt[i]=0;
4471: }
4472: /* for (z1=1; z1<= nqfveff; z1++) { */
4473: /* meanq[z1]+=0.; */
4474: /* for(m=1;m<=lastpass;m++){ */
4475: /* meanqt[m][z1]=0.; */
4476: /* } */
4477: /* } */
4478:
4479: /* dateintsum=0; */
4480: /* k2cpt=0; */
4481:
1.265 brouard 4482: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4483: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4484: bool=1;
4485: if(j !=0){
4486: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4487: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4488: /* for (z1=1; z1<= nqfveff; z1++) { */
4489: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4490: /* } */
4491: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4492: /* if(Tvaraff[z1] ==-20){ */
4493: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4494: /* }else if(Tvaraff[z1] ==-10){ */
4495: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4496: /* }else */
4497: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4498: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4499: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4500: /* 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",
4501: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4502: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4503: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4504: } /* Onlyf fixed */
4505: } /* end z1 */
4506: } /* cptcovn > 0 */
4507: } /* end any */
4508: }/* end j==0 */
1.265 brouard 4509: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4510: /* for(m=firstpass; m<=lastpass; m++){ */
4511: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4512: m=mw[mi][iind];
4513: if(j!=0){
4514: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4515: for (z1=1; z1<=cptcoveff; z1++) {
4516: if( Fixed[Tmodelind[z1]]==1){
4517: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4518: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4519: value is -1, we don't select. It differs from the
4520: constant and age model which counts them. */
4521: bool=0; /* not selected */
4522: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4523: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4524: bool=0;
4525: }
4526: }
4527: }
4528: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4529: } /* end j==0 */
4530: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4531: if(bool==1){
4532: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4533: and mw[mi+1][iind]. dh depends on stepm. */
4534: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4535: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4536: if(m >=firstpass && m <=lastpass){
4537: k2=anint[m][iind]+(mint[m][iind]/12.);
4538: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4539: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4540: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4541: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4542: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4543: if (m<lastpass) {
4544: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4545: /* 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]); */
4546: if(s[m][iind]==-1)
4547: 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.));
4548: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4549: /* if((int)agev[m][iind] == 55) */
4550: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4551: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4552: 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 4553: }
1.251 brouard 4554: } /* end if between passes */
4555: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4556: dateintsum=dateintsum+k2; /* on all covariates ?*/
4557: k2cpt++;
4558: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4559: }
1.251 brouard 4560: }else{
4561: bool=1;
4562: }/* end bool 2 */
4563: } /* end m */
4564: } /* end bool */
4565: } /* end iind = 1 to imx */
4566: /* prop[s][age] is feeded for any initial and valid live state as well as
4567: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4568:
4569:
4570: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4571: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4572: pstamp(ficresp);
1.251 brouard 4573: if (cptcoveff>0 && j!=0){
1.265 brouard 4574: pstamp(ficresp);
1.251 brouard 4575: printf( "\n#********** Variable ");
4576: fprintf(ficresp, "\n#********** Variable ");
4577: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4578: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4579: fprintf(ficlog, "\n#********** Variable ");
4580: for (z1=1; z1<=cptcoveff; z1++){
4581: if(!FixedV[Tvaraff[z1]]){
4582: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4583: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4584: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4585: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4586: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4587: }else{
1.251 brouard 4588: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4589: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4590: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4591: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4592: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4593: }
4594: }
4595: printf( "**********\n#");
4596: fprintf(ficresp, "**********\n#");
4597: fprintf(ficresphtm, "**********</h3>\n");
4598: fprintf(ficresphtmfr, "**********</h3>\n");
4599: fprintf(ficlog, "**********\n");
4600: }
4601: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4602: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4603: fprintf(ficresp, " Age");
4604: 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 4605: for(i=1; i<=nlstate;i++) {
1.265 brouard 4606: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4607: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4608: }
1.265 brouard 4609: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4610: fprintf(ficresphtm, "\n");
4611:
4612: /* Header of frequency table by age */
4613: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4614: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4615: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4616: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4617: if(s2!=0 && m!=0)
4618: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4619: }
1.226 brouard 4620: }
1.251 brouard 4621: fprintf(ficresphtmfr, "\n");
4622:
4623: /* For each age */
4624: for(iage=iagemin; iage <= iagemax+3; iage++){
4625: fprintf(ficresphtm,"<tr>");
4626: if(iage==iagemax+1){
4627: fprintf(ficlog,"1");
4628: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4629: }else if(iage==iagemax+2){
4630: fprintf(ficlog,"0");
4631: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4632: }else if(iage==iagemax+3){
4633: fprintf(ficlog,"Total");
4634: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4635: }else{
1.240 brouard 4636: if(first==1){
1.251 brouard 4637: first=0;
4638: printf("See log file for details...\n");
4639: }
4640: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4641: fprintf(ficlog,"Age %d", iage);
4642: }
1.265 brouard 4643: for(s1=1; s1 <=nlstate ; s1++){
4644: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4645: pp[s1] += freq[s1][m][iage];
1.251 brouard 4646: }
1.265 brouard 4647: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4648: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4649: pos += freq[s1][m][iage];
4650: if(pp[s1]>=1.e-10){
1.251 brouard 4651: if(first==1){
1.265 brouard 4652: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4653: }
1.265 brouard 4654: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4655: }else{
4656: if(first==1)
1.265 brouard 4657: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4658: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4659: }
4660: }
4661:
1.265 brouard 4662: for(s1=1; s1 <=nlstate ; s1++){
4663: /* posprop[s1]=0; */
4664: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4665: pp[s1] += freq[s1][m][iage];
4666: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4667:
4668: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4669: pos += pp[s1]; /* pos is the total number of transitions until this age */
4670: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4671: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4672: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4673: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4674: }
4675:
4676: /* Writing ficresp */
4677: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4678: if( iage <= iagemax){
4679: fprintf(ficresp," %d",iage);
4680: }
4681: }else if( nj==2){
4682: if( iage <= iagemax){
4683: fprintf(ficresp," %d",iage);
4684: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4685: }
1.240 brouard 4686: }
1.265 brouard 4687: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4688: if(pos>=1.e-5){
1.251 brouard 4689: if(first==1)
1.265 brouard 4690: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4691: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4692: }else{
4693: if(first==1)
1.265 brouard 4694: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4695: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4696: }
4697: if( iage <= iagemax){
4698: if(pos>=1.e-5){
1.265 brouard 4699: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4700: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4701: }else if( nj==2){
4702: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4703: }
4704: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4705: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4706: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4707: } else{
4708: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4709: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4710: }
1.240 brouard 4711: }
1.265 brouard 4712: pospropt[s1] +=posprop[s1];
4713: } /* end loop s1 */
1.251 brouard 4714: /* pospropt=0.; */
1.265 brouard 4715: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4716: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4717: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4718: if(first==1){
1.265 brouard 4719: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4720: }
1.265 brouard 4721: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4722: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4723: }
1.265 brouard 4724: if(s1!=0 && m!=0)
4725: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4726: }
1.265 brouard 4727: } /* end loop s1 */
1.251 brouard 4728: posproptt=0.;
1.265 brouard 4729: for(s1=1; s1 <=nlstate; s1++){
4730: posproptt += pospropt[s1];
1.251 brouard 4731: }
4732: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4733: fprintf(ficresphtm,"</tr>\n");
4734: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4735: if(iage <= iagemax)
4736: fprintf(ficresp,"\n");
1.240 brouard 4737: }
1.251 brouard 4738: if(first==1)
4739: printf("Others in log...\n");
4740: fprintf(ficlog,"\n");
4741: } /* end loop age iage */
1.265 brouard 4742:
1.251 brouard 4743: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4744: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4745: if(posproptt < 1.e-5){
1.265 brouard 4746: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4747: }else{
1.265 brouard 4748: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4749: }
1.226 brouard 4750: }
1.251 brouard 4751: fprintf(ficresphtm,"</tr>\n");
4752: fprintf(ficresphtm,"</table>\n");
4753: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4754: if(posproptt < 1.e-5){
1.251 brouard 4755: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4756: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4757: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4758: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4759: invalidvarcomb[j1]=1;
1.226 brouard 4760: }else{
1.251 brouard 4761: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4762: invalidvarcomb[j1]=0;
1.226 brouard 4763: }
1.251 brouard 4764: fprintf(ficresphtmfr,"</table>\n");
4765: fprintf(ficlog,"\n");
4766: if(j!=0){
4767: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4768: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4769: for(k=1; k <=(nlstate+ndeath); k++){
4770: if (k != i) {
1.265 brouard 4771: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4772: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4773: if(j1==1){ /* All dummy covariates to zero */
4774: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4775: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4776: printf("%d%d ",i,k);
4777: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4778: 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]));
4779: 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]));
4780: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4781: }
1.253 brouard 4782: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4783: for(iage=iagemin; iage <= iagemax+3; iage++){
4784: x[iage]= (double)iage;
4785: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4786: /* 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 4787: }
1.268 brouard 4788: /* Some are not finite, but linreg will ignore these ages */
4789: no=0;
1.253 brouard 4790: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4791: pstart[s1]=b;
4792: pstart[s1-1]=a;
1.252 brouard 4793: }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 */
4794: 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]);
4795: 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 4796: 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 4797: printf("%d%d ",i,k);
4798: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4799: 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 4800: }else{ /* Other cases, like quantitative fixed or varying covariates */
4801: ;
4802: }
4803: /* printf("%12.7f )", param[i][jj][k]); */
4804: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4805: s1++;
1.251 brouard 4806: } /* end jj */
4807: } /* end k!= i */
4808: } /* end k */
1.265 brouard 4809: } /* end i, s1 */
1.251 brouard 4810: } /* end j !=0 */
4811: } /* end selected combination of covariate j1 */
4812: if(j==0){ /* We can estimate starting values from the occurences in each case */
4813: printf("#Freqsummary: Starting values for the constants:\n");
4814: fprintf(ficlog,"\n");
1.265 brouard 4815: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4816: for(k=1; k <=(nlstate+ndeath); k++){
4817: if (k != i) {
4818: printf("%d%d ",i,k);
4819: fprintf(ficlog,"%d%d ",i,k);
4820: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4821: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4822: if(jj==1){ /* Age has to be done */
1.265 brouard 4823: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4824: 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]));
4825: 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 4826: }
4827: /* printf("%12.7f )", param[i][jj][k]); */
4828: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4829: s1++;
1.250 brouard 4830: }
1.251 brouard 4831: printf("\n");
4832: fprintf(ficlog,"\n");
1.250 brouard 4833: }
4834: }
4835: }
1.251 brouard 4836: printf("#Freqsummary\n");
4837: fprintf(ficlog,"\n");
1.265 brouard 4838: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4839: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4840: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4841: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4842: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4843: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4844: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4845: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4846: /* } */
4847: }
1.265 brouard 4848: } /* end loop s1 */
1.251 brouard 4849:
4850: printf("\n");
4851: fprintf(ficlog,"\n");
4852: } /* end j=0 */
1.249 brouard 4853: } /* end j */
1.252 brouard 4854:
1.253 brouard 4855: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4856: for(i=1, jk=1; i <=nlstate; i++){
4857: for(j=1; j <=nlstate+ndeath; j++){
4858: if(j!=i){
4859: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4860: printf("%1d%1d",i,j);
4861: fprintf(ficparo,"%1d%1d",i,j);
4862: for(k=1; k<=ncovmodel;k++){
4863: /* printf(" %lf",param[i][j][k]); */
4864: /* fprintf(ficparo," %lf",param[i][j][k]); */
4865: p[jk]=pstart[jk];
4866: printf(" %f ",pstart[jk]);
4867: fprintf(ficparo," %f ",pstart[jk]);
4868: jk++;
4869: }
4870: printf("\n");
4871: fprintf(ficparo,"\n");
4872: }
4873: }
4874: }
4875: } /* end mle=-2 */
1.226 brouard 4876: dateintmean=dateintsum/k2cpt;
1.240 brouard 4877:
1.226 brouard 4878: fclose(ficresp);
4879: fclose(ficresphtm);
4880: fclose(ficresphtmfr);
4881: free_vector(meanq,1,nqfveff);
4882: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4883: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4884: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4885: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4886: free_vector(pospropt,1,nlstate);
4887: free_vector(posprop,1,nlstate);
1.251 brouard 4888: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4889: free_vector(pp,1,nlstate);
4890: /* End of freqsummary */
4891: }
1.126 brouard 4892:
1.268 brouard 4893: /* Simple linear regression */
4894: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4895:
4896: /* y=a+bx regression */
4897: double sumx = 0.0; /* sum of x */
4898: double sumx2 = 0.0; /* sum of x**2 */
4899: double sumxy = 0.0; /* sum of x * y */
4900: double sumy = 0.0; /* sum of y */
4901: double sumy2 = 0.0; /* sum of y**2 */
4902: double sume2 = 0.0; /* sum of square or residuals */
4903: double yhat;
4904:
4905: double denom=0;
4906: int i;
4907: int ne=*no;
4908:
4909: for ( i=ifi, ne=0;i<=ila;i++) {
4910: if(!isfinite(x[i]) || !isfinite(y[i])){
4911: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4912: continue;
4913: }
4914: ne=ne+1;
4915: sumx += x[i];
4916: sumx2 += x[i]*x[i];
4917: sumxy += x[i] * y[i];
4918: sumy += y[i];
4919: sumy2 += y[i]*y[i];
4920: denom = (ne * sumx2 - sumx*sumx);
4921: /* 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); */
4922: }
4923:
4924: denom = (ne * sumx2 - sumx*sumx);
4925: if (denom == 0) {
4926: // vertical, slope m is infinity
4927: *b = INFINITY;
4928: *a = 0;
4929: if (r) *r = 0;
4930: return 1;
4931: }
4932:
4933: *b = (ne * sumxy - sumx * sumy) / denom;
4934: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4935: if (r!=NULL) {
4936: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4937: sqrt((sumx2 - sumx*sumx/ne) *
4938: (sumy2 - sumy*sumy/ne));
4939: }
4940: *no=ne;
4941: for ( i=ifi, ne=0;i<=ila;i++) {
4942: if(!isfinite(x[i]) || !isfinite(y[i])){
4943: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4944: continue;
4945: }
4946: ne=ne+1;
4947: yhat = y[i] - *a -*b* x[i];
4948: sume2 += yhat * yhat ;
4949:
4950: denom = (ne * sumx2 - sumx*sumx);
4951: /* 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); */
4952: }
4953: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
4954: *sa= *sb * sqrt(sumx2/ne);
4955:
4956: return 0;
4957: }
4958:
1.126 brouard 4959: /************ Prevalence ********************/
1.227 brouard 4960: 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)
4961: {
4962: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4963: in each health status at the date of interview (if between dateprev1 and dateprev2).
4964: We still use firstpass and lastpass as another selection.
4965: */
1.126 brouard 4966:
1.227 brouard 4967: int i, m, jk, j1, bool, z1,j, iv;
4968: int mi; /* Effective wave */
4969: int iage;
4970: double agebegin, ageend;
4971:
4972: double **prop;
4973: double posprop;
4974: double y2; /* in fractional years */
4975: int iagemin, iagemax;
4976: int first; /** to stop verbosity which is redirected to log file */
4977:
4978: iagemin= (int) agemin;
4979: iagemax= (int) agemax;
4980: /*pp=vector(1,nlstate);*/
1.251 brouard 4981: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4982: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4983: j1=0;
1.222 brouard 4984:
1.227 brouard 4985: /*j=cptcoveff;*/
4986: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4987:
1.227 brouard 4988: first=1;
4989: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4990: for (i=1; i<=nlstate; i++)
1.251 brouard 4991: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4992: prop[i][iage]=0.0;
4993: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4994: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4995: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4996:
4997: for (i=1; i<=imx; i++) { /* Each individual */
4998: bool=1;
4999: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5000: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5001: m=mw[mi][i];
5002: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5003: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5004: for (z1=1; z1<=cptcoveff; z1++){
5005: if( Fixed[Tmodelind[z1]]==1){
5006: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5007: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5008: bool=0;
5009: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5010: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5011: bool=0;
5012: }
5013: }
5014: if(bool==1){ /* Otherwise we skip that wave/person */
5015: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5016: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5017: if(m >=firstpass && m <=lastpass){
5018: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5019: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5020: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5021: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5022: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5023: 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);
5024: exit(1);
5025: }
5026: if (s[m][i]>0 && s[m][i]<=nlstate) {
5027: /*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]]);*/
5028: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5029: prop[s[m][i]][iagemax+3] += weight[i];
5030: } /* end valid statuses */
5031: } /* end selection of dates */
5032: } /* end selection of waves */
5033: } /* end bool */
5034: } /* end wave */
5035: } /* end individual */
5036: for(i=iagemin; i <= iagemax+3; i++){
5037: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5038: posprop += prop[jk][i];
5039: }
5040:
5041: for(jk=1; jk <=nlstate ; jk++){
5042: if( i <= iagemax){
5043: if(posprop>=1.e-5){
5044: probs[i][jk][j1]= prop[jk][i]/posprop;
5045: } else{
5046: if(first==1){
5047: first=0;
1.266 brouard 5048: 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]);
5049: 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]);
5050: }else{
5051: 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 5052: }
5053: }
5054: }
5055: }/* end jk */
5056: }/* end i */
1.222 brouard 5057: /*} *//* end i1 */
1.227 brouard 5058: } /* end j1 */
1.222 brouard 5059:
1.227 brouard 5060: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5061: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5062: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5063: } /* End of prevalence */
1.126 brouard 5064:
5065: /************* Waves Concatenation ***************/
5066:
5067: 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)
5068: {
5069: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5070: Death is a valid wave (if date is known).
5071: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5072: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5073: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5074: */
1.126 brouard 5075:
1.224 brouard 5076: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5077: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5078: double sum=0., jmean=0.;*/
1.224 brouard 5079: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5080: int j, k=0,jk, ju, jl;
5081: double sum=0.;
5082: first=0;
1.214 brouard 5083: firstwo=0;
1.217 brouard 5084: firsthree=0;
1.218 brouard 5085: firstfour=0;
1.164 brouard 5086: jmin=100000;
1.126 brouard 5087: jmax=-1;
5088: jmean=0.;
1.224 brouard 5089:
5090: /* Treating live states */
1.214 brouard 5091: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5092: mi=0; /* First valid wave */
1.227 brouard 5093: mli=0; /* Last valid wave */
1.126 brouard 5094: m=firstpass;
1.214 brouard 5095: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5096: 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 */
5097: mli=m-1;/* mw[++mi][i]=m-1; */
5098: }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 */
5099: mw[++mi][i]=m;
5100: mli=m;
1.224 brouard 5101: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5102: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5103: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5104: }
1.227 brouard 5105: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5106: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5107: break;
1.224 brouard 5108: #else
1.227 brouard 5109: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5110: if(firsthree == 0){
1.262 brouard 5111: 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 5112: firsthree=1;
5113: }
1.262 brouard 5114: 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 5115: mw[++mi][i]=m;
5116: mli=m;
5117: }
5118: if(s[m][i]==-2){ /* Vital status is really unknown */
5119: nbwarn++;
5120: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5121: 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);
5122: 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);
5123: }
5124: break;
5125: }
5126: break;
1.224 brouard 5127: #endif
1.227 brouard 5128: }/* End m >= lastpass */
1.126 brouard 5129: }/* end while */
1.224 brouard 5130:
1.227 brouard 5131: /* 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 5132: /* After last pass */
1.224 brouard 5133: /* Treating death states */
1.214 brouard 5134: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5135: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5136: /* } */
1.126 brouard 5137: mi++; /* Death is another wave */
5138: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5139: /* Only death is a correct wave */
1.126 brouard 5140: mw[mi][i]=m;
1.257 brouard 5141: } /* else not in a death state */
1.224 brouard 5142: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5143: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5144: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5145: 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 */
5146: nbwarn++;
5147: if(firstfiv==0){
5148: 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 );
5149: firstfiv=1;
5150: }else{
5151: 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 );
5152: }
5153: }else{ /* Death occured afer last wave potential bias */
5154: nberr++;
5155: if(firstwo==0){
1.257 brouard 5156: 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 5157: firstwo=1;
5158: }
1.257 brouard 5159: 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 5160: }
1.257 brouard 5161: }else{ /* if date of interview is unknown */
1.227 brouard 5162: /* death is known but not confirmed by death status at any wave */
5163: if(firstfour==0){
5164: 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 );
5165: firstfour=1;
5166: }
5167: 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 5168: }
1.224 brouard 5169: } /* end if date of death is known */
5170: #endif
5171: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5172: /* wav[i]=mw[mi][i]; */
1.126 brouard 5173: if(mi==0){
5174: nbwarn++;
5175: if(first==0){
1.227 brouard 5176: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5177: first=1;
1.126 brouard 5178: }
5179: if(first==1){
1.227 brouard 5180: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5181: }
5182: } /* end mi==0 */
5183: } /* End individuals */
1.214 brouard 5184: /* wav and mw are no more changed */
1.223 brouard 5185:
1.214 brouard 5186:
1.126 brouard 5187: for(i=1; i<=imx; i++){
5188: for(mi=1; mi<wav[i];mi++){
5189: if (stepm <=0)
1.227 brouard 5190: dh[mi][i]=1;
1.126 brouard 5191: else{
1.260 brouard 5192: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5193: if (agedc[i] < 2*AGESUP) {
5194: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5195: if(j==0) j=1; /* Survives at least one month after exam */
5196: else if(j<0){
5197: nberr++;
5198: 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]);
5199: j=1; /* Temporary Dangerous patch */
5200: 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);
5201: 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]);
5202: 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);
5203: }
5204: k=k+1;
5205: if (j >= jmax){
5206: jmax=j;
5207: ijmax=i;
5208: }
5209: if (j <= jmin){
5210: jmin=j;
5211: ijmin=i;
5212: }
5213: sum=sum+j;
5214: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5215: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5216: }
5217: }
5218: else{
5219: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5220: /* 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 5221:
1.227 brouard 5222: k=k+1;
5223: if (j >= jmax) {
5224: jmax=j;
5225: ijmax=i;
5226: }
5227: else if (j <= jmin){
5228: jmin=j;
5229: ijmin=i;
5230: }
5231: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5232: /*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]);*/
5233: if(j<0){
5234: nberr++;
5235: 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]);
5236: 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]);
5237: }
5238: sum=sum+j;
5239: }
5240: jk= j/stepm;
5241: jl= j -jk*stepm;
5242: ju= j -(jk+1)*stepm;
5243: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5244: if(jl==0){
5245: dh[mi][i]=jk;
5246: bh[mi][i]=0;
5247: }else{ /* We want a negative bias in order to only have interpolation ie
5248: * to avoid the price of an extra matrix product in likelihood */
5249: dh[mi][i]=jk+1;
5250: bh[mi][i]=ju;
5251: }
5252: }else{
5253: if(jl <= -ju){
5254: dh[mi][i]=jk;
5255: bh[mi][i]=jl; /* bias is positive if real duration
5256: * is higher than the multiple of stepm and negative otherwise.
5257: */
5258: }
5259: else{
5260: dh[mi][i]=jk+1;
5261: bh[mi][i]=ju;
5262: }
5263: if(dh[mi][i]==0){
5264: dh[mi][i]=1; /* At least one step */
5265: bh[mi][i]=ju; /* At least one step */
5266: /* 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);*/
5267: }
5268: } /* end if mle */
1.126 brouard 5269: }
5270: } /* end wave */
5271: }
5272: jmean=sum/k;
5273: 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 5274: 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 5275: }
1.126 brouard 5276:
5277: /*********** Tricode ****************************/
1.220 brouard 5278: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5279: {
5280: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5281: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5282: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5283: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5284: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5285: */
1.130 brouard 5286:
1.242 brouard 5287: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5288: int modmaxcovj=0; /* Modality max of covariates j */
5289: int cptcode=0; /* Modality max of covariates j */
5290: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5291:
5292:
1.242 brouard 5293: /* cptcoveff=0; */
5294: /* *cptcov=0; */
1.126 brouard 5295:
1.242 brouard 5296: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5297:
1.242 brouard 5298: /* Loop on covariates without age and products and no quantitative variable */
5299: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5300: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5301: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5302: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5303: switch(Fixed[k]) {
5304: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5305: 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*/
5306: ij=(int)(covar[Tvar[k]][i]);
5307: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5308: * If product of Vn*Vm, still boolean *:
5309: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5310: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5311: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5312: modality of the nth covariate of individual i. */
5313: if (ij > modmaxcovj)
5314: modmaxcovj=ij;
5315: else if (ij < modmincovj)
5316: modmincovj=ij;
5317: if ((ij < -1) && (ij > NCOVMAX)){
5318: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5319: exit(1);
5320: }else
5321: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5322: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5323: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5324: /* getting the maximum value of the modality of the covariate
5325: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5326: female ies 1, then modmaxcovj=1.
5327: */
5328: } /* end for loop on individuals i */
5329: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5330: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5331: cptcode=modmaxcovj;
5332: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5333: /*for (i=0; i<=cptcode; i++) {*/
5334: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5335: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5336: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5337: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5338: if( j != -1){
5339: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5340: covariate for which somebody answered excluding
5341: undefined. Usually 2: 0 and 1. */
5342: }
5343: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5344: covariate for which somebody answered including
5345: undefined. Usually 3: -1, 0 and 1. */
5346: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5347: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5348: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5349:
1.242 brouard 5350: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5351: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5352: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5353: /* modmincovj=3; modmaxcovj = 7; */
5354: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5355: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5356: /* defining two dummy variables: variables V1_1 and V1_2.*/
5357: /* nbcode[Tvar[j]][ij]=k; */
5358: /* nbcode[Tvar[j]][1]=0; */
5359: /* nbcode[Tvar[j]][2]=1; */
5360: /* nbcode[Tvar[j]][3]=2; */
5361: /* To be continued (not working yet). */
5362: ij=0; /* ij is similar to i but can jump over null modalities */
5363: 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*/
5364: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5365: break;
5366: }
5367: ij++;
5368: 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*/
5369: cptcode = ij; /* New max modality for covar j */
5370: } /* end of loop on modality i=-1 to 1 or more */
5371: break;
5372: case 1: /* Testing on varying covariate, could be simple and
5373: * should look at waves or product of fixed *
5374: * varying. No time to test -1, assuming 0 and 1 only */
5375: ij=0;
5376: for(i=0; i<=1;i++){
5377: nbcode[Tvar[k]][++ij]=i;
5378: }
5379: break;
5380: default:
5381: break;
5382: } /* end switch */
5383: } /* end dummy test */
5384:
5385: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5386: /* /\*recode from 0 *\/ */
5387: /* k is a modality. If we have model=V1+V1*sex */
5388: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5389: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5390: /* } */
5391: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5392: /* if (ij > ncodemax[j]) { */
5393: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5394: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5395: /* break; */
5396: /* } */
5397: /* } /\* end of loop on modality k *\/ */
5398: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5399:
5400: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5401: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5402: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5403: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5404: 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 */
5405: 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 */
5406: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5407: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5408:
5409: ij=0;
5410: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5411: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5412: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5413: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5414: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5415: /* If product not in single variable we don't print results */
5416: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5417: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5418: 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*/
5419: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5420: 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 */
5421: if(Fixed[k]!=0)
5422: anyvaryingduminmodel=1;
5423: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5424: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5425: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5426: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5427: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5428: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5429: }
5430: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5431: /* ij--; */
5432: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5433: *cptcov=ij; /*Number of total real effective covariates: effective
5434: * because they can be excluded from the model and real
5435: * if in the model but excluded because missing values, but how to get k from ij?*/
5436: for(j=ij+1; j<= cptcovt; j++){
5437: Tvaraff[j]=0;
5438: Tmodelind[j]=0;
5439: }
5440: for(j=ntveff+1; j<= cptcovt; j++){
5441: TmodelInvind[j]=0;
5442: }
5443: /* To be sorted */
5444: ;
5445: }
1.126 brouard 5446:
1.145 brouard 5447:
1.126 brouard 5448: /*********** Health Expectancies ****************/
5449:
1.235 brouard 5450: 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 5451:
5452: {
5453: /* Health expectancies, no variances */
1.164 brouard 5454: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5455: int nhstepma, nstepma; /* Decreasing with age */
5456: double age, agelim, hf;
5457: double ***p3mat;
5458: double eip;
5459:
1.238 brouard 5460: /* pstamp(ficreseij); */
1.126 brouard 5461: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5462: fprintf(ficreseij,"# Age");
5463: for(i=1; i<=nlstate;i++){
5464: for(j=1; j<=nlstate;j++){
5465: fprintf(ficreseij," e%1d%1d ",i,j);
5466: }
5467: fprintf(ficreseij," e%1d. ",i);
5468: }
5469: fprintf(ficreseij,"\n");
5470:
5471:
5472: if(estepm < stepm){
5473: printf ("Problem %d lower than %d\n",estepm, stepm);
5474: }
5475: else hstepm=estepm;
5476: /* We compute the life expectancy from trapezoids spaced every estepm months
5477: * This is mainly to measure the difference between two models: for example
5478: * if stepm=24 months pijx are given only every 2 years and by summing them
5479: * we are calculating an estimate of the Life Expectancy assuming a linear
5480: * progression in between and thus overestimating or underestimating according
5481: * to the curvature of the survival function. If, for the same date, we
5482: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5483: * to compare the new estimate of Life expectancy with the same linear
5484: * hypothesis. A more precise result, taking into account a more precise
5485: * curvature will be obtained if estepm is as small as stepm. */
5486:
5487: /* For example we decided to compute the life expectancy with the smallest unit */
5488: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5489: nhstepm is the number of hstepm from age to agelim
5490: nstepm is the number of stepm from age to agelin.
1.270 brouard 5491: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5492: and note for a fixed period like estepm months */
5493: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5494: survival function given by stepm (the optimization length). Unfortunately it
5495: means that if the survival funtion is printed only each two years of age and if
5496: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5497: results. So we changed our mind and took the option of the best precision.
5498: */
5499: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5500:
5501: agelim=AGESUP;
5502: /* If stepm=6 months */
5503: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5504: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5505:
5506: /* nhstepm age range expressed in number of stepm */
5507: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5508: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5509: /* if (stepm >= YEARM) hstepm=1;*/
5510: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5511: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5512:
5513: for (age=bage; age<=fage; age ++){
5514: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5515: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5516: /* if (stepm >= YEARM) hstepm=1;*/
5517: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5518:
5519: /* If stepm=6 months */
5520: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5521: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5522:
1.235 brouard 5523: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5524:
5525: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5526:
5527: printf("%d|",(int)age);fflush(stdout);
5528: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5529:
5530: /* Computing expectancies */
5531: for(i=1; i<=nlstate;i++)
5532: for(j=1; j<=nlstate;j++)
5533: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5534: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5535:
5536: /* 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]);*/
5537:
5538: }
5539:
5540: fprintf(ficreseij,"%3.0f",age );
5541: for(i=1; i<=nlstate;i++){
5542: eip=0;
5543: for(j=1; j<=nlstate;j++){
5544: eip +=eij[i][j][(int)age];
5545: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5546: }
5547: fprintf(ficreseij,"%9.4f", eip );
5548: }
5549: fprintf(ficreseij,"\n");
5550:
5551: }
5552: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5553: printf("\n");
5554: fprintf(ficlog,"\n");
5555:
5556: }
5557:
1.235 brouard 5558: 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 5559:
5560: {
5561: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5562: to initial status i, ei. .
1.126 brouard 5563: */
5564: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5565: int nhstepma, nstepma; /* Decreasing with age */
5566: double age, agelim, hf;
5567: double ***p3matp, ***p3matm, ***varhe;
5568: double **dnewm,**doldm;
5569: double *xp, *xm;
5570: double **gp, **gm;
5571: double ***gradg, ***trgradg;
5572: int theta;
5573:
5574: double eip, vip;
5575:
5576: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5577: xp=vector(1,npar);
5578: xm=vector(1,npar);
5579: dnewm=matrix(1,nlstate*nlstate,1,npar);
5580: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5581:
5582: pstamp(ficresstdeij);
5583: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5584: fprintf(ficresstdeij,"# Age");
5585: for(i=1; i<=nlstate;i++){
5586: for(j=1; j<=nlstate;j++)
5587: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5588: fprintf(ficresstdeij," e%1d. ",i);
5589: }
5590: fprintf(ficresstdeij,"\n");
5591:
5592: pstamp(ficrescveij);
5593: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5594: fprintf(ficrescveij,"# Age");
5595: for(i=1; i<=nlstate;i++)
5596: for(j=1; j<=nlstate;j++){
5597: cptj= (j-1)*nlstate+i;
5598: for(i2=1; i2<=nlstate;i2++)
5599: for(j2=1; j2<=nlstate;j2++){
5600: cptj2= (j2-1)*nlstate+i2;
5601: if(cptj2 <= cptj)
5602: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5603: }
5604: }
5605: fprintf(ficrescveij,"\n");
5606:
5607: if(estepm < stepm){
5608: printf ("Problem %d lower than %d\n",estepm, stepm);
5609: }
5610: else hstepm=estepm;
5611: /* We compute the life expectancy from trapezoids spaced every estepm months
5612: * This is mainly to measure the difference between two models: for example
5613: * if stepm=24 months pijx are given only every 2 years and by summing them
5614: * we are calculating an estimate of the Life Expectancy assuming a linear
5615: * progression in between and thus overestimating or underestimating according
5616: * to the curvature of the survival function. If, for the same date, we
5617: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5618: * to compare the new estimate of Life expectancy with the same linear
5619: * hypothesis. A more precise result, taking into account a more precise
5620: * curvature will be obtained if estepm is as small as stepm. */
5621:
5622: /* For example we decided to compute the life expectancy with the smallest unit */
5623: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5624: nhstepm is the number of hstepm from age to agelim
5625: nstepm is the number of stepm from age to agelin.
5626: Look at hpijx to understand the reason of that which relies in memory size
5627: and note for a fixed period like estepm months */
5628: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5629: survival function given by stepm (the optimization length). Unfortunately it
5630: means that if the survival funtion is printed only each two years of age and if
5631: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5632: results. So we changed our mind and took the option of the best precision.
5633: */
5634: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5635:
5636: /* If stepm=6 months */
5637: /* nhstepm age range expressed in number of stepm */
5638: agelim=AGESUP;
5639: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5640: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5641: /* if (stepm >= YEARM) hstepm=1;*/
5642: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5643:
5644: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5645: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5646: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5647: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5648: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5649: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5650:
5651: for (age=bage; age<=fage; age ++){
5652: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5653: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5654: /* if (stepm >= YEARM) hstepm=1;*/
5655: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5656:
1.126 brouard 5657: /* If stepm=6 months */
5658: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5659: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5660:
5661: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5662:
1.126 brouard 5663: /* Computing Variances of health expectancies */
5664: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5665: decrease memory allocation */
5666: for(theta=1; theta <=npar; theta++){
5667: for(i=1; i<=npar; i++){
1.222 brouard 5668: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5669: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5670: }
1.235 brouard 5671: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5672: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5673:
1.126 brouard 5674: for(j=1; j<= nlstate; j++){
1.222 brouard 5675: for(i=1; i<=nlstate; i++){
5676: for(h=0; h<=nhstepm-1; h++){
5677: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5678: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5679: }
5680: }
1.126 brouard 5681: }
1.218 brouard 5682:
1.126 brouard 5683: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5684: for(h=0; h<=nhstepm-1; h++){
5685: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5686: }
1.126 brouard 5687: }/* End theta */
5688:
5689:
5690: for(h=0; h<=nhstepm-1; h++)
5691: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5692: for(theta=1; theta <=npar; theta++)
5693: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5694:
1.218 brouard 5695:
1.222 brouard 5696: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5697: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5698: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5699:
1.222 brouard 5700: printf("%d|",(int)age);fflush(stdout);
5701: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5702: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5703: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5704: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5705: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5706: for(ij=1;ij<=nlstate*nlstate;ij++)
5707: for(ji=1;ji<=nlstate*nlstate;ji++)
5708: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5709: }
5710: }
1.218 brouard 5711:
1.126 brouard 5712: /* Computing expectancies */
1.235 brouard 5713: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5714: for(i=1; i<=nlstate;i++)
5715: for(j=1; j<=nlstate;j++)
1.222 brouard 5716: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5717: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5718:
1.222 brouard 5719: /* 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 5720:
1.222 brouard 5721: }
1.269 brouard 5722:
5723: /* Standard deviation of expectancies ij */
1.126 brouard 5724: fprintf(ficresstdeij,"%3.0f",age );
5725: for(i=1; i<=nlstate;i++){
5726: eip=0.;
5727: vip=0.;
5728: for(j=1; j<=nlstate;j++){
1.222 brouard 5729: eip += eij[i][j][(int)age];
5730: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5731: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5732: 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 5733: }
5734: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5735: }
5736: fprintf(ficresstdeij,"\n");
1.218 brouard 5737:
1.269 brouard 5738: /* Variance of expectancies ij */
1.126 brouard 5739: fprintf(ficrescveij,"%3.0f",age );
5740: for(i=1; i<=nlstate;i++)
5741: for(j=1; j<=nlstate;j++){
1.222 brouard 5742: cptj= (j-1)*nlstate+i;
5743: for(i2=1; i2<=nlstate;i2++)
5744: for(j2=1; j2<=nlstate;j2++){
5745: cptj2= (j2-1)*nlstate+i2;
5746: if(cptj2 <= cptj)
5747: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5748: }
1.126 brouard 5749: }
5750: fprintf(ficrescveij,"\n");
1.218 brouard 5751:
1.126 brouard 5752: }
5753: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5754: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5755: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5756: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5757: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5758: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5759: printf("\n");
5760: fprintf(ficlog,"\n");
1.218 brouard 5761:
1.126 brouard 5762: free_vector(xm,1,npar);
5763: free_vector(xp,1,npar);
5764: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5765: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5766: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5767: }
1.218 brouard 5768:
1.126 brouard 5769: /************ Variance ******************/
1.235 brouard 5770: 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 5771: {
5772: /* Variance of health expectancies */
5773: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5774: /* double **newm;*/
5775: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5776:
5777: /* int movingaverage(); */
5778: double **dnewm,**doldm;
5779: double **dnewmp,**doldmp;
5780: int i, j, nhstepm, hstepm, h, nstepm ;
5781: int k;
5782: double *xp;
5783: double **gp, **gm; /* for var eij */
5784: double ***gradg, ***trgradg; /*for var eij */
5785: double **gradgp, **trgradgp; /* for var p point j */
5786: double *gpp, *gmp; /* for var p point j */
5787: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5788: double ***p3mat;
5789: double age,agelim, hf;
5790: /* double ***mobaverage; */
5791: int theta;
5792: char digit[4];
5793: char digitp[25];
5794:
5795: char fileresprobmorprev[FILENAMELENGTH];
5796:
5797: if(popbased==1){
5798: if(mobilav!=0)
5799: strcpy(digitp,"-POPULBASED-MOBILAV_");
5800: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5801: }
5802: else
5803: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5804:
1.218 brouard 5805: /* if (mobilav!=0) { */
5806: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5807: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5808: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5809: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5810: /* } */
5811: /* } */
5812:
5813: strcpy(fileresprobmorprev,"PRMORPREV-");
5814: sprintf(digit,"%-d",ij);
5815: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5816: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5817: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5818: strcat(fileresprobmorprev,fileresu);
5819: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5820: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5821: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5822: }
5823: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5824: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5825: pstamp(ficresprobmorprev);
5826: 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 5827: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5828: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5829: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5830: }
5831: for(j=1;j<=cptcoveff;j++)
5832: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5833: fprintf(ficresprobmorprev,"\n");
5834:
1.218 brouard 5835: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5836: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5837: fprintf(ficresprobmorprev," p.%-d SE",j);
5838: for(i=1; i<=nlstate;i++)
5839: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5840: }
5841: fprintf(ficresprobmorprev,"\n");
5842:
5843: fprintf(ficgp,"\n# Routine varevsij");
5844: fprintf(ficgp,"\nunset title \n");
5845: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5846: 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");
5847: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5848: /* } */
5849: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5850: pstamp(ficresvij);
5851: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5852: if(popbased==1)
5853: 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);
5854: else
5855: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5856: fprintf(ficresvij,"# Age");
5857: for(i=1; i<=nlstate;i++)
5858: for(j=1; j<=nlstate;j++)
5859: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5860: fprintf(ficresvij,"\n");
5861:
5862: xp=vector(1,npar);
5863: dnewm=matrix(1,nlstate,1,npar);
5864: doldm=matrix(1,nlstate,1,nlstate);
5865: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5866: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5867:
5868: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5869: gpp=vector(nlstate+1,nlstate+ndeath);
5870: gmp=vector(nlstate+1,nlstate+ndeath);
5871: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5872:
1.218 brouard 5873: if(estepm < stepm){
5874: printf ("Problem %d lower than %d\n",estepm, stepm);
5875: }
5876: else hstepm=estepm;
5877: /* For example we decided to compute the life expectancy with the smallest unit */
5878: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5879: nhstepm is the number of hstepm from age to agelim
5880: nstepm is the number of stepm from age to agelim.
5881: Look at function hpijx to understand why because of memory size limitations,
5882: we decided (b) to get a life expectancy respecting the most precise curvature of the
5883: survival function given by stepm (the optimization length). Unfortunately it
5884: means that if the survival funtion is printed every two years of age and if
5885: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5886: results. So we changed our mind and took the option of the best precision.
5887: */
5888: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5889: agelim = AGESUP;
5890: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5891: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5892: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5893: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5894: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5895: gp=matrix(0,nhstepm,1,nlstate);
5896: gm=matrix(0,nhstepm,1,nlstate);
5897:
5898:
5899: for(theta=1; theta <=npar; theta++){
5900: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5901: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5902: }
5903:
1.242 brouard 5904: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5905:
5906: if (popbased==1) {
5907: if(mobilav ==0){
5908: for(i=1; i<=nlstate;i++)
5909: prlim[i][i]=probs[(int)age][i][ij];
5910: }else{ /* mobilav */
5911: for(i=1; i<=nlstate;i++)
5912: prlim[i][i]=mobaverage[(int)age][i][ij];
5913: }
5914: }
5915:
1.235 brouard 5916: 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 5917: for(j=1; j<= nlstate; j++){
5918: for(h=0; h<=nhstepm; h++){
5919: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5920: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5921: }
5922: }
5923: /* Next for computing probability of death (h=1 means
5924: computed over hstepm matrices product = hstepm*stepm months)
5925: as a weighted average of prlim.
5926: */
5927: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5928: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5929: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5930: }
5931: /* end probability of death */
5932:
5933: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5934: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5935:
1.242 brouard 5936: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5937:
5938: if (popbased==1) {
5939: if(mobilav ==0){
5940: for(i=1; i<=nlstate;i++)
5941: prlim[i][i]=probs[(int)age][i][ij];
5942: }else{ /* mobilav */
5943: for(i=1; i<=nlstate;i++)
5944: prlim[i][i]=mobaverage[(int)age][i][ij];
5945: }
5946: }
5947:
1.235 brouard 5948: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5949:
5950: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5951: for(h=0; h<=nhstepm; h++){
5952: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5953: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5954: }
5955: }
5956: /* This for computing probability of death (h=1 means
5957: computed over hstepm matrices product = hstepm*stepm months)
5958: as a weighted average of prlim.
5959: */
5960: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5961: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5962: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5963: }
5964: /* end probability of death */
5965:
5966: for(j=1; j<= nlstate; j++) /* vareij */
5967: for(h=0; h<=nhstepm; h++){
5968: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5969: }
5970:
5971: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5972: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5973: }
5974:
5975: } /* End theta */
5976:
5977: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5978:
5979: for(h=0; h<=nhstepm; h++) /* veij */
5980: for(j=1; j<=nlstate;j++)
5981: for(theta=1; theta <=npar; theta++)
5982: trgradg[h][j][theta]=gradg[h][theta][j];
5983:
5984: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5985: for(theta=1; theta <=npar; theta++)
5986: trgradgp[j][theta]=gradgp[theta][j];
5987:
5988:
5989: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5990: for(i=1;i<=nlstate;i++)
5991: for(j=1;j<=nlstate;j++)
5992: vareij[i][j][(int)age] =0.;
5993:
5994: for(h=0;h<=nhstepm;h++){
5995: for(k=0;k<=nhstepm;k++){
5996: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5997: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5998: for(i=1;i<=nlstate;i++)
5999: for(j=1;j<=nlstate;j++)
6000: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6001: }
6002: }
6003:
6004: /* pptj */
6005: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6006: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6007: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6008: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6009: varppt[j][i]=doldmp[j][i];
6010: /* end ppptj */
6011: /* x centered again */
6012:
1.242 brouard 6013: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6014:
6015: if (popbased==1) {
6016: if(mobilav ==0){
6017: for(i=1; i<=nlstate;i++)
6018: prlim[i][i]=probs[(int)age][i][ij];
6019: }else{ /* mobilav */
6020: for(i=1; i<=nlstate;i++)
6021: prlim[i][i]=mobaverage[(int)age][i][ij];
6022: }
6023: }
6024:
6025: /* This for computing probability of death (h=1 means
6026: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6027: as a weighted average of prlim.
6028: */
1.235 brouard 6029: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6030: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6031: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6032: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6033: }
6034: /* end probability of death */
6035:
6036: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6037: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6038: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6039: for(i=1; i<=nlstate;i++){
6040: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6041: }
6042: }
6043: fprintf(ficresprobmorprev,"\n");
6044:
6045: fprintf(ficresvij,"%.0f ",age );
6046: for(i=1; i<=nlstate;i++)
6047: for(j=1; j<=nlstate;j++){
6048: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6049: }
6050: fprintf(ficresvij,"\n");
6051: free_matrix(gp,0,nhstepm,1,nlstate);
6052: free_matrix(gm,0,nhstepm,1,nlstate);
6053: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6054: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6055: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6056: } /* End age */
6057: free_vector(gpp,nlstate+1,nlstate+ndeath);
6058: free_vector(gmp,nlstate+1,nlstate+ndeath);
6059: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6060: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6061: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6062: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6063: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6064: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6065: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6066: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6067: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6068: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6069: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6070: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6071: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6072: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6073: 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);
6074: /* 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 6075: */
1.218 brouard 6076: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6077: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6078:
1.218 brouard 6079: free_vector(xp,1,npar);
6080: free_matrix(doldm,1,nlstate,1,nlstate);
6081: free_matrix(dnewm,1,nlstate,1,npar);
6082: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6083: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6084: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6085: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6086: fclose(ficresprobmorprev);
6087: fflush(ficgp);
6088: fflush(fichtm);
6089: } /* end varevsij */
1.126 brouard 6090:
6091: /************ Variance of prevlim ******************/
1.269 brouard 6092: void varprevlim(char fileresvpl[], FILE *ficresvpl, double **varpl, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **prlim, double ftolpl, int *ncvyearp, int ij, char strstart[], int nres)
1.126 brouard 6093: {
1.205 brouard 6094: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6095: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6096:
1.268 brouard 6097: double **dnewmpar,**doldm;
1.126 brouard 6098: int i, j, nhstepm, hstepm;
6099: double *xp;
6100: double *gp, *gm;
6101: double **gradg, **trgradg;
1.208 brouard 6102: double **mgm, **mgp;
1.126 brouard 6103: double age,agelim;
6104: int theta;
6105:
6106: pstamp(ficresvpl);
6107: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6108: fprintf(ficresvpl,"# Age ");
6109: if(nresult >=1)
6110: fprintf(ficresvpl," Result# ");
1.126 brouard 6111: for(i=1; i<=nlstate;i++)
6112: fprintf(ficresvpl," %1d-%1d",i,i);
6113: fprintf(ficresvpl,"\n");
6114:
6115: xp=vector(1,npar);
1.268 brouard 6116: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6117: doldm=matrix(1,nlstate,1,nlstate);
6118:
6119: hstepm=1*YEARM; /* Every year of age */
6120: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6121: agelim = AGESUP;
6122: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6123: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6124: if (stepm >= YEARM) hstepm=1;
6125: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6126: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6127: mgp=matrix(1,npar,1,nlstate);
6128: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6129: gp=vector(1,nlstate);
6130: gm=vector(1,nlstate);
6131:
6132: for(theta=1; theta <=npar; theta++){
6133: for(i=1; i<=npar; i++){ /* Computes gradient */
6134: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6135: }
1.209 brouard 6136: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6137: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6138: else
1.235 brouard 6139: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6140: for(i=1;i<=nlstate;i++){
1.126 brouard 6141: gp[i] = prlim[i][i];
1.208 brouard 6142: mgp[theta][i] = prlim[i][i];
6143: }
1.126 brouard 6144: for(i=1; i<=npar; i++) /* Computes gradient */
6145: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6146: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6147: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6148: else
1.235 brouard 6149: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6150: for(i=1;i<=nlstate;i++){
1.126 brouard 6151: gm[i] = prlim[i][i];
1.208 brouard 6152: mgm[theta][i] = prlim[i][i];
6153: }
1.126 brouard 6154: for(i=1;i<=nlstate;i++)
6155: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6156: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6157: } /* End theta */
6158:
6159: trgradg =matrix(1,nlstate,1,npar);
6160:
6161: for(j=1; j<=nlstate;j++)
6162: for(theta=1; theta <=npar; theta++)
6163: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6164: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6165: /* printf("\nmgm mgp %d ",(int)age); */
6166: /* for(j=1; j<=nlstate;j++){ */
6167: /* printf(" %d ",j); */
6168: /* for(theta=1; theta <=npar; theta++) */
6169: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6170: /* printf("\n "); */
6171: /* } */
6172: /* } */
6173: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6174: /* printf("\n gradg %d ",(int)age); */
6175: /* for(j=1; j<=nlstate;j++){ */
6176: /* printf("%d ",j); */
6177: /* for(theta=1; theta <=npar; theta++) */
6178: /* printf("%d %lf ",theta,gradg[theta][j]); */
6179: /* printf("\n "); */
6180: /* } */
6181: /* } */
1.126 brouard 6182:
6183: for(i=1;i<=nlstate;i++)
6184: varpl[i][(int)age] =0.;
1.209 brouard 6185: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6186: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6187: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6188: }else{
1.268 brouard 6189: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6190: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6191: }
1.126 brouard 6192: for(i=1;i<=nlstate;i++)
6193: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6194:
6195: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6196: if(nresult >=1)
6197: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6198: for(i=1; i<=nlstate;i++)
6199: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6200: fprintf(ficresvpl,"\n");
6201: free_vector(gp,1,nlstate);
6202: free_vector(gm,1,nlstate);
1.208 brouard 6203: free_matrix(mgm,1,npar,1,nlstate);
6204: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6205: free_matrix(gradg,1,npar,1,nlstate);
6206: free_matrix(trgradg,1,nlstate,1,npar);
6207: } /* End age */
6208:
6209: free_vector(xp,1,npar);
6210: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6211: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6212:
6213: }
6214:
6215:
6216: /************ Variance of backprevalence limit ******************/
1.269 brouard 6217: void varbrevlim(char fileresvbl[], FILE *ficresvbl, double **varbpl, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **bprlim, double ftolpl, int mobilavproj, int *ncvyearp, int ij, char strstart[], int nres)
1.268 brouard 6218: {
6219: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6220: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6221:
6222: double **dnewmpar,**doldm;
6223: int i, j, nhstepm, hstepm;
6224: double *xp;
6225: double *gp, *gm;
6226: double **gradg, **trgradg;
6227: double **mgm, **mgp;
6228: double age,agelim;
6229: int theta;
6230:
6231: pstamp(ficresvbl);
6232: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6233: fprintf(ficresvbl,"# Age ");
6234: if(nresult >=1)
6235: fprintf(ficresvbl," Result# ");
6236: for(i=1; i<=nlstate;i++)
6237: fprintf(ficresvbl," %1d-%1d",i,i);
6238: fprintf(ficresvbl,"\n");
6239:
6240: xp=vector(1,npar);
6241: dnewmpar=matrix(1,nlstate,1,npar);
6242: doldm=matrix(1,nlstate,1,nlstate);
6243:
6244: hstepm=1*YEARM; /* Every year of age */
6245: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6246: agelim = AGEINF;
6247: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6248: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6249: if (stepm >= YEARM) hstepm=1;
6250: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6251: gradg=matrix(1,npar,1,nlstate);
6252: mgp=matrix(1,npar,1,nlstate);
6253: mgm=matrix(1,npar,1,nlstate);
6254: gp=vector(1,nlstate);
6255: gm=vector(1,nlstate);
6256:
6257: for(theta=1; theta <=npar; theta++){
6258: for(i=1; i<=npar; i++){ /* Computes gradient */
6259: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6260: }
6261: if(mobilavproj > 0 )
6262: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6263: else
6264: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6265: for(i=1;i<=nlstate;i++){
6266: gp[i] = bprlim[i][i];
6267: mgp[theta][i] = bprlim[i][i];
6268: }
6269: for(i=1; i<=npar; i++) /* Computes gradient */
6270: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6271: if(mobilavproj > 0 )
6272: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6273: else
6274: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6275: for(i=1;i<=nlstate;i++){
6276: gm[i] = bprlim[i][i];
6277: mgm[theta][i] = bprlim[i][i];
6278: }
6279: for(i=1;i<=nlstate;i++)
6280: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6281: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6282: } /* End theta */
6283:
6284: trgradg =matrix(1,nlstate,1,npar);
6285:
6286: for(j=1; j<=nlstate;j++)
6287: for(theta=1; theta <=npar; theta++)
6288: trgradg[j][theta]=gradg[theta][j];
6289: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6290: /* printf("\nmgm mgp %d ",(int)age); */
6291: /* for(j=1; j<=nlstate;j++){ */
6292: /* printf(" %d ",j); */
6293: /* for(theta=1; theta <=npar; theta++) */
6294: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6295: /* printf("\n "); */
6296: /* } */
6297: /* } */
6298: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6299: /* printf("\n gradg %d ",(int)age); */
6300: /* for(j=1; j<=nlstate;j++){ */
6301: /* printf("%d ",j); */
6302: /* for(theta=1; theta <=npar; theta++) */
6303: /* printf("%d %lf ",theta,gradg[theta][j]); */
6304: /* printf("\n "); */
6305: /* } */
6306: /* } */
6307:
6308: for(i=1;i<=nlstate;i++)
6309: varbpl[i][(int)age] =0.;
6310: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6311: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6312: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6313: }else{
6314: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6315: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6316: }
6317: for(i=1;i<=nlstate;i++)
6318: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6319:
6320: fprintf(ficresvbl,"%.0f ",age );
6321: if(nresult >=1)
6322: fprintf(ficresvbl,"%d ",nres );
6323: for(i=1; i<=nlstate;i++)
6324: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6325: fprintf(ficresvbl,"\n");
6326: free_vector(gp,1,nlstate);
6327: free_vector(gm,1,nlstate);
6328: free_matrix(mgm,1,npar,1,nlstate);
6329: free_matrix(mgp,1,npar,1,nlstate);
6330: free_matrix(gradg,1,npar,1,nlstate);
6331: free_matrix(trgradg,1,nlstate,1,npar);
6332: } /* End age */
6333:
6334: free_vector(xp,1,npar);
6335: free_matrix(doldm,1,nlstate,1,npar);
6336: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6337:
6338: }
6339:
6340: /************ Variance of one-step probabilities ******************/
6341: 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 6342: {
6343: int i, j=0, k1, l1, tj;
6344: int k2, l2, j1, z1;
6345: int k=0, l;
6346: int first=1, first1, first2;
6347: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6348: double **dnewm,**doldm;
6349: double *xp;
6350: double *gp, *gm;
6351: double **gradg, **trgradg;
6352: double **mu;
6353: double age, cov[NCOVMAX+1];
6354: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6355: int theta;
6356: char fileresprob[FILENAMELENGTH];
6357: char fileresprobcov[FILENAMELENGTH];
6358: char fileresprobcor[FILENAMELENGTH];
6359: double ***varpij;
6360:
6361: strcpy(fileresprob,"PROB_");
6362: strcat(fileresprob,fileres);
6363: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6364: printf("Problem with resultfile: %s\n", fileresprob);
6365: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6366: }
6367: strcpy(fileresprobcov,"PROBCOV_");
6368: strcat(fileresprobcov,fileresu);
6369: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6370: printf("Problem with resultfile: %s\n", fileresprobcov);
6371: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6372: }
6373: strcpy(fileresprobcor,"PROBCOR_");
6374: strcat(fileresprobcor,fileresu);
6375: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6376: printf("Problem with resultfile: %s\n", fileresprobcor);
6377: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6378: }
6379: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6380: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6381: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6382: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6383: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6384: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6385: pstamp(ficresprob);
6386: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6387: fprintf(ficresprob,"# Age");
6388: pstamp(ficresprobcov);
6389: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6390: fprintf(ficresprobcov,"# Age");
6391: pstamp(ficresprobcor);
6392: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6393: fprintf(ficresprobcor,"# Age");
1.126 brouard 6394:
6395:
1.222 brouard 6396: for(i=1; i<=nlstate;i++)
6397: for(j=1; j<=(nlstate+ndeath);j++){
6398: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6399: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6400: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6401: }
6402: /* fprintf(ficresprob,"\n");
6403: fprintf(ficresprobcov,"\n");
6404: fprintf(ficresprobcor,"\n");
6405: */
6406: xp=vector(1,npar);
6407: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6408: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6409: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6410: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6411: first=1;
6412: fprintf(ficgp,"\n# Routine varprob");
6413: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6414: fprintf(fichtm,"\n");
6415:
1.266 brouard 6416: 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 6417: 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);
6418: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6419: and drawn. It helps understanding how is the covariance between two incidences.\
6420: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6421: 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 6422: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6423: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6424: standard deviations wide on each axis. <br>\
6425: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6426: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6427: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6428:
1.222 brouard 6429: cov[1]=1;
6430: /* tj=cptcoveff; */
1.225 brouard 6431: tj = (int) pow(2,cptcoveff);
1.222 brouard 6432: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6433: j1=0;
1.224 brouard 6434: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6435: if (cptcovn>0) {
6436: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6437: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6438: fprintf(ficresprob, "**********\n#\n");
6439: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6440: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6441: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6442:
1.222 brouard 6443: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6444: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6445: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6446:
6447:
1.222 brouard 6448: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6449: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6450: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6451:
1.222 brouard 6452: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6453: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6454: fprintf(ficresprobcor, "**********\n#");
6455: if(invalidvarcomb[j1]){
6456: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6457: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6458: continue;
6459: }
6460: }
6461: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6462: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6463: gp=vector(1,(nlstate)*(nlstate+ndeath));
6464: gm=vector(1,(nlstate)*(nlstate+ndeath));
6465: for (age=bage; age<=fage; age ++){
6466: cov[2]=age;
6467: if(nagesqr==1)
6468: cov[3]= age*age;
6469: for (k=1; k<=cptcovn;k++) {
6470: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6471: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6472: * 1 1 1 1 1
6473: * 2 2 1 1 1
6474: * 3 1 2 1 1
6475: */
6476: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6477: }
6478: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6479: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6480: for (k=1; k<=cptcovprod;k++)
6481: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6482:
6483:
1.222 brouard 6484: for(theta=1; theta <=npar; theta++){
6485: for(i=1; i<=npar; i++)
6486: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6487:
1.222 brouard 6488: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6489:
1.222 brouard 6490: k=0;
6491: for(i=1; i<= (nlstate); i++){
6492: for(j=1; j<=(nlstate+ndeath);j++){
6493: k=k+1;
6494: gp[k]=pmmij[i][j];
6495: }
6496: }
1.220 brouard 6497:
1.222 brouard 6498: for(i=1; i<=npar; i++)
6499: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6500:
1.222 brouard 6501: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6502: k=0;
6503: for(i=1; i<=(nlstate); i++){
6504: for(j=1; j<=(nlstate+ndeath);j++){
6505: k=k+1;
6506: gm[k]=pmmij[i][j];
6507: }
6508: }
1.220 brouard 6509:
1.222 brouard 6510: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6511: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6512: }
1.126 brouard 6513:
1.222 brouard 6514: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6515: for(theta=1; theta <=npar; theta++)
6516: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6517:
1.222 brouard 6518: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6519: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6520:
1.222 brouard 6521: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6522:
1.222 brouard 6523: k=0;
6524: for(i=1; i<=(nlstate); i++){
6525: for(j=1; j<=(nlstate+ndeath);j++){
6526: k=k+1;
6527: mu[k][(int) age]=pmmij[i][j];
6528: }
6529: }
6530: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6531: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6532: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6533:
1.222 brouard 6534: /*printf("\n%d ",(int)age);
6535: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6536: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6537: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6538: }*/
1.220 brouard 6539:
1.222 brouard 6540: fprintf(ficresprob,"\n%d ",(int)age);
6541: fprintf(ficresprobcov,"\n%d ",(int)age);
6542: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6543:
1.222 brouard 6544: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6545: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6546: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6547: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6548: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6549: }
6550: i=0;
6551: for (k=1; k<=(nlstate);k++){
6552: for (l=1; l<=(nlstate+ndeath);l++){
6553: i++;
6554: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6555: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6556: for (j=1; j<=i;j++){
6557: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6558: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6559: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6560: }
6561: }
6562: }/* end of loop for state */
6563: } /* end of loop for age */
6564: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6565: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6566: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6567: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6568:
6569: /* Confidence intervalle of pij */
6570: /*
6571: fprintf(ficgp,"\nunset parametric;unset label");
6572: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6573: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6574: 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);
6575: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6576: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6577: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6578: */
6579:
6580: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6581: first1=1;first2=2;
6582: for (k2=1; k2<=(nlstate);k2++){
6583: for (l2=1; l2<=(nlstate+ndeath);l2++){
6584: if(l2==k2) continue;
6585: j=(k2-1)*(nlstate+ndeath)+l2;
6586: for (k1=1; k1<=(nlstate);k1++){
6587: for (l1=1; l1<=(nlstate+ndeath);l1++){
6588: if(l1==k1) continue;
6589: i=(k1-1)*(nlstate+ndeath)+l1;
6590: if(i<=j) continue;
6591: for (age=bage; age<=fage; age ++){
6592: if ((int)age %5==0){
6593: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6594: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6595: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6596: mu1=mu[i][(int) age]/stepm*YEARM ;
6597: mu2=mu[j][(int) age]/stepm*YEARM;
6598: c12=cv12/sqrt(v1*v2);
6599: /* Computing eigen value of matrix of covariance */
6600: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6601: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6602: if ((lc2 <0) || (lc1 <0) ){
6603: if(first2==1){
6604: first1=0;
6605: 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);
6606: }
6607: 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);
6608: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6609: /* lc2=fabs(lc2); */
6610: }
1.220 brouard 6611:
1.222 brouard 6612: /* Eigen vectors */
6613: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6614: /*v21=sqrt(1.-v11*v11); *//* error */
6615: v21=(lc1-v1)/cv12*v11;
6616: v12=-v21;
6617: v22=v11;
6618: tnalp=v21/v11;
6619: if(first1==1){
6620: first1=0;
6621: 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);
6622: }
6623: 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);
6624: /*printf(fignu*/
6625: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6626: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6627: if(first==1){
6628: first=0;
6629: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6630: fprintf(ficgp,"\nset parametric;unset label");
6631: 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);
6632: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6633: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6634: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6635: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6636: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6637: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6638: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6639: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6640: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6641: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6642: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6643: 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 6644: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6645: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6646: }else{
6647: first=0;
6648: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6649: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6650: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6651: 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 6652: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6653: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6654: }/* if first */
6655: } /* age mod 5 */
6656: } /* end loop age */
6657: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6658: first=1;
6659: } /*l12 */
6660: } /* k12 */
6661: } /*l1 */
6662: }/* k1 */
6663: } /* loop on combination of covariates j1 */
6664: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6665: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6666: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6667: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6668: free_vector(xp,1,npar);
6669: fclose(ficresprob);
6670: fclose(ficresprobcov);
6671: fclose(ficresprobcor);
6672: fflush(ficgp);
6673: fflush(fichtmcov);
6674: }
1.126 brouard 6675:
6676:
6677: /******************* Printing html file ***********/
1.201 brouard 6678: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6679: int lastpass, int stepm, int weightopt, char model[],\
6680: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6681: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6682: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6683: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6684: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6685:
6686: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6687: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6688: </ul>");
1.237 brouard 6689: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6690: </ul>", model);
1.214 brouard 6691: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6692: 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",
6693: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6694: 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 6695: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6696: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6697: fprintf(fichtm,"\
6698: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6699: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6700: fprintf(fichtm,"\
1.217 brouard 6701: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6702: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6703: fprintf(fichtm,"\
1.126 brouard 6704: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6705: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6706: fprintf(fichtm,"\
1.217 brouard 6707: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6708: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6709: fprintf(fichtm,"\
1.211 brouard 6710: - (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 6711: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6712: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6713: if(prevfcast==1){
6714: fprintf(fichtm,"\
6715: - Prevalence projections by age and states: \
1.201 brouard 6716: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6717: }
1.126 brouard 6718:
6719:
1.225 brouard 6720: m=pow(2,cptcoveff);
1.222 brouard 6721: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6722:
1.264 brouard 6723: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6724:
6725: jj1=0;
6726:
6727: fprintf(fichtm," \n<ul>");
6728: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6729: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6730: if(m != 1 && TKresult[nres]!= k1)
6731: continue;
6732: jj1++;
6733: if (cptcovn > 0) {
6734: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6735: for (cpt=1; cpt<=cptcoveff;cpt++){
6736: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6737: }
6738: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6739: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6740: }
6741: fprintf(fichtm,"\">");
6742:
6743: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6744: fprintf(fichtm,"************ Results for covariates");
6745: for (cpt=1; cpt<=cptcoveff;cpt++){
6746: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6747: }
6748: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6749: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6750: }
6751: if(invalidvarcomb[k1]){
6752: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6753: continue;
6754: }
6755: fprintf(fichtm,"</a></li>");
6756: } /* cptcovn >0 */
6757: }
6758: fprintf(fichtm," \n</ul>");
6759:
1.222 brouard 6760: jj1=0;
1.237 brouard 6761:
6762: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6763: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6764: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6765: continue;
1.220 brouard 6766:
1.222 brouard 6767: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6768: jj1++;
6769: if (cptcovn > 0) {
1.264 brouard 6770: fprintf(fichtm,"\n<p><a name=\"rescov");
6771: for (cpt=1; cpt<=cptcoveff;cpt++){
6772: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6773: }
6774: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6775: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6776: }
6777: fprintf(fichtm,"\"</a>");
6778:
1.222 brouard 6779: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6780: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6781: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6782: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6783: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6784: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6785: }
1.237 brouard 6786: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6787: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6788: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6789: }
6790:
1.230 brouard 6791: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6792: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6793: if(invalidvarcomb[k1]){
6794: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6795: printf("\nCombination (%d) ignored because no cases \n",k1);
6796: continue;
6797: }
6798: }
6799: /* aij, bij */
1.259 brouard 6800: 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 6801: <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 6802: /* Pij */
1.241 brouard 6803: 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> \
6804: <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 6805: /* Quasi-incidences */
6806: 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 6807: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6808: 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 6809: 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> \
6810: <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 6811: /* Survival functions (period) in state j */
6812: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6813: 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> \
6814: <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 6815: }
6816: /* State specific survival functions (period) */
6817: for(cpt=1; cpt<=nlstate;cpt++){
6818: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6819: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6820: <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 6821: }
6822: /* Period (stable) prevalence in each health state */
6823: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6824: 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> \
6825: <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 6826: }
6827: if(backcast==1){
6828: /* Period (stable) back prevalence in each health state */
6829: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6830: 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 6831: <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 6832: }
1.217 brouard 6833: }
1.222 brouard 6834: if(prevfcast==1){
6835: /* Projection of prevalence up to period (stable) prevalence in each health state */
6836: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6837: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), from year %.1f up to year %.1f tending to period (stable) prevalence in state %d. Or probability to be in state %d being in an observed weighted state (from 1 to %d). <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6838: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateproj1, dateproj2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6839: }
6840: }
1.268 brouard 6841: if(backcast==1){
6842: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6843: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6844: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6845: from year %.1f up to year %.1f (probably close to stable [mixed] back prevalence in state %d (randomness in cross-sectional prevalence is not taken into \
6846: account but can visually be appreciated). Or probability to have been in an state %d, knowing that the person was in either state (1 or %d) \
6847: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6848: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateback1, dateback2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 6849: }
6850: }
1.220 brouard 6851:
1.222 brouard 6852: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6853: 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> \
6854: <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 6855: }
6856: /* } /\* end i1 *\/ */
6857: }/* End k1 */
6858: fprintf(fichtm,"</ul>");
1.126 brouard 6859:
1.222 brouard 6860: fprintf(fichtm,"\
1.126 brouard 6861: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6862: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6863: - 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 6864: But because parameters are usually highly correlated (a higher incidence of disability \
6865: and a higher incidence of recovery can give very close observed transition) it might \
6866: be very useful to look not only at linear confidence intervals estimated from the \
6867: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6868: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6869: covariance matrix of the one-step probabilities. \
6870: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6871:
1.222 brouard 6872: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6873: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6874: fprintf(fichtm,"\
1.126 brouard 6875: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6876: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6877:
1.222 brouard 6878: fprintf(fichtm,"\
1.126 brouard 6879: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6880: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6881: fprintf(fichtm,"\
1.126 brouard 6882: - 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): \
6883: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6884: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6885: fprintf(fichtm,"\
1.126 brouard 6886: - (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): \
6887: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6888: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6889: fprintf(fichtm,"\
1.128 brouard 6890: - 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 6891: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6892: fprintf(fichtm,"\
1.128 brouard 6893: - 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 6894: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6895: fprintf(fichtm,"\
1.126 brouard 6896: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6897: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6898:
6899: /* if(popforecast==1) fprintf(fichtm,"\n */
6900: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6901: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6902: /* <br>",fileres,fileres,fileres,fileres); */
6903: /* else */
6904: /* 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 6905: fflush(fichtm);
6906: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6907:
1.225 brouard 6908: m=pow(2,cptcoveff);
1.222 brouard 6909: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6910:
1.222 brouard 6911: jj1=0;
1.237 brouard 6912:
1.241 brouard 6913: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6914: for(k1=1; k1<=m;k1++){
1.253 brouard 6915: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6916: continue;
1.222 brouard 6917: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6918: jj1++;
1.126 brouard 6919: if (cptcovn > 0) {
6920: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6921: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6922: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6923: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6924: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6925: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6926: }
6927:
1.126 brouard 6928: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6929:
1.222 brouard 6930: if(invalidvarcomb[k1]){
6931: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6932: continue;
6933: }
1.126 brouard 6934: }
6935: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6936: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6937: 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 6938: <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 6939: }
6940: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6941: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6942: true period expectancies (those weighted with period prevalences are also\
6943: drawn in addition to the population based expectancies computed using\
1.241 brouard 6944: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6945: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6946: /* } /\* end i1 *\/ */
6947: }/* End k1 */
1.241 brouard 6948: }/* End nres */
1.222 brouard 6949: fprintf(fichtm,"</ul>");
6950: fflush(fichtm);
1.126 brouard 6951: }
6952:
6953: /******************* Gnuplot file **************/
1.270 brouard 6954: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double bage, double fage , int prevfcast, int backcast, char pathc[], double p[], int offyear, int offbyear){
1.126 brouard 6955:
6956: char dirfileres[132],optfileres[132];
1.264 brouard 6957: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6958: 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 6959: int lv=0, vlv=0, kl=0;
1.130 brouard 6960: int ng=0;
1.201 brouard 6961: int vpopbased;
1.223 brouard 6962: int ioffset; /* variable offset for columns */
1.270 brouard 6963: int iyearc=1; /* variable column for year of projection */
6964: int iagec=1; /* variable column for age of projection */
1.235 brouard 6965: int nres=0; /* Index of resultline */
1.266 brouard 6966: int istart=1; /* For starting graphs in projections */
1.219 brouard 6967:
1.126 brouard 6968: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6969: /* printf("Problem with file %s",optionfilegnuplot); */
6970: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6971: /* } */
6972:
6973: /*#ifdef windows */
6974: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6975: /*#endif */
1.225 brouard 6976: m=pow(2,cptcoveff);
1.126 brouard 6977:
1.274 brouard 6978: /* diagram of the model */
6979: fprintf(ficgp,"\n#Diagram of the model \n");
6980: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
6981: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
6982: fprintf(ficgp,"\n#Peripheral arrows\nset for [i=1:%d] for [j=1:%d] arrow i*10+j from cos(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.95*(cos(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) - cos(pi*((1-(%d/2)*2./%d)/2+(j-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta2:0)), -0.95*(sin(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) - sin(pi*((1-(%d/2)*2./%d)/2+(j-1)*2./%d))+( i!=j?(i-j)/abs(i-j)*delta2:0)) ls (i < j? 1:2)\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
6983:
6984: fprintf(ficgp,"\n#Centripete arrows (turning in other direction (1-i) instead of (i-1)) \nset for [i=1:%d] arrow (%d+1)*10+i from cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.80*(cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) ), -0.80*(sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) + yoff ) ls 4\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
6985: fprintf(ficgp,"\n#show arrow\nunset label\n");
6986: fprintf(ficgp,"\n#States labels, starting from 2 (2-i) instead of (1-i), was (i-1)\nset for [i=1:%d] label i sprintf(\"State %%d\",i) center at cos(pi*((1-(%d/2)*2./%d)/2+(2-i)*2./%d)), yoff+sin(pi*((1-(%d/2)*2./%d)/2+(2-i)*2./%d)) font \"helvetica, 16\" tc rgbcolor \"blue\"\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
6987: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
6988: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
6989: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
6990: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
6991:
1.202 brouard 6992: /* Contribution to likelihood */
6993: /* Plot the probability implied in the likelihood */
1.223 brouard 6994: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6995: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6996: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6997: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6998: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6999: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7000: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7001: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7002: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7003: 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));
7004: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7005: 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));
7006: for (i=1; i<= nlstate ; i ++) {
7007: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7008: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7009: 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);
7010: for (j=2; j<= nlstate+ndeath ; j ++) {
7011: 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);
7012: }
7013: fprintf(ficgp,";\nset out; unset ylabel;\n");
7014: }
7015: /* 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 */
7016: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7017: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7018: fprintf(ficgp,"\nset out;unset log\n");
7019: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7020:
1.126 brouard 7021: strcpy(dirfileres,optionfilefiname);
7022: strcpy(optfileres,"vpl");
1.223 brouard 7023: /* 1eme*/
1.238 brouard 7024: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7025: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7026: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7027: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7028: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7029: continue;
7030: /* We are interested in selected combination by the resultline */
1.246 brouard 7031: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7032: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7033: strcpy(gplotlabel,"(");
1.238 brouard 7034: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7035: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7036: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7037: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7038: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7039: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7040: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7041: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7042: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7043: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7044: }
7045: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7046: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7047: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7048: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7049: }
7050: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7051: /* printf("\n#\n"); */
1.238 brouard 7052: fprintf(ficgp,"\n#\n");
7053: if(invalidvarcomb[k1]){
1.260 brouard 7054: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7055: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7056: continue;
7057: }
1.235 brouard 7058:
1.241 brouard 7059: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7060: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7061: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7062: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7063: 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);
7064: /* 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); */
7065: /* k1-1 error should be nres-1*/
1.238 brouard 7066: for (i=1; i<= nlstate ; i ++) {
7067: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7068: else fprintf(ficgp," %%*lf (%%*lf)");
7069: }
1.260 brouard 7070: 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 7071: for (i=1; i<= nlstate ; i ++) {
7072: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7073: else fprintf(ficgp," %%*lf (%%*lf)");
7074: }
1.260 brouard 7075: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7076: for (i=1; i<= nlstate ; i ++) {
7077: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7078: else fprintf(ficgp," %%*lf (%%*lf)");
7079: }
1.265 brouard 7080: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7081:
7082: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7083: if(cptcoveff ==0){
1.271 brouard 7084: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7085: }else{
7086: kl=0;
7087: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7088: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7089: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7090: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7091: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7092: vlv= nbcode[Tvaraff[k]][lv];
7093: kl++;
7094: /* 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 *\/ */
7095: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7096: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7097: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7098: if(k==cptcoveff){
7099: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7100: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7101: }else{
7102: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7103: kl++;
7104: }
7105: } /* end covariate */
7106: } /* end if no covariate */
7107:
1.238 brouard 7108: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7109: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7110: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7111: if(cptcoveff ==0){
1.245 brouard 7112: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7113: }else{
7114: kl=0;
7115: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7116: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7117: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7118: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7119: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7120: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7121: kl++;
1.238 brouard 7122: /* 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 *\/ */
7123: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7124: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7125: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7126: if(k==cptcoveff){
1.245 brouard 7127: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7128: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7129: }else{
7130: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7131: kl++;
7132: }
7133: } /* end covariate */
7134: } /* end if no covariate */
1.268 brouard 7135: if(backcast == 1){
7136: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7137: /* k1-1 error should be nres-1*/
7138: for (i=1; i<= nlstate ; i ++) {
7139: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7140: else fprintf(ficgp," %%*lf (%%*lf)");
7141: }
1.271 brouard 7142: fprintf(ficgp,"\" t\"Backward (stable) prevalence\" w l lt 6 dt 3,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 7143: for (i=1; i<= nlstate ; i ++) {
7144: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7145: else fprintf(ficgp," %%*lf (%%*lf)");
7146: }
1.276 brouard 7147: fprintf(ficgp,"\" t\"95%% CI\" w l lt 4,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 7148: for (i=1; i<= nlstate ; i ++) {
7149: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7150: else fprintf(ficgp," %%*lf (%%*lf)");
7151: }
1.274 brouard 7152: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7153: } /* end if backprojcast */
1.238 brouard 7154: } /* end if backcast */
1.276 brouard 7155: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7156: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7157: } /* nres */
1.201 brouard 7158: } /* k1 */
7159: } /* cpt */
1.235 brouard 7160:
7161:
1.126 brouard 7162: /*2 eme*/
1.238 brouard 7163: for (k1=1; k1<= m ; k1 ++){
7164: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7165: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7166: continue;
7167: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7168: strcpy(gplotlabel,"(");
1.238 brouard 7169: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7170: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7171: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7172: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7173: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7174: vlv= nbcode[Tvaraff[k]][lv];
7175: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7176: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7177: }
1.237 brouard 7178: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7179: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7180: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7181: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7182: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7183: }
1.264 brouard 7184: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7185: fprintf(ficgp,"\n#\n");
1.223 brouard 7186: if(invalidvarcomb[k1]){
7187: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7188: continue;
7189: }
1.219 brouard 7190:
1.241 brouard 7191: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7192: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7193: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7194: if(vpopbased==0){
1.238 brouard 7195: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7196: }else
1.238 brouard 7197: fprintf(ficgp,"\nreplot ");
7198: for (i=1; i<= nlstate+1 ; i ++) {
7199: k=2*i;
1.261 brouard 7200: 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 7201: for (j=1; j<= nlstate+1 ; j ++) {
7202: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7203: else fprintf(ficgp," %%*lf (%%*lf)");
7204: }
7205: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7206: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7207: 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 7208: for (j=1; j<= nlstate+1 ; j ++) {
7209: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7210: else fprintf(ficgp," %%*lf (%%*lf)");
7211: }
7212: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7213: 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 7214: for (j=1; j<= nlstate+1 ; j ++) {
7215: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7216: else fprintf(ficgp," %%*lf (%%*lf)");
7217: }
7218: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7219: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7220: } /* state */
7221: } /* vpopbased */
1.264 brouard 7222: 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 7223: } /* end nres */
7224: } /* k1 end 2 eme*/
7225:
7226:
7227: /*3eme*/
7228: for (k1=1; k1<= m ; k1 ++){
7229: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7230: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7231: continue;
7232:
7233: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7234: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7235: strcpy(gplotlabel,"(");
1.238 brouard 7236: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7237: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7238: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7239: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7240: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7241: vlv= nbcode[Tvaraff[k]][lv];
7242: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7243: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7244: }
7245: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7246: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7247: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7248: }
1.264 brouard 7249: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7250: fprintf(ficgp,"\n#\n");
7251: if(invalidvarcomb[k1]){
7252: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7253: continue;
7254: }
7255:
7256: /* k=2+nlstate*(2*cpt-2); */
7257: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7258: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7259: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7260: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7261: 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 7262: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7263: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7264: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7265: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7266: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7267: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7268:
1.238 brouard 7269: */
7270: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7271: 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 7272: /* 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 7273:
1.238 brouard 7274: }
1.261 brouard 7275: 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 7276: }
1.264 brouard 7277: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7278: } /* end nres */
7279: } /* end kl 3eme */
1.126 brouard 7280:
1.223 brouard 7281: /* 4eme */
1.201 brouard 7282: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7283: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7284: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7285: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7286: continue;
1.238 brouard 7287: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7288: strcpy(gplotlabel,"(");
1.238 brouard 7289: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7290: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7291: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7292: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7293: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7294: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7295: vlv= nbcode[Tvaraff[k]][lv];
7296: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7297: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7298: }
7299: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7300: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7301: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7302: }
1.264 brouard 7303: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7304: fprintf(ficgp,"\n#\n");
7305: if(invalidvarcomb[k1]){
7306: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7307: continue;
1.223 brouard 7308: }
1.238 brouard 7309:
1.241 brouard 7310: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7311: 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 7312: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7313: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7314: k=3;
7315: for (i=1; i<= nlstate ; i ++){
7316: if(i==1){
7317: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7318: }else{
7319: fprintf(ficgp,", '' ");
7320: }
7321: l=(nlstate+ndeath)*(i-1)+1;
7322: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7323: for (j=2; j<= nlstate+ndeath ; j ++)
7324: fprintf(ficgp,"+$%d",k+l+j-1);
7325: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7326: } /* nlstate */
1.264 brouard 7327: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7328: } /* end cpt state*/
7329: } /* end nres */
7330: } /* end covariate k1 */
7331:
1.220 brouard 7332: /* 5eme */
1.201 brouard 7333: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7334: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7335: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7336: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7337: continue;
1.238 brouard 7338: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7339: strcpy(gplotlabel,"(");
1.238 brouard 7340: 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);
7341: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7342: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7343: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7344: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7345: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7346: vlv= nbcode[Tvaraff[k]][lv];
7347: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7348: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7349: }
7350: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7351: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7352: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7353: }
1.264 brouard 7354: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7355: fprintf(ficgp,"\n#\n");
7356: if(invalidvarcomb[k1]){
7357: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7358: continue;
7359: }
1.227 brouard 7360:
1.241 brouard 7361: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7362: 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 7363: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7364: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7365: k=3;
7366: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7367: if(j==1)
7368: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7369: else
7370: fprintf(ficgp,", '' ");
7371: l=(nlstate+ndeath)*(cpt-1) +j;
7372: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7373: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7374: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7375: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7376: } /* nlstate */
7377: fprintf(ficgp,", '' ");
7378: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7379: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7380: l=(nlstate+ndeath)*(cpt-1) +j;
7381: if(j < nlstate)
7382: fprintf(ficgp,"$%d +",k+l);
7383: else
7384: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7385: }
1.264 brouard 7386: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7387: } /* end cpt state*/
7388: } /* end covariate */
7389: } /* end nres */
1.227 brouard 7390:
1.220 brouard 7391: /* 6eme */
1.202 brouard 7392: /* CV preval stable (period) for each covariate */
1.237 brouard 7393: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7394: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7395: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7396: continue;
1.255 brouard 7397: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7398: strcpy(gplotlabel,"(");
1.211 brouard 7399: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7400: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7401: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7402: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7403: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7404: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7405: vlv= nbcode[Tvaraff[k]][lv];
7406: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7407: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7408: }
1.237 brouard 7409: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7410: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7411: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7412: }
1.264 brouard 7413: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7414: fprintf(ficgp,"\n#\n");
1.223 brouard 7415: if(invalidvarcomb[k1]){
1.227 brouard 7416: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7417: continue;
1.223 brouard 7418: }
1.227 brouard 7419:
1.241 brouard 7420: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7421: 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 7422: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7423: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7424: k=3; /* Offset */
1.255 brouard 7425: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7426: if(i==1)
7427: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7428: else
7429: fprintf(ficgp,", '' ");
1.255 brouard 7430: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7431: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7432: for (j=2; j<= nlstate ; j ++)
7433: fprintf(ficgp,"+$%d",k+l+j-1);
7434: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7435: } /* nlstate */
1.264 brouard 7436: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7437: } /* end cpt state*/
7438: } /* end covariate */
1.227 brouard 7439:
7440:
1.220 brouard 7441: /* 7eme */
1.218 brouard 7442: if(backcast == 1){
1.217 brouard 7443: /* CV back preval stable (period) for each covariate */
1.237 brouard 7444: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7445: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7446: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7447: continue;
1.268 brouard 7448: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7449: strcpy(gplotlabel,"(");
7450: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7451: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7452: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7453: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7454: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7455: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7456: vlv= nbcode[Tvaraff[k]][lv];
7457: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7458: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7459: }
1.237 brouard 7460: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7461: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7462: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7463: }
1.264 brouard 7464: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7465: fprintf(ficgp,"\n#\n");
7466: if(invalidvarcomb[k1]){
7467: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7468: continue;
7469: }
7470:
1.241 brouard 7471: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7472: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 7473: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7474: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7475: k=3; /* Offset */
1.268 brouard 7476: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7477: if(i==1)
7478: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7479: else
7480: fprintf(ficgp,", '' ");
7481: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7482: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7483: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7484: /* 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 7485: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7486: /* for (j=2; j<= nlstate ; j ++) */
7487: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7488: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7489: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7490: } /* nlstate */
1.264 brouard 7491: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7492: } /* end cpt state*/
7493: } /* end covariate */
7494: } /* End if backcast */
7495:
1.223 brouard 7496: /* 8eme */
1.218 brouard 7497: if(prevfcast==1){
7498: /* Projection from cross-sectional to stable (period) for each covariate */
7499:
1.237 brouard 7500: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7501: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7502: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7503: continue;
1.211 brouard 7504: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7505: strcpy(gplotlabel,"(");
1.227 brouard 7506: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7507: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7508: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7509: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7510: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7511: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7512: vlv= nbcode[Tvaraff[k]][lv];
7513: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7514: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7515: }
1.237 brouard 7516: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7517: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7518: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7519: }
1.264 brouard 7520: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7521: fprintf(ficgp,"\n#\n");
7522: if(invalidvarcomb[k1]){
7523: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7524: continue;
7525: }
7526:
7527: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7528: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7529: 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 7530: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7531: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7532:
7533: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7534: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7535: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7536: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7537: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7538: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7539: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7540: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7541: if(i==istart){
1.227 brouard 7542: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7543: }else{
7544: fprintf(ficgp,",\\\n '' ");
7545: }
7546: if(cptcoveff ==0){ /* No covariate */
7547: ioffset=2; /* Age is in 2 */
7548: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7549: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7550: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7551: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7552: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7553: if(i==nlstate+1){
1.270 brouard 7554: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7555: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7556: fprintf(ficgp,",\\\n '' ");
7557: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7558: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7559: offyear, \
1.268 brouard 7560: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7561: }else
1.227 brouard 7562: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7563: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7564: }else{ /* more than 2 covariates */
1.270 brouard 7565: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7566: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7567: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7568: iyearc=ioffset-1;
7569: iagec=ioffset;
1.227 brouard 7570: fprintf(ficgp," u %d:(",ioffset);
7571: kl=0;
7572: strcpy(gplotcondition,"(");
7573: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7574: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7575: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7576: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7577: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7578: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7579: kl++;
7580: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7581: kl++;
7582: if(k <cptcoveff && cptcoveff>1)
7583: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7584: }
7585: strcpy(gplotcondition+strlen(gplotcondition),")");
7586: /* 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 *\/ */
7587: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7588: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7589: /* '' 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*/
7590: if(i==nlstate+1){
1.270 brouard 7591: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7592: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7593: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7594: fprintf(ficgp," u %d:(",iagec);
7595: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7596: iyearc, iagec, offyear, \
7597: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7598: /* '' 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 7599: }else{
7600: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7601: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7602: }
7603: } /* end if covariate */
7604: } /* nlstate */
1.264 brouard 7605: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7606: } /* end cpt state*/
7607: } /* end covariate */
7608: } /* End if prevfcast */
1.227 brouard 7609:
1.268 brouard 7610: if(backcast==1){
7611: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7612:
7613: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7614: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7615: if(m != 1 && TKresult[nres]!= k1)
7616: continue;
7617: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7618: strcpy(gplotlabel,"(");
7619: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7620: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7621: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7622: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7623: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7624: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7625: vlv= nbcode[Tvaraff[k]][lv];
7626: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7627: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7628: }
7629: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7630: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7631: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7632: }
7633: strcpy(gplotlabel+strlen(gplotlabel),")");
7634: fprintf(ficgp,"\n#\n");
7635: if(invalidvarcomb[k1]){
7636: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7637: continue;
7638: }
7639:
7640: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7641: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7642: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7643: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7644: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7645:
7646: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7647: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7648: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7649: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7650: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7651: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7652: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7653: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7654: if(i==istart){
7655: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7656: }else{
7657: fprintf(ficgp,",\\\n '' ");
7658: }
7659: if(cptcoveff ==0){ /* No covariate */
7660: ioffset=2; /* Age is in 2 */
7661: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7662: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7663: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7664: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7665: fprintf(ficgp," u %d:(", ioffset);
7666: if(i==nlstate+1){
1.270 brouard 7667: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7668: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7669: fprintf(ficgp,",\\\n '' ");
7670: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7671: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7672: offbyear, \
7673: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7674: }else
7675: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7676: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7677: }else{ /* more than 2 covariates */
1.270 brouard 7678: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7679: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7680: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7681: iyearc=ioffset-1;
7682: iagec=ioffset;
1.268 brouard 7683: fprintf(ficgp," u %d:(",ioffset);
7684: kl=0;
7685: strcpy(gplotcondition,"(");
7686: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7687: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7688: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7689: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7690: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7691: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7692: kl++;
7693: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7694: kl++;
7695: if(k <cptcoveff && cptcoveff>1)
7696: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7697: }
7698: strcpy(gplotcondition+strlen(gplotcondition),")");
7699: /* 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 *\/ */
7700: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7701: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7702: /* '' 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*/
7703: if(i==nlstate+1){
1.270 brouard 7704: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7705: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7706: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7707: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7708: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7709: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7710: iyearc,iagec,offbyear, \
7711: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7712: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7713: }else{
7714: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7715: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7716: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7717: }
7718: } /* end if covariate */
7719: } /* nlstate */
7720: fprintf(ficgp,"\nset out; unset label;\n");
7721: } /* end cpt state*/
7722: } /* end covariate */
7723: } /* End if backcast */
7724:
1.227 brouard 7725:
1.238 brouard 7726: /* 9eme writing MLE parameters */
7727: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7728: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7729: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7730: for(k=1; k <=(nlstate+ndeath); k++){
7731: if (k != i) {
1.227 brouard 7732: fprintf(ficgp,"# current state %d\n",k);
7733: for(j=1; j <=ncovmodel; j++){
7734: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7735: jk++;
7736: }
7737: fprintf(ficgp,"\n");
1.126 brouard 7738: }
7739: }
1.223 brouard 7740: }
1.187 brouard 7741: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7742:
1.145 brouard 7743: /*goto avoid;*/
1.238 brouard 7744: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7745: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7746: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7747: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7748: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7749: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7750: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7751: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7752: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7753: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7754: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7755: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7756: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7757: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7758: fprintf(ficgp,"#\n");
1.223 brouard 7759: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7760: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7761: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7762: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7763: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7764: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7765: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7766: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7767: continue;
1.264 brouard 7768: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7769: strcpy(gplotlabel,"(");
1.276 brouard 7770: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7771: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7772: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7773: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7774: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7775: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7776: vlv= nbcode[Tvaraff[k]][lv];
7777: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7778: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7779: }
1.237 brouard 7780: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7781: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7782: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7783: }
1.264 brouard 7784: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7785: fprintf(ficgp,"\n#\n");
1.264 brouard 7786: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7787: fprintf(ficgp,"\nset key outside ");
7788: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7789: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7790: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7791: if (ng==1){
7792: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7793: fprintf(ficgp,"\nunset log y");
7794: }else if (ng==2){
7795: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7796: fprintf(ficgp,"\nset log y");
7797: }else if (ng==3){
7798: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7799: fprintf(ficgp,"\nset log y");
7800: }else
7801: fprintf(ficgp,"\nunset title ");
7802: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7803: i=1;
7804: for(k2=1; k2<=nlstate; k2++) {
7805: k3=i;
7806: for(k=1; k<=(nlstate+ndeath); k++) {
7807: if (k != k2){
7808: switch( ng) {
7809: case 1:
7810: if(nagesqr==0)
7811: fprintf(ficgp," p%d+p%d*x",i,i+1);
7812: else /* nagesqr =1 */
7813: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7814: break;
7815: case 2: /* ng=2 */
7816: if(nagesqr==0)
7817: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7818: else /* nagesqr =1 */
7819: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7820: break;
7821: case 3:
7822: if(nagesqr==0)
7823: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7824: else /* nagesqr =1 */
7825: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7826: break;
7827: }
7828: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7829: ijp=1; /* product no age */
7830: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7831: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7832: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7833: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7834: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7835: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7836: if(DummyV[j]==0){
7837: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7838: }else{ /* quantitative */
7839: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7840: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7841: }
7842: ij++;
1.237 brouard 7843: }
1.268 brouard 7844: }
7845: }else if(cptcovprod >0){
7846: if(j==Tprod[ijp]) { /* */
7847: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7848: if(ijp <=cptcovprod) { /* Product */
7849: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7850: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7851: /* 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)]); */
7852: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7853: }else{ /* Vn is dummy and Vm is quanti */
7854: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7855: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7856: }
7857: }else{ /* Vn*Vm Vn is quanti */
7858: if(DummyV[Tvard[ijp][2]]==0){
7859: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7860: }else{ /* Both quanti */
7861: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7862: }
1.237 brouard 7863: }
1.268 brouard 7864: ijp++;
1.237 brouard 7865: }
1.268 brouard 7866: } /* end Tprod */
1.237 brouard 7867: } else{ /* simple covariate */
1.264 brouard 7868: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7869: if(Dummy[j]==0){
7870: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7871: }else{ /* quantitative */
7872: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7873: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7874: }
1.237 brouard 7875: } /* end simple */
7876: } /* end j */
1.223 brouard 7877: }else{
7878: i=i-ncovmodel;
7879: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7880: fprintf(ficgp," (1.");
7881: }
1.227 brouard 7882:
1.223 brouard 7883: if(ng != 1){
7884: fprintf(ficgp,")/(1");
1.227 brouard 7885:
1.264 brouard 7886: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7887: if(nagesqr==0)
1.264 brouard 7888: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7889: else /* nagesqr =1 */
1.264 brouard 7890: 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 7891:
1.223 brouard 7892: ij=1;
7893: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7894: if(cptcovage >0){
7895: if((j-2)==Tage[ij]) { /* Bug valgrind */
7896: if(ij <=cptcovage) { /* Bug valgrind */
7897: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7898: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7899: ij++;
7900: }
7901: }
7902: }else
7903: 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 7904: }
7905: fprintf(ficgp,")");
7906: }
7907: fprintf(ficgp,")");
7908: if(ng ==2)
1.276 brouard 7909: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"p%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 7910: else /* ng= 3 */
1.276 brouard 7911: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 7912: }else{ /* end ng <> 1 */
7913: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 7914: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"logit(p%d%d)\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 7915: }
7916: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7917: fprintf(ficgp,",");
7918: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7919: fprintf(ficgp,",");
7920: i=i+ncovmodel;
7921: } /* end k */
7922: } /* end k2 */
1.276 brouard 7923: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
7924: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 7925: } /* end k1 */
1.223 brouard 7926: } /* end ng */
7927: /* avoid: */
7928: fflush(ficgp);
1.126 brouard 7929: } /* end gnuplot */
7930:
7931:
7932: /*************** Moving average **************/
1.219 brouard 7933: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7934: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7935:
1.222 brouard 7936: int i, cpt, cptcod;
7937: int modcovmax =1;
7938: int mobilavrange, mob;
7939: int iage=0;
7940:
1.266 brouard 7941: double sum=0., sumr=0.;
1.222 brouard 7942: double age;
1.266 brouard 7943: double *sumnewp, *sumnewm, *sumnewmr;
7944: double *agemingood, *agemaxgood;
7945: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7946:
7947:
1.278 ! brouard 7948: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
! 7949: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 7950:
7951: sumnewp = vector(1,ncovcombmax);
7952: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7953: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7954: agemingood = vector(1,ncovcombmax);
1.266 brouard 7955: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7956: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7957: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7958:
7959: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7960: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7961: sumnewp[cptcod]=0.;
1.266 brouard 7962: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
7963: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7964: }
7965: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7966:
1.266 brouard 7967: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7968: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7969: else mobilavrange=mobilav;
7970: for (age=bage; age<=fage; age++)
7971: for (i=1; i<=nlstate;i++)
7972: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7973: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7974: /* We keep the original values on the extreme ages bage, fage and for
7975: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7976: we use a 5 terms etc. until the borders are no more concerned.
7977: */
7978: for (mob=3;mob <=mobilavrange;mob=mob+2){
7979: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 7980: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7981: sumnewm[cptcod]=0.;
7982: for (i=1; i<=nlstate;i++){
1.222 brouard 7983: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7984: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7985: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7986: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7987: }
7988: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 7989: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7990: } /* end i */
7991: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
7992: } /* end cptcod */
1.222 brouard 7993: }/* end age */
7994: }/* end mob */
1.266 brouard 7995: }else{
7996: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 7997: return -1;
1.266 brouard 7998: }
7999:
8000: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8001: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8002: if(invalidvarcomb[cptcod]){
8003: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8004: continue;
8005: }
1.219 brouard 8006:
1.266 brouard 8007: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8008: sumnewm[cptcod]=0.;
8009: sumnewmr[cptcod]=0.;
8010: for (i=1; i<=nlstate;i++){
8011: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8012: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8013: }
8014: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8015: agemingoodr[cptcod]=age;
8016: }
8017: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8018: agemingood[cptcod]=age;
8019: }
8020: } /* age */
8021: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8022: sumnewm[cptcod]=0.;
1.266 brouard 8023: sumnewmr[cptcod]=0.;
1.222 brouard 8024: for (i=1; i<=nlstate;i++){
8025: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8026: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8027: }
8028: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8029: agemaxgoodr[cptcod]=age;
1.222 brouard 8030: }
8031: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8032: agemaxgood[cptcod]=age;
8033: }
8034: } /* age */
8035: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8036: /* but they will change */
8037: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8038: sumnewm[cptcod]=0.;
8039: sumnewmr[cptcod]=0.;
8040: for (i=1; i<=nlstate;i++){
8041: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8042: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8043: }
8044: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8045: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8046: agemaxgoodr[cptcod]=age; /* age min */
8047: for (i=1; i<=nlstate;i++)
8048: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8049: }else{ /* bad we change the value with the values of good ages */
8050: for (i=1; i<=nlstate;i++){
8051: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8052: } /* i */
8053: } /* end bad */
8054: }else{
8055: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8056: agemaxgood[cptcod]=age;
8057: }else{ /* bad we change the value with the values of good ages */
8058: for (i=1; i<=nlstate;i++){
8059: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8060: } /* i */
8061: } /* end bad */
8062: }/* end else */
8063: sum=0.;sumr=0.;
8064: for (i=1; i<=nlstate;i++){
8065: sum+=mobaverage[(int)age][i][cptcod];
8066: sumr+=probs[(int)age][i][cptcod];
8067: }
8068: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8069: printf("Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.266 brouard 8070: } /* end bad */
8071: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8072: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8073: printf("Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.222 brouard 8074: } /* end bad */
8075: }/* age */
1.266 brouard 8076:
8077: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8078: sumnewm[cptcod]=0.;
1.266 brouard 8079: sumnewmr[cptcod]=0.;
1.222 brouard 8080: for (i=1; i<=nlstate;i++){
8081: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8082: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8083: }
8084: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8085: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8086: agemingoodr[cptcod]=age;
8087: for (i=1; i<=nlstate;i++)
8088: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8089: }else{ /* bad we change the value with the values of good ages */
8090: for (i=1; i<=nlstate;i++){
8091: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8092: } /* i */
8093: } /* end bad */
8094: }else{
8095: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8096: agemingood[cptcod]=age;
8097: }else{ /* bad */
8098: for (i=1; i<=nlstate;i++){
8099: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8100: } /* i */
8101: } /* end bad */
8102: }/* end else */
8103: sum=0.;sumr=0.;
8104: for (i=1; i<=nlstate;i++){
8105: sum+=mobaverage[(int)age][i][cptcod];
8106: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8107: }
1.266 brouard 8108: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8109: printf("Moving average B1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you decrease fage=%d?\n",cptcod, sum, (int) age, (int)fage);
1.266 brouard 8110: } /* end bad */
8111: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8112: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8113: printf("Moving average B2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase fage=%d\n",cptcod,sumr, (int)age, (int)fage);
1.222 brouard 8114: } /* end bad */
8115: }/* age */
1.266 brouard 8116:
1.222 brouard 8117:
8118: for (age=bage; age<=fage; age++){
1.235 brouard 8119: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8120: sumnewp[cptcod]=0.;
8121: sumnewm[cptcod]=0.;
8122: for (i=1; i<=nlstate;i++){
8123: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8124: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8125: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8126: }
8127: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8128: }
8129: /* printf("\n"); */
8130: /* } */
1.266 brouard 8131:
1.222 brouard 8132: /* brutal averaging */
1.266 brouard 8133: /* for (i=1; i<=nlstate;i++){ */
8134: /* for (age=1; age<=bage; age++){ */
8135: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8136: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8137: /* } */
8138: /* for (age=fage; age<=AGESUP; age++){ */
8139: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8140: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8141: /* } */
8142: /* } /\* end i status *\/ */
8143: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8144: /* for (age=1; age<=AGESUP; age++){ */
8145: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8146: /* mobaverage[(int)age][i][cptcod]=0.; */
8147: /* } */
8148: /* } */
1.222 brouard 8149: }/* end cptcod */
1.266 brouard 8150: free_vector(agemaxgoodr,1, ncovcombmax);
8151: free_vector(agemaxgood,1, ncovcombmax);
8152: free_vector(agemingood,1, ncovcombmax);
8153: free_vector(agemingoodr,1, ncovcombmax);
8154: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8155: free_vector(sumnewm,1, ncovcombmax);
8156: free_vector(sumnewp,1, ncovcombmax);
8157: return 0;
8158: }/* End movingaverage */
1.218 brouard 8159:
1.126 brouard 8160:
8161: /************** Forecasting ******************/
1.269 brouard 8162: void prevforecast(char fileres[], double anproj1, double mproj1, double jproj1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double anproj2, double p[], int cptcoveff){
1.126 brouard 8163: /* proj1, year, month, day of starting projection
8164: agemin, agemax range of age
8165: dateprev1 dateprev2 range of dates during which prevalence is computed
8166: anproj2 year of en of projection (same day and month as proj1).
8167: */
1.267 brouard 8168: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8169: double agec; /* generic age */
8170: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8171: double *popeffectif,*popcount;
8172: double ***p3mat;
1.218 brouard 8173: /* double ***mobaverage; */
1.126 brouard 8174: char fileresf[FILENAMELENGTH];
8175:
8176: agelim=AGESUP;
1.211 brouard 8177: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8178: in each health status at the date of interview (if between dateprev1 and dateprev2).
8179: We still use firstpass and lastpass as another selection.
8180: */
1.214 brouard 8181: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8182: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8183:
1.201 brouard 8184: strcpy(fileresf,"F_");
8185: strcat(fileresf,fileresu);
1.126 brouard 8186: if((ficresf=fopen(fileresf,"w"))==NULL) {
8187: printf("Problem with forecast resultfile: %s\n", fileresf);
8188: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8189: }
1.235 brouard 8190: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8191: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8192:
1.225 brouard 8193: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8194:
8195:
8196: stepsize=(int) (stepm+YEARM-1)/YEARM;
8197: if (stepm<=12) stepsize=1;
8198: if(estepm < stepm){
8199: printf ("Problem %d lower than %d\n",estepm, stepm);
8200: }
1.270 brouard 8201: else{
8202: hstepm=estepm;
8203: }
8204: if(estepm > stepm){ /* Yes every two year */
8205: stepsize=2;
8206: }
1.126 brouard 8207:
8208: hstepm=hstepm/stepm;
8209: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8210: fractional in yp1 */
8211: anprojmean=yp;
8212: yp2=modf((yp1*12),&yp);
8213: mprojmean=yp;
8214: yp1=modf((yp2*30.5),&yp);
8215: jprojmean=yp;
8216: if(jprojmean==0) jprojmean=1;
8217: if(mprojmean==0) jprojmean=1;
8218:
1.227 brouard 8219: i1=pow(2,cptcoveff);
1.126 brouard 8220: if (cptcovn < 1){i1=1;}
8221:
8222: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8223:
8224: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8225:
1.126 brouard 8226: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8227: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8228: for(k=1; k<=i1;k++){
1.253 brouard 8229: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8230: continue;
1.227 brouard 8231: if(invalidvarcomb[k]){
8232: printf("\nCombination (%d) projection ignored because no cases \n",k);
8233: continue;
8234: }
8235: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8236: for(j=1;j<=cptcoveff;j++) {
8237: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8238: }
1.235 brouard 8239: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8240: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8241: }
1.227 brouard 8242: fprintf(ficresf," yearproj age");
8243: for(j=1; j<=nlstate+ndeath;j++){
8244: for(i=1; i<=nlstate;i++)
8245: fprintf(ficresf," p%d%d",i,j);
8246: fprintf(ficresf," wp.%d",j);
8247: }
8248: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8249: fprintf(ficresf,"\n");
8250: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8251: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8252: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8253: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8254: nhstepm = nhstepm/hstepm;
8255: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8256: oldm=oldms;savm=savms;
1.268 brouard 8257: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8258: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8259: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8260: for (h=0; h<=nhstepm; h++){
8261: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8262: break;
8263: }
8264: }
8265: fprintf(ficresf,"\n");
8266: for(j=1;j<=cptcoveff;j++)
8267: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8268: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8269:
8270: for(j=1; j<=nlstate+ndeath;j++) {
8271: ppij=0.;
8272: for(i=1; i<=nlstate;i++) {
1.278 ! brouard 8273: if (mobilav>=1)
! 8274: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
! 8275: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
! 8276: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
! 8277: }
1.268 brouard 8278: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8279: } /* end i */
8280: fprintf(ficresf," %.3f", ppij);
8281: }/* end j */
1.227 brouard 8282: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8283: } /* end agec */
1.266 brouard 8284: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8285: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8286: } /* end yearp */
8287: } /* end k */
1.219 brouard 8288:
1.126 brouard 8289: fclose(ficresf);
1.215 brouard 8290: printf("End of Computing forecasting \n");
8291: fprintf(ficlog,"End of Computing forecasting\n");
8292:
1.126 brouard 8293: }
8294:
1.269 brouard 8295: /************** Back Forecasting ******************/
8296: void prevbackforecast(char fileres[], double ***prevacurrent, double anback1, double mback1, double jback1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anback2, double p[], int cptcoveff){
1.267 brouard 8297: /* back1, year, month, day of starting backection
8298: agemin, agemax range of age
8299: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8300: anback2 year of end of backprojection (same day and month as back1).
8301: prevacurrent and prev are prevalences.
1.267 brouard 8302: */
8303: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8304: double agec; /* generic age */
1.268 brouard 8305: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8306: double *popeffectif,*popcount;
8307: double ***p3mat;
8308: /* double ***mobaverage; */
8309: char fileresfb[FILENAMELENGTH];
8310:
1.268 brouard 8311: agelim=AGEINF;
1.267 brouard 8312: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8313: in each health status at the date of interview (if between dateprev1 and dateprev2).
8314: We still use firstpass and lastpass as another selection.
8315: */
8316: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8317: /* firstpass, lastpass, stepm, weightopt, model); */
8318:
8319: /*Do we need to compute prevalence again?*/
8320:
8321: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8322:
8323: strcpy(fileresfb,"FB_");
8324: strcat(fileresfb,fileresu);
8325: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8326: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8327: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8328: }
8329: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8330: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8331:
8332: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8333:
8334:
8335: stepsize=(int) (stepm+YEARM-1)/YEARM;
8336: if (stepm<=12) stepsize=1;
8337: if(estepm < stepm){
8338: printf ("Problem %d lower than %d\n",estepm, stepm);
8339: }
1.270 brouard 8340: else{
8341: hstepm=estepm;
8342: }
8343: if(estepm >= stepm){ /* Yes every two year */
8344: stepsize=2;
8345: }
1.267 brouard 8346:
8347: hstepm=hstepm/stepm;
8348: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8349: fractional in yp1 */
8350: anprojmean=yp;
8351: yp2=modf((yp1*12),&yp);
8352: mprojmean=yp;
8353: yp1=modf((yp2*30.5),&yp);
8354: jprojmean=yp;
8355: if(jprojmean==0) jprojmean=1;
8356: if(mprojmean==0) jprojmean=1;
8357:
8358: i1=pow(2,cptcoveff);
8359: if (cptcovn < 1){i1=1;}
8360:
8361: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8362: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8363:
8364: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8365:
8366: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8367: for(k=1; k<=i1;k++){
8368: if(i1 != 1 && TKresult[nres]!= k)
8369: continue;
8370: if(invalidvarcomb[k]){
8371: printf("\nCombination (%d) projection ignored because no cases \n",k);
8372: continue;
8373: }
1.268 brouard 8374: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8375: for(j=1;j<=cptcoveff;j++) {
8376: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8377: }
8378: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8379: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8380: }
8381: fprintf(ficresfb," yearbproj age");
8382: for(j=1; j<=nlstate+ndeath;j++){
8383: for(i=1; i<=nlstate;i++)
1.268 brouard 8384: fprintf(ficresfb," b%d%d",i,j);
8385: fprintf(ficresfb," b.%d",j);
1.267 brouard 8386: }
8387: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8388: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8389: fprintf(ficresfb,"\n");
8390: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8391: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8392: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8393: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8394: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8395: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8396: nhstepm = nhstepm/hstepm;
8397: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8398: oldm=oldms;savm=savms;
1.268 brouard 8399: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8400: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8401: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8402: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8403: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8404: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8405: for (h=0; h<=nhstepm; h++){
1.268 brouard 8406: if (h*hstepm/YEARM*stepm ==-yearp) {
8407: break;
8408: }
8409: }
8410: fprintf(ficresfb,"\n");
8411: for(j=1;j<=cptcoveff;j++)
8412: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8413: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8414: for(i=1; i<=nlstate+ndeath;i++) {
8415: ppij=0.;ppi=0.;
8416: for(j=1; j<=nlstate;j++) {
8417: /* if (mobilav==1) */
1.269 brouard 8418: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8419: ppi=ppi+prevacurrent[(int)agec][j][k];
8420: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8421: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8422: /* else { */
8423: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8424: /* } */
1.268 brouard 8425: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8426: } /* end j */
8427: if(ppi <0.99){
8428: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8429: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8430: }
8431: fprintf(ficresfb," %.3f", ppij);
8432: }/* end j */
1.267 brouard 8433: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8434: } /* end agec */
8435: } /* end yearp */
8436: } /* end k */
1.217 brouard 8437:
1.267 brouard 8438: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8439:
1.267 brouard 8440: fclose(ficresfb);
8441: printf("End of Computing Back forecasting \n");
8442: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8443:
1.267 brouard 8444: }
1.217 brouard 8445:
1.269 brouard 8446: /* Variance of prevalence limit: varprlim */
8447: void varprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **prlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
8448: /*------- Variance of period (stable) prevalence------*/
8449:
8450: char fileresvpl[FILENAMELENGTH];
8451: FILE *ficresvpl;
8452: double **oldm, **savm;
8453: double **varpl; /* Variances of prevalence limits by age */
8454: int i1, k, nres, j ;
8455:
8456: strcpy(fileresvpl,"VPL_");
8457: strcat(fileresvpl,fileresu);
8458: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8459: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8460: exit(0);
8461: }
8462: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8463: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8464:
8465: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8466: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8467:
8468: i1=pow(2,cptcoveff);
8469: if (cptcovn < 1){i1=1;}
8470:
8471: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8472: for(k=1; k<=i1;k++){
8473: if(i1 != 1 && TKresult[nres]!= k)
8474: continue;
8475: fprintf(ficresvpl,"\n#****** ");
8476: printf("\n#****** ");
8477: fprintf(ficlog,"\n#****** ");
8478: for(j=1;j<=cptcoveff;j++) {
8479: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8480: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8481: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8482: }
8483: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8484: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8485: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8486: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8487: }
8488: fprintf(ficresvpl,"******\n");
8489: printf("******\n");
8490: fprintf(ficlog,"******\n");
8491:
8492: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8493: oldm=oldms;savm=savms;
8494: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8495: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8496: /*}*/
8497: }
8498:
8499: fclose(ficresvpl);
8500: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8501: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8502:
8503: }
8504: /* Variance of back prevalence: varbprlim */
8505: void varbprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **bprlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
8506: /*------- Variance of back (stable) prevalence------*/
8507:
8508: char fileresvbl[FILENAMELENGTH];
8509: FILE *ficresvbl;
8510:
8511: double **oldm, **savm;
8512: double **varbpl; /* Variances of back prevalence limits by age */
8513: int i1, k, nres, j ;
8514:
8515: strcpy(fileresvbl,"VBL_");
8516: strcat(fileresvbl,fileresu);
8517: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8518: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8519: exit(0);
8520: }
8521: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8522: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8523:
8524:
8525: i1=pow(2,cptcoveff);
8526: if (cptcovn < 1){i1=1;}
8527:
8528: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8529: for(k=1; k<=i1;k++){
8530: if(i1 != 1 && TKresult[nres]!= k)
8531: continue;
8532: fprintf(ficresvbl,"\n#****** ");
8533: printf("\n#****** ");
8534: fprintf(ficlog,"\n#****** ");
8535: for(j=1;j<=cptcoveff;j++) {
8536: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8537: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8538: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8539: }
8540: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8541: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8542: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8543: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8544: }
8545: fprintf(ficresvbl,"******\n");
8546: printf("******\n");
8547: fprintf(ficlog,"******\n");
8548:
8549: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8550: oldm=oldms;savm=savms;
8551:
8552: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8553: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8554: /*}*/
8555: }
8556:
8557: fclose(ficresvbl);
8558: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8559: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8560:
8561: } /* End of varbprlim */
8562:
1.126 brouard 8563: /************** Forecasting *****not tested NB*************/
1.227 brouard 8564: /* 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 8565:
1.227 brouard 8566: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8567: /* int *popage; */
8568: /* double calagedatem, agelim, kk1, kk2; */
8569: /* double *popeffectif,*popcount; */
8570: /* double ***p3mat,***tabpop,***tabpopprev; */
8571: /* /\* double ***mobaverage; *\/ */
8572: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8573:
1.227 brouard 8574: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8575: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8576: /* agelim=AGESUP; */
8577: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8578:
1.227 brouard 8579: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8580:
8581:
1.227 brouard 8582: /* strcpy(filerespop,"POP_"); */
8583: /* strcat(filerespop,fileresu); */
8584: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8585: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8586: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8587: /* } */
8588: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8589: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8590:
1.227 brouard 8591: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8592:
1.227 brouard 8593: /* /\* if (mobilav!=0) { *\/ */
8594: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8595: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8596: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8597: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8598: /* /\* } *\/ */
8599: /* /\* } *\/ */
1.126 brouard 8600:
1.227 brouard 8601: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8602: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8603:
1.227 brouard 8604: /* agelim=AGESUP; */
1.126 brouard 8605:
1.227 brouard 8606: /* hstepm=1; */
8607: /* hstepm=hstepm/stepm; */
1.218 brouard 8608:
1.227 brouard 8609: /* if (popforecast==1) { */
8610: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8611: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8612: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8613: /* } */
8614: /* popage=ivector(0,AGESUP); */
8615: /* popeffectif=vector(0,AGESUP); */
8616: /* popcount=vector(0,AGESUP); */
1.126 brouard 8617:
1.227 brouard 8618: /* i=1; */
8619: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8620:
1.227 brouard 8621: /* imx=i; */
8622: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8623: /* } */
1.218 brouard 8624:
1.227 brouard 8625: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8626: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8627: /* k=k+1; */
8628: /* fprintf(ficrespop,"\n#******"); */
8629: /* for(j=1;j<=cptcoveff;j++) { */
8630: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8631: /* } */
8632: /* fprintf(ficrespop,"******\n"); */
8633: /* fprintf(ficrespop,"# Age"); */
8634: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8635: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8636:
1.227 brouard 8637: /* for (cpt=0; cpt<=0;cpt++) { */
8638: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8639:
1.227 brouard 8640: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8641: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8642: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8643:
1.227 brouard 8644: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8645: /* oldm=oldms;savm=savms; */
8646: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8647:
1.227 brouard 8648: /* for (h=0; h<=nhstepm; h++){ */
8649: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8650: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8651: /* } */
8652: /* for(j=1; j<=nlstate+ndeath;j++) { */
8653: /* kk1=0.;kk2=0; */
8654: /* for(i=1; i<=nlstate;i++) { */
8655: /* if (mobilav==1) */
8656: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8657: /* else { */
8658: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8659: /* } */
8660: /* } */
8661: /* if (h==(int)(calagedatem+12*cpt)){ */
8662: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8663: /* /\*fprintf(ficrespop," %.3f", kk1); */
8664: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8665: /* } */
8666: /* } */
8667: /* for(i=1; i<=nlstate;i++){ */
8668: /* kk1=0.; */
8669: /* for(j=1; j<=nlstate;j++){ */
8670: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8671: /* } */
8672: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8673: /* } */
1.218 brouard 8674:
1.227 brouard 8675: /* if (h==(int)(calagedatem+12*cpt)) */
8676: /* for(j=1; j<=nlstate;j++) */
8677: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8678: /* } */
8679: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8680: /* } */
8681: /* } */
1.218 brouard 8682:
1.227 brouard 8683: /* /\******\/ */
1.218 brouard 8684:
1.227 brouard 8685: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8686: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8687: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8688: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8689: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8690:
1.227 brouard 8691: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8692: /* oldm=oldms;savm=savms; */
8693: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8694: /* for (h=0; h<=nhstepm; h++){ */
8695: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8696: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8697: /* } */
8698: /* for(j=1; j<=nlstate+ndeath;j++) { */
8699: /* kk1=0.;kk2=0; */
8700: /* for(i=1; i<=nlstate;i++) { */
8701: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8702: /* } */
8703: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8704: /* } */
8705: /* } */
8706: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8707: /* } */
8708: /* } */
8709: /* } */
8710: /* } */
1.218 brouard 8711:
1.227 brouard 8712: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8713:
1.227 brouard 8714: /* if (popforecast==1) { */
8715: /* free_ivector(popage,0,AGESUP); */
8716: /* free_vector(popeffectif,0,AGESUP); */
8717: /* free_vector(popcount,0,AGESUP); */
8718: /* } */
8719: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8720: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8721: /* fclose(ficrespop); */
8722: /* } /\* End of popforecast *\/ */
1.218 brouard 8723:
1.126 brouard 8724: int fileappend(FILE *fichier, char *optionfich)
8725: {
8726: if((fichier=fopen(optionfich,"a"))==NULL) {
8727: printf("Problem with file: %s\n", optionfich);
8728: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8729: return (0);
8730: }
8731: fflush(fichier);
8732: return (1);
8733: }
8734:
8735:
8736: /**************** function prwizard **********************/
8737: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8738: {
8739:
8740: /* Wizard to print covariance matrix template */
8741:
1.164 brouard 8742: char ca[32], cb[32];
8743: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8744: int numlinepar;
8745:
8746: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8747: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8748: for(i=1; i <=nlstate; i++){
8749: jj=0;
8750: for(j=1; j <=nlstate+ndeath; j++){
8751: if(j==i) continue;
8752: jj++;
8753: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8754: printf("%1d%1d",i,j);
8755: fprintf(ficparo,"%1d%1d",i,j);
8756: for(k=1; k<=ncovmodel;k++){
8757: /* printf(" %lf",param[i][j][k]); */
8758: /* fprintf(ficparo," %lf",param[i][j][k]); */
8759: printf(" 0.");
8760: fprintf(ficparo," 0.");
8761: }
8762: printf("\n");
8763: fprintf(ficparo,"\n");
8764: }
8765: }
8766: printf("# Scales (for hessian or gradient estimation)\n");
8767: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8768: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8769: for(i=1; i <=nlstate; i++){
8770: jj=0;
8771: for(j=1; j <=nlstate+ndeath; j++){
8772: if(j==i) continue;
8773: jj++;
8774: fprintf(ficparo,"%1d%1d",i,j);
8775: printf("%1d%1d",i,j);
8776: fflush(stdout);
8777: for(k=1; k<=ncovmodel;k++){
8778: /* printf(" %le",delti3[i][j][k]); */
8779: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8780: printf(" 0.");
8781: fprintf(ficparo," 0.");
8782: }
8783: numlinepar++;
8784: printf("\n");
8785: fprintf(ficparo,"\n");
8786: }
8787: }
8788: printf("# Covariance matrix\n");
8789: /* # 121 Var(a12)\n\ */
8790: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8791: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8792: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8793: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8794: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8795: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8796: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8797: fflush(stdout);
8798: fprintf(ficparo,"# Covariance matrix\n");
8799: /* # 121 Var(a12)\n\ */
8800: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8801: /* # ...\n\ */
8802: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8803:
8804: for(itimes=1;itimes<=2;itimes++){
8805: jj=0;
8806: for(i=1; i <=nlstate; i++){
8807: for(j=1; j <=nlstate+ndeath; j++){
8808: if(j==i) continue;
8809: for(k=1; k<=ncovmodel;k++){
8810: jj++;
8811: ca[0]= k+'a'-1;ca[1]='\0';
8812: if(itimes==1){
8813: printf("#%1d%1d%d",i,j,k);
8814: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8815: }else{
8816: printf("%1d%1d%d",i,j,k);
8817: fprintf(ficparo,"%1d%1d%d",i,j,k);
8818: /* printf(" %.5le",matcov[i][j]); */
8819: }
8820: ll=0;
8821: for(li=1;li <=nlstate; li++){
8822: for(lj=1;lj <=nlstate+ndeath; lj++){
8823: if(lj==li) continue;
8824: for(lk=1;lk<=ncovmodel;lk++){
8825: ll++;
8826: if(ll<=jj){
8827: cb[0]= lk +'a'-1;cb[1]='\0';
8828: if(ll<jj){
8829: if(itimes==1){
8830: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8831: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8832: }else{
8833: printf(" 0.");
8834: fprintf(ficparo," 0.");
8835: }
8836: }else{
8837: if(itimes==1){
8838: printf(" Var(%s%1d%1d)",ca,i,j);
8839: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8840: }else{
8841: printf(" 0.");
8842: fprintf(ficparo," 0.");
8843: }
8844: }
8845: }
8846: } /* end lk */
8847: } /* end lj */
8848: } /* end li */
8849: printf("\n");
8850: fprintf(ficparo,"\n");
8851: numlinepar++;
8852: } /* end k*/
8853: } /*end j */
8854: } /* end i */
8855: } /* end itimes */
8856:
8857: } /* end of prwizard */
8858: /******************* Gompertz Likelihood ******************************/
8859: double gompertz(double x[])
8860: {
8861: double A,B,L=0.0,sump=0.,num=0.;
8862: int i,n=0; /* n is the size of the sample */
8863:
1.220 brouard 8864: for (i=1;i<=imx ; i++) {
1.126 brouard 8865: sump=sump+weight[i];
8866: /* sump=sump+1;*/
8867: num=num+1;
8868: }
8869:
8870:
8871: /* for (i=0; i<=imx; i++)
8872: 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]);*/
8873:
8874: for (i=1;i<=imx ; i++)
8875: {
8876: if (cens[i] == 1 && wav[i]>1)
8877: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8878:
8879: if (cens[i] == 0 && wav[i]>1)
8880: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8881: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8882:
8883: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8884: if (wav[i] > 1 ) { /* ??? */
8885: L=L+A*weight[i];
8886: /* 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]);*/
8887: }
8888: }
8889:
8890: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8891:
8892: return -2*L*num/sump;
8893: }
8894:
1.136 brouard 8895: #ifdef GSL
8896: /******************* Gompertz_f Likelihood ******************************/
8897: double gompertz_f(const gsl_vector *v, void *params)
8898: {
8899: double A,B,LL=0.0,sump=0.,num=0.;
8900: double *x= (double *) v->data;
8901: int i,n=0; /* n is the size of the sample */
8902:
8903: for (i=0;i<=imx-1 ; i++) {
8904: sump=sump+weight[i];
8905: /* sump=sump+1;*/
8906: num=num+1;
8907: }
8908:
8909:
8910: /* for (i=0; i<=imx; i++)
8911: 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]);*/
8912: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8913: for (i=1;i<=imx ; i++)
8914: {
8915: if (cens[i] == 1 && wav[i]>1)
8916: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8917:
8918: if (cens[i] == 0 && wav[i]>1)
8919: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8920: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8921:
8922: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8923: if (wav[i] > 1 ) { /* ??? */
8924: LL=LL+A*weight[i];
8925: /* 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]);*/
8926: }
8927: }
8928:
8929: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8930: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8931:
8932: return -2*LL*num/sump;
8933: }
8934: #endif
8935:
1.126 brouard 8936: /******************* Printing html file ***********/
1.201 brouard 8937: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8938: int lastpass, int stepm, int weightopt, char model[],\
8939: int imx, double p[],double **matcov,double agemortsup){
8940: int i,k;
8941:
8942: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8943: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8944: for (i=1;i<=2;i++)
8945: 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 8946: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8947: fprintf(fichtm,"</ul>");
8948:
8949: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8950:
8951: 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>");
8952:
8953: for (k=agegomp;k<(agemortsup-2);k++)
8954: 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]);
8955:
8956:
8957: fflush(fichtm);
8958: }
8959:
8960: /******************* Gnuplot file **************/
1.201 brouard 8961: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8962:
8963: char dirfileres[132],optfileres[132];
1.164 brouard 8964:
1.126 brouard 8965: int ng;
8966:
8967:
8968: /*#ifdef windows */
8969: fprintf(ficgp,"cd \"%s\" \n",pathc);
8970: /*#endif */
8971:
8972:
8973: strcpy(dirfileres,optionfilefiname);
8974: strcpy(optfileres,"vpl");
1.199 brouard 8975: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8976: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8977: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8978: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8979: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8980:
8981: }
8982:
1.136 brouard 8983: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8984: {
1.126 brouard 8985:
1.136 brouard 8986: /*-------- data file ----------*/
8987: FILE *fic;
8988: char dummy[]=" ";
1.240 brouard 8989: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8990: int lstra;
1.136 brouard 8991: int linei, month, year,iout;
8992: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8993: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8994: char *stratrunc;
1.223 brouard 8995:
1.240 brouard 8996: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8997: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8998:
1.240 brouard 8999: for(v=1; v <=ncovcol;v++){
9000: DummyV[v]=0;
9001: FixedV[v]=0;
9002: }
9003: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9004: DummyV[v]=1;
9005: FixedV[v]=0;
9006: }
9007: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9008: DummyV[v]=0;
9009: FixedV[v]=1;
9010: }
9011: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9012: DummyV[v]=1;
9013: FixedV[v]=1;
9014: }
9015: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9016: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9017: 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]);
9018: }
1.126 brouard 9019:
1.136 brouard 9020: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9021: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9022: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9023: }
1.126 brouard 9024:
1.136 brouard 9025: i=1;
9026: linei=0;
9027: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9028: linei=linei+1;
9029: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9030: if(line[j] == '\t')
9031: line[j] = ' ';
9032: }
9033: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9034: ;
9035: };
9036: line[j+1]=0; /* Trims blanks at end of line */
9037: if(line[0]=='#'){
9038: fprintf(ficlog,"Comment line\n%s\n",line);
9039: printf("Comment line\n%s\n",line);
9040: continue;
9041: }
9042: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9043: strcpy(line, linetmp);
1.223 brouard 9044:
9045: /* Loops on waves */
9046: for (j=maxwav;j>=1;j--){
9047: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9048: cutv(stra, strb, line, ' ');
9049: if(strb[0]=='.') { /* Missing value */
9050: lval=-1;
9051: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9052: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9053: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9054: 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);
9055: 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);
9056: return 1;
9057: }
9058: }else{
9059: errno=0;
9060: /* what_kind_of_number(strb); */
9061: dval=strtod(strb,&endptr);
9062: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9063: /* if(strb != endptr && *endptr == '\0') */
9064: /* dval=dlval; */
9065: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9066: if( strb[0]=='\0' || (*endptr != '\0')){
9067: 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);
9068: 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);
9069: return 1;
9070: }
9071: cotqvar[j][iv][i]=dval;
9072: cotvar[j][ntv+iv][i]=dval;
9073: }
9074: strcpy(line,stra);
1.223 brouard 9075: }/* end loop ntqv */
1.225 brouard 9076:
1.223 brouard 9077: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9078: cutv(stra, strb, line, ' ');
9079: if(strb[0]=='.') { /* Missing value */
9080: lval=-1;
9081: }else{
9082: errno=0;
9083: lval=strtol(strb,&endptr,10);
9084: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9085: if( strb[0]=='\0' || (*endptr != '\0')){
9086: 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);
9087: 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);
9088: return 1;
9089: }
9090: }
9091: if(lval <-1 || lval >1){
9092: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9093: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9094: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9095: For example, for multinomial values like 1, 2 and 3,\n \
9096: build V1=0 V2=0 for the reference value (1),\n \
9097: V1=1 V2=0 for (2) \n \
1.223 brouard 9098: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9099: output of IMaCh is often meaningless.\n \
1.223 brouard 9100: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9101: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9102: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9103: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9104: For example, for multinomial values like 1, 2 and 3,\n \
9105: build V1=0 V2=0 for the reference value (1),\n \
9106: V1=1 V2=0 for (2) \n \
1.223 brouard 9107: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9108: output of IMaCh is often meaningless.\n \
1.223 brouard 9109: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9110: return 1;
9111: }
9112: cotvar[j][iv][i]=(double)(lval);
9113: strcpy(line,stra);
1.223 brouard 9114: }/* end loop ntv */
1.225 brouard 9115:
1.223 brouard 9116: /* Statuses at wave */
1.137 brouard 9117: cutv(stra, strb, line, ' ');
1.223 brouard 9118: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9119: lval=-1;
1.136 brouard 9120: }else{
1.238 brouard 9121: errno=0;
9122: lval=strtol(strb,&endptr,10);
9123: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9124: if( strb[0]=='\0' || (*endptr != '\0')){
9125: 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);
9126: 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);
9127: return 1;
9128: }
1.136 brouard 9129: }
1.225 brouard 9130:
1.136 brouard 9131: s[j][i]=lval;
1.225 brouard 9132:
1.223 brouard 9133: /* Date of Interview */
1.136 brouard 9134: strcpy(line,stra);
9135: cutv(stra, strb,line,' ');
1.169 brouard 9136: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9137: }
1.169 brouard 9138: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9139: month=99;
9140: year=9999;
1.136 brouard 9141: }else{
1.225 brouard 9142: 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);
9143: 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);
9144: return 1;
1.136 brouard 9145: }
9146: anint[j][i]= (double) year;
9147: mint[j][i]= (double)month;
9148: strcpy(line,stra);
1.223 brouard 9149: } /* End loop on waves */
1.225 brouard 9150:
1.223 brouard 9151: /* Date of death */
1.136 brouard 9152: cutv(stra, strb,line,' ');
1.169 brouard 9153: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9154: }
1.169 brouard 9155: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9156: month=99;
9157: year=9999;
9158: }else{
1.141 brouard 9159: 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 9160: 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);
9161: return 1;
1.136 brouard 9162: }
9163: andc[i]=(double) year;
9164: moisdc[i]=(double) month;
9165: strcpy(line,stra);
9166:
1.223 brouard 9167: /* Date of birth */
1.136 brouard 9168: cutv(stra, strb,line,' ');
1.169 brouard 9169: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9170: }
1.169 brouard 9171: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9172: month=99;
9173: year=9999;
9174: }else{
1.141 brouard 9175: 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);
9176: 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 9177: return 1;
1.136 brouard 9178: }
9179: if (year==9999) {
1.141 brouard 9180: 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);
9181: 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 9182: return 1;
9183:
1.136 brouard 9184: }
9185: annais[i]=(double)(year);
9186: moisnais[i]=(double)(month);
9187: strcpy(line,stra);
1.225 brouard 9188:
1.223 brouard 9189: /* Sample weight */
1.136 brouard 9190: cutv(stra, strb,line,' ');
9191: errno=0;
9192: dval=strtod(strb,&endptr);
9193: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9194: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9195: 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 9196: fflush(ficlog);
9197: return 1;
9198: }
9199: weight[i]=dval;
9200: strcpy(line,stra);
1.225 brouard 9201:
1.223 brouard 9202: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9203: cutv(stra, strb, line, ' ');
9204: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9205: lval=-1;
1.223 brouard 9206: }else{
1.225 brouard 9207: errno=0;
9208: /* what_kind_of_number(strb); */
9209: dval=strtod(strb,&endptr);
9210: /* if(strb != endptr && *endptr == '\0') */
9211: /* dval=dlval; */
9212: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9213: if( strb[0]=='\0' || (*endptr != '\0')){
9214: 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);
9215: 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);
9216: return 1;
9217: }
9218: coqvar[iv][i]=dval;
1.226 brouard 9219: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9220: }
9221: strcpy(line,stra);
9222: }/* end loop nqv */
1.136 brouard 9223:
1.223 brouard 9224: /* Covariate values */
1.136 brouard 9225: for (j=ncovcol;j>=1;j--){
9226: cutv(stra, strb,line,' ');
1.223 brouard 9227: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9228: lval=-1;
1.136 brouard 9229: }else{
1.225 brouard 9230: errno=0;
9231: lval=strtol(strb,&endptr,10);
9232: if( strb[0]=='\0' || (*endptr != '\0')){
9233: 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);
9234: 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);
9235: return 1;
9236: }
1.136 brouard 9237: }
9238: if(lval <-1 || lval >1){
1.225 brouard 9239: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9240: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9241: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9242: For example, for multinomial values like 1, 2 and 3,\n \
9243: build V1=0 V2=0 for the reference value (1),\n \
9244: V1=1 V2=0 for (2) \n \
1.136 brouard 9245: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9246: output of IMaCh is often meaningless.\n \
1.136 brouard 9247: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9248: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9249: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9250: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9251: For example, for multinomial values like 1, 2 and 3,\n \
9252: build V1=0 V2=0 for the reference value (1),\n \
9253: V1=1 V2=0 for (2) \n \
1.136 brouard 9254: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9255: output of IMaCh is often meaningless.\n \
1.136 brouard 9256: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9257: return 1;
1.136 brouard 9258: }
9259: covar[j][i]=(double)(lval);
9260: strcpy(line,stra);
9261: }
9262: lstra=strlen(stra);
1.225 brouard 9263:
1.136 brouard 9264: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9265: stratrunc = &(stra[lstra-9]);
9266: num[i]=atol(stratrunc);
9267: }
9268: else
9269: num[i]=atol(stra);
9270: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9271: 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;}*/
9272:
9273: i=i+1;
9274: } /* End loop reading data */
1.225 brouard 9275:
1.136 brouard 9276: *imax=i-1; /* Number of individuals */
9277: fclose(fic);
1.225 brouard 9278:
1.136 brouard 9279: return (0);
1.164 brouard 9280: /* endread: */
1.225 brouard 9281: printf("Exiting readdata: ");
9282: fclose(fic);
9283: return (1);
1.223 brouard 9284: }
1.126 brouard 9285:
1.234 brouard 9286: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9287: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9288: while (*p2 == ' ')
1.234 brouard 9289: p2++;
9290: /* while ((*p1++ = *p2++) !=0) */
9291: /* ; */
9292: /* do */
9293: /* while (*p2 == ' ') */
9294: /* p2++; */
9295: /* while (*p1++ == *p2++); */
9296: *stri=p2;
1.145 brouard 9297: }
9298:
1.235 brouard 9299: int decoderesult ( char resultline[], int nres)
1.230 brouard 9300: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9301: {
1.235 brouard 9302: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9303: char resultsav[MAXLINE];
1.234 brouard 9304: int resultmodel[MAXLINE];
9305: int modelresult[MAXLINE];
1.230 brouard 9306: char stra[80], strb[80], strc[80], strd[80],stre[80];
9307:
1.234 brouard 9308: removefirstspace(&resultline);
1.233 brouard 9309: printf("decoderesult:%s\n",resultline);
1.230 brouard 9310:
9311: if (strstr(resultline,"v") !=0){
9312: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9313: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9314: return 1;
9315: }
9316: trimbb(resultsav, resultline);
9317: if (strlen(resultsav) >1){
9318: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9319: }
1.253 brouard 9320: if(j == 0){ /* Resultline but no = */
9321: TKresult[nres]=0; /* Combination for the nresult and the model */
9322: return (0);
9323: }
9324:
1.234 brouard 9325: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9326: 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);
9327: 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);
9328: }
9329: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9330: if(nbocc(resultsav,'=') >1){
9331: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9332: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9333: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9334: }else
9335: cutl(strc,strd,resultsav,'=');
1.230 brouard 9336: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9337:
1.230 brouard 9338: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9339: Tvarsel[k]=atoi(strc);
9340: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9341: /* cptcovsel++; */
9342: if (nbocc(stra,'=') >0)
9343: strcpy(resultsav,stra); /* and analyzes it */
9344: }
1.235 brouard 9345: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9346: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9347: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9348: match=0;
1.236 brouard 9349: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9350: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9351: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9352: match=1;
9353: break;
9354: }
9355: }
9356: if(match == 0){
9357: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9358: }
9359: }
9360: }
1.235 brouard 9361: /* Checking for missing or useless values in comparison of current model needs */
9362: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9363: match=0;
1.235 brouard 9364: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9365: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9366: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9367: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9368: ++match;
9369: }
9370: }
9371: }
9372: if(match == 0){
9373: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9374: }else if(match > 1){
9375: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9376: }
9377: }
1.235 brouard 9378:
1.234 brouard 9379: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9380: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9381: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9382: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9383: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9384: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9385: /* 1 0 0 0 */
9386: /* 2 1 0 0 */
9387: /* 3 0 1 0 */
9388: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9389: /* 5 0 0 1 */
9390: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9391: /* 7 0 1 1 */
9392: /* 8 1 1 1 */
1.237 brouard 9393: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9394: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9395: /* V5*age V5 known which value for nres? */
9396: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9397: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9398: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9399: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9400: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9401: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9402: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9403: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9404: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9405: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9406: k4++;;
9407: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9408: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9409: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9410: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9411: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9412: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9413: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9414: k4q++;;
9415: }
9416: }
1.234 brouard 9417:
1.235 brouard 9418: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9419: return (0);
9420: }
1.235 brouard 9421:
1.230 brouard 9422: int decodemodel( char model[], int lastobs)
9423: /**< This routine decodes the model and returns:
1.224 brouard 9424: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9425: * - nagesqr = 1 if age*age in the model, otherwise 0.
9426: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9427: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9428: * - cptcovage number of covariates with age*products =2
9429: * - cptcovs number of simple covariates
9430: * - 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
9431: * which is a new column after the 9 (ncovcol) variables.
9432: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9433: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9434: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9435: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9436: */
1.136 brouard 9437: {
1.238 brouard 9438: int i, j, k, ks, v;
1.227 brouard 9439: int j1, k1, k2, k3, k4;
1.136 brouard 9440: char modelsav[80];
1.145 brouard 9441: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9442: char *strpt;
1.136 brouard 9443:
1.145 brouard 9444: /*removespace(model);*/
1.136 brouard 9445: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9446: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9447: if (strstr(model,"AGE") !=0){
1.192 brouard 9448: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9449: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9450: return 1;
9451: }
1.141 brouard 9452: if (strstr(model,"v") !=0){
9453: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9454: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9455: return 1;
9456: }
1.187 brouard 9457: strcpy(modelsav,model);
9458: if ((strpt=strstr(model,"age*age")) !=0){
9459: printf(" strpt=%s, model=%s\n",strpt, model);
9460: if(strpt != model){
1.234 brouard 9461: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9462: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9463: corresponding column of parameters.\n",model);
1.234 brouard 9464: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9465: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9466: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9467: return 1;
1.225 brouard 9468: }
1.187 brouard 9469: nagesqr=1;
9470: if (strstr(model,"+age*age") !=0)
1.234 brouard 9471: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9472: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9473: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9474: else
1.234 brouard 9475: substrchaine(modelsav, model, "age*age");
1.187 brouard 9476: }else
9477: nagesqr=0;
9478: if (strlen(modelsav) >1){
9479: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9480: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9481: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9482: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9483: * cst, age and age*age
9484: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9485: /* including age products which are counted in cptcovage.
9486: * but the covariates which are products must be treated
9487: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9488: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9489: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9490:
9491:
1.187 brouard 9492: /* Design
9493: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9494: * < ncovcol=8 >
9495: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9496: * k= 1 2 3 4 5 6 7 8
9497: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9498: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9499: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9500: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9501: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9502: * Tage[++cptcovage]=k
9503: * if products, new covar are created after ncovcol with k1
9504: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9505: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9506: * 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
9507: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9508: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9509: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9510: * < ncovcol=8 >
9511: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9512: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9513: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9514: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9515: * p Tprod[1]@2={ 6, 5}
9516: *p Tvard[1][1]@4= {7, 8, 5, 6}
9517: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9518: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9519: *How to reorganize?
9520: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9521: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9522: * {2, 1, 4, 8, 5, 6, 3, 7}
9523: * Struct []
9524: */
1.225 brouard 9525:
1.187 brouard 9526: /* This loop fills the array Tvar from the string 'model'.*/
9527: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9528: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9529: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9530: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9531: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9532: /* k=1 Tvar[1]=2 (from V2) */
9533: /* k=5 Tvar[5] */
9534: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9535: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9536: /* } */
1.198 brouard 9537: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9538: /*
9539: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9540: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9541: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9542: }
1.187 brouard 9543: cptcovage=0;
9544: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9545: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9546: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9547: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9548: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9549: /*scanf("%d",i);*/
9550: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9551: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9552: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9553: /* covar is not filled and then is empty */
9554: cptcovprod--;
9555: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9556: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9557: Typevar[k]=1; /* 1 for age product */
9558: cptcovage++; /* Sums the number of covariates which include age as a product */
9559: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9560: /*printf("stre=%s ", stre);*/
9561: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9562: cptcovprod--;
9563: cutl(stre,strb,strc,'V');
9564: Tvar[k]=atoi(stre);
9565: Typevar[k]=1; /* 1 for age product */
9566: cptcovage++;
9567: Tage[cptcovage]=k;
9568: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9569: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9570: cptcovn++;
9571: cptcovprodnoage++;k1++;
9572: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9573: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9574: because this model-covariate is a construction we invent a new column
9575: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9576: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9577: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9578: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9579: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9580: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9581: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9582: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9583: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9584: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9585: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9586: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9587: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9588: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9589: for (i=1; i<=lastobs;i++){
9590: /* Computes the new covariate which is a product of
9591: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9592: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9593: }
9594: } /* End age is not in the model */
9595: } /* End if model includes a product */
9596: else { /* no more sum */
9597: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9598: /* scanf("%d",i);*/
9599: cutl(strd,strc,strb,'V');
9600: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9601: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9602: Tvar[k]=atoi(strd);
9603: Typevar[k]=0; /* 0 for simple covariates */
9604: }
9605: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9606: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9607: scanf("%d",i);*/
1.187 brouard 9608: } /* end of loop + on total covariates */
9609: } /* end if strlen(modelsave == 0) age*age might exist */
9610: } /* end if strlen(model == 0) */
1.136 brouard 9611:
9612: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9613: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9614:
1.136 brouard 9615: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9616: printf("cptcovprod=%d ", cptcovprod);
9617: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9618: scanf("%d ",i);*/
9619:
9620:
1.230 brouard 9621: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9622: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9623: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9624: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9625: k = 1 2 3 4 5 6 7 8 9
9626: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9627: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9628: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9629: Dummy[k] 1 0 0 0 3 1 1 2 3
9630: Tmodelind[combination of covar]=k;
1.225 brouard 9631: */
9632: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9633: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9634: /* 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 9635: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9636: printf("Model=%s\n\
9637: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9638: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9639: 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);
9640: fprintf(ficlog,"Model=%s\n\
9641: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9642: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9643: 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 9644: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9645: 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 */
9646: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9647: Fixed[k]= 0;
9648: Dummy[k]= 0;
1.225 brouard 9649: ncoveff++;
1.232 brouard 9650: ncovf++;
1.234 brouard 9651: nsd++;
9652: modell[k].maintype= FTYPE;
9653: TvarsD[nsd]=Tvar[k];
9654: TvarsDind[nsd]=k;
9655: TvarF[ncovf]=Tvar[k];
9656: TvarFind[ncovf]=k;
9657: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9658: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9659: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9660: Fixed[k]= 0;
9661: Dummy[k]= 0;
9662: ncoveff++;
9663: ncovf++;
9664: modell[k].maintype= FTYPE;
9665: TvarF[ncovf]=Tvar[k];
9666: TvarFind[ncovf]=k;
1.230 brouard 9667: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9668: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9669: }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 9670: Fixed[k]= 0;
9671: Dummy[k]= 1;
1.230 brouard 9672: nqfveff++;
1.234 brouard 9673: modell[k].maintype= FTYPE;
9674: modell[k].subtype= FQ;
9675: nsq++;
9676: TvarsQ[nsq]=Tvar[k];
9677: TvarsQind[nsq]=k;
1.232 brouard 9678: ncovf++;
1.234 brouard 9679: TvarF[ncovf]=Tvar[k];
9680: TvarFind[ncovf]=k;
1.231 brouard 9681: 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 9682: 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 9683: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9684: Fixed[k]= 1;
9685: Dummy[k]= 0;
1.225 brouard 9686: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9687: modell[k].maintype= VTYPE;
9688: modell[k].subtype= VD;
9689: nsd++;
9690: TvarsD[nsd]=Tvar[k];
9691: TvarsDind[nsd]=k;
9692: ncovv++; /* Only simple time varying variables */
9693: TvarV[ncovv]=Tvar[k];
1.242 brouard 9694: 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 9695: 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 */
9696: 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 9697: 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);
9698: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9699: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9700: Fixed[k]= 1;
9701: Dummy[k]= 1;
9702: nqtveff++;
9703: modell[k].maintype= VTYPE;
9704: modell[k].subtype= VQ;
9705: ncovv++; /* Only simple time varying variables */
9706: nsq++;
9707: TvarsQ[nsq]=Tvar[k];
9708: TvarsQind[nsq]=k;
9709: TvarV[ncovv]=Tvar[k];
1.242 brouard 9710: 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 9711: 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 */
9712: 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 9713: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9714: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9715: 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 9716: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9717: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9718: ncova++;
9719: TvarA[ncova]=Tvar[k];
9720: TvarAind[ncova]=k;
1.231 brouard 9721: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9722: Fixed[k]= 2;
9723: Dummy[k]= 2;
9724: modell[k].maintype= ATYPE;
9725: modell[k].subtype= APFD;
9726: /* ncoveff++; */
1.227 brouard 9727: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9728: Fixed[k]= 2;
9729: Dummy[k]= 3;
9730: modell[k].maintype= ATYPE;
9731: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9732: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9733: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9734: Fixed[k]= 3;
9735: Dummy[k]= 2;
9736: modell[k].maintype= ATYPE;
9737: modell[k].subtype= APVD; /* Product age * varying dummy */
9738: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9739: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9740: Fixed[k]= 3;
9741: Dummy[k]= 3;
9742: modell[k].maintype= ATYPE;
9743: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9744: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9745: }
9746: }else if (Typevar[k] == 2) { /* product without age */
9747: k1=Tposprod[k];
9748: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9749: if(Tvard[k1][2] <=ncovcol){
9750: Fixed[k]= 1;
9751: Dummy[k]= 0;
9752: modell[k].maintype= FTYPE;
9753: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9754: ncovf++; /* Fixed variables without age */
9755: TvarF[ncovf]=Tvar[k];
9756: TvarFind[ncovf]=k;
9757: }else if(Tvard[k1][2] <=ncovcol+nqv){
9758: Fixed[k]= 0; /* or 2 ?*/
9759: Dummy[k]= 1;
9760: modell[k].maintype= FTYPE;
9761: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9762: ncovf++; /* Varying variables without age */
9763: TvarF[ncovf]=Tvar[k];
9764: TvarFind[ncovf]=k;
9765: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9766: Fixed[k]= 1;
9767: Dummy[k]= 0;
9768: modell[k].maintype= VTYPE;
9769: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9770: ncovv++; /* Varying variables without age */
9771: TvarV[ncovv]=Tvar[k];
9772: TvarVind[ncovv]=k;
9773: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9774: Fixed[k]= 1;
9775: Dummy[k]= 1;
9776: modell[k].maintype= VTYPE;
9777: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9778: ncovv++; /* Varying variables without age */
9779: TvarV[ncovv]=Tvar[k];
9780: TvarVind[ncovv]=k;
9781: }
1.227 brouard 9782: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9783: if(Tvard[k1][2] <=ncovcol){
9784: Fixed[k]= 0; /* or 2 ?*/
9785: Dummy[k]= 1;
9786: modell[k].maintype= FTYPE;
9787: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9788: ncovf++; /* Fixed variables without age */
9789: TvarF[ncovf]=Tvar[k];
9790: TvarFind[ncovf]=k;
9791: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9792: Fixed[k]= 1;
9793: Dummy[k]= 1;
9794: modell[k].maintype= VTYPE;
9795: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9796: ncovv++; /* Varying variables without age */
9797: TvarV[ncovv]=Tvar[k];
9798: TvarVind[ncovv]=k;
9799: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9800: Fixed[k]= 1;
9801: Dummy[k]= 1;
9802: modell[k].maintype= VTYPE;
9803: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9804: ncovv++; /* Varying variables without age */
9805: TvarV[ncovv]=Tvar[k];
9806: TvarVind[ncovv]=k;
9807: ncovv++; /* Varying variables without age */
9808: TvarV[ncovv]=Tvar[k];
9809: TvarVind[ncovv]=k;
9810: }
1.227 brouard 9811: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9812: if(Tvard[k1][2] <=ncovcol){
9813: Fixed[k]= 1;
9814: Dummy[k]= 1;
9815: modell[k].maintype= VTYPE;
9816: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9817: ncovv++; /* Varying variables without age */
9818: TvarV[ncovv]=Tvar[k];
9819: TvarVind[ncovv]=k;
9820: }else if(Tvard[k1][2] <=ncovcol+nqv){
9821: Fixed[k]= 1;
9822: Dummy[k]= 1;
9823: modell[k].maintype= VTYPE;
9824: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9825: ncovv++; /* Varying variables without age */
9826: TvarV[ncovv]=Tvar[k];
9827: TvarVind[ncovv]=k;
9828: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9829: Fixed[k]= 1;
9830: Dummy[k]= 0;
9831: modell[k].maintype= VTYPE;
9832: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9833: ncovv++; /* Varying variables without age */
9834: TvarV[ncovv]=Tvar[k];
9835: TvarVind[ncovv]=k;
9836: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9837: Fixed[k]= 1;
9838: Dummy[k]= 1;
9839: modell[k].maintype= VTYPE;
9840: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9841: ncovv++; /* Varying variables without age */
9842: TvarV[ncovv]=Tvar[k];
9843: TvarVind[ncovv]=k;
9844: }
1.227 brouard 9845: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9846: if(Tvard[k1][2] <=ncovcol){
9847: Fixed[k]= 1;
9848: Dummy[k]= 1;
9849: modell[k].maintype= VTYPE;
9850: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9851: ncovv++; /* Varying variables without age */
9852: TvarV[ncovv]=Tvar[k];
9853: TvarVind[ncovv]=k;
9854: }else if(Tvard[k1][2] <=ncovcol+nqv){
9855: Fixed[k]= 1;
9856: Dummy[k]= 1;
9857: modell[k].maintype= VTYPE;
9858: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9859: ncovv++; /* Varying variables without age */
9860: TvarV[ncovv]=Tvar[k];
9861: TvarVind[ncovv]=k;
9862: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9863: Fixed[k]= 1;
9864: Dummy[k]= 1;
9865: modell[k].maintype= VTYPE;
9866: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9867: ncovv++; /* Varying variables without age */
9868: TvarV[ncovv]=Tvar[k];
9869: TvarVind[ncovv]=k;
9870: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9871: Fixed[k]= 1;
9872: Dummy[k]= 1;
9873: modell[k].maintype= VTYPE;
9874: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9875: ncovv++; /* Varying variables without age */
9876: TvarV[ncovv]=Tvar[k];
9877: TvarVind[ncovv]=k;
9878: }
1.227 brouard 9879: }else{
1.240 brouard 9880: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9881: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9882: } /*end k1*/
1.225 brouard 9883: }else{
1.226 brouard 9884: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9885: 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 9886: }
1.227 brouard 9887: 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 9888: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9889: 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]);
9890: }
9891: /* Searching for doublons in the model */
9892: for(k1=1; k1<= cptcovt;k1++){
9893: for(k2=1; k2 <k1;k2++){
9894: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9895: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9896: if(Tvar[k1]==Tvar[k2]){
9897: 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]]);
9898: 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);
9899: return(1);
9900: }
9901: }else if (Typevar[k1] ==2){
9902: k3=Tposprod[k1];
9903: k4=Tposprod[k2];
9904: 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])) ){
9905: 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]]);
9906: 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);
9907: return(1);
9908: }
9909: }
1.227 brouard 9910: }
9911: }
1.225 brouard 9912: }
9913: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9914: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9915: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9916: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9917: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9918: /*endread:*/
1.225 brouard 9919: printf("Exiting decodemodel: ");
9920: return (1);
1.136 brouard 9921: }
9922:
1.169 brouard 9923: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9924: {/* Check ages at death */
1.136 brouard 9925: int i, m;
1.218 brouard 9926: int firstone=0;
9927:
1.136 brouard 9928: for (i=1; i<=imx; i++) {
9929: for(m=2; (m<= maxwav); m++) {
9930: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9931: anint[m][i]=9999;
1.216 brouard 9932: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9933: s[m][i]=-1;
1.136 brouard 9934: }
9935: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9936: *nberr = *nberr + 1;
1.218 brouard 9937: if(firstone == 0){
9938: firstone=1;
1.260 brouard 9939: 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 9940: }
1.262 brouard 9941: 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 9942: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9943: }
9944: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9945: (*nberr)++;
1.259 brouard 9946: 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 9947: 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 9948: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9949: }
9950: }
9951: }
9952:
9953: for (i=1; i<=imx; i++) {
9954: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9955: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9956: 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 9957: if (s[m][i] >= nlstate+1) {
1.169 brouard 9958: if(agedc[i]>0){
9959: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9960: agev[m][i]=agedc[i];
1.214 brouard 9961: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9962: }else {
1.136 brouard 9963: if ((int)andc[i]!=9999){
9964: nbwarn++;
9965: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9966: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9967: agev[m][i]=-1;
9968: }
9969: }
1.169 brouard 9970: } /* agedc > 0 */
1.214 brouard 9971: } /* end if */
1.136 brouard 9972: else if(s[m][i] !=9){ /* Standard case, age in fractional
9973: years but with the precision of a month */
9974: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9975: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9976: agev[m][i]=1;
9977: else if(agev[m][i] < *agemin){
9978: *agemin=agev[m][i];
9979: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9980: }
9981: else if(agev[m][i] >*agemax){
9982: *agemax=agev[m][i];
1.156 brouard 9983: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9984: }
9985: /*agev[m][i]=anint[m][i]-annais[i];*/
9986: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9987: } /* en if 9*/
1.136 brouard 9988: else { /* =9 */
1.214 brouard 9989: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9990: agev[m][i]=1;
9991: s[m][i]=-1;
9992: }
9993: }
1.214 brouard 9994: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9995: agev[m][i]=1;
1.214 brouard 9996: else{
9997: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9998: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9999: agev[m][i]=0;
10000: }
10001: } /* End for lastpass */
10002: }
1.136 brouard 10003:
10004: for (i=1; i<=imx; i++) {
10005: for(m=firstpass; (m<=lastpass); m++){
10006: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10007: (*nberr)++;
1.136 brouard 10008: 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);
10009: 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);
10010: return 1;
10011: }
10012: }
10013: }
10014:
10015: /*for (i=1; i<=imx; i++){
10016: for (m=firstpass; (m<lastpass); m++){
10017: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10018: }
10019:
10020: }*/
10021:
10022:
1.139 brouard 10023: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10024: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10025:
10026: return (0);
1.164 brouard 10027: /* endread:*/
1.136 brouard 10028: printf("Exiting calandcheckages: ");
10029: return (1);
10030: }
10031:
1.172 brouard 10032: #if defined(_MSC_VER)
10033: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10034: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10035: //#include "stdafx.h"
10036: //#include <stdio.h>
10037: //#include <tchar.h>
10038: //#include <windows.h>
10039: //#include <iostream>
10040: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10041:
10042: LPFN_ISWOW64PROCESS fnIsWow64Process;
10043:
10044: BOOL IsWow64()
10045: {
10046: BOOL bIsWow64 = FALSE;
10047:
10048: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10049: // (HANDLE, PBOOL);
10050:
10051: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10052:
10053: HMODULE module = GetModuleHandle(_T("kernel32"));
10054: const char funcName[] = "IsWow64Process";
10055: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10056: GetProcAddress(module, funcName);
10057:
10058: if (NULL != fnIsWow64Process)
10059: {
10060: if (!fnIsWow64Process(GetCurrentProcess(),
10061: &bIsWow64))
10062: //throw std::exception("Unknown error");
10063: printf("Unknown error\n");
10064: }
10065: return bIsWow64 != FALSE;
10066: }
10067: #endif
1.177 brouard 10068:
1.191 brouard 10069: void syscompilerinfo(int logged)
1.167 brouard 10070: {
10071: /* #include "syscompilerinfo.h"*/
1.185 brouard 10072: /* command line Intel compiler 32bit windows, XP compatible:*/
10073: /* /GS /W3 /Gy
10074: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10075: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10076: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10077: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10078: */
10079: /* 64 bits */
1.185 brouard 10080: /*
10081: /GS /W3 /Gy
10082: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10083: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10084: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10085: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10086: /* Optimization are useless and O3 is slower than O2 */
10087: /*
10088: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10089: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10090: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10091: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10092: */
1.186 brouard 10093: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10094: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10095: /PDB:"visual studio
10096: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10097: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10098: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10099: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10100: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10101: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10102: uiAccess='false'"
10103: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10104: /NOLOGO /TLBID:1
10105: */
1.177 brouard 10106: #if defined __INTEL_COMPILER
1.178 brouard 10107: #if defined(__GNUC__)
10108: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10109: #endif
1.177 brouard 10110: #elif defined(__GNUC__)
1.179 brouard 10111: #ifndef __APPLE__
1.174 brouard 10112: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10113: #endif
1.177 brouard 10114: struct utsname sysInfo;
1.178 brouard 10115: int cross = CROSS;
10116: if (cross){
10117: printf("Cross-");
1.191 brouard 10118: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10119: }
1.174 brouard 10120: #endif
10121:
1.171 brouard 10122: #include <stdint.h>
1.178 brouard 10123:
1.191 brouard 10124: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10125: #if defined(__clang__)
1.191 brouard 10126: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10127: #endif
10128: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10129: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10130: #endif
10131: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10132: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10133: #endif
10134: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10135: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10136: #endif
10137: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10138: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10139: #endif
10140: #if defined(_MSC_VER)
1.191 brouard 10141: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10142: #endif
10143: #if defined(__PGI)
1.191 brouard 10144: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10145: #endif
10146: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10147: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10148: #endif
1.191 brouard 10149: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10150:
1.167 brouard 10151: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10152: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10153: // Windows (x64 and x86)
1.191 brouard 10154: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10155: #elif __unix__ // all unices, not all compilers
10156: // Unix
1.191 brouard 10157: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10158: #elif __linux__
10159: // linux
1.191 brouard 10160: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10161: #elif __APPLE__
1.174 brouard 10162: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10163: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10164: #endif
10165:
10166: /* __MINGW32__ */
10167: /* __CYGWIN__ */
10168: /* __MINGW64__ */
10169: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10170: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10171: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10172: /* _WIN64 // Defined for applications for Win64. */
10173: /* _M_X64 // Defined for compilations that target x64 processors. */
10174: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10175:
1.167 brouard 10176: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10177: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10178: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10179: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10180: #else
1.191 brouard 10181: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10182: #endif
10183:
1.169 brouard 10184: #if defined(__GNUC__)
10185: # if defined(__GNUC_PATCHLEVEL__)
10186: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10187: + __GNUC_MINOR__ * 100 \
10188: + __GNUC_PATCHLEVEL__)
10189: # else
10190: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10191: + __GNUC_MINOR__ * 100)
10192: # endif
1.174 brouard 10193: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10194: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10195:
10196: if (uname(&sysInfo) != -1) {
10197: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10198: 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 10199: }
10200: else
10201: perror("uname() error");
1.179 brouard 10202: //#ifndef __INTEL_COMPILER
10203: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10204: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10205: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10206: #endif
1.169 brouard 10207: #endif
1.172 brouard 10208:
10209: // void main()
10210: // {
1.169 brouard 10211: #if defined(_MSC_VER)
1.174 brouard 10212: if (IsWow64()){
1.191 brouard 10213: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10214: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10215: }
10216: else{
1.191 brouard 10217: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10218: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10219: }
1.172 brouard 10220: // printf("\nPress Enter to continue...");
10221: // getchar();
10222: // }
10223:
1.169 brouard 10224: #endif
10225:
1.167 brouard 10226:
1.219 brouard 10227: }
1.136 brouard 10228:
1.219 brouard 10229: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10230: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10231: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10232: /* double ftolpl = 1.e-10; */
1.180 brouard 10233: double age, agebase, agelim;
1.203 brouard 10234: double tot;
1.180 brouard 10235:
1.202 brouard 10236: strcpy(filerespl,"PL_");
10237: strcat(filerespl,fileresu);
10238: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10239: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10240: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10241: }
1.227 brouard 10242: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10243: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10244: pstamp(ficrespl);
1.203 brouard 10245: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10246: fprintf(ficrespl,"#Age ");
10247: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10248: fprintf(ficrespl,"\n");
1.180 brouard 10249:
1.219 brouard 10250: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10251:
1.219 brouard 10252: agebase=ageminpar;
10253: agelim=agemaxpar;
1.180 brouard 10254:
1.227 brouard 10255: /* i1=pow(2,ncoveff); */
1.234 brouard 10256: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10257: if (cptcovn < 1){i1=1;}
1.180 brouard 10258:
1.238 brouard 10259: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10260: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10261: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10262: continue;
1.235 brouard 10263:
1.238 brouard 10264: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10265: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10266: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10267: /* k=k+1; */
10268: /* to clean */
10269: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10270: fprintf(ficrespl,"#******");
10271: printf("#******");
10272: fprintf(ficlog,"#******");
10273: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10274: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10275: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10276: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10277: }
10278: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10279: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10280: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10281: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10282: }
10283: fprintf(ficrespl,"******\n");
10284: printf("******\n");
10285: fprintf(ficlog,"******\n");
10286: if(invalidvarcomb[k]){
10287: printf("\nCombination (%d) ignored because no case \n",k);
10288: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10289: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10290: continue;
10291: }
1.219 brouard 10292:
1.238 brouard 10293: fprintf(ficrespl,"#Age ");
10294: for(j=1;j<=cptcoveff;j++) {
10295: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10296: }
10297: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10298: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10299:
1.238 brouard 10300: for (age=agebase; age<=agelim; age++){
10301: /* for (age=agebase; age<=agebase; age++){ */
10302: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10303: fprintf(ficrespl,"%.0f ",age );
10304: for(j=1;j<=cptcoveff;j++)
10305: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10306: tot=0.;
10307: for(i=1; i<=nlstate;i++){
10308: tot += prlim[i][i];
10309: fprintf(ficrespl," %.5f", prlim[i][i]);
10310: }
10311: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10312: } /* Age */
10313: /* was end of cptcod */
10314: } /* cptcov */
10315: } /* nres */
1.219 brouard 10316: return 0;
1.180 brouard 10317: }
10318:
1.218 brouard 10319: 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){
10320: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10321:
10322: /* Computes the back prevalence limit for any combination of covariate values
10323: * at any age between ageminpar and agemaxpar
10324: */
1.235 brouard 10325: int i, j, k, i1, nres=0 ;
1.217 brouard 10326: /* double ftolpl = 1.e-10; */
10327: double age, agebase, agelim;
10328: double tot;
1.218 brouard 10329: /* double ***mobaverage; */
10330: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10331:
10332: strcpy(fileresplb,"PLB_");
10333: strcat(fileresplb,fileresu);
10334: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10335: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10336: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10337: }
10338: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10339: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10340: pstamp(ficresplb);
10341: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10342: fprintf(ficresplb,"#Age ");
10343: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10344: fprintf(ficresplb,"\n");
10345:
1.218 brouard 10346:
10347: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10348:
10349: agebase=ageminpar;
10350: agelim=agemaxpar;
10351:
10352:
1.227 brouard 10353: i1=pow(2,cptcoveff);
1.218 brouard 10354: if (cptcovn < 1){i1=1;}
1.227 brouard 10355:
1.238 brouard 10356: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10357: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10358: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10359: continue;
10360: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10361: fprintf(ficresplb,"#******");
10362: printf("#******");
10363: fprintf(ficlog,"#******");
10364: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10365: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10366: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10367: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10368: }
10369: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10370: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10371: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10372: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10373: }
10374: fprintf(ficresplb,"******\n");
10375: printf("******\n");
10376: fprintf(ficlog,"******\n");
10377: if(invalidvarcomb[k]){
10378: printf("\nCombination (%d) ignored because no cases \n",k);
10379: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10380: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10381: continue;
10382: }
1.218 brouard 10383:
1.238 brouard 10384: fprintf(ficresplb,"#Age ");
10385: for(j=1;j<=cptcoveff;j++) {
10386: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10387: }
10388: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10389: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10390:
10391:
1.238 brouard 10392: for (age=agebase; age<=agelim; age++){
10393: /* for (age=agebase; age<=agebase; age++){ */
10394: if(mobilavproj > 0){
10395: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10396: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10397: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10398: }else if (mobilavproj == 0){
10399: 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);
10400: 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);
10401: exit(1);
10402: }else{
10403: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10404: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10405: /* printf("TOTOT\n"); */
10406: /* exit(1); */
1.238 brouard 10407: }
10408: fprintf(ficresplb,"%.0f ",age );
10409: for(j=1;j<=cptcoveff;j++)
10410: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10411: tot=0.;
10412: for(i=1; i<=nlstate;i++){
10413: tot += bprlim[i][i];
10414: fprintf(ficresplb," %.5f", bprlim[i][i]);
10415: }
10416: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10417: } /* Age */
10418: /* was end of cptcod */
1.255 brouard 10419: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10420: } /* end of any combination */
10421: } /* end of nres */
1.218 brouard 10422: /* hBijx(p, bage, fage); */
10423: /* fclose(ficrespijb); */
10424:
10425: return 0;
1.217 brouard 10426: }
1.218 brouard 10427:
1.180 brouard 10428: int hPijx(double *p, int bage, int fage){
10429: /*------------- h Pij x at various ages ------------*/
10430:
10431: int stepsize;
10432: int agelim;
10433: int hstepm;
10434: int nhstepm;
1.235 brouard 10435: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10436:
10437: double agedeb;
10438: double ***p3mat;
10439:
1.201 brouard 10440: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10441: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10442: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10443: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10444: }
10445: printf("Computing pij: result on file '%s' \n", filerespij);
10446: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10447:
10448: stepsize=(int) (stepm+YEARM-1)/YEARM;
10449: /*if (stepm<=24) stepsize=2;*/
10450:
10451: agelim=AGESUP;
10452: hstepm=stepsize*YEARM; /* Every year of age */
10453: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10454:
1.180 brouard 10455: /* hstepm=1; aff par mois*/
10456: pstamp(ficrespij);
10457: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10458: i1= pow(2,cptcoveff);
1.218 brouard 10459: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10460: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10461: /* k=k+1; */
1.235 brouard 10462: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10463: for(k=1; k<=i1;k++){
1.253 brouard 10464: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10465: continue;
1.183 brouard 10466: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10467: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10468: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10469: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10470: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10471: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10472: }
1.183 brouard 10473: fprintf(ficrespij,"******\n");
10474:
10475: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10476: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10477: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10478:
10479: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10480:
1.183 brouard 10481: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10482: oldm=oldms;savm=savms;
1.235 brouard 10483: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10484: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10485: for(i=1; i<=nlstate;i++)
10486: for(j=1; j<=nlstate+ndeath;j++)
10487: fprintf(ficrespij," %1d-%1d",i,j);
10488: fprintf(ficrespij,"\n");
10489: for (h=0; h<=nhstepm; h++){
10490: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10491: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10492: for(i=1; i<=nlstate;i++)
10493: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10494: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10495: fprintf(ficrespij,"\n");
10496: }
1.183 brouard 10497: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10498: fprintf(ficrespij,"\n");
10499: }
1.180 brouard 10500: /*}*/
10501: }
1.218 brouard 10502: return 0;
1.180 brouard 10503: }
1.218 brouard 10504:
10505: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10506: /*------------- h Bij x at various ages ------------*/
10507:
10508: int stepsize;
1.218 brouard 10509: /* int agelim; */
10510: int ageminl;
1.217 brouard 10511: int hstepm;
10512: int nhstepm;
1.238 brouard 10513: int h, i, i1, j, k, nres;
1.218 brouard 10514:
1.217 brouard 10515: double agedeb;
10516: double ***p3mat;
1.218 brouard 10517:
10518: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10519: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10520: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10521: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10522: }
10523: printf("Computing pij back: result on file '%s' \n", filerespijb);
10524: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10525:
10526: stepsize=(int) (stepm+YEARM-1)/YEARM;
10527: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10528:
1.218 brouard 10529: /* agelim=AGESUP; */
10530: ageminl=30;
10531: hstepm=stepsize*YEARM; /* Every year of age */
10532: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10533:
10534: /* hstepm=1; aff par mois*/
10535: pstamp(ficrespijb);
1.255 brouard 10536: 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 10537: i1= pow(2,cptcoveff);
1.218 brouard 10538: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10539: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10540: /* k=k+1; */
1.238 brouard 10541: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10542: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10543: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10544: continue;
10545: fprintf(ficrespijb,"\n#****** ");
10546: for(j=1;j<=cptcoveff;j++)
10547: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10548: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10549: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10550: }
10551: fprintf(ficrespijb,"******\n");
1.264 brouard 10552: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10553: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10554: continue;
10555: }
10556:
10557: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10558: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10559: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10560: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10561: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10562:
10563: /* nhstepm=nhstepm*YEARM; aff par mois*/
10564:
1.266 brouard 10565: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10566: /* and memory limitations if stepm is small */
10567:
1.238 brouard 10568: /* oldm=oldms;savm=savms; */
10569: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10570: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10571: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10572: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10573: for(i=1; i<=nlstate;i++)
10574: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10575: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10576: fprintf(ficrespijb,"\n");
1.238 brouard 10577: for (h=0; h<=nhstepm; h++){
10578: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10579: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10580: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10581: for(i=1; i<=nlstate;i++)
10582: for(j=1; j<=nlstate+ndeath;j++)
10583: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10584: fprintf(ficrespijb,"\n");
10585: }
10586: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10587: fprintf(ficrespijb,"\n");
10588: } /* end age deb */
10589: } /* end combination */
10590: } /* end nres */
1.218 brouard 10591: return 0;
10592: } /* hBijx */
1.217 brouard 10593:
1.180 brouard 10594:
1.136 brouard 10595: /***********************************************/
10596: /**************** Main Program *****************/
10597: /***********************************************/
10598:
10599: int main(int argc, char *argv[])
10600: {
10601: #ifdef GSL
10602: const gsl_multimin_fminimizer_type *T;
10603: size_t iteri = 0, it;
10604: int rval = GSL_CONTINUE;
10605: int status = GSL_SUCCESS;
10606: double ssval;
10607: #endif
10608: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10609: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10610: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10611: int jj, ll, li, lj, lk;
1.136 brouard 10612: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10613: int num_filled;
1.136 brouard 10614: int itimes;
10615: int NDIM=2;
10616: int vpopbased=0;
1.235 brouard 10617: int nres=0;
1.258 brouard 10618: int endishere=0;
1.277 brouard 10619: int noffset=0;
1.274 brouard 10620: int ncurrv=0; /* Temporary variable */
10621:
1.164 brouard 10622: char ca[32], cb[32];
1.136 brouard 10623: /* FILE *fichtm; *//* Html File */
10624: /* FILE *ficgp;*/ /*Gnuplot File */
10625: struct stat info;
1.191 brouard 10626: double agedeb=0.;
1.194 brouard 10627:
10628: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10629: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10630:
1.165 brouard 10631: double fret;
1.191 brouard 10632: double dum=0.; /* Dummy variable */
1.136 brouard 10633: double ***p3mat;
1.218 brouard 10634: /* double ***mobaverage; */
1.164 brouard 10635:
10636: char line[MAXLINE];
1.197 brouard 10637: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10638:
1.234 brouard 10639: char modeltemp[MAXLINE];
1.230 brouard 10640: char resultline[MAXLINE];
10641:
1.136 brouard 10642: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10643: char *tok, *val; /* pathtot */
1.136 brouard 10644: int firstobs=1, lastobs=10;
1.195 brouard 10645: int c, h , cpt, c2;
1.191 brouard 10646: int jl=0;
10647: int i1, j1, jk, stepsize=0;
1.194 brouard 10648: int count=0;
10649:
1.164 brouard 10650: int *tab;
1.136 brouard 10651: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10652: int backcast=0;
1.136 brouard 10653: int mobilav=0,popforecast=0;
1.191 brouard 10654: int hstepm=0, nhstepm=0;
1.136 brouard 10655: int agemortsup;
10656: float sumlpop=0.;
10657: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10658: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10659:
1.191 brouard 10660: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10661: double ftolpl=FTOL;
10662: double **prlim;
1.217 brouard 10663: double **bprlim;
1.136 brouard 10664: double ***param; /* Matrix of parameters */
1.251 brouard 10665: double ***paramstart; /* Matrix of starting parameter values */
10666: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10667: double **matcov; /* Matrix of covariance */
1.203 brouard 10668: double **hess; /* Hessian matrix */
1.136 brouard 10669: double ***delti3; /* Scale */
10670: double *delti; /* Scale */
10671: double ***eij, ***vareij;
10672: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10673:
1.136 brouard 10674: double *epj, vepp;
1.164 brouard 10675:
1.273 brouard 10676: double dateprev1, dateprev2;
10677: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10678: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10679:
1.136 brouard 10680: double **ximort;
1.145 brouard 10681: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10682: int *dcwave;
10683:
1.164 brouard 10684: char z[1]="c";
1.136 brouard 10685:
10686: /*char *strt;*/
10687: char strtend[80];
1.126 brouard 10688:
1.164 brouard 10689:
1.126 brouard 10690: /* setlocale (LC_ALL, ""); */
10691: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10692: /* textdomain (PACKAGE); */
10693: /* setlocale (LC_CTYPE, ""); */
10694: /* setlocale (LC_MESSAGES, ""); */
10695:
10696: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10697: rstart_time = time(NULL);
10698: /* (void) gettimeofday(&start_time,&tzp);*/
10699: start_time = *localtime(&rstart_time);
1.126 brouard 10700: curr_time=start_time;
1.157 brouard 10701: /*tml = *localtime(&start_time.tm_sec);*/
10702: /* strcpy(strstart,asctime(&tml)); */
10703: strcpy(strstart,asctime(&start_time));
1.126 brouard 10704:
10705: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10706: /* tp.tm_sec = tp.tm_sec +86400; */
10707: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10708: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10709: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10710: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10711: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10712: /* strt=asctime(&tmg); */
10713: /* printf("Time(after) =%s",strstart); */
10714: /* (void) time (&time_value);
10715: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10716: * tm = *localtime(&time_value);
10717: * strstart=asctime(&tm);
10718: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10719: */
10720:
10721: nberr=0; /* Number of errors and warnings */
10722: nbwarn=0;
1.184 brouard 10723: #ifdef WIN32
10724: _getcwd(pathcd, size);
10725: #else
1.126 brouard 10726: getcwd(pathcd, size);
1.184 brouard 10727: #endif
1.191 brouard 10728: syscompilerinfo(0);
1.196 brouard 10729: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10730: if(argc <=1){
10731: printf("\nEnter the parameter file name: ");
1.205 brouard 10732: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10733: printf("ERROR Empty parameter file name\n");
10734: goto end;
10735: }
1.126 brouard 10736: i=strlen(pathr);
10737: if(pathr[i-1]=='\n')
10738: pathr[i-1]='\0';
1.156 brouard 10739: i=strlen(pathr);
1.205 brouard 10740: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10741: pathr[i-1]='\0';
1.205 brouard 10742: }
10743: i=strlen(pathr);
10744: if( i==0 ){
10745: printf("ERROR Empty parameter file name\n");
10746: goto end;
10747: }
10748: for (tok = pathr; tok != NULL; ){
1.126 brouard 10749: printf("Pathr |%s|\n",pathr);
10750: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10751: printf("val= |%s| pathr=%s\n",val,pathr);
10752: strcpy (pathtot, val);
10753: if(pathr[0] == '\0') break; /* Dirty */
10754: }
10755: }
10756: else{
10757: strcpy(pathtot,argv[1]);
10758: }
10759: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10760: /*cygwin_split_path(pathtot,path,optionfile);
10761: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10762: /* cutv(path,optionfile,pathtot,'\\');*/
10763:
10764: /* Split argv[0], imach program to get pathimach */
10765: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10766: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10767: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10768: /* strcpy(pathimach,argv[0]); */
10769: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10770: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10771: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10772: #ifdef WIN32
10773: _chdir(path); /* Can be a relative path */
10774: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10775: #else
1.126 brouard 10776: chdir(path); /* Can be a relative path */
1.184 brouard 10777: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10778: #endif
10779: printf("Current directory %s!\n",pathcd);
1.126 brouard 10780: strcpy(command,"mkdir ");
10781: strcat(command,optionfilefiname);
10782: if((outcmd=system(command)) != 0){
1.169 brouard 10783: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10784: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10785: /* fclose(ficlog); */
10786: /* exit(1); */
10787: }
10788: /* if((imk=mkdir(optionfilefiname))<0){ */
10789: /* perror("mkdir"); */
10790: /* } */
10791:
10792: /*-------- arguments in the command line --------*/
10793:
1.186 brouard 10794: /* Main Log file */
1.126 brouard 10795: strcat(filelog, optionfilefiname);
10796: strcat(filelog,".log"); /* */
10797: if((ficlog=fopen(filelog,"w"))==NULL) {
10798: printf("Problem with logfile %s\n",filelog);
10799: goto end;
10800: }
10801: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10802: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10803: fprintf(ficlog,"\nEnter the parameter file name: \n");
10804: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10805: path=%s \n\
10806: optionfile=%s\n\
10807: optionfilext=%s\n\
1.156 brouard 10808: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10809:
1.197 brouard 10810: syscompilerinfo(1);
1.167 brouard 10811:
1.126 brouard 10812: printf("Local time (at start):%s",strstart);
10813: fprintf(ficlog,"Local time (at start): %s",strstart);
10814: fflush(ficlog);
10815: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10816: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10817:
10818: /* */
10819: strcpy(fileres,"r");
10820: strcat(fileres, optionfilefiname);
1.201 brouard 10821: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10822: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10823: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10824:
1.186 brouard 10825: /* Main ---------arguments file --------*/
1.126 brouard 10826:
10827: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10828: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10829: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10830: fflush(ficlog);
1.149 brouard 10831: /* goto end; */
10832: exit(70);
1.126 brouard 10833: }
10834:
10835:
10836:
10837: strcpy(filereso,"o");
1.201 brouard 10838: strcat(filereso,fileresu);
1.126 brouard 10839: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10840: printf("Problem with Output resultfile: %s\n", filereso);
10841: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10842: fflush(ficlog);
10843: goto end;
10844: }
1.278 ! brouard 10845: /*-------- Rewriting parameter file ----------*/
! 10846: strcpy(rfileres,"r"); /* "Rparameterfile */
! 10847: strcat(rfileres,optionfilefiname); /* Parameter file first name */
! 10848: strcat(rfileres,"."); /* */
! 10849: strcat(rfileres,optionfilext); /* Other files have txt extension */
! 10850: if((ficres =fopen(rfileres,"w"))==NULL) {
! 10851: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
! 10852: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
! 10853: fflush(ficlog);
! 10854: goto end;
! 10855: }
! 10856: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 10857:
1.278 ! brouard 10858:
1.126 brouard 10859: /* Reads comments: lines beginning with '#' */
10860: numlinepar=0;
1.277 brouard 10861: /* Is it a BOM UTF-8 Windows file? */
10862: /* First parameter line */
1.197 brouard 10863: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 10864: noffset=0;
10865: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10866: {
10867: noffset=noffset+3;
10868: printf("# File is an UTF8 Bom.\n"); // 0xBF
10869: }
10870: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
10871: {
10872: noffset=noffset+2;
10873: printf("# File is an UTF16BE BOM file\n");
10874: }
10875: else if( line[0] == 0 && line[1] == 0)
10876: {
10877: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10878: noffset=noffset+4;
10879: printf("# File is an UTF16BE BOM file\n");
10880: }
10881: } else{
10882: ;/*printf(" Not a BOM file\n");*/
10883: }
10884:
1.197 brouard 10885: /* If line starts with a # it is a comment */
1.277 brouard 10886: if (line[noffset] == '#') {
1.197 brouard 10887: numlinepar++;
10888: fputs(line,stdout);
10889: fputs(line,ficparo);
1.278 ! brouard 10890: fputs(line,ficres);
1.197 brouard 10891: fputs(line,ficlog);
10892: continue;
10893: }else
10894: break;
10895: }
10896: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10897: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10898: if (num_filled != 5) {
10899: printf("Should be 5 parameters\n");
10900: }
1.126 brouard 10901: numlinepar++;
1.197 brouard 10902: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10903: }
10904: /* Second parameter line */
10905: while(fgets(line, MAXLINE, ficpar)) {
10906: /* If line starts with a # it is a comment */
10907: if (line[0] == '#') {
10908: numlinepar++;
10909: fputs(line,stdout);
10910: fputs(line,ficparo);
1.278 ! brouard 10911: fputs(line,ficres);
1.197 brouard 10912: fputs(line,ficlog);
10913: continue;
10914: }else
10915: break;
10916: }
1.223 brouard 10917: 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", \
10918: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10919: if (num_filled != 11) {
10920: 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 10921: printf("but line=%s\n",line);
1.197 brouard 10922: }
1.223 brouard 10923: 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 10924: }
1.203 brouard 10925: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10926: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10927: /* Third parameter line */
10928: while(fgets(line, MAXLINE, ficpar)) {
10929: /* If line starts with a # it is a comment */
10930: if (line[0] == '#') {
10931: numlinepar++;
10932: fputs(line,stdout);
10933: fputs(line,ficparo);
1.278 ! brouard 10934: fputs(line,ficres);
1.197 brouard 10935: fputs(line,ficlog);
10936: continue;
10937: }else
10938: break;
10939: }
1.201 brouard 10940: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 brouard 10941: if (num_filled == 0){
10942: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10943: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10944: model[0]='\0';
10945: goto end;
10946: } else if (num_filled != 1){
1.197 brouard 10947: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10948: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10949: model[0]='\0';
10950: goto end;
10951: }
10952: else{
10953: if (model[0]=='+'){
10954: for(i=1; i<=strlen(model);i++)
10955: modeltemp[i-1]=model[i];
1.201 brouard 10956: strcpy(model,modeltemp);
1.197 brouard 10957: }
10958: }
1.199 brouard 10959: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10960: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10961: }
10962: /* 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); */
10963: /* numlinepar=numlinepar+3; /\* In general *\/ */
10964: /* 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 10965: 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);
10966: 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 10967: fflush(ficlog);
1.190 brouard 10968: /* if(model[0]=='#'|| model[0]== '\0'){ */
10969: if(model[0]=='#'){
1.187 brouard 10970: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10971: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10972: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10973: if(mle != -1){
10974: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10975: exit(1);
10976: }
10977: }
1.126 brouard 10978: while((c=getc(ficpar))=='#' && c!= EOF){
10979: ungetc(c,ficpar);
10980: fgets(line, MAXLINE, ficpar);
10981: numlinepar++;
1.195 brouard 10982: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10983: z[0]=line[1];
10984: }
10985: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10986: fputs(line, stdout);
10987: //puts(line);
1.126 brouard 10988: fputs(line,ficparo);
10989: fputs(line,ficlog);
10990: }
10991: ungetc(c,ficpar);
10992:
10993:
1.145 brouard 10994: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 10995: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
10996: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
10997: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 10998: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10999: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11000: v1+v2*age+v2*v3 makes cptcovn = 3
11001: */
11002: if (strlen(model)>1)
1.187 brouard 11003: 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 11004: else
1.187 brouard 11005: ncovmodel=2; /* Constant and age */
1.133 brouard 11006: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11007: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11008: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11009: 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);
11010: 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);
11011: fflush(stdout);
11012: fclose (ficlog);
11013: goto end;
11014: }
1.126 brouard 11015: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11016: delti=delti3[1][1];
11017: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11018: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11019: /* We could also provide initial parameters values giving by simple logistic regression
11020: * only one way, that is without matrix product. We will have nlstate maximizations */
11021: /* for(i=1;i<nlstate;i++){ */
11022: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11023: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11024: /* } */
1.126 brouard 11025: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11026: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11027: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11028: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11029: fclose (ficparo);
11030: fclose (ficlog);
11031: goto end;
11032: exit(0);
1.220 brouard 11033: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11034: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11035: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11036: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11037: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11038: matcov=matrix(1,npar,1,npar);
1.203 brouard 11039: hess=matrix(1,npar,1,npar);
1.220 brouard 11040: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11041: /* Read guessed parameters */
1.126 brouard 11042: /* Reads comments: lines beginning with '#' */
11043: while((c=getc(ficpar))=='#' && c!= EOF){
11044: ungetc(c,ficpar);
11045: fgets(line, MAXLINE, ficpar);
11046: numlinepar++;
1.141 brouard 11047: fputs(line,stdout);
1.126 brouard 11048: fputs(line,ficparo);
11049: fputs(line,ficlog);
11050: }
11051: ungetc(c,ficpar);
11052:
11053: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11054: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11055: for(i=1; i <=nlstate; i++){
1.234 brouard 11056: j=0;
1.126 brouard 11057: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11058: if(jj==i) continue;
11059: j++;
11060: fscanf(ficpar,"%1d%1d",&i1,&j1);
11061: if ((i1 != i) || (j1 != jj)){
11062: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11063: It might be a problem of design; if ncovcol and the model are correct\n \
11064: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11065: exit(1);
11066: }
11067: fprintf(ficparo,"%1d%1d",i1,j1);
11068: if(mle==1)
11069: printf("%1d%1d",i,jj);
11070: fprintf(ficlog,"%1d%1d",i,jj);
11071: for(k=1; k<=ncovmodel;k++){
11072: fscanf(ficpar," %lf",¶m[i][j][k]);
11073: if(mle==1){
11074: printf(" %lf",param[i][j][k]);
11075: fprintf(ficlog," %lf",param[i][j][k]);
11076: }
11077: else
11078: fprintf(ficlog," %lf",param[i][j][k]);
11079: fprintf(ficparo," %lf",param[i][j][k]);
11080: }
11081: fscanf(ficpar,"\n");
11082: numlinepar++;
11083: if(mle==1)
11084: printf("\n");
11085: fprintf(ficlog,"\n");
11086: fprintf(ficparo,"\n");
1.126 brouard 11087: }
11088: }
11089: fflush(ficlog);
1.234 brouard 11090:
1.251 brouard 11091: /* Reads parameters values */
1.126 brouard 11092: p=param[1][1];
1.251 brouard 11093: pstart=paramstart[1][1];
1.126 brouard 11094:
11095: /* Reads comments: lines beginning with '#' */
11096: while((c=getc(ficpar))=='#' && c!= EOF){
11097: ungetc(c,ficpar);
11098: fgets(line, MAXLINE, ficpar);
11099: numlinepar++;
1.141 brouard 11100: fputs(line,stdout);
1.126 brouard 11101: fputs(line,ficparo);
11102: fputs(line,ficlog);
11103: }
11104: ungetc(c,ficpar);
11105:
11106: for(i=1; i <=nlstate; i++){
11107: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11108: fscanf(ficpar,"%1d%1d",&i1,&j1);
11109: if ( (i1-i) * (j1-j) != 0){
11110: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11111: exit(1);
11112: }
11113: printf("%1d%1d",i,j);
11114: fprintf(ficparo,"%1d%1d",i1,j1);
11115: fprintf(ficlog,"%1d%1d",i1,j1);
11116: for(k=1; k<=ncovmodel;k++){
11117: fscanf(ficpar,"%le",&delti3[i][j][k]);
11118: printf(" %le",delti3[i][j][k]);
11119: fprintf(ficparo," %le",delti3[i][j][k]);
11120: fprintf(ficlog," %le",delti3[i][j][k]);
11121: }
11122: fscanf(ficpar,"\n");
11123: numlinepar++;
11124: printf("\n");
11125: fprintf(ficparo,"\n");
11126: fprintf(ficlog,"\n");
1.126 brouard 11127: }
11128: }
11129: fflush(ficlog);
1.234 brouard 11130:
1.145 brouard 11131: /* Reads covariance matrix */
1.126 brouard 11132: delti=delti3[1][1];
1.220 brouard 11133:
11134:
1.126 brouard 11135: /* 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 11136:
1.126 brouard 11137: /* Reads comments: lines beginning with '#' */
11138: while((c=getc(ficpar))=='#' && c!= EOF){
11139: ungetc(c,ficpar);
11140: fgets(line, MAXLINE, ficpar);
11141: numlinepar++;
1.141 brouard 11142: fputs(line,stdout);
1.126 brouard 11143: fputs(line,ficparo);
11144: fputs(line,ficlog);
11145: }
11146: ungetc(c,ficpar);
1.220 brouard 11147:
1.126 brouard 11148: matcov=matrix(1,npar,1,npar);
1.203 brouard 11149: hess=matrix(1,npar,1,npar);
1.131 brouard 11150: for(i=1; i <=npar; i++)
11151: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11152:
1.194 brouard 11153: /* Scans npar lines */
1.126 brouard 11154: for(i=1; i <=npar; i++){
1.226 brouard 11155: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11156: if(count != 3){
1.226 brouard 11157: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11158: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11159: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11160: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11161: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11162: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11163: exit(1);
1.220 brouard 11164: }else{
1.226 brouard 11165: if(mle==1)
11166: printf("%1d%1d%d",i1,j1,jk);
11167: }
11168: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11169: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11170: for(j=1; j <=i; j++){
1.226 brouard 11171: fscanf(ficpar," %le",&matcov[i][j]);
11172: if(mle==1){
11173: printf(" %.5le",matcov[i][j]);
11174: }
11175: fprintf(ficlog," %.5le",matcov[i][j]);
11176: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11177: }
11178: fscanf(ficpar,"\n");
11179: numlinepar++;
11180: if(mle==1)
1.220 brouard 11181: printf("\n");
1.126 brouard 11182: fprintf(ficlog,"\n");
11183: fprintf(ficparo,"\n");
11184: }
1.194 brouard 11185: /* End of read covariance matrix npar lines */
1.126 brouard 11186: for(i=1; i <=npar; i++)
11187: for(j=i+1;j<=npar;j++)
1.226 brouard 11188: matcov[i][j]=matcov[j][i];
1.126 brouard 11189:
11190: if(mle==1)
11191: printf("\n");
11192: fprintf(ficlog,"\n");
11193:
11194: fflush(ficlog);
11195:
11196: } /* End of mle != -3 */
1.218 brouard 11197:
1.186 brouard 11198: /* Main data
11199: */
1.126 brouard 11200: n= lastobs;
11201: num=lvector(1,n);
11202: moisnais=vector(1,n);
11203: annais=vector(1,n);
11204: moisdc=vector(1,n);
11205: andc=vector(1,n);
1.220 brouard 11206: weight=vector(1,n);
1.126 brouard 11207: agedc=vector(1,n);
11208: cod=ivector(1,n);
1.220 brouard 11209: for(i=1;i<=n;i++){
1.234 brouard 11210: num[i]=0;
11211: moisnais[i]=0;
11212: annais[i]=0;
11213: moisdc[i]=0;
11214: andc[i]=0;
11215: agedc[i]=0;
11216: cod[i]=0;
11217: weight[i]=1.0; /* Equal weights, 1 by default */
11218: }
1.126 brouard 11219: mint=matrix(1,maxwav,1,n);
11220: anint=matrix(1,maxwav,1,n);
1.131 brouard 11221: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11222: tab=ivector(1,NCOVMAX);
1.144 brouard 11223: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11224: 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 11225:
1.136 brouard 11226: /* Reads data from file datafile */
11227: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11228: goto end;
11229:
11230: /* Calculation of the number of parameters from char model */
1.234 brouard 11231: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11232: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11233: k=3 V4 Tvar[k=3]= 4 (from V4)
11234: k=2 V1 Tvar[k=2]= 1 (from V1)
11235: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11236: */
11237:
11238: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11239: TvarsDind=ivector(1,NCOVMAX); /* */
11240: TvarsD=ivector(1,NCOVMAX); /* */
11241: TvarsQind=ivector(1,NCOVMAX); /* */
11242: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11243: TvarF=ivector(1,NCOVMAX); /* */
11244: TvarFind=ivector(1,NCOVMAX); /* */
11245: TvarV=ivector(1,NCOVMAX); /* */
11246: TvarVind=ivector(1,NCOVMAX); /* */
11247: TvarA=ivector(1,NCOVMAX); /* */
11248: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11249: TvarFD=ivector(1,NCOVMAX); /* */
11250: TvarFDind=ivector(1,NCOVMAX); /* */
11251: TvarFQ=ivector(1,NCOVMAX); /* */
11252: TvarFQind=ivector(1,NCOVMAX); /* */
11253: TvarVD=ivector(1,NCOVMAX); /* */
11254: TvarVDind=ivector(1,NCOVMAX); /* */
11255: TvarVQ=ivector(1,NCOVMAX); /* */
11256: TvarVQind=ivector(1,NCOVMAX); /* */
11257:
1.230 brouard 11258: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11259: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11260: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11261: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11262: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11263: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11264: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11265: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11266: */
11267: /* For model-covariate k tells which data-covariate to use but
11268: because this model-covariate is a construction we invent a new column
11269: ncovcol + k1
11270: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11271: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11272: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11273: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11274: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11275: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11276: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11277: */
1.145 brouard 11278: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11279: 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 11280: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11281: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11282: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11283: 4 covariates (3 plus signs)
11284: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11285: */
1.230 brouard 11286: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11287: * individual dummy, fixed or varying:
11288: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11289: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11290: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11291: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11292: * Tmodelind[1]@9={9,0,3,2,}*/
11293: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11294: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11295: * individual quantitative, fixed or varying:
11296: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11297: * 3, 1, 0, 0, 0, 0, 0, 0},
11298: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11299: /* Main decodemodel */
11300:
1.187 brouard 11301:
1.223 brouard 11302: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11303: goto end;
11304:
1.137 brouard 11305: if((double)(lastobs-imx)/(double)imx > 1.10){
11306: nbwarn++;
11307: 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);
11308: 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);
11309: }
1.136 brouard 11310: /* if(mle==1){*/
1.137 brouard 11311: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11312: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11313: }
11314:
11315: /*-calculation of age at interview from date of interview and age at death -*/
11316: agev=matrix(1,maxwav,1,imx);
11317:
11318: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11319: goto end;
11320:
1.126 brouard 11321:
1.136 brouard 11322: agegomp=(int)agemin;
11323: free_vector(moisnais,1,n);
11324: free_vector(annais,1,n);
1.126 brouard 11325: /* free_matrix(mint,1,maxwav,1,n);
11326: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11327: /* free_vector(moisdc,1,n); */
11328: /* free_vector(andc,1,n); */
1.145 brouard 11329: /* */
11330:
1.126 brouard 11331: wav=ivector(1,imx);
1.214 brouard 11332: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11333: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11334: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11335: 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.*/
11336: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11337: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11338:
11339: /* Concatenates waves */
1.214 brouard 11340: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11341: Death is a valid wave (if date is known).
11342: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11343: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11344: and mw[mi+1][i]. dh depends on stepm.
11345: */
11346:
1.126 brouard 11347: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11348: /* Concatenates waves */
1.145 brouard 11349:
1.215 brouard 11350: free_vector(moisdc,1,n);
11351: free_vector(andc,1,n);
11352:
1.126 brouard 11353: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11354: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11355: ncodemax[1]=1;
1.145 brouard 11356: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11357: cptcoveff=0;
1.220 brouard 11358: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11359: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11360: }
11361:
11362: ncovcombmax=pow(2,cptcoveff);
11363: invalidvarcomb=ivector(1, ncovcombmax);
11364: for(i=1;i<ncovcombmax;i++)
11365: invalidvarcomb[i]=0;
11366:
1.211 brouard 11367: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11368: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11369: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11370:
1.200 brouard 11371: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11372: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11373: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11374: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11375: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11376: * (currently 0 or 1) in the data.
11377: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11378: * corresponding modality (h,j).
11379: */
11380:
1.145 brouard 11381: h=0;
11382: /*if (cptcovn > 0) */
1.126 brouard 11383: m=pow(2,cptcoveff);
11384:
1.144 brouard 11385: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11386: * For k=4 covariates, h goes from 1 to m=2**k
11387: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11388: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11389: * h\k 1 2 3 4
1.143 brouard 11390: *______________________________
11391: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11392: * 2 2 1 1 1
11393: * 3 i=2 1 2 1 1
11394: * 4 2 2 1 1
11395: * 5 i=3 1 i=2 1 2 1
11396: * 6 2 1 2 1
11397: * 7 i=4 1 2 2 1
11398: * 8 2 2 2 1
1.197 brouard 11399: * 9 i=5 1 i=3 1 i=2 1 2
11400: * 10 2 1 1 2
11401: * 11 i=6 1 2 1 2
11402: * 12 2 2 1 2
11403: * 13 i=7 1 i=4 1 2 2
11404: * 14 2 1 2 2
11405: * 15 i=8 1 2 2 2
11406: * 16 2 2 2 2
1.143 brouard 11407: */
1.212 brouard 11408: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11409: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11410: * and the value of each covariate?
11411: * V1=1, V2=1, V3=2, V4=1 ?
11412: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11413: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11414: * In order to get the real value in the data, we use nbcode
11415: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11416: * We are keeping this crazy system in order to be able (in the future?)
11417: * to have more than 2 values (0 or 1) for a covariate.
11418: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11419: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11420: * bbbbbbbb
11421: * 76543210
11422: * h-1 00000101 (6-1=5)
1.219 brouard 11423: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11424: * &
11425: * 1 00000001 (1)
1.219 brouard 11426: * 00000000 = 1 & ((h-1) >> (k-1))
11427: * +1= 00000001 =1
1.211 brouard 11428: *
11429: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11430: * h' 1101 =2^3+2^2+0x2^1+2^0
11431: * >>k' 11
11432: * & 00000001
11433: * = 00000001
11434: * +1 = 00000010=2 = codtabm(14,3)
11435: * Reverse h=6 and m=16?
11436: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11437: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11438: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11439: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11440: * V3=decodtabm(14,3,2**4)=2
11441: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11442: *(h-1) >> (j-1) 0011 =13 >> 2
11443: * &1 000000001
11444: * = 000000001
11445: * +1= 000000010 =2
11446: * 2211
11447: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11448: * V3=2
1.220 brouard 11449: * codtabm and decodtabm are identical
1.211 brouard 11450: */
11451:
1.145 brouard 11452:
11453: free_ivector(Ndum,-1,NCOVMAX);
11454:
11455:
1.126 brouard 11456:
1.186 brouard 11457: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11458: strcpy(optionfilegnuplot,optionfilefiname);
11459: if(mle==-3)
1.201 brouard 11460: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11461: strcat(optionfilegnuplot,".gp");
11462:
11463: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11464: printf("Problem with file %s",optionfilegnuplot);
11465: }
11466: else{
1.204 brouard 11467: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11468: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11469: //fprintf(ficgp,"set missing 'NaNq'\n");
11470: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11471: }
11472: /* fclose(ficgp);*/
1.186 brouard 11473:
11474:
11475: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11476:
11477: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11478: if(mle==-3)
1.201 brouard 11479: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11480: strcat(optionfilehtm,".htm");
11481: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11482: printf("Problem with %s \n",optionfilehtm);
11483: exit(0);
1.126 brouard 11484: }
11485:
11486: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11487: strcat(optionfilehtmcov,"-cov.htm");
11488: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11489: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11490: }
11491: else{
11492: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11493: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11494: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11495: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11496: }
11497:
1.213 brouard 11498: 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 11499: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11500: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11501: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11502: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11503: \n\
11504: <hr size=\"2\" color=\"#EC5E5E\">\
11505: <ul><li><h4>Parameter files</h4>\n\
11506: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11507: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11508: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11509: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11510: - Date and time at start: %s</ul>\n",\
11511: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11512: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11513: fileres,fileres,\
11514: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11515: fflush(fichtm);
11516:
11517: strcpy(pathr,path);
11518: strcat(pathr,optionfilefiname);
1.184 brouard 11519: #ifdef WIN32
11520: _chdir(optionfilefiname); /* Move to directory named optionfile */
11521: #else
1.126 brouard 11522: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11523: #endif
11524:
1.126 brouard 11525:
1.220 brouard 11526: /* Calculates basic frequencies. Computes observed prevalence at single age
11527: and for any valid combination of covariates
1.126 brouard 11528: and prints on file fileres'p'. */
1.251 brouard 11529: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11530: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11531:
11532: fprintf(fichtm,"\n");
1.274 brouard 11533: fprintf(fichtm,"<h4>Parameter line 2</h4><ul><li>Tolerance for the convergence of the likelihood: ftol=%f \n<li>Interval for the elementary matrix (in month): stepm=%d",\
11534: ftol, stepm);
11535: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11536: ncurrv=1;
11537: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11538: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11539: ncurrv=i;
11540: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
11541: fprintf(fichtm,"\n<li> Number of time varying (wave varying) covariates: ntv=%d ", ntv);
11542: ncurrv=i;
11543: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
11544: fprintf(fichtm,"\n<li>Number of quantitative time varying covariates: nqtv=%d ", nqtv);
11545: ncurrv=i;
11546: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11547: fprintf(fichtm,"\n<li>Weights column \n<br>Number of alive states: nlstate=%d <br>Number of death states (not really implemented): ndeath=%d \n<li>Number of waves: maxwav=%d \n<li>Parameter for maximization (1), using parameter values (0), for design of parameters and variance-covariance matrix: mle=%d \n<li>Does the weight column be taken into account (1), or not (0): weight=%d</ul>\n", \
11548: nlstate, ndeath, maxwav, mle, weightopt);
11549:
11550: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11551: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11552:
11553:
11554: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11555: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11556: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11557: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11558: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11559: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11560: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11561: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11562: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11563:
1.126 brouard 11564: /* For Powell, parameters are in a vector p[] starting at p[1]
11565: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11566: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11567:
11568: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11569: /* For mortality only */
1.126 brouard 11570: if (mle==-3){
1.136 brouard 11571: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11572: for(i=1;i<=NDIM;i++)
11573: for(j=1;j<=NDIM;j++)
11574: ximort[i][j]=0.;
1.186 brouard 11575: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11576: cens=ivector(1,n);
11577: ageexmed=vector(1,n);
11578: agecens=vector(1,n);
11579: dcwave=ivector(1,n);
1.223 brouard 11580:
1.126 brouard 11581: for (i=1; i<=imx; i++){
11582: dcwave[i]=-1;
11583: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11584: if (s[m][i]>nlstate) {
11585: dcwave[i]=m;
11586: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11587: break;
11588: }
1.126 brouard 11589: }
1.226 brouard 11590:
1.126 brouard 11591: for (i=1; i<=imx; i++) {
11592: if (wav[i]>0){
1.226 brouard 11593: ageexmed[i]=agev[mw[1][i]][i];
11594: j=wav[i];
11595: agecens[i]=1.;
11596:
11597: if (ageexmed[i]> 1 && wav[i] > 0){
11598: agecens[i]=agev[mw[j][i]][i];
11599: cens[i]= 1;
11600: }else if (ageexmed[i]< 1)
11601: cens[i]= -1;
11602: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11603: cens[i]=0 ;
1.126 brouard 11604: }
11605: else cens[i]=-1;
11606: }
11607:
11608: for (i=1;i<=NDIM;i++) {
11609: for (j=1;j<=NDIM;j++)
1.226 brouard 11610: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11611: }
11612:
1.145 brouard 11613: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11614: /*printf("%lf %lf", p[1], p[2]);*/
11615:
11616:
1.136 brouard 11617: #ifdef GSL
11618: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11619: #else
1.126 brouard 11620: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11621: #endif
1.201 brouard 11622: strcpy(filerespow,"POW-MORT_");
11623: strcat(filerespow,fileresu);
1.126 brouard 11624: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11625: printf("Problem with resultfile: %s\n", filerespow);
11626: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11627: }
1.136 brouard 11628: #ifdef GSL
11629: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11630: #else
1.126 brouard 11631: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11632: #endif
1.126 brouard 11633: /* for (i=1;i<=nlstate;i++)
11634: for(j=1;j<=nlstate+ndeath;j++)
11635: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11636: */
11637: fprintf(ficrespow,"\n");
1.136 brouard 11638: #ifdef GSL
11639: /* gsl starts here */
11640: T = gsl_multimin_fminimizer_nmsimplex;
11641: gsl_multimin_fminimizer *sfm = NULL;
11642: gsl_vector *ss, *x;
11643: gsl_multimin_function minex_func;
11644:
11645: /* Initial vertex size vector */
11646: ss = gsl_vector_alloc (NDIM);
11647:
11648: if (ss == NULL){
11649: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11650: }
11651: /* Set all step sizes to 1 */
11652: gsl_vector_set_all (ss, 0.001);
11653:
11654: /* Starting point */
1.126 brouard 11655:
1.136 brouard 11656: x = gsl_vector_alloc (NDIM);
11657:
11658: if (x == NULL){
11659: gsl_vector_free(ss);
11660: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11661: }
11662:
11663: /* Initialize method and iterate */
11664: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11665: /* gsl_vector_set(x, 0, 0.0268); */
11666: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11667: gsl_vector_set(x, 0, p[1]);
11668: gsl_vector_set(x, 1, p[2]);
11669:
11670: minex_func.f = &gompertz_f;
11671: minex_func.n = NDIM;
11672: minex_func.params = (void *)&p; /* ??? */
11673:
11674: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11675: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11676:
11677: printf("Iterations beginning .....\n\n");
11678: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11679:
11680: iteri=0;
11681: while (rval == GSL_CONTINUE){
11682: iteri++;
11683: status = gsl_multimin_fminimizer_iterate(sfm);
11684:
11685: if (status) printf("error: %s\n", gsl_strerror (status));
11686: fflush(0);
11687:
11688: if (status)
11689: break;
11690:
11691: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11692: ssval = gsl_multimin_fminimizer_size (sfm);
11693:
11694: if (rval == GSL_SUCCESS)
11695: printf ("converged to a local maximum at\n");
11696:
11697: printf("%5d ", iteri);
11698: for (it = 0; it < NDIM; it++){
11699: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11700: }
11701: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11702: }
11703:
11704: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11705:
11706: gsl_vector_free(x); /* initial values */
11707: gsl_vector_free(ss); /* inital step size */
11708: for (it=0; it<NDIM; it++){
11709: p[it+1]=gsl_vector_get(sfm->x,it);
11710: fprintf(ficrespow," %.12lf", p[it]);
11711: }
11712: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11713: #endif
11714: #ifdef POWELL
11715: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11716: #endif
1.126 brouard 11717: fclose(ficrespow);
11718:
1.203 brouard 11719: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11720:
11721: for(i=1; i <=NDIM; i++)
11722: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11723: matcov[i][j]=matcov[j][i];
1.126 brouard 11724:
11725: printf("\nCovariance matrix\n ");
1.203 brouard 11726: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11727: for(i=1; i <=NDIM; i++) {
11728: for(j=1;j<=NDIM;j++){
1.220 brouard 11729: printf("%f ",matcov[i][j]);
11730: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11731: }
1.203 brouard 11732: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11733: }
11734:
11735: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11736: for (i=1;i<=NDIM;i++) {
1.126 brouard 11737: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11738: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11739: }
1.126 brouard 11740: lsurv=vector(1,AGESUP);
11741: lpop=vector(1,AGESUP);
11742: tpop=vector(1,AGESUP);
11743: lsurv[agegomp]=100000;
11744:
11745: for (k=agegomp;k<=AGESUP;k++) {
11746: agemortsup=k;
11747: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11748: }
11749:
11750: for (k=agegomp;k<agemortsup;k++)
11751: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11752:
11753: for (k=agegomp;k<agemortsup;k++){
11754: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11755: sumlpop=sumlpop+lpop[k];
11756: }
11757:
11758: tpop[agegomp]=sumlpop;
11759: for (k=agegomp;k<(agemortsup-3);k++){
11760: /* tpop[k+1]=2;*/
11761: tpop[k+1]=tpop[k]-lpop[k];
11762: }
11763:
11764:
11765: printf("\nAge lx qx dx Lx Tx e(x)\n");
11766: for (k=agegomp;k<(agemortsup-2);k++)
11767: 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]);
11768:
11769:
11770: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11771: ageminpar=50;
11772: agemaxpar=100;
1.194 brouard 11773: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11774: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11775: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11776: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11777: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11778: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11779: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11780: }else{
11781: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11782: 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 11783: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11784: }
1.201 brouard 11785: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11786: stepm, weightopt,\
11787: model,imx,p,matcov,agemortsup);
11788:
11789: free_vector(lsurv,1,AGESUP);
11790: free_vector(lpop,1,AGESUP);
11791: free_vector(tpop,1,AGESUP);
1.220 brouard 11792: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11793: free_ivector(cens,1,n);
11794: free_vector(agecens,1,n);
11795: free_ivector(dcwave,1,n);
1.220 brouard 11796: #ifdef GSL
1.136 brouard 11797: #endif
1.186 brouard 11798: } /* Endof if mle==-3 mortality only */
1.205 brouard 11799: /* Standard */
11800: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11801: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11802: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11803: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11804: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11805: for (k=1; k<=npar;k++)
11806: printf(" %d %8.5f",k,p[k]);
11807: printf("\n");
1.205 brouard 11808: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11809: /* mlikeli uses func not funcone */
1.247 brouard 11810: /* for(i=1;i<nlstate;i++){ */
11811: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11812: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11813: /* } */
1.205 brouard 11814: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11815: }
11816: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11817: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11818: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11819: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11820: }
11821: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11822: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11823: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11824: for (k=1; k<=npar;k++)
11825: printf(" %d %8.5f",k,p[k]);
11826: printf("\n");
11827:
11828: /*--------- results files --------------*/
1.224 brouard 11829: 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 11830:
11831:
11832: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11833: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11834: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11835: for(i=1,jk=1; i <=nlstate; i++){
11836: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11837: if (k != i) {
11838: printf("%d%d ",i,k);
11839: fprintf(ficlog,"%d%d ",i,k);
11840: fprintf(ficres,"%1d%1d ",i,k);
11841: for(j=1; j <=ncovmodel; j++){
11842: printf("%12.7f ",p[jk]);
11843: fprintf(ficlog,"%12.7f ",p[jk]);
11844: fprintf(ficres,"%12.7f ",p[jk]);
11845: jk++;
11846: }
11847: printf("\n");
11848: fprintf(ficlog,"\n");
11849: fprintf(ficres,"\n");
11850: }
1.126 brouard 11851: }
11852: }
1.203 brouard 11853: if(mle != 0){
11854: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11855: ftolhess=ftol; /* Usually correct */
1.203 brouard 11856: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11857: 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");
11858: 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");
11859: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11860: for(k=1; k <=(nlstate+ndeath); k++){
11861: if (k != i) {
11862: printf("%d%d ",i,k);
11863: fprintf(ficlog,"%d%d ",i,k);
11864: for(j=1; j <=ncovmodel; j++){
11865: 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]));
11866: 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]));
11867: jk++;
11868: }
11869: printf("\n");
11870: fprintf(ficlog,"\n");
11871: }
11872: }
1.193 brouard 11873: }
1.203 brouard 11874: } /* end of hesscov and Wald tests */
1.225 brouard 11875:
1.203 brouard 11876: /* */
1.126 brouard 11877: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11878: printf("# Scales (for hessian or gradient estimation)\n");
11879: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11880: for(i=1,jk=1; i <=nlstate; i++){
11881: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11882: if (j!=i) {
11883: fprintf(ficres,"%1d%1d",i,j);
11884: printf("%1d%1d",i,j);
11885: fprintf(ficlog,"%1d%1d",i,j);
11886: for(k=1; k<=ncovmodel;k++){
11887: printf(" %.5e",delti[jk]);
11888: fprintf(ficlog," %.5e",delti[jk]);
11889: fprintf(ficres," %.5e",delti[jk]);
11890: jk++;
11891: }
11892: printf("\n");
11893: fprintf(ficlog,"\n");
11894: fprintf(ficres,"\n");
11895: }
1.126 brouard 11896: }
11897: }
11898:
11899: 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 11900: if(mle >= 1) /* To big for the screen */
1.126 brouard 11901: 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");
11902: 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");
11903: /* # 121 Var(a12)\n\ */
11904: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11905: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11906: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11907: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11908: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11909: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11910: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11911:
11912:
11913: /* Just to have a covariance matrix which will be more understandable
11914: even is we still don't want to manage dictionary of variables
11915: */
11916: for(itimes=1;itimes<=2;itimes++){
11917: jj=0;
11918: for(i=1; i <=nlstate; i++){
1.225 brouard 11919: for(j=1; j <=nlstate+ndeath; j++){
11920: if(j==i) continue;
11921: for(k=1; k<=ncovmodel;k++){
11922: jj++;
11923: ca[0]= k+'a'-1;ca[1]='\0';
11924: if(itimes==1){
11925: if(mle>=1)
11926: printf("#%1d%1d%d",i,j,k);
11927: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11928: fprintf(ficres,"#%1d%1d%d",i,j,k);
11929: }else{
11930: if(mle>=1)
11931: printf("%1d%1d%d",i,j,k);
11932: fprintf(ficlog,"%1d%1d%d",i,j,k);
11933: fprintf(ficres,"%1d%1d%d",i,j,k);
11934: }
11935: ll=0;
11936: for(li=1;li <=nlstate; li++){
11937: for(lj=1;lj <=nlstate+ndeath; lj++){
11938: if(lj==li) continue;
11939: for(lk=1;lk<=ncovmodel;lk++){
11940: ll++;
11941: if(ll<=jj){
11942: cb[0]= lk +'a'-1;cb[1]='\0';
11943: if(ll<jj){
11944: if(itimes==1){
11945: if(mle>=1)
11946: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11947: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11948: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11949: }else{
11950: if(mle>=1)
11951: printf(" %.5e",matcov[jj][ll]);
11952: fprintf(ficlog," %.5e",matcov[jj][ll]);
11953: fprintf(ficres," %.5e",matcov[jj][ll]);
11954: }
11955: }else{
11956: if(itimes==1){
11957: if(mle>=1)
11958: printf(" Var(%s%1d%1d)",ca,i,j);
11959: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11960: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11961: }else{
11962: if(mle>=1)
11963: printf(" %.7e",matcov[jj][ll]);
11964: fprintf(ficlog," %.7e",matcov[jj][ll]);
11965: fprintf(ficres," %.7e",matcov[jj][ll]);
11966: }
11967: }
11968: }
11969: } /* end lk */
11970: } /* end lj */
11971: } /* end li */
11972: if(mle>=1)
11973: printf("\n");
11974: fprintf(ficlog,"\n");
11975: fprintf(ficres,"\n");
11976: numlinepar++;
11977: } /* end k*/
11978: } /*end j */
1.126 brouard 11979: } /* end i */
11980: } /* end itimes */
11981:
11982: fflush(ficlog);
11983: fflush(ficres);
1.225 brouard 11984: while(fgets(line, MAXLINE, ficpar)) {
11985: /* If line starts with a # it is a comment */
11986: if (line[0] == '#') {
11987: numlinepar++;
11988: fputs(line,stdout);
11989: fputs(line,ficparo);
11990: fputs(line,ficlog);
11991: continue;
11992: }else
11993: break;
11994: }
11995:
1.209 brouard 11996: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11997: /* ungetc(c,ficpar); */
11998: /* fgets(line, MAXLINE, ficpar); */
11999: /* fputs(line,stdout); */
12000: /* fputs(line,ficparo); */
12001: /* } */
12002: /* ungetc(c,ficpar); */
1.126 brouard 12003:
12004: estepm=0;
1.209 brouard 12005: 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 12006:
12007: if (num_filled != 6) {
12008: 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);
12009: 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);
12010: goto end;
12011: }
12012: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12013: }
12014: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12015: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12016:
1.209 brouard 12017: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12018: if (estepm==0 || estepm < stepm) estepm=stepm;
12019: if (fage <= 2) {
12020: bage = ageminpar;
12021: fage = agemaxpar;
12022: }
12023:
12024: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12025: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12026: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12027:
1.186 brouard 12028: /* Other stuffs, more or less useful */
1.254 brouard 12029: while(fgets(line, MAXLINE, ficpar)) {
12030: /* If line starts with a # it is a comment */
12031: if (line[0] == '#') {
12032: numlinepar++;
12033: fputs(line,stdout);
12034: fputs(line,ficparo);
12035: fputs(line,ficlog);
12036: continue;
12037: }else
12038: break;
12039: }
12040:
12041: 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){
12042:
12043: if (num_filled != 7) {
12044: 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);
12045: 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);
12046: goto end;
12047: }
12048: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12049: 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);
12050: 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);
12051: 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 12052: }
1.254 brouard 12053:
12054: while(fgets(line, MAXLINE, ficpar)) {
12055: /* If line starts with a # it is a comment */
12056: if (line[0] == '#') {
12057: numlinepar++;
12058: fputs(line,stdout);
12059: fputs(line,ficparo);
12060: fputs(line,ficlog);
12061: continue;
12062: }else
12063: break;
1.126 brouard 12064: }
12065:
12066:
12067: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12068: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12069:
1.254 brouard 12070: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12071: if (num_filled != 1) {
12072: 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);
12073: 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);
12074: goto end;
12075: }
12076: printf("pop_based=%d\n",popbased);
12077: fprintf(ficlog,"pop_based=%d\n",popbased);
12078: fprintf(ficparo,"pop_based=%d\n",popbased);
12079: fprintf(ficres,"pop_based=%d\n",popbased);
12080: }
12081:
1.258 brouard 12082: /* Results */
12083: nresult=0;
12084: do{
12085: if(!fgets(line, MAXLINE, ficpar)){
12086: endishere=1;
12087: parameterline=14;
12088: }else if (line[0] == '#') {
12089: /* If line starts with a # it is a comment */
1.254 brouard 12090: numlinepar++;
12091: fputs(line,stdout);
12092: fputs(line,ficparo);
12093: fputs(line,ficlog);
12094: continue;
1.258 brouard 12095: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12096: parameterline=11;
12097: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12098: parameterline=12;
12099: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12100: parameterline=13;
12101: else{
12102: parameterline=14;
1.254 brouard 12103: }
1.258 brouard 12104: switch (parameterline){
12105: case 11:
12106: 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){
12107: if (num_filled != 8) {
12108: 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);
12109: 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);
12110: goto end;
12111: }
12112: 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);
12113: 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);
12114: 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);
12115: 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);
12116: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12117: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12118: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12119:
1.258 brouard 12120: }
1.254 brouard 12121: break;
1.258 brouard 12122: case 12:
12123: /*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);*/
12124: 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){
12125: if (num_filled != 8) {
1.262 brouard 12126: 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);
12127: 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 12128: goto end;
12129: }
12130: 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);
12131: 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);
12132: 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);
12133: 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);
12134: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12135: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12136: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12137: }
1.230 brouard 12138: break;
1.258 brouard 12139: case 13:
12140: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12141: if (num_filled == 0){
12142: resultline[0]='\0';
12143: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12144: 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);
12145: break;
12146: } else if (num_filled != 1){
12147: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12148: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12149: }
12150: nresult++; /* Sum of resultlines */
12151: printf("Result %d: result=%s\n",nresult, resultline);
12152: if(nresult > MAXRESULTLINES){
12153: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12154: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12155: goto end;
12156: }
12157: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12158: fprintf(ficparo,"result: %s\n",resultline);
12159: fprintf(ficres,"result: %s\n",resultline);
12160: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12161: break;
1.258 brouard 12162: case 14:
1.259 brouard 12163: if(ncovmodel >2 && nresult==0 ){
12164: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12165: goto end;
12166: }
1.259 brouard 12167: break;
1.258 brouard 12168: default:
12169: nresult=1;
12170: decoderesult(".",nresult ); /* No covariate */
12171: }
12172: } /* End switch parameterline */
12173: }while(endishere==0); /* End do */
1.126 brouard 12174:
1.230 brouard 12175: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12176: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12177:
12178: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12179: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12180: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12181: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12182: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12183: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12184: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12185: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12186: }else{
1.270 brouard 12187: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12188: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12189: }
12190: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12191: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12192: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12193:
1.225 brouard 12194: /*------------ free_vector -------------*/
12195: /* chdir(path); */
1.220 brouard 12196:
1.215 brouard 12197: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12198: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12199: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12200: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12201: free_lvector(num,1,n);
12202: free_vector(agedc,1,n);
12203: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12204: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12205: fclose(ficparo);
12206: fclose(ficres);
1.220 brouard 12207:
12208:
1.186 brouard 12209: /* Other results (useful)*/
1.220 brouard 12210:
12211:
1.126 brouard 12212: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12213: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12214: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12215: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12216: fclose(ficrespl);
12217:
12218: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12219: /*#include "hpijx.h"*/
12220: hPijx(p, bage, fage);
1.145 brouard 12221: fclose(ficrespij);
1.227 brouard 12222:
1.220 brouard 12223: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12224: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12225: k=1;
1.126 brouard 12226: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12227:
1.269 brouard 12228: /* Prevalence for each covariate combination in probs[age][status][cov] */
12229: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12230: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12231: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12232: for(k=1;k<=ncovcombmax;k++)
12233: probs[i][j][k]=0.;
1.269 brouard 12234: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12235: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12236: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12237: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12238: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12239: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12240: for(k=1;k<=ncovcombmax;k++)
12241: mobaverages[i][j][k]=0.;
1.219 brouard 12242: mobaverage=mobaverages;
12243: if (mobilav!=0) {
1.235 brouard 12244: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12245: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12246: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12247: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12248: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12249: }
1.269 brouard 12250: } else if (mobilavproj !=0) {
1.235 brouard 12251: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12252: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12253: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12254: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12255: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12256: }
1.269 brouard 12257: }else{
12258: printf("Internal error moving average\n");
12259: fflush(stdout);
12260: exit(1);
1.219 brouard 12261: }
12262: }/* end if moving average */
1.227 brouard 12263:
1.126 brouard 12264: /*---------- Forecasting ------------------*/
12265: if(prevfcast==1){
12266: /* if(stepm ==1){*/
1.269 brouard 12267: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12268: }
1.269 brouard 12269:
12270: /* Backcasting */
1.217 brouard 12271: if(backcast==1){
1.219 brouard 12272: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12273: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12274: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12275:
12276: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12277:
12278: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12279:
1.219 brouard 12280: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12281: fclose(ficresplb);
12282:
1.222 brouard 12283: hBijx(p, bage, fage, mobaverage);
12284: fclose(ficrespijb);
1.219 brouard 12285:
1.269 brouard 12286: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12287: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12288: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12289:
12290:
1.269 brouard 12291: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12292: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12293: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12294: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12295: } /* end Backcasting */
1.268 brouard 12296:
1.186 brouard 12297:
12298: /* ------ Other prevalence ratios------------ */
1.126 brouard 12299:
1.215 brouard 12300: free_ivector(wav,1,imx);
12301: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12302: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12303: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12304:
12305:
1.127 brouard 12306: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12307:
1.201 brouard 12308: strcpy(filerese,"E_");
12309: strcat(filerese,fileresu);
1.126 brouard 12310: if((ficreseij=fopen(filerese,"w"))==NULL) {
12311: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12312: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12313: }
1.208 brouard 12314: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12315: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12316:
12317: pstamp(ficreseij);
1.219 brouard 12318:
1.235 brouard 12319: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12320: if (cptcovn < 1){i1=1;}
12321:
12322: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12323: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12324: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12325: continue;
1.219 brouard 12326: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12327: printf("\n#****** ");
1.225 brouard 12328: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12329: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12330: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12331: }
12332: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12333: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12334: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12335: }
12336: fprintf(ficreseij,"******\n");
1.235 brouard 12337: printf("******\n");
1.219 brouard 12338:
12339: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12340: oldm=oldms;savm=savms;
1.235 brouard 12341: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12342:
1.219 brouard 12343: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12344: }
12345: fclose(ficreseij);
1.208 brouard 12346: printf("done evsij\n");fflush(stdout);
12347: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12348:
1.218 brouard 12349:
1.227 brouard 12350: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12351:
1.201 brouard 12352: strcpy(filerest,"T_");
12353: strcat(filerest,fileresu);
1.127 brouard 12354: if((ficrest=fopen(filerest,"w"))==NULL) {
12355: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12356: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12357: }
1.208 brouard 12358: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12359: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12360: strcpy(fileresstde,"STDE_");
12361: strcat(fileresstde,fileresu);
1.126 brouard 12362: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12363: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12364: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12365: }
1.227 brouard 12366: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12367: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12368:
1.201 brouard 12369: strcpy(filerescve,"CVE_");
12370: strcat(filerescve,fileresu);
1.126 brouard 12371: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12372: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12373: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12374: }
1.227 brouard 12375: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12376: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12377:
1.201 brouard 12378: strcpy(fileresv,"V_");
12379: strcat(fileresv,fileresu);
1.126 brouard 12380: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12381: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12382: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12383: }
1.227 brouard 12384: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12385: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12386:
1.235 brouard 12387: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12388: if (cptcovn < 1){i1=1;}
12389:
12390: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12391: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12392: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12393: continue;
1.242 brouard 12394: printf("\n#****** Result for:");
12395: fprintf(ficrest,"\n#****** Result for:");
12396: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12397: for(j=1;j<=cptcoveff;j++){
12398: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12399: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12400: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12401: }
1.235 brouard 12402: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12403: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12404: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12405: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12406: }
1.208 brouard 12407: fprintf(ficrest,"******\n");
1.227 brouard 12408: fprintf(ficlog,"******\n");
12409: printf("******\n");
1.208 brouard 12410:
12411: fprintf(ficresstdeij,"\n#****** ");
12412: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12413: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12414: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12415: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12416: }
1.235 brouard 12417: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12418: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12419: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12420: }
1.208 brouard 12421: fprintf(ficresstdeij,"******\n");
12422: fprintf(ficrescveij,"******\n");
12423:
12424: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12425: /* pstamp(ficresvij); */
1.225 brouard 12426: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12427: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12428: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12429: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12430: }
1.208 brouard 12431: fprintf(ficresvij,"******\n");
12432:
12433: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12434: oldm=oldms;savm=savms;
1.235 brouard 12435: printf(" cvevsij ");
12436: fprintf(ficlog, " cvevsij ");
12437: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12438: printf(" end cvevsij \n ");
12439: fprintf(ficlog, " end cvevsij \n ");
12440:
12441: /*
12442: */
12443: /* goto endfree; */
12444:
12445: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12446: pstamp(ficrest);
12447:
1.269 brouard 12448: epj=vector(1,nlstate+1);
1.208 brouard 12449: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12450: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12451: cptcod= 0; /* To be deleted */
12452: printf("varevsij vpopbased=%d \n",vpopbased);
12453: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12454: 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 12455: 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 ");
12456: if(vpopbased==1)
12457: 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);
12458: else
12459: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12460: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12461: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12462: fprintf(ficrest,"\n");
12463: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12464: printf("Computing age specific period (stable) prevalences in each health state \n");
12465: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12466: for(age=bage; age <=fage ;age++){
1.235 brouard 12467: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12468: if (vpopbased==1) {
12469: if(mobilav ==0){
12470: for(i=1; i<=nlstate;i++)
12471: prlim[i][i]=probs[(int)age][i][k];
12472: }else{ /* mobilav */
12473: for(i=1; i<=nlstate;i++)
12474: prlim[i][i]=mobaverage[(int)age][i][k];
12475: }
12476: }
1.219 brouard 12477:
1.227 brouard 12478: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12479: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12480: /* printf(" age %4.0f ",age); */
12481: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12482: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12483: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12484: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12485: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12486: }
12487: epj[nlstate+1] +=epj[j];
12488: }
12489: /* printf(" age %4.0f \n",age); */
1.219 brouard 12490:
1.227 brouard 12491: for(i=1, vepp=0.;i <=nlstate;i++)
12492: for(j=1;j <=nlstate;j++)
12493: vepp += vareij[i][j][(int)age];
12494: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12495: for(j=1;j <=nlstate;j++){
12496: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12497: }
12498: fprintf(ficrest,"\n");
12499: }
1.208 brouard 12500: } /* End vpopbased */
1.269 brouard 12501: free_vector(epj,1,nlstate+1);
1.208 brouard 12502: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12503: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12504: printf("done selection\n");fflush(stdout);
12505: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12506:
1.235 brouard 12507: } /* End k selection */
1.227 brouard 12508:
12509: printf("done State-specific expectancies\n");fflush(stdout);
12510: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12511:
1.269 brouard 12512: /* variance-covariance of period prevalence*/
12513: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12514:
1.227 brouard 12515:
12516: free_vector(weight,1,n);
12517: free_imatrix(Tvard,1,NCOVMAX,1,2);
12518: free_imatrix(s,1,maxwav+1,1,n);
12519: free_matrix(anint,1,maxwav,1,n);
12520: free_matrix(mint,1,maxwav,1,n);
12521: free_ivector(cod,1,n);
12522: free_ivector(tab,1,NCOVMAX);
12523: fclose(ficresstdeij);
12524: fclose(ficrescveij);
12525: fclose(ficresvij);
12526: fclose(ficrest);
12527: fclose(ficpar);
12528:
12529:
1.126 brouard 12530: /*---------- End : free ----------------*/
1.219 brouard 12531: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12532: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12533: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12534: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12535: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12536: } /* mle==-3 arrives here for freeing */
1.227 brouard 12537: /* endfree:*/
12538: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12539: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12540: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12541: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12542: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12543: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12544: free_matrix(covar,0,NCOVMAX,1,n);
12545: free_matrix(matcov,1,npar,1,npar);
12546: free_matrix(hess,1,npar,1,npar);
12547: /*free_vector(delti,1,npar);*/
12548: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12549: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12550: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12551: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12552:
12553: free_ivector(ncodemax,1,NCOVMAX);
12554: free_ivector(ncodemaxwundef,1,NCOVMAX);
12555: free_ivector(Dummy,-1,NCOVMAX);
12556: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12557: free_ivector(DummyV,1,NCOVMAX);
12558: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12559: free_ivector(Typevar,-1,NCOVMAX);
12560: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12561: free_ivector(TvarsQ,1,NCOVMAX);
12562: free_ivector(TvarsQind,1,NCOVMAX);
12563: free_ivector(TvarsD,1,NCOVMAX);
12564: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12565: free_ivector(TvarFD,1,NCOVMAX);
12566: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12567: free_ivector(TvarF,1,NCOVMAX);
12568: free_ivector(TvarFind,1,NCOVMAX);
12569: free_ivector(TvarV,1,NCOVMAX);
12570: free_ivector(TvarVind,1,NCOVMAX);
12571: free_ivector(TvarA,1,NCOVMAX);
12572: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12573: free_ivector(TvarFQ,1,NCOVMAX);
12574: free_ivector(TvarFQind,1,NCOVMAX);
12575: free_ivector(TvarVD,1,NCOVMAX);
12576: free_ivector(TvarVDind,1,NCOVMAX);
12577: free_ivector(TvarVQ,1,NCOVMAX);
12578: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12579: free_ivector(Tvarsel,1,NCOVMAX);
12580: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12581: free_ivector(Tposprod,1,NCOVMAX);
12582: free_ivector(Tprod,1,NCOVMAX);
12583: free_ivector(Tvaraff,1,NCOVMAX);
12584: free_ivector(invalidvarcomb,1,ncovcombmax);
12585: free_ivector(Tage,1,NCOVMAX);
12586: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12587: free_ivector(TmodelInvind,1,NCOVMAX);
12588: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12589:
12590: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12591: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12592: fflush(fichtm);
12593: fflush(ficgp);
12594:
1.227 brouard 12595:
1.126 brouard 12596: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12597: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12598: 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 12599: }else{
12600: printf("End of Imach\n");
12601: fprintf(ficlog,"End of Imach\n");
12602: }
12603: printf("See log file on %s\n",filelog);
12604: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12605: /*(void) gettimeofday(&end_time,&tzp);*/
12606: rend_time = time(NULL);
12607: end_time = *localtime(&rend_time);
12608: /* tml = *localtime(&end_time.tm_sec); */
12609: strcpy(strtend,asctime(&end_time));
1.126 brouard 12610: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12611: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12612: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12613:
1.157 brouard 12614: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12615: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12616: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12617: /* printf("Total time was %d uSec.\n", total_usecs);*/
12618: /* if(fileappend(fichtm,optionfilehtm)){ */
12619: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12620: fclose(fichtm);
12621: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12622: fclose(fichtmcov);
12623: fclose(ficgp);
12624: fclose(ficlog);
12625: /*------ End -----------*/
1.227 brouard 12626:
12627:
12628: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12629: #ifdef WIN32
1.227 brouard 12630: if (_chdir(pathcd) != 0)
12631: printf("Can't move to directory %s!\n",path);
12632: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12633: #else
1.227 brouard 12634: if(chdir(pathcd) != 0)
12635: printf("Can't move to directory %s!\n", path);
12636: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12637: #endif
1.126 brouard 12638: printf("Current directory %s!\n",pathcd);
12639: /*strcat(plotcmd,CHARSEPARATOR);*/
12640: sprintf(plotcmd,"gnuplot");
1.157 brouard 12641: #ifdef _WIN32
1.126 brouard 12642: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12643: #endif
12644: if(!stat(plotcmd,&info)){
1.158 brouard 12645: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12646: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12647: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12648: }else
12649: strcpy(pplotcmd,plotcmd);
1.157 brouard 12650: #ifdef __unix
1.126 brouard 12651: strcpy(plotcmd,GNUPLOTPROGRAM);
12652: if(!stat(plotcmd,&info)){
1.158 brouard 12653: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12654: }else
12655: strcpy(pplotcmd,plotcmd);
12656: #endif
12657: }else
12658: strcpy(pplotcmd,plotcmd);
12659:
12660: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12661: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12662:
1.126 brouard 12663: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12664: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12665: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12666: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12667: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12668: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12669: }
1.158 brouard 12670: printf(" Successful, please wait...");
1.126 brouard 12671: while (z[0] != 'q') {
12672: /* chdir(path); */
1.154 brouard 12673: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12674: scanf("%s",z);
12675: /* if (z[0] == 'c') system("./imach"); */
12676: if (z[0] == 'e') {
1.158 brouard 12677: #ifdef __APPLE__
1.152 brouard 12678: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12679: #elif __linux
12680: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12681: #else
1.152 brouard 12682: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12683: #endif
12684: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12685: system(pplotcmd);
1.126 brouard 12686: }
12687: else if (z[0] == 'g') system(plotcmd);
12688: else if (z[0] == 'q') exit(0);
12689: }
1.227 brouard 12690: end:
1.126 brouard 12691: while (z[0] != 'q') {
1.195 brouard 12692: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12693: scanf("%s",z);
12694: }
12695: }
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