Annotation of imach/src/imach.c, revision 1.276
1.276 ! brouard 1: /* $Id: imach.c,v 1.275 2017/06/30 13:39:33 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.276 ! brouard 4: Revision 1.275 2017/06/30 13:39:33 brouard
! 5: Summary: Saito's color
! 6:
1.275 brouard 7: Revision 1.274 2017/06/29 09:47:08 brouard
8: Summary: Version 0.99r14
9:
1.274 brouard 10: Revision 1.273 2017/06/27 11:06:02 brouard
11: Summary: More documentation on projections
12:
1.273 brouard 13: Revision 1.272 2017/06/27 10:22:40 brouard
14: Summary: Color of backprojection changed from 6 to 5(yellow)
15:
1.272 brouard 16: Revision 1.271 2017/06/27 10:17:50 brouard
17: Summary: Some bug with rint
18:
1.271 brouard 19: Revision 1.270 2017/05/24 05:45:29 brouard
20: *** empty log message ***
21:
1.270 brouard 22: Revision 1.269 2017/05/23 08:39:25 brouard
23: Summary: Code into subroutine, cleanings
24:
1.269 brouard 25: Revision 1.268 2017/05/18 20:09:32 brouard
26: Summary: backprojection and confidence intervals of backprevalence
27:
1.268 brouard 28: Revision 1.267 2017/05/13 10:25:05 brouard
29: Summary: temporary save for backprojection
30:
1.267 brouard 31: Revision 1.266 2017/05/13 07:26:12 brouard
32: Summary: Version 0.99r13 (improvements and bugs fixed)
33:
1.266 brouard 34: Revision 1.265 2017/04/26 16:22:11 brouard
35: Summary: imach 0.99r13 Some bugs fixed
36:
1.265 brouard 37: Revision 1.264 2017/04/26 06:01:29 brouard
38: Summary: Labels in graphs
39:
1.264 brouard 40: Revision 1.263 2017/04/24 15:23:15 brouard
41: Summary: to save
42:
1.263 brouard 43: Revision 1.262 2017/04/18 16:48:12 brouard
44: *** empty log message ***
45:
1.262 brouard 46: Revision 1.261 2017/04/05 10:14:09 brouard
47: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
48:
1.261 brouard 49: Revision 1.260 2017/04/04 17:46:59 brouard
50: Summary: Gnuplot indexations fixed (humm)
51:
1.260 brouard 52: Revision 1.259 2017/04/04 13:01:16 brouard
53: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
54:
1.259 brouard 55: Revision 1.258 2017/04/03 10:17:47 brouard
56: Summary: Version 0.99r12
57:
58: Some cleanings, conformed with updated documentation.
59:
1.258 brouard 60: Revision 1.257 2017/03/29 16:53:30 brouard
61: Summary: Temp
62:
1.257 brouard 63: Revision 1.256 2017/03/27 05:50:23 brouard
64: Summary: Temporary
65:
1.256 brouard 66: Revision 1.255 2017/03/08 16:02:28 brouard
67: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
68:
1.255 brouard 69: Revision 1.254 2017/03/08 07:13:00 brouard
70: Summary: Fixing data parameter line
71:
1.254 brouard 72: Revision 1.253 2016/12/15 11:59:41 brouard
73: Summary: 0.99 in progress
74:
1.253 brouard 75: Revision 1.252 2016/09/15 21:15:37 brouard
76: *** empty log message ***
77:
1.252 brouard 78: Revision 1.251 2016/09/15 15:01:13 brouard
79: Summary: not working
80:
1.251 brouard 81: Revision 1.250 2016/09/08 16:07:27 brouard
82: Summary: continue
83:
1.250 brouard 84: Revision 1.249 2016/09/07 17:14:18 brouard
85: Summary: Starting values from frequencies
86:
1.249 brouard 87: Revision 1.248 2016/09/07 14:10:18 brouard
88: *** empty log message ***
89:
1.248 brouard 90: Revision 1.247 2016/09/02 11:11:21 brouard
91: *** empty log message ***
92:
1.247 brouard 93: Revision 1.246 2016/09/02 08:49:22 brouard
94: *** empty log message ***
95:
1.246 brouard 96: Revision 1.245 2016/09/02 07:25:01 brouard
97: *** empty log message ***
98:
1.245 brouard 99: Revision 1.244 2016/09/02 07:17:34 brouard
100: *** empty log message ***
101:
1.244 brouard 102: Revision 1.243 2016/09/02 06:45:35 brouard
103: *** empty log message ***
104:
1.243 brouard 105: Revision 1.242 2016/08/30 15:01:20 brouard
106: Summary: Fixing a lots
107:
1.242 brouard 108: Revision 1.241 2016/08/29 17:17:25 brouard
109: Summary: gnuplot problem in Back projection to fix
110:
1.241 brouard 111: Revision 1.240 2016/08/29 07:53:18 brouard
112: Summary: Better
113:
1.240 brouard 114: Revision 1.239 2016/08/26 15:51:03 brouard
115: Summary: Improvement in Powell output in order to copy and paste
116:
117: Author:
118:
1.239 brouard 119: Revision 1.238 2016/08/26 14:23:35 brouard
120: Summary: Starting tests of 0.99
121:
1.238 brouard 122: Revision 1.237 2016/08/26 09:20:19 brouard
123: Summary: to valgrind
124:
1.237 brouard 125: Revision 1.236 2016/08/25 10:50:18 brouard
126: *** empty log message ***
127:
1.236 brouard 128: Revision 1.235 2016/08/25 06:59:23 brouard
129: *** empty log message ***
130:
1.235 brouard 131: Revision 1.234 2016/08/23 16:51:20 brouard
132: *** empty log message ***
133:
1.234 brouard 134: Revision 1.233 2016/08/23 07:40:50 brouard
135: Summary: not working
136:
1.233 brouard 137: Revision 1.232 2016/08/22 14:20:21 brouard
138: Summary: not working
139:
1.232 brouard 140: Revision 1.231 2016/08/22 07:17:15 brouard
141: Summary: not working
142:
1.231 brouard 143: Revision 1.230 2016/08/22 06:55:53 brouard
144: Summary: Not working
145:
1.230 brouard 146: Revision 1.229 2016/07/23 09:45:53 brouard
147: Summary: Completing for func too
148:
1.229 brouard 149: Revision 1.228 2016/07/22 17:45:30 brouard
150: Summary: Fixing some arrays, still debugging
151:
1.227 brouard 152: Revision 1.226 2016/07/12 18:42:34 brouard
153: Summary: temp
154:
1.226 brouard 155: Revision 1.225 2016/07/12 08:40:03 brouard
156: Summary: saving but not running
157:
1.225 brouard 158: Revision 1.224 2016/07/01 13:16:01 brouard
159: Summary: Fixes
160:
1.224 brouard 161: Revision 1.223 2016/02/19 09:23:35 brouard
162: Summary: temporary
163:
1.223 brouard 164: Revision 1.222 2016/02/17 08:14:50 brouard
165: Summary: Probably last 0.98 stable version 0.98r6
166:
1.222 brouard 167: Revision 1.221 2016/02/15 23:35:36 brouard
168: Summary: minor bug
169:
1.220 brouard 170: Revision 1.219 2016/02/15 00:48:12 brouard
171: *** empty log message ***
172:
1.219 brouard 173: Revision 1.218 2016/02/12 11:29:23 brouard
174: Summary: 0.99 Back projections
175:
1.218 brouard 176: Revision 1.217 2015/12/23 17:18:31 brouard
177: Summary: Experimental backcast
178:
1.217 brouard 179: Revision 1.216 2015/12/18 17:32:11 brouard
180: Summary: 0.98r4 Warning and status=-2
181:
182: Version 0.98r4 is now:
183: - displaying an error when status is -1, date of interview unknown and date of death known;
184: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
185: Older changes concerning s=-2, dating from 2005 have been supersed.
186:
1.216 brouard 187: Revision 1.215 2015/12/16 08:52:24 brouard
188: Summary: 0.98r4 working
189:
1.215 brouard 190: Revision 1.214 2015/12/16 06:57:54 brouard
191: Summary: temporary not working
192:
1.214 brouard 193: Revision 1.213 2015/12/11 18:22:17 brouard
194: Summary: 0.98r4
195:
1.213 brouard 196: Revision 1.212 2015/11/21 12:47:24 brouard
197: Summary: minor typo
198:
1.212 brouard 199: Revision 1.211 2015/11/21 12:41:11 brouard
200: Summary: 0.98r3 with some graph of projected cross-sectional
201:
202: Author: Nicolas Brouard
203:
1.211 brouard 204: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 205: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 206: Summary: Adding ftolpl parameter
207: Author: N Brouard
208:
209: We had difficulties to get smoothed confidence intervals. It was due
210: to the period prevalence which wasn't computed accurately. The inner
211: parameter ftolpl is now an outer parameter of the .imach parameter
212: file after estepm. If ftolpl is small 1.e-4 and estepm too,
213: computation are long.
214:
1.209 brouard 215: Revision 1.208 2015/11/17 14:31:57 brouard
216: Summary: temporary
217:
1.208 brouard 218: Revision 1.207 2015/10/27 17:36:57 brouard
219: *** empty log message ***
220:
1.207 brouard 221: Revision 1.206 2015/10/24 07:14:11 brouard
222: *** empty log message ***
223:
1.206 brouard 224: Revision 1.205 2015/10/23 15:50:53 brouard
225: Summary: 0.98r3 some clarification for graphs on likelihood contributions
226:
1.205 brouard 227: Revision 1.204 2015/10/01 16:20:26 brouard
228: Summary: Some new graphs of contribution to likelihood
229:
1.204 brouard 230: Revision 1.203 2015/09/30 17:45:14 brouard
231: Summary: looking at better estimation of the hessian
232:
233: Also a better criteria for convergence to the period prevalence And
234: therefore adding the number of years needed to converge. (The
235: prevalence in any alive state shold sum to one
236:
1.203 brouard 237: Revision 1.202 2015/09/22 19:45:16 brouard
238: Summary: Adding some overall graph on contribution to likelihood. Might change
239:
1.202 brouard 240: Revision 1.201 2015/09/15 17:34:58 brouard
241: Summary: 0.98r0
242:
243: - Some new graphs like suvival functions
244: - Some bugs fixed like model=1+age+V2.
245:
1.201 brouard 246: Revision 1.200 2015/09/09 16:53:55 brouard
247: Summary: Big bug thanks to Flavia
248:
249: Even model=1+age+V2. did not work anymore
250:
1.200 brouard 251: Revision 1.199 2015/09/07 14:09:23 brouard
252: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
253:
1.199 brouard 254: Revision 1.198 2015/09/03 07:14:39 brouard
255: Summary: 0.98q5 Flavia
256:
1.198 brouard 257: Revision 1.197 2015/09/01 18:24:39 brouard
258: *** empty log message ***
259:
1.197 brouard 260: Revision 1.196 2015/08/18 23:17:52 brouard
261: Summary: 0.98q5
262:
1.196 brouard 263: Revision 1.195 2015/08/18 16:28:39 brouard
264: Summary: Adding a hack for testing purpose
265:
266: After reading the title, ftol and model lines, if the comment line has
267: a q, starting with #q, the answer at the end of the run is quit. It
268: permits to run test files in batch with ctest. The former workaround was
269: $ echo q | imach foo.imach
270:
1.195 brouard 271: Revision 1.194 2015/08/18 13:32:00 brouard
272: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
273:
1.194 brouard 274: Revision 1.193 2015/08/04 07:17:42 brouard
275: Summary: 0.98q4
276:
1.193 brouard 277: Revision 1.192 2015/07/16 16:49:02 brouard
278: Summary: Fixing some outputs
279:
1.192 brouard 280: Revision 1.191 2015/07/14 10:00:33 brouard
281: Summary: Some fixes
282:
1.191 brouard 283: Revision 1.190 2015/05/05 08:51:13 brouard
284: Summary: Adding digits in output parameters (7 digits instead of 6)
285:
286: Fix 1+age+.
287:
1.190 brouard 288: Revision 1.189 2015/04/30 14:45:16 brouard
289: Summary: 0.98q2
290:
1.189 brouard 291: Revision 1.188 2015/04/30 08:27:53 brouard
292: *** empty log message ***
293:
1.188 brouard 294: Revision 1.187 2015/04/29 09:11:15 brouard
295: *** empty log message ***
296:
1.187 brouard 297: Revision 1.186 2015/04/23 12:01:52 brouard
298: Summary: V1*age is working now, version 0.98q1
299:
300: Some codes had been disabled in order to simplify and Vn*age was
301: working in the optimization phase, ie, giving correct MLE parameters,
302: but, as usual, outputs were not correct and program core dumped.
303:
1.186 brouard 304: Revision 1.185 2015/03/11 13:26:42 brouard
305: Summary: Inclusion of compile and links command line for Intel Compiler
306:
1.185 brouard 307: Revision 1.184 2015/03/11 11:52:39 brouard
308: Summary: Back from Windows 8. Intel Compiler
309:
1.184 brouard 310: Revision 1.183 2015/03/10 20:34:32 brouard
311: Summary: 0.98q0, trying with directest, mnbrak fixed
312:
313: We use directest instead of original Powell test; probably no
314: incidence on the results, but better justifications;
315: We fixed Numerical Recipes mnbrak routine which was wrong and gave
316: wrong results.
317:
1.183 brouard 318: Revision 1.182 2015/02/12 08:19:57 brouard
319: Summary: Trying to keep directest which seems simpler and more general
320: Author: Nicolas Brouard
321:
1.182 brouard 322: Revision 1.181 2015/02/11 23:22:24 brouard
323: Summary: Comments on Powell added
324:
325: Author:
326:
1.181 brouard 327: Revision 1.180 2015/02/11 17:33:45 brouard
328: Summary: Finishing move from main to function (hpijx and prevalence_limit)
329:
1.180 brouard 330: Revision 1.179 2015/01/04 09:57:06 brouard
331: Summary: back to OS/X
332:
1.179 brouard 333: Revision 1.178 2015/01/04 09:35:48 brouard
334: *** empty log message ***
335:
1.178 brouard 336: Revision 1.177 2015/01/03 18:40:56 brouard
337: Summary: Still testing ilc32 on OSX
338:
1.177 brouard 339: Revision 1.176 2015/01/03 16:45:04 brouard
340: *** empty log message ***
341:
1.176 brouard 342: Revision 1.175 2015/01/03 16:33:42 brouard
343: *** empty log message ***
344:
1.175 brouard 345: Revision 1.174 2015/01/03 16:15:49 brouard
346: Summary: Still in cross-compilation
347:
1.174 brouard 348: Revision 1.173 2015/01/03 12:06:26 brouard
349: Summary: trying to detect cross-compilation
350:
1.173 brouard 351: Revision 1.172 2014/12/27 12:07:47 brouard
352: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
353:
1.172 brouard 354: Revision 1.171 2014/12/23 13:26:59 brouard
355: Summary: Back from Visual C
356:
357: Still problem with utsname.h on Windows
358:
1.171 brouard 359: Revision 1.170 2014/12/23 11:17:12 brouard
360: Summary: Cleaning some \%% back to %%
361:
362: The escape was mandatory for a specific compiler (which one?), but too many warnings.
363:
1.170 brouard 364: Revision 1.169 2014/12/22 23:08:31 brouard
365: Summary: 0.98p
366:
367: Outputs some informations on compiler used, OS etc. Testing on different platforms.
368:
1.169 brouard 369: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 370: Summary: update
1.169 brouard 371:
1.168 brouard 372: Revision 1.167 2014/12/22 13:50:56 brouard
373: Summary: Testing uname and compiler version and if compiled 32 or 64
374:
375: Testing on Linux 64
376:
1.167 brouard 377: Revision 1.166 2014/12/22 11:40:47 brouard
378: *** empty log message ***
379:
1.166 brouard 380: Revision 1.165 2014/12/16 11:20:36 brouard
381: Summary: After compiling on Visual C
382:
383: * imach.c (Module): Merging 1.61 to 1.162
384:
1.165 brouard 385: Revision 1.164 2014/12/16 10:52:11 brouard
386: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
387:
388: * imach.c (Module): Merging 1.61 to 1.162
389:
1.164 brouard 390: Revision 1.163 2014/12/16 10:30:11 brouard
391: * imach.c (Module): Merging 1.61 to 1.162
392:
1.163 brouard 393: Revision 1.162 2014/09/25 11:43:39 brouard
394: Summary: temporary backup 0.99!
395:
1.162 brouard 396: Revision 1.1 2014/09/16 11:06:58 brouard
397: Summary: With some code (wrong) for nlopt
398:
399: Author:
400:
401: Revision 1.161 2014/09/15 20:41:41 brouard
402: Summary: Problem with macro SQR on Intel compiler
403:
1.161 brouard 404: Revision 1.160 2014/09/02 09:24:05 brouard
405: *** empty log message ***
406:
1.160 brouard 407: Revision 1.159 2014/09/01 10:34:10 brouard
408: Summary: WIN32
409: Author: Brouard
410:
1.159 brouard 411: Revision 1.158 2014/08/27 17:11:51 brouard
412: *** empty log message ***
413:
1.158 brouard 414: Revision 1.157 2014/08/27 16:26:55 brouard
415: Summary: Preparing windows Visual studio version
416: Author: Brouard
417:
418: In order to compile on Visual studio, time.h is now correct and time_t
419: and tm struct should be used. difftime should be used but sometimes I
420: just make the differences in raw time format (time(&now).
421: Trying to suppress #ifdef LINUX
422: Add xdg-open for __linux in order to open default browser.
423:
1.157 brouard 424: Revision 1.156 2014/08/25 20:10:10 brouard
425: *** empty log message ***
426:
1.156 brouard 427: Revision 1.155 2014/08/25 18:32:34 brouard
428: Summary: New compile, minor changes
429: Author: Brouard
430:
1.155 brouard 431: Revision 1.154 2014/06/20 17:32:08 brouard
432: Summary: Outputs now all graphs of convergence to period prevalence
433:
1.154 brouard 434: Revision 1.153 2014/06/20 16:45:46 brouard
435: Summary: If 3 live state, convergence to period prevalence on same graph
436: Author: Brouard
437:
1.153 brouard 438: Revision 1.152 2014/06/18 17:54:09 brouard
439: Summary: open browser, use gnuplot on same dir than imach if not found in the path
440:
1.152 brouard 441: Revision 1.151 2014/06/18 16:43:30 brouard
442: *** empty log message ***
443:
1.151 brouard 444: Revision 1.150 2014/06/18 16:42:35 brouard
445: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
446: Author: brouard
447:
1.150 brouard 448: Revision 1.149 2014/06/18 15:51:14 brouard
449: Summary: Some fixes in parameter files errors
450: Author: Nicolas Brouard
451:
1.149 brouard 452: Revision 1.148 2014/06/17 17:38:48 brouard
453: Summary: Nothing new
454: Author: Brouard
455:
456: Just a new packaging for OS/X version 0.98nS
457:
1.148 brouard 458: Revision 1.147 2014/06/16 10:33:11 brouard
459: *** empty log message ***
460:
1.147 brouard 461: Revision 1.146 2014/06/16 10:20:28 brouard
462: Summary: Merge
463: Author: Brouard
464:
465: Merge, before building revised version.
466:
1.146 brouard 467: Revision 1.145 2014/06/10 21:23:15 brouard
468: Summary: Debugging with valgrind
469: Author: Nicolas Brouard
470:
471: Lot of changes in order to output the results with some covariates
472: After the Edimburgh REVES conference 2014, it seems mandatory to
473: improve the code.
474: No more memory valgrind error but a lot has to be done in order to
475: continue the work of splitting the code into subroutines.
476: Also, decodemodel has been improved. Tricode is still not
477: optimal. nbcode should be improved. Documentation has been added in
478: the source code.
479:
1.144 brouard 480: Revision 1.143 2014/01/26 09:45:38 brouard
481: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
482:
483: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
484: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
485:
1.143 brouard 486: Revision 1.142 2014/01/26 03:57:36 brouard
487: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
488:
489: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
490:
1.142 brouard 491: Revision 1.141 2014/01/26 02:42:01 brouard
492: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
493:
1.141 brouard 494: Revision 1.140 2011/09/02 10:37:54 brouard
495: Summary: times.h is ok with mingw32 now.
496:
1.140 brouard 497: Revision 1.139 2010/06/14 07:50:17 brouard
498: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
499: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
500:
1.139 brouard 501: Revision 1.138 2010/04/30 18:19:40 brouard
502: *** empty log message ***
503:
1.138 brouard 504: Revision 1.137 2010/04/29 18:11:38 brouard
505: (Module): Checking covariates for more complex models
506: than V1+V2. A lot of change to be done. Unstable.
507:
1.137 brouard 508: Revision 1.136 2010/04/26 20:30:53 brouard
509: (Module): merging some libgsl code. Fixing computation
510: of likelione (using inter/intrapolation if mle = 0) in order to
511: get same likelihood as if mle=1.
512: Some cleaning of code and comments added.
513:
1.136 brouard 514: Revision 1.135 2009/10/29 15:33:14 brouard
515: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
516:
1.135 brouard 517: Revision 1.134 2009/10/29 13:18:53 brouard
518: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
519:
1.134 brouard 520: Revision 1.133 2009/07/06 10:21:25 brouard
521: just nforces
522:
1.133 brouard 523: Revision 1.132 2009/07/06 08:22:05 brouard
524: Many tings
525:
1.132 brouard 526: Revision 1.131 2009/06/20 16:22:47 brouard
527: Some dimensions resccaled
528:
1.131 brouard 529: Revision 1.130 2009/05/26 06:44:34 brouard
530: (Module): Max Covariate is now set to 20 instead of 8. A
531: lot of cleaning with variables initialized to 0. Trying to make
532: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
533:
1.130 brouard 534: Revision 1.129 2007/08/31 13:49:27 lievre
535: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
536:
1.129 lievre 537: Revision 1.128 2006/06/30 13:02:05 brouard
538: (Module): Clarifications on computing e.j
539:
1.128 brouard 540: Revision 1.127 2006/04/28 18:11:50 brouard
541: (Module): Yes the sum of survivors was wrong since
542: imach-114 because nhstepm was no more computed in the age
543: loop. Now we define nhstepma in the age loop.
544: (Module): In order to speed up (in case of numerous covariates) we
545: compute health expectancies (without variances) in a first step
546: and then all the health expectancies with variances or standard
547: deviation (needs data from the Hessian matrices) which slows the
548: computation.
549: In the future we should be able to stop the program is only health
550: expectancies and graph are needed without standard deviations.
551:
1.127 brouard 552: Revision 1.126 2006/04/28 17:23:28 brouard
553: (Module): Yes the sum of survivors was wrong since
554: imach-114 because nhstepm was no more computed in the age
555: loop. Now we define nhstepma in the age loop.
556: Version 0.98h
557:
1.126 brouard 558: Revision 1.125 2006/04/04 15:20:31 lievre
559: Errors in calculation of health expectancies. Age was not initialized.
560: Forecasting file added.
561:
562: Revision 1.124 2006/03/22 17:13:53 lievre
563: Parameters are printed with %lf instead of %f (more numbers after the comma).
564: The log-likelihood is printed in the log file
565:
566: Revision 1.123 2006/03/20 10:52:43 brouard
567: * imach.c (Module): <title> changed, corresponds to .htm file
568: name. <head> headers where missing.
569:
570: * imach.c (Module): Weights can have a decimal point as for
571: English (a comma might work with a correct LC_NUMERIC environment,
572: otherwise the weight is truncated).
573: Modification of warning when the covariates values are not 0 or
574: 1.
575: Version 0.98g
576:
577: Revision 1.122 2006/03/20 09:45:41 brouard
578: (Module): Weights can have a decimal point as for
579: English (a comma might work with a correct LC_NUMERIC environment,
580: otherwise the weight is truncated).
581: Modification of warning when the covariates values are not 0 or
582: 1.
583: Version 0.98g
584:
585: Revision 1.121 2006/03/16 17:45:01 lievre
586: * imach.c (Module): Comments concerning covariates added
587:
588: * imach.c (Module): refinements in the computation of lli if
589: status=-2 in order to have more reliable computation if stepm is
590: not 1 month. Version 0.98f
591:
592: Revision 1.120 2006/03/16 15:10:38 lievre
593: (Module): refinements in the computation of lli if
594: status=-2 in order to have more reliable computation if stepm is
595: not 1 month. Version 0.98f
596:
597: Revision 1.119 2006/03/15 17:42:26 brouard
598: (Module): Bug if status = -2, the loglikelihood was
599: computed as likelihood omitting the logarithm. Version O.98e
600:
601: Revision 1.118 2006/03/14 18:20:07 brouard
602: (Module): varevsij Comments added explaining the second
603: table of variances if popbased=1 .
604: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
605: (Module): Function pstamp added
606: (Module): Version 0.98d
607:
608: Revision 1.117 2006/03/14 17:16:22 brouard
609: (Module): varevsij Comments added explaining the second
610: table of variances if popbased=1 .
611: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
612: (Module): Function pstamp added
613: (Module): Version 0.98d
614:
615: Revision 1.116 2006/03/06 10:29:27 brouard
616: (Module): Variance-covariance wrong links and
617: varian-covariance of ej. is needed (Saito).
618:
619: Revision 1.115 2006/02/27 12:17:45 brouard
620: (Module): One freematrix added in mlikeli! 0.98c
621:
622: Revision 1.114 2006/02/26 12:57:58 brouard
623: (Module): Some improvements in processing parameter
624: filename with strsep.
625:
626: Revision 1.113 2006/02/24 14:20:24 brouard
627: (Module): Memory leaks checks with valgrind and:
628: datafile was not closed, some imatrix were not freed and on matrix
629: allocation too.
630:
631: Revision 1.112 2006/01/30 09:55:26 brouard
632: (Module): Back to gnuplot.exe instead of wgnuplot.exe
633:
634: Revision 1.111 2006/01/25 20:38:18 brouard
635: (Module): Lots of cleaning and bugs added (Gompertz)
636: (Module): Comments can be added in data file. Missing date values
637: can be a simple dot '.'.
638:
639: Revision 1.110 2006/01/25 00:51:50 brouard
640: (Module): Lots of cleaning and bugs added (Gompertz)
641:
642: Revision 1.109 2006/01/24 19:37:15 brouard
643: (Module): Comments (lines starting with a #) are allowed in data.
644:
645: Revision 1.108 2006/01/19 18:05:42 lievre
646: Gnuplot problem appeared...
647: To be fixed
648:
649: Revision 1.107 2006/01/19 16:20:37 brouard
650: Test existence of gnuplot in imach path
651:
652: Revision 1.106 2006/01/19 13:24:36 brouard
653: Some cleaning and links added in html output
654:
655: Revision 1.105 2006/01/05 20:23:19 lievre
656: *** empty log message ***
657:
658: Revision 1.104 2005/09/30 16:11:43 lievre
659: (Module): sump fixed, loop imx fixed, and simplifications.
660: (Module): If the status is missing at the last wave but we know
661: that the person is alive, then we can code his/her status as -2
662: (instead of missing=-1 in earlier versions) and his/her
663: contributions to the likelihood is 1 - Prob of dying from last
664: health status (= 1-p13= p11+p12 in the easiest case of somebody in
665: the healthy state at last known wave). Version is 0.98
666:
667: Revision 1.103 2005/09/30 15:54:49 lievre
668: (Module): sump fixed, loop imx fixed, and simplifications.
669:
670: Revision 1.102 2004/09/15 17:31:30 brouard
671: Add the possibility to read data file including tab characters.
672:
673: Revision 1.101 2004/09/15 10:38:38 brouard
674: Fix on curr_time
675:
676: Revision 1.100 2004/07/12 18:29:06 brouard
677: Add version for Mac OS X. Just define UNIX in Makefile
678:
679: Revision 1.99 2004/06/05 08:57:40 brouard
680: *** empty log message ***
681:
682: Revision 1.98 2004/05/16 15:05:56 brouard
683: New version 0.97 . First attempt to estimate force of mortality
684: directly from the data i.e. without the need of knowing the health
685: state at each age, but using a Gompertz model: log u =a + b*age .
686: This is the basic analysis of mortality and should be done before any
687: other analysis, in order to test if the mortality estimated from the
688: cross-longitudinal survey is different from the mortality estimated
689: from other sources like vital statistic data.
690:
691: The same imach parameter file can be used but the option for mle should be -3.
692:
1.133 brouard 693: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 694: former routines in order to include the new code within the former code.
695:
696: The output is very simple: only an estimate of the intercept and of
697: the slope with 95% confident intervals.
698:
699: Current limitations:
700: A) Even if you enter covariates, i.e. with the
701: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
702: B) There is no computation of Life Expectancy nor Life Table.
703:
704: Revision 1.97 2004/02/20 13:25:42 lievre
705: Version 0.96d. Population forecasting command line is (temporarily)
706: suppressed.
707:
708: Revision 1.96 2003/07/15 15:38:55 brouard
709: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
710: rewritten within the same printf. Workaround: many printfs.
711:
712: Revision 1.95 2003/07/08 07:54:34 brouard
713: * imach.c (Repository):
714: (Repository): Using imachwizard code to output a more meaningful covariance
715: matrix (cov(a12,c31) instead of numbers.
716:
717: Revision 1.94 2003/06/27 13:00:02 brouard
718: Just cleaning
719:
720: Revision 1.93 2003/06/25 16:33:55 brouard
721: (Module): On windows (cygwin) function asctime_r doesn't
722: exist so I changed back to asctime which exists.
723: (Module): Version 0.96b
724:
725: Revision 1.92 2003/06/25 16:30:45 brouard
726: (Module): On windows (cygwin) function asctime_r doesn't
727: exist so I changed back to asctime which exists.
728:
729: Revision 1.91 2003/06/25 15:30:29 brouard
730: * imach.c (Repository): Duplicated warning errors corrected.
731: (Repository): Elapsed time after each iteration is now output. It
732: helps to forecast when convergence will be reached. Elapsed time
733: is stamped in powell. We created a new html file for the graphs
734: concerning matrix of covariance. It has extension -cov.htm.
735:
736: Revision 1.90 2003/06/24 12:34:15 brouard
737: (Module): Some bugs corrected for windows. Also, when
738: mle=-1 a template is output in file "or"mypar.txt with the design
739: of the covariance matrix to be input.
740:
741: Revision 1.89 2003/06/24 12:30:52 brouard
742: (Module): Some bugs corrected for windows. Also, when
743: mle=-1 a template is output in file "or"mypar.txt with the design
744: of the covariance matrix to be input.
745:
746: Revision 1.88 2003/06/23 17:54:56 brouard
747: * 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.
748:
749: Revision 1.87 2003/06/18 12:26:01 brouard
750: Version 0.96
751:
752: Revision 1.86 2003/06/17 20:04:08 brouard
753: (Module): Change position of html and gnuplot routines and added
754: routine fileappend.
755:
756: Revision 1.85 2003/06/17 13:12:43 brouard
757: * imach.c (Repository): Check when date of death was earlier that
758: current date of interview. It may happen when the death was just
759: prior to the death. In this case, dh was negative and likelihood
760: was wrong (infinity). We still send an "Error" but patch by
761: assuming that the date of death was just one stepm after the
762: interview.
763: (Repository): Because some people have very long ID (first column)
764: we changed int to long in num[] and we added a new lvector for
765: memory allocation. But we also truncated to 8 characters (left
766: truncation)
767: (Repository): No more line truncation errors.
768:
769: Revision 1.84 2003/06/13 21:44:43 brouard
770: * imach.c (Repository): Replace "freqsummary" at a correct
771: place. It differs from routine "prevalence" which may be called
772: many times. Probs is memory consuming and must be used with
773: parcimony.
774: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
775:
776: Revision 1.83 2003/06/10 13:39:11 lievre
777: *** empty log message ***
778:
779: Revision 1.82 2003/06/05 15:57:20 brouard
780: Add log in imach.c and fullversion number is now printed.
781:
782: */
783: /*
784: Interpolated Markov Chain
785:
786: Short summary of the programme:
787:
1.227 brouard 788: This program computes Healthy Life Expectancies or State-specific
789: (if states aren't health statuses) Expectancies from
790: cross-longitudinal data. Cross-longitudinal data consist in:
791:
792: -1- a first survey ("cross") where individuals from different ages
793: are interviewed on their health status or degree of disability (in
794: the case of a health survey which is our main interest)
795:
796: -2- at least a second wave of interviews ("longitudinal") which
797: measure each change (if any) in individual health status. Health
798: expectancies are computed from the time spent in each health state
799: according to a model. More health states you consider, more time is
800: necessary to reach the Maximum Likelihood of the parameters involved
801: in the model. The simplest model is the multinomial logistic model
802: where pij is the probability to be observed in state j at the second
803: wave conditional to be observed in state i at the first
804: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
805: etc , where 'age' is age and 'sex' is a covariate. If you want to
806: have a more complex model than "constant and age", you should modify
807: the program where the markup *Covariates have to be included here
808: again* invites you to do it. More covariates you add, slower the
1.126 brouard 809: convergence.
810:
811: The advantage of this computer programme, compared to a simple
812: multinomial logistic model, is clear when the delay between waves is not
813: identical for each individual. Also, if a individual missed an
814: intermediate interview, the information is lost, but taken into
815: account using an interpolation or extrapolation.
816:
817: hPijx is the probability to be observed in state i at age x+h
818: conditional to the observed state i at age x. The delay 'h' can be
819: split into an exact number (nh*stepm) of unobserved intermediate
820: states. This elementary transition (by month, quarter,
821: semester or year) is modelled as a multinomial logistic. The hPx
822: matrix is simply the matrix product of nh*stepm elementary matrices
823: and the contribution of each individual to the likelihood is simply
824: hPijx.
825:
826: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 827: of the life expectancies. It also computes the period (stable) prevalence.
828:
829: Back prevalence and projections:
1.227 brouard 830:
831: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
832: double agemaxpar, double ftolpl, int *ncvyearp, double
833: dateprev1,double dateprev2, int firstpass, int lastpass, int
834: mobilavproj)
835:
836: Computes the back prevalence limit for any combination of
837: covariate values k at any age between ageminpar and agemaxpar and
838: returns it in **bprlim. In the loops,
839:
840: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
841: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
842:
843: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 844: Computes for any combination of covariates k and any age between bage and fage
845: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
846: oldm=oldms;savm=savms;
1.227 brouard 847:
1.267 brouard 848: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 849: Computes the transition matrix starting at age 'age' over
850: 'nhstepm*hstepm*stepm' months (i.e. until
851: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 852: nhstepm*hstepm matrices.
853:
854: Returns p3mat[i][j][h] after calling
855: p3mat[i][j][h]=matprod2(newm,
856: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
857: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
858: oldm);
1.226 brouard 859:
860: Important routines
861:
862: - func (or funcone), computes logit (pij) distinguishing
863: o fixed variables (single or product dummies or quantitative);
864: o varying variables by:
865: (1) wave (single, product dummies, quantitative),
866: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
867: % fixed dummy (treated) or quantitative (not done because time-consuming);
868: % varying dummy (not done) or quantitative (not done);
869: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
870: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
871: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
872: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
873: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 874:
1.226 brouard 875:
876:
1.133 brouard 877: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
878: Institut national d'études démographiques, Paris.
1.126 brouard 879: This software have been partly granted by Euro-REVES, a concerted action
880: from the European Union.
881: It is copyrighted identically to a GNU software product, ie programme and
882: software can be distributed freely for non commercial use. Latest version
883: can be accessed at http://euroreves.ined.fr/imach .
884:
885: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
886: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
887:
888: **********************************************************************/
889: /*
890: main
891: read parameterfile
892: read datafile
893: concatwav
894: freqsummary
895: if (mle >= 1)
896: mlikeli
897: print results files
898: if mle==1
899: computes hessian
900: read end of parameter file: agemin, agemax, bage, fage, estepm
901: begin-prev-date,...
902: open gnuplot file
903: open html file
1.145 brouard 904: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
905: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
906: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
907: freexexit2 possible for memory heap.
908:
909: h Pij x | pij_nom ficrestpij
910: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
911: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
912: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
913:
914: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
915: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
916: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
917: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
918: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
919:
1.126 brouard 920: forecasting if prevfcast==1 prevforecast call prevalence()
921: health expectancies
922: Variance-covariance of DFLE
923: prevalence()
924: movingaverage()
925: varevsij()
926: if popbased==1 varevsij(,popbased)
927: total life expectancies
928: Variance of period (stable) prevalence
929: end
930: */
931:
1.187 brouard 932: /* #define DEBUG */
933: /* #define DEBUGBRENT */
1.203 brouard 934: /* #define DEBUGLINMIN */
935: /* #define DEBUGHESS */
936: #define DEBUGHESSIJ
1.224 brouard 937: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 938: #define POWELL /* Instead of NLOPT */
1.224 brouard 939: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 940: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
941: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 942:
943: #include <math.h>
944: #include <stdio.h>
945: #include <stdlib.h>
946: #include <string.h>
1.226 brouard 947: #include <ctype.h>
1.159 brouard 948:
949: #ifdef _WIN32
950: #include <io.h>
1.172 brouard 951: #include <windows.h>
952: #include <tchar.h>
1.159 brouard 953: #else
1.126 brouard 954: #include <unistd.h>
1.159 brouard 955: #endif
1.126 brouard 956:
957: #include <limits.h>
958: #include <sys/types.h>
1.171 brouard 959:
960: #if defined(__GNUC__)
961: #include <sys/utsname.h> /* Doesn't work on Windows */
962: #endif
963:
1.126 brouard 964: #include <sys/stat.h>
965: #include <errno.h>
1.159 brouard 966: /* extern int errno; */
1.126 brouard 967:
1.157 brouard 968: /* #ifdef LINUX */
969: /* #include <time.h> */
970: /* #include "timeval.h" */
971: /* #else */
972: /* #include <sys/time.h> */
973: /* #endif */
974:
1.126 brouard 975: #include <time.h>
976:
1.136 brouard 977: #ifdef GSL
978: #include <gsl/gsl_errno.h>
979: #include <gsl/gsl_multimin.h>
980: #endif
981:
1.167 brouard 982:
1.162 brouard 983: #ifdef NLOPT
984: #include <nlopt.h>
985: typedef struct {
986: double (* function)(double [] );
987: } myfunc_data ;
988: #endif
989:
1.126 brouard 990: /* #include <libintl.h> */
991: /* #define _(String) gettext (String) */
992:
1.251 brouard 993: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 994:
995: #define GNUPLOTPROGRAM "gnuplot"
996: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
997: #define FILENAMELENGTH 132
998:
999: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1000: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1001:
1.144 brouard 1002: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1003: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1004:
1005: #define NINTERVMAX 8
1.144 brouard 1006: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1007: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1008: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1009: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1010: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1011: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1012: #define MAXN 20000
1.144 brouard 1013: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1014: /* #define AGESUP 130 */
1015: #define AGESUP 150
1.268 brouard 1016: #define AGEINF 0
1.218 brouard 1017: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1018: #define AGEBASE 40
1.194 brouard 1019: #define AGEOVERFLOW 1.e20
1.164 brouard 1020: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1021: #ifdef _WIN32
1022: #define DIRSEPARATOR '\\'
1023: #define CHARSEPARATOR "\\"
1024: #define ODIRSEPARATOR '/'
1025: #else
1.126 brouard 1026: #define DIRSEPARATOR '/'
1027: #define CHARSEPARATOR "/"
1028: #define ODIRSEPARATOR '\\'
1029: #endif
1030:
1.276 ! brouard 1031: /* $Id: imach.c,v 1.275 2017/06/30 13:39:33 brouard Exp $ */
1.126 brouard 1032: /* $State: Exp $ */
1.196 brouard 1033: #include "version.h"
1034: char version[]=__IMACH_VERSION__;
1.224 brouard 1035: 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";
1.276 ! brouard 1036: char fullversion[]="$Revision: 1.275 $ $Date: 2017/06/30 13:39:33 $";
1.126 brouard 1037: char strstart[80];
1038: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1039: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1040: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1041: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1042: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1043: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1044: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1045: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1046: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1047: int cptcovprodnoage=0; /**< Number of covariate products without age */
1048: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1049: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1050: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1051: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1052: int nsd=0; /**< Total number of single dummy variables (output) */
1053: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1054: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1055: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1056: int ntveff=0; /**< ntveff number of effective time varying variables */
1057: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1058: int cptcov=0; /* Working variable */
1.218 brouard 1059: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1060: int npar=NPARMAX;
1061: int nlstate=2; /* Number of live states */
1062: int ndeath=1; /* Number of dead states */
1.130 brouard 1063: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1064: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1065: int popbased=0;
1066:
1067: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1068: int maxwav=0; /* Maxim number of waves */
1069: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1070: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1071: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1072: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1073: int mle=1, weightopt=0;
1.126 brouard 1074: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1075: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1076: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1077: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1078: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1079: int selected(int kvar); /* Is covariate kvar selected for printing results */
1080:
1.130 brouard 1081: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1082: double **matprod2(); /* test */
1.126 brouard 1083: double **oldm, **newm, **savm; /* Working pointers to matrices */
1084: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1085: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1086:
1.136 brouard 1087: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1088: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1089: FILE *ficlog, *ficrespow;
1.130 brouard 1090: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1091: double fretone; /* Only one call to likelihood */
1.130 brouard 1092: long ipmx=0; /* Number of contributions */
1.126 brouard 1093: double sw; /* Sum of weights */
1094: char filerespow[FILENAMELENGTH];
1095: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1096: FILE *ficresilk;
1097: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1098: FILE *ficresprobmorprev;
1099: FILE *fichtm, *fichtmcov; /* Html File */
1100: FILE *ficreseij;
1101: char filerese[FILENAMELENGTH];
1102: FILE *ficresstdeij;
1103: char fileresstde[FILENAMELENGTH];
1104: FILE *ficrescveij;
1105: char filerescve[FILENAMELENGTH];
1106: FILE *ficresvij;
1107: char fileresv[FILENAMELENGTH];
1.269 brouard 1108:
1.126 brouard 1109: char title[MAXLINE];
1.234 brouard 1110: char model[MAXLINE]; /**< The model line */
1.217 brouard 1111: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1112: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1113: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1114: char command[FILENAMELENGTH];
1115: int outcmd=0;
1116:
1.217 brouard 1117: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1118: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1119: char filelog[FILENAMELENGTH]; /* Log file */
1120: char filerest[FILENAMELENGTH];
1121: char fileregp[FILENAMELENGTH];
1122: char popfile[FILENAMELENGTH];
1123:
1124: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1125:
1.157 brouard 1126: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1127: /* struct timezone tzp; */
1128: /* extern int gettimeofday(); */
1129: struct tm tml, *gmtime(), *localtime();
1130:
1131: extern time_t time();
1132:
1133: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1134: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1135: struct tm tm;
1136:
1.126 brouard 1137: char strcurr[80], strfor[80];
1138:
1139: char *endptr;
1140: long lval;
1141: double dval;
1142:
1143: #define NR_END 1
1144: #define FREE_ARG char*
1145: #define FTOL 1.0e-10
1146:
1147: #define NRANSI
1.240 brouard 1148: #define ITMAX 200
1149: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1150:
1151: #define TOL 2.0e-4
1152:
1153: #define CGOLD 0.3819660
1154: #define ZEPS 1.0e-10
1155: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1156:
1157: #define GOLD 1.618034
1158: #define GLIMIT 100.0
1159: #define TINY 1.0e-20
1160:
1161: static double maxarg1,maxarg2;
1162: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1163: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1164:
1165: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1166: #define rint(a) floor(a+0.5)
1.166 brouard 1167: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1168: #define mytinydouble 1.0e-16
1.166 brouard 1169: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1170: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1171: /* static double dsqrarg; */
1172: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1173: static double sqrarg;
1174: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1175: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1176: int agegomp= AGEGOMP;
1177:
1178: int imx;
1179: int stepm=1;
1180: /* Stepm, step in month: minimum step interpolation*/
1181:
1182: int estepm;
1183: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1184:
1185: int m,nb;
1186: long *num;
1.197 brouard 1187: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1188: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1189: covariate for which somebody answered excluding
1190: undefined. Usually 2: 0 and 1. */
1191: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1192: covariate for which somebody answered including
1193: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1194: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1195: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1196: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1197: double *ageexmed,*agecens;
1198: double dateintmean=0;
1199:
1200: double *weight;
1201: int **s; /* Status */
1.141 brouard 1202: double *agedc;
1.145 brouard 1203: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1204: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1205: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1206: double **coqvar; /* Fixed quantitative covariate nqv */
1207: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1208: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1209: double idx;
1210: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1211: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1212: /*k 1 2 3 4 5 6 7 8 9 */
1213: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1214: /* Tndvar[k] 1 2 3 4 5 */
1215: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1216: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1217: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1218: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1219: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1220: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1221: /* Tprod[i]=k 4 7 */
1222: /* Tage[i]=k 5 8 */
1223: /* */
1224: /* Type */
1225: /* V 1 2 3 4 5 */
1226: /* F F V V V */
1227: /* D Q D D Q */
1228: /* */
1229: int *TvarsD;
1230: int *TvarsDind;
1231: int *TvarsQ;
1232: int *TvarsQind;
1233:
1.235 brouard 1234: #define MAXRESULTLINES 10
1235: int nresult=0;
1.258 brouard 1236: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1237: int TKresult[MAXRESULTLINES];
1.237 brouard 1238: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1239: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1240: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1241: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1242: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1243: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1244:
1.234 brouard 1245: /* 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 1246: 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 */
1247: 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 */
1248: 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 */
1249: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1250: 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 */
1251: 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 1252: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1253: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1254: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1255: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1256: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1257: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1258: 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 */
1259: 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 */
1260:
1.230 brouard 1261: int *Tvarsel; /**< Selected covariates for output */
1262: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1263: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1264: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1265: 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 1266: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1267: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1268: int *Tage;
1.227 brouard 1269: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1270: 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 1271: 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*/
1272: 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 1273: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1274: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1275: int **Tvard;
1276: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1277: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1278: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1279: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1280: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1281: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1282: double *lsurv, *lpop, *tpop;
1283:
1.231 brouard 1284: #define FD 1; /* Fixed dummy covariate */
1285: #define FQ 2; /* Fixed quantitative covariate */
1286: #define FP 3; /* Fixed product covariate */
1287: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1288: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1289: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1290: #define VD 10; /* Varying dummy covariate */
1291: #define VQ 11; /* Varying quantitative covariate */
1292: #define VP 12; /* Varying product covariate */
1293: #define VPDD 13; /* Varying product dummy*dummy covariate */
1294: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1295: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1296: #define APFD 16; /* Age product * fixed dummy covariate */
1297: #define APFQ 17; /* Age product * fixed quantitative covariate */
1298: #define APVD 18; /* Age product * varying dummy covariate */
1299: #define APVQ 19; /* Age product * varying quantitative covariate */
1300:
1301: #define FTYPE 1; /* Fixed covariate */
1302: #define VTYPE 2; /* Varying covariate (loop in wave) */
1303: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1304:
1305: struct kmodel{
1306: int maintype; /* main type */
1307: int subtype; /* subtype */
1308: };
1309: struct kmodel modell[NCOVMAX];
1310:
1.143 brouard 1311: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1312: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1313:
1314: /**************** split *************************/
1315: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1316: {
1317: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1318: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1319: */
1320: char *ss; /* pointer */
1.186 brouard 1321: int l1=0, l2=0; /* length counters */
1.126 brouard 1322:
1323: l1 = strlen(path ); /* length of path */
1324: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1325: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1326: if ( ss == NULL ) { /* no directory, so determine current directory */
1327: strcpy( name, path ); /* we got the fullname name because no directory */
1328: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1329: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1330: /* get current working directory */
1331: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1332: #ifdef WIN32
1333: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1334: #else
1335: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1336: #endif
1.126 brouard 1337: return( GLOCK_ERROR_GETCWD );
1338: }
1339: /* got dirc from getcwd*/
1340: printf(" DIRC = %s \n",dirc);
1.205 brouard 1341: } else { /* strip directory from path */
1.126 brouard 1342: ss++; /* after this, the filename */
1343: l2 = strlen( ss ); /* length of filename */
1344: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1345: strcpy( name, ss ); /* save file name */
1346: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1347: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1348: printf(" DIRC2 = %s \n",dirc);
1349: }
1350: /* We add a separator at the end of dirc if not exists */
1351: l1 = strlen( dirc ); /* length of directory */
1352: if( dirc[l1-1] != DIRSEPARATOR ){
1353: dirc[l1] = DIRSEPARATOR;
1354: dirc[l1+1] = 0;
1355: printf(" DIRC3 = %s \n",dirc);
1356: }
1357: ss = strrchr( name, '.' ); /* find last / */
1358: if (ss >0){
1359: ss++;
1360: strcpy(ext,ss); /* save extension */
1361: l1= strlen( name);
1362: l2= strlen(ss)+1;
1363: strncpy( finame, name, l1-l2);
1364: finame[l1-l2]= 0;
1365: }
1366:
1367: return( 0 ); /* we're done */
1368: }
1369:
1370:
1371: /******************************************/
1372:
1373: void replace_back_to_slash(char *s, char*t)
1374: {
1375: int i;
1376: int lg=0;
1377: i=0;
1378: lg=strlen(t);
1379: for(i=0; i<= lg; i++) {
1380: (s[i] = t[i]);
1381: if (t[i]== '\\') s[i]='/';
1382: }
1383: }
1384:
1.132 brouard 1385: char *trimbb(char *out, char *in)
1.137 brouard 1386: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1387: char *s;
1388: s=out;
1389: while (*in != '\0'){
1.137 brouard 1390: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1391: in++;
1392: }
1393: *out++ = *in++;
1394: }
1395: *out='\0';
1396: return s;
1397: }
1398:
1.187 brouard 1399: /* char *substrchaine(char *out, char *in, char *chain) */
1400: /* { */
1401: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1402: /* char *s, *t; */
1403: /* t=in;s=out; */
1404: /* while ((*in != *chain) && (*in != '\0')){ */
1405: /* *out++ = *in++; */
1406: /* } */
1407:
1408: /* /\* *in matches *chain *\/ */
1409: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1410: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1411: /* } */
1412: /* in--; chain--; */
1413: /* while ( (*in != '\0')){ */
1414: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1415: /* *out++ = *in++; */
1416: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1417: /* } */
1418: /* *out='\0'; */
1419: /* out=s; */
1420: /* return out; */
1421: /* } */
1422: char *substrchaine(char *out, char *in, char *chain)
1423: {
1424: /* Substract chain 'chain' from 'in', return and output 'out' */
1425: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1426:
1427: char *strloc;
1428:
1429: strcpy (out, in);
1430: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1431: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1432: if(strloc != NULL){
1433: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1434: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1435: /* strcpy (strloc, strloc +strlen(chain));*/
1436: }
1437: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1438: return out;
1439: }
1440:
1441:
1.145 brouard 1442: char *cutl(char *blocc, char *alocc, char *in, char occ)
1443: {
1.187 brouard 1444: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1445: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1446: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1447: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1448: */
1.160 brouard 1449: char *s, *t;
1.145 brouard 1450: t=in;s=in;
1451: while ((*in != occ) && (*in != '\0')){
1452: *alocc++ = *in++;
1453: }
1454: if( *in == occ){
1455: *(alocc)='\0';
1456: s=++in;
1457: }
1458:
1459: if (s == t) {/* occ not found */
1460: *(alocc-(in-s))='\0';
1461: in=s;
1462: }
1463: while ( *in != '\0'){
1464: *blocc++ = *in++;
1465: }
1466:
1467: *blocc='\0';
1468: return t;
1469: }
1.137 brouard 1470: char *cutv(char *blocc, char *alocc, char *in, char occ)
1471: {
1.187 brouard 1472: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1473: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1474: gives blocc="abcdef2ghi" and alocc="j".
1475: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1476: */
1477: char *s, *t;
1478: t=in;s=in;
1479: while (*in != '\0'){
1480: while( *in == occ){
1481: *blocc++ = *in++;
1482: s=in;
1483: }
1484: *blocc++ = *in++;
1485: }
1486: if (s == t) /* occ not found */
1487: *(blocc-(in-s))='\0';
1488: else
1489: *(blocc-(in-s)-1)='\0';
1490: in=s;
1491: while ( *in != '\0'){
1492: *alocc++ = *in++;
1493: }
1494:
1495: *alocc='\0';
1496: return s;
1497: }
1498:
1.126 brouard 1499: int nbocc(char *s, char occ)
1500: {
1501: int i,j=0;
1502: int lg=20;
1503: i=0;
1504: lg=strlen(s);
1505: for(i=0; i<= lg; i++) {
1.234 brouard 1506: if (s[i] == occ ) j++;
1.126 brouard 1507: }
1508: return j;
1509: }
1510:
1.137 brouard 1511: /* void cutv(char *u,char *v, char*t, char occ) */
1512: /* { */
1513: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1514: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1515: /* gives u="abcdef2ghi" and v="j" *\/ */
1516: /* int i,lg,j,p=0; */
1517: /* i=0; */
1518: /* lg=strlen(t); */
1519: /* for(j=0; j<=lg-1; j++) { */
1520: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1521: /* } */
1.126 brouard 1522:
1.137 brouard 1523: /* for(j=0; j<p; j++) { */
1524: /* (u[j] = t[j]); */
1525: /* } */
1526: /* u[p]='\0'; */
1.126 brouard 1527:
1.137 brouard 1528: /* for(j=0; j<= lg; j++) { */
1529: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1530: /* } */
1531: /* } */
1.126 brouard 1532:
1.160 brouard 1533: #ifdef _WIN32
1534: char * strsep(char **pp, const char *delim)
1535: {
1536: char *p, *q;
1537:
1538: if ((p = *pp) == NULL)
1539: return 0;
1540: if ((q = strpbrk (p, delim)) != NULL)
1541: {
1542: *pp = q + 1;
1543: *q = '\0';
1544: }
1545: else
1546: *pp = 0;
1547: return p;
1548: }
1549: #endif
1550:
1.126 brouard 1551: /********************** nrerror ********************/
1552:
1553: void nrerror(char error_text[])
1554: {
1555: fprintf(stderr,"ERREUR ...\n");
1556: fprintf(stderr,"%s\n",error_text);
1557: exit(EXIT_FAILURE);
1558: }
1559: /*********************** vector *******************/
1560: double *vector(int nl, int nh)
1561: {
1562: double *v;
1563: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1564: if (!v) nrerror("allocation failure in vector");
1565: return v-nl+NR_END;
1566: }
1567:
1568: /************************ free vector ******************/
1569: void free_vector(double*v, int nl, int nh)
1570: {
1571: free((FREE_ARG)(v+nl-NR_END));
1572: }
1573:
1574: /************************ivector *******************************/
1575: int *ivector(long nl,long nh)
1576: {
1577: int *v;
1578: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1579: if (!v) nrerror("allocation failure in ivector");
1580: return v-nl+NR_END;
1581: }
1582:
1583: /******************free ivector **************************/
1584: void free_ivector(int *v, long nl, long nh)
1585: {
1586: free((FREE_ARG)(v+nl-NR_END));
1587: }
1588:
1589: /************************lvector *******************************/
1590: long *lvector(long nl,long nh)
1591: {
1592: long *v;
1593: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1594: if (!v) nrerror("allocation failure in ivector");
1595: return v-nl+NR_END;
1596: }
1597:
1598: /******************free lvector **************************/
1599: void free_lvector(long *v, long nl, long nh)
1600: {
1601: free((FREE_ARG)(v+nl-NR_END));
1602: }
1603:
1604: /******************* imatrix *******************************/
1605: int **imatrix(long nrl, long nrh, long ncl, long nch)
1606: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1607: {
1608: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1609: int **m;
1610:
1611: /* allocate pointers to rows */
1612: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1613: if (!m) nrerror("allocation failure 1 in matrix()");
1614: m += NR_END;
1615: m -= nrl;
1616:
1617:
1618: /* allocate rows and set pointers to them */
1619: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1620: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1621: m[nrl] += NR_END;
1622: m[nrl] -= ncl;
1623:
1624: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1625:
1626: /* return pointer to array of pointers to rows */
1627: return m;
1628: }
1629:
1630: /****************** free_imatrix *************************/
1631: void free_imatrix(m,nrl,nrh,ncl,nch)
1632: int **m;
1633: long nch,ncl,nrh,nrl;
1634: /* free an int matrix allocated by imatrix() */
1635: {
1636: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1637: free((FREE_ARG) (m+nrl-NR_END));
1638: }
1639:
1640: /******************* matrix *******************************/
1641: double **matrix(long nrl, long nrh, long ncl, long nch)
1642: {
1643: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1644: double **m;
1645:
1646: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1647: if (!m) nrerror("allocation failure 1 in matrix()");
1648: m += NR_END;
1649: m -= nrl;
1650:
1651: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1652: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1653: m[nrl] += NR_END;
1654: m[nrl] -= ncl;
1655:
1656: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1657: return m;
1.145 brouard 1658: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1659: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1660: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1661: */
1662: }
1663:
1664: /*************************free matrix ************************/
1665: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1666: {
1667: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1668: free((FREE_ARG)(m+nrl-NR_END));
1669: }
1670:
1671: /******************* ma3x *******************************/
1672: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1673: {
1674: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1675: double ***m;
1676:
1677: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1678: if (!m) nrerror("allocation failure 1 in matrix()");
1679: m += NR_END;
1680: m -= nrl;
1681:
1682: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1683: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1684: m[nrl] += NR_END;
1685: m[nrl] -= ncl;
1686:
1687: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1688:
1689: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1690: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1691: m[nrl][ncl] += NR_END;
1692: m[nrl][ncl] -= nll;
1693: for (j=ncl+1; j<=nch; j++)
1694: m[nrl][j]=m[nrl][j-1]+nlay;
1695:
1696: for (i=nrl+1; i<=nrh; i++) {
1697: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1698: for (j=ncl+1; j<=nch; j++)
1699: m[i][j]=m[i][j-1]+nlay;
1700: }
1701: return m;
1702: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1703: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1704: */
1705: }
1706:
1707: /*************************free ma3x ************************/
1708: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1709: {
1710: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1711: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1712: free((FREE_ARG)(m+nrl-NR_END));
1713: }
1714:
1715: /*************** function subdirf ***********/
1716: char *subdirf(char fileres[])
1717: {
1718: /* Caution optionfilefiname is hidden */
1719: strcpy(tmpout,optionfilefiname);
1720: strcat(tmpout,"/"); /* Add to the right */
1721: strcat(tmpout,fileres);
1722: return tmpout;
1723: }
1724:
1725: /*************** function subdirf2 ***********/
1726: char *subdirf2(char fileres[], char *preop)
1727: {
1728:
1729: /* Caution optionfilefiname is hidden */
1730: strcpy(tmpout,optionfilefiname);
1731: strcat(tmpout,"/");
1732: strcat(tmpout,preop);
1733: strcat(tmpout,fileres);
1734: return tmpout;
1735: }
1736:
1737: /*************** function subdirf3 ***********/
1738: char *subdirf3(char fileres[], char *preop, char *preop2)
1739: {
1740:
1741: /* Caution optionfilefiname is hidden */
1742: strcpy(tmpout,optionfilefiname);
1743: strcat(tmpout,"/");
1744: strcat(tmpout,preop);
1745: strcat(tmpout,preop2);
1746: strcat(tmpout,fileres);
1747: return tmpout;
1748: }
1.213 brouard 1749:
1750: /*************** function subdirfext ***********/
1751: char *subdirfext(char fileres[], char *preop, char *postop)
1752: {
1753:
1754: strcpy(tmpout,preop);
1755: strcat(tmpout,fileres);
1756: strcat(tmpout,postop);
1757: return tmpout;
1758: }
1.126 brouard 1759:
1.213 brouard 1760: /*************** function subdirfext3 ***********/
1761: char *subdirfext3(char fileres[], char *preop, char *postop)
1762: {
1763:
1764: /* Caution optionfilefiname is hidden */
1765: strcpy(tmpout,optionfilefiname);
1766: strcat(tmpout,"/");
1767: strcat(tmpout,preop);
1768: strcat(tmpout,fileres);
1769: strcat(tmpout,postop);
1770: return tmpout;
1771: }
1772:
1.162 brouard 1773: char *asc_diff_time(long time_sec, char ascdiff[])
1774: {
1775: long sec_left, days, hours, minutes;
1776: days = (time_sec) / (60*60*24);
1777: sec_left = (time_sec) % (60*60*24);
1778: hours = (sec_left) / (60*60) ;
1779: sec_left = (sec_left) %(60*60);
1780: minutes = (sec_left) /60;
1781: sec_left = (sec_left) % (60);
1782: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1783: return ascdiff;
1784: }
1785:
1.126 brouard 1786: /***************** f1dim *************************/
1787: extern int ncom;
1788: extern double *pcom,*xicom;
1789: extern double (*nrfunc)(double []);
1790:
1791: double f1dim(double x)
1792: {
1793: int j;
1794: double f;
1795: double *xt;
1796:
1797: xt=vector(1,ncom);
1798: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1799: f=(*nrfunc)(xt);
1800: free_vector(xt,1,ncom);
1801: return f;
1802: }
1803:
1804: /*****************brent *************************/
1805: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1806: {
1807: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1808: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1809: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1810: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1811: * returned function value.
1812: */
1.126 brouard 1813: int iter;
1814: double a,b,d,etemp;
1.159 brouard 1815: double fu=0,fv,fw,fx;
1.164 brouard 1816: double ftemp=0.;
1.126 brouard 1817: double p,q,r,tol1,tol2,u,v,w,x,xm;
1818: double e=0.0;
1819:
1820: a=(ax < cx ? ax : cx);
1821: b=(ax > cx ? ax : cx);
1822: x=w=v=bx;
1823: fw=fv=fx=(*f)(x);
1824: for (iter=1;iter<=ITMAX;iter++) {
1825: xm=0.5*(a+b);
1826: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1827: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1828: printf(".");fflush(stdout);
1829: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1830: #ifdef DEBUGBRENT
1.126 brouard 1831: 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);
1832: 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);
1833: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1834: #endif
1835: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1836: *xmin=x;
1837: return fx;
1838: }
1839: ftemp=fu;
1840: if (fabs(e) > tol1) {
1841: r=(x-w)*(fx-fv);
1842: q=(x-v)*(fx-fw);
1843: p=(x-v)*q-(x-w)*r;
1844: q=2.0*(q-r);
1845: if (q > 0.0) p = -p;
1846: q=fabs(q);
1847: etemp=e;
1848: e=d;
1849: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1850: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1851: else {
1.224 brouard 1852: d=p/q;
1853: u=x+d;
1854: if (u-a < tol2 || b-u < tol2)
1855: d=SIGN(tol1,xm-x);
1.126 brouard 1856: }
1857: } else {
1858: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1859: }
1860: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1861: fu=(*f)(u);
1862: if (fu <= fx) {
1863: if (u >= x) a=x; else b=x;
1864: SHFT(v,w,x,u)
1.183 brouard 1865: SHFT(fv,fw,fx,fu)
1866: } else {
1867: if (u < x) a=u; else b=u;
1868: if (fu <= fw || w == x) {
1.224 brouard 1869: v=w;
1870: w=u;
1871: fv=fw;
1872: fw=fu;
1.183 brouard 1873: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1874: v=u;
1875: fv=fu;
1.183 brouard 1876: }
1877: }
1.126 brouard 1878: }
1879: nrerror("Too many iterations in brent");
1880: *xmin=x;
1881: return fx;
1882: }
1883:
1884: /****************** mnbrak ***********************/
1885:
1886: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1887: double (*func)(double))
1.183 brouard 1888: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1889: the downhill direction (defined by the function as evaluated at the initial points) and returns
1890: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1891: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1892: */
1.126 brouard 1893: double ulim,u,r,q, dum;
1894: double fu;
1.187 brouard 1895:
1896: double scale=10.;
1897: int iterscale=0;
1898:
1899: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1900: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1901:
1902:
1903: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1904: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1905: /* *bx = *ax - (*ax - *bx)/scale; */
1906: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1907: /* } */
1908:
1.126 brouard 1909: if (*fb > *fa) {
1910: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1911: SHFT(dum,*fb,*fa,dum)
1912: }
1.126 brouard 1913: *cx=(*bx)+GOLD*(*bx-*ax);
1914: *fc=(*func)(*cx);
1.183 brouard 1915: #ifdef DEBUG
1.224 brouard 1916: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1917: 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 1918: #endif
1.224 brouard 1919: 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 1920: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1921: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1922: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1923: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1924: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1925: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1926: fu=(*func)(u);
1.163 brouard 1927: #ifdef DEBUG
1928: /* f(x)=A(x-u)**2+f(u) */
1929: double A, fparabu;
1930: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1931: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1932: 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);
1933: 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 1934: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1935: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1936: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1937: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1938: #endif
1.184 brouard 1939: #ifdef MNBRAKORIGINAL
1.183 brouard 1940: #else
1.191 brouard 1941: /* if (fu > *fc) { */
1942: /* #ifdef DEBUG */
1943: /* printf("mnbrak4 fu > fc \n"); */
1944: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1945: /* #endif */
1946: /* /\* 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 *\\/ *\/ */
1947: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1948: /* dum=u; /\* Shifting c and u *\/ */
1949: /* u = *cx; */
1950: /* *cx = dum; */
1951: /* dum = fu; */
1952: /* fu = *fc; */
1953: /* *fc =dum; */
1954: /* } else { /\* end *\/ */
1955: /* #ifdef DEBUG */
1956: /* printf("mnbrak3 fu < fc \n"); */
1957: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1958: /* #endif */
1959: /* dum=u; /\* Shifting c and u *\/ */
1960: /* u = *cx; */
1961: /* *cx = dum; */
1962: /* dum = fu; */
1963: /* fu = *fc; */
1964: /* *fc =dum; */
1965: /* } */
1.224 brouard 1966: #ifdef DEBUGMNBRAK
1967: double A, fparabu;
1968: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1969: fparabu= *fa - A*(*ax-u)*(*ax-u);
1970: 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);
1971: 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 1972: #endif
1.191 brouard 1973: dum=u; /* Shifting c and u */
1974: u = *cx;
1975: *cx = dum;
1976: dum = fu;
1977: fu = *fc;
1978: *fc =dum;
1.183 brouard 1979: #endif
1.162 brouard 1980: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1981: #ifdef DEBUG
1.224 brouard 1982: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1983: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1984: #endif
1.126 brouard 1985: fu=(*func)(u);
1986: if (fu < *fc) {
1.183 brouard 1987: #ifdef DEBUG
1.224 brouard 1988: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1989: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1990: #endif
1991: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1992: SHFT(*fb,*fc,fu,(*func)(u))
1993: #ifdef DEBUG
1994: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1995: #endif
1996: }
1.162 brouard 1997: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1998: #ifdef DEBUG
1.224 brouard 1999: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2000: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2001: #endif
1.126 brouard 2002: u=ulim;
2003: fu=(*func)(u);
1.183 brouard 2004: } else { /* u could be left to b (if r > q parabola has a maximum) */
2005: #ifdef DEBUG
1.224 brouard 2006: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2007: 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 2008: #endif
1.126 brouard 2009: u=(*cx)+GOLD*(*cx-*bx);
2010: fu=(*func)(u);
1.224 brouard 2011: #ifdef DEBUG
2012: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2013: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2014: #endif
1.183 brouard 2015: } /* end tests */
1.126 brouard 2016: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2017: SHFT(*fa,*fb,*fc,fu)
2018: #ifdef DEBUG
1.224 brouard 2019: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2020: 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 2021: #endif
2022: } /* 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 2023: }
2024:
2025: /*************** linmin ************************/
1.162 brouard 2026: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2027: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2028: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2029: the value of func at the returned location p . This is actually all accomplished by calling the
2030: routines mnbrak and brent .*/
1.126 brouard 2031: int ncom;
2032: double *pcom,*xicom;
2033: double (*nrfunc)(double []);
2034:
1.224 brouard 2035: #ifdef LINMINORIGINAL
1.126 brouard 2036: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2037: #else
2038: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2039: #endif
1.126 brouard 2040: {
2041: double brent(double ax, double bx, double cx,
2042: double (*f)(double), double tol, double *xmin);
2043: double f1dim(double x);
2044: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2045: double *fc, double (*func)(double));
2046: int j;
2047: double xx,xmin,bx,ax;
2048: double fx,fb,fa;
1.187 brouard 2049:
1.203 brouard 2050: #ifdef LINMINORIGINAL
2051: #else
2052: double scale=10., axs, xxs; /* Scale added for infinity */
2053: #endif
2054:
1.126 brouard 2055: ncom=n;
2056: pcom=vector(1,n);
2057: xicom=vector(1,n);
2058: nrfunc=func;
2059: for (j=1;j<=n;j++) {
2060: pcom[j]=p[j];
1.202 brouard 2061: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2062: }
1.187 brouard 2063:
1.203 brouard 2064: #ifdef LINMINORIGINAL
2065: xx=1.;
2066: #else
2067: axs=0.0;
2068: xxs=1.;
2069: do{
2070: xx= xxs;
2071: #endif
1.187 brouard 2072: ax=0.;
2073: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2074: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2075: /* 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)) */
2076: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2077: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2078: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2079: /* 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 2080: #ifdef LINMINORIGINAL
2081: #else
2082: if (fx != fx){
1.224 brouard 2083: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2084: printf("|");
2085: fprintf(ficlog,"|");
1.203 brouard 2086: #ifdef DEBUGLINMIN
1.224 brouard 2087: 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 2088: #endif
2089: }
1.224 brouard 2090: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2091: #endif
2092:
1.191 brouard 2093: #ifdef DEBUGLINMIN
2094: 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 2095: 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 2096: #endif
1.224 brouard 2097: #ifdef LINMINORIGINAL
2098: #else
2099: if(fb == fx){ /* Flat function in the direction */
2100: xmin=xx;
2101: *flat=1;
2102: }else{
2103: *flat=0;
2104: #endif
2105: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2106: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2107: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2108: /* fmin = f(p[j] + xmin * xi[j]) */
2109: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2110: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2111: #ifdef DEBUG
1.224 brouard 2112: 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);
2113: 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);
2114: #endif
2115: #ifdef LINMINORIGINAL
2116: #else
2117: }
1.126 brouard 2118: #endif
1.191 brouard 2119: #ifdef DEBUGLINMIN
2120: printf("linmin end ");
1.202 brouard 2121: fprintf(ficlog,"linmin end ");
1.191 brouard 2122: #endif
1.126 brouard 2123: for (j=1;j<=n;j++) {
1.203 brouard 2124: #ifdef LINMINORIGINAL
2125: xi[j] *= xmin;
2126: #else
2127: #ifdef DEBUGLINMIN
2128: if(xxs <1.0)
2129: printf(" before xi[%d]=%12.8f", j,xi[j]);
2130: #endif
2131: 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) */
2132: #ifdef DEBUGLINMIN
2133: if(xxs <1.0)
2134: 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 );
2135: #endif
2136: #endif
1.187 brouard 2137: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2138: }
1.191 brouard 2139: #ifdef DEBUGLINMIN
1.203 brouard 2140: printf("\n");
1.191 brouard 2141: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2142: 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 2143: for (j=1;j<=n;j++) {
1.202 brouard 2144: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2145: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2146: if(j % ncovmodel == 0){
1.191 brouard 2147: printf("\n");
1.202 brouard 2148: fprintf(ficlog,"\n");
2149: }
1.191 brouard 2150: }
1.203 brouard 2151: #else
1.191 brouard 2152: #endif
1.126 brouard 2153: free_vector(xicom,1,n);
2154: free_vector(pcom,1,n);
2155: }
2156:
2157:
2158: /*************** powell ************************/
1.162 brouard 2159: /*
2160: Minimization of a function func of n variables. Input consists of an initial starting point
2161: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2162: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2163: such that failure to decrease by more than this amount on one iteration signals doneness. On
2164: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2165: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2166: */
1.224 brouard 2167: #ifdef LINMINORIGINAL
2168: #else
2169: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2170: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2171: #endif
1.126 brouard 2172: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2173: double (*func)(double []))
2174: {
1.224 brouard 2175: #ifdef LINMINORIGINAL
2176: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2177: double (*func)(double []));
1.224 brouard 2178: #else
1.241 brouard 2179: void linmin(double p[], double xi[], int n, double *fret,
2180: double (*func)(double []),int *flat);
1.224 brouard 2181: #endif
1.239 brouard 2182: int i,ibig,j,jk,k;
1.126 brouard 2183: double del,t,*pt,*ptt,*xit;
1.181 brouard 2184: double directest;
1.126 brouard 2185: double fp,fptt;
2186: double *xits;
2187: int niterf, itmp;
1.224 brouard 2188: #ifdef LINMINORIGINAL
2189: #else
2190:
2191: flatdir=ivector(1,n);
2192: for (j=1;j<=n;j++) flatdir[j]=0;
2193: #endif
1.126 brouard 2194:
2195: pt=vector(1,n);
2196: ptt=vector(1,n);
2197: xit=vector(1,n);
2198: xits=vector(1,n);
2199: *fret=(*func)(p);
2200: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2201: rcurr_time = time(NULL);
1.126 brouard 2202: for (*iter=1;;++(*iter)) {
1.187 brouard 2203: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2204: ibig=0;
2205: del=0.0;
1.157 brouard 2206: rlast_time=rcurr_time;
2207: /* (void) gettimeofday(&curr_time,&tzp); */
2208: rcurr_time = time(NULL);
2209: curr_time = *localtime(&rcurr_time);
2210: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2211: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2212: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2213: for (i=1;i<=n;i++) {
1.126 brouard 2214: fprintf(ficrespow," %.12lf", p[i]);
2215: }
1.239 brouard 2216: fprintf(ficrespow,"\n");fflush(ficrespow);
2217: printf("\n#model= 1 + age ");
2218: fprintf(ficlog,"\n#model= 1 + age ");
2219: if(nagesqr==1){
1.241 brouard 2220: printf(" + age*age ");
2221: fprintf(ficlog," + age*age ");
1.239 brouard 2222: }
2223: for(j=1;j <=ncovmodel-2;j++){
2224: if(Typevar[j]==0) {
2225: printf(" + V%d ",Tvar[j]);
2226: fprintf(ficlog," + V%d ",Tvar[j]);
2227: }else if(Typevar[j]==1) {
2228: printf(" + V%d*age ",Tvar[j]);
2229: fprintf(ficlog," + V%d*age ",Tvar[j]);
2230: }else if(Typevar[j]==2) {
2231: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2232: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2233: }
2234: }
1.126 brouard 2235: printf("\n");
1.239 brouard 2236: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2237: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2238: fprintf(ficlog,"\n");
1.239 brouard 2239: for(i=1,jk=1; i <=nlstate; i++){
2240: for(k=1; k <=(nlstate+ndeath); k++){
2241: if (k != i) {
2242: printf("%d%d ",i,k);
2243: fprintf(ficlog,"%d%d ",i,k);
2244: for(j=1; j <=ncovmodel; j++){
2245: printf("%12.7f ",p[jk]);
2246: fprintf(ficlog,"%12.7f ",p[jk]);
2247: jk++;
2248: }
2249: printf("\n");
2250: fprintf(ficlog,"\n");
2251: }
2252: }
2253: }
1.241 brouard 2254: if(*iter <=3 && *iter >1){
1.157 brouard 2255: tml = *localtime(&rcurr_time);
2256: strcpy(strcurr,asctime(&tml));
2257: rforecast_time=rcurr_time;
1.126 brouard 2258: itmp = strlen(strcurr);
2259: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2260: strcurr[itmp-1]='\0';
1.162 brouard 2261: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2262: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2263: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2264: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2265: forecast_time = *localtime(&rforecast_time);
2266: strcpy(strfor,asctime(&forecast_time));
2267: itmp = strlen(strfor);
2268: if(strfor[itmp-1]=='\n')
2269: strfor[itmp-1]='\0';
2270: 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);
2271: 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 2272: }
2273: }
1.187 brouard 2274: for (i=1;i<=n;i++) { /* For each direction i */
2275: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2276: fptt=(*fret);
2277: #ifdef DEBUG
1.203 brouard 2278: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2279: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2280: #endif
1.203 brouard 2281: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2282: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2283: #ifdef LINMINORIGINAL
1.188 brouard 2284: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2285: #else
2286: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2287: flatdir[i]=flat; /* Function is vanishing in that direction i */
2288: #endif
2289: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2290: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2291: /* because that direction will be replaced unless the gain del is small */
2292: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2293: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2294: /* with the new direction. */
2295: del=fabs(fptt-(*fret));
2296: ibig=i;
1.126 brouard 2297: }
2298: #ifdef DEBUG
2299: printf("%d %.12e",i,(*fret));
2300: fprintf(ficlog,"%d %.12e",i,(*fret));
2301: for (j=1;j<=n;j++) {
1.224 brouard 2302: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2303: printf(" x(%d)=%.12e",j,xit[j]);
2304: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2305: }
2306: for(j=1;j<=n;j++) {
1.225 brouard 2307: printf(" p(%d)=%.12e",j,p[j]);
2308: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2309: }
2310: printf("\n");
2311: fprintf(ficlog,"\n");
2312: #endif
1.187 brouard 2313: } /* end loop on each direction i */
2314: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2315: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2316: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2317: for(j=1;j<=n;j++) {
1.225 brouard 2318: if(flatdir[j] >0){
2319: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2320: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2321: }
2322: /* printf("\n"); */
2323: /* fprintf(ficlog,"\n"); */
2324: }
1.243 brouard 2325: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2326: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2327: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2328: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2329: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2330: /* decreased of more than 3.84 */
2331: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2332: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2333: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2334:
1.188 brouard 2335: /* Starting the program with initial values given by a former maximization will simply change */
2336: /* the scales of the directions and the directions, because the are reset to canonical directions */
2337: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2338: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2339: #ifdef DEBUG
2340: int k[2],l;
2341: k[0]=1;
2342: k[1]=-1;
2343: printf("Max: %.12e",(*func)(p));
2344: fprintf(ficlog,"Max: %.12e",(*func)(p));
2345: for (j=1;j<=n;j++) {
2346: printf(" %.12e",p[j]);
2347: fprintf(ficlog," %.12e",p[j]);
2348: }
2349: printf("\n");
2350: fprintf(ficlog,"\n");
2351: for(l=0;l<=1;l++) {
2352: for (j=1;j<=n;j++) {
2353: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2354: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2355: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2356: }
2357: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2358: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2359: }
2360: #endif
2361:
1.224 brouard 2362: #ifdef LINMINORIGINAL
2363: #else
2364: free_ivector(flatdir,1,n);
2365: #endif
1.126 brouard 2366: free_vector(xit,1,n);
2367: free_vector(xits,1,n);
2368: free_vector(ptt,1,n);
2369: free_vector(pt,1,n);
2370: return;
1.192 brouard 2371: } /* enough precision */
1.240 brouard 2372: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2373: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2374: ptt[j]=2.0*p[j]-pt[j];
2375: xit[j]=p[j]-pt[j];
2376: pt[j]=p[j];
2377: }
1.181 brouard 2378: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2379: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2380: if (*iter <=4) {
1.225 brouard 2381: #else
2382: #endif
1.224 brouard 2383: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2384: #else
1.161 brouard 2385: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2386: #endif
1.162 brouard 2387: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2388: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2389: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2390: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2391: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2392: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2393: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2394: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2395: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2396: /* Even if f3 <f1, directest can be negative and t >0 */
2397: /* mu² and del² are equal when f3=f1 */
2398: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2399: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2400: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2401: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2402: #ifdef NRCORIGINAL
2403: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2404: #else
2405: 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 2406: t= t- del*SQR(fp-fptt);
1.183 brouard 2407: #endif
1.202 brouard 2408: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2409: #ifdef DEBUG
1.181 brouard 2410: 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);
2411: 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 2412: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2413: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2414: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2415: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2416: 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);
2417: 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);
2418: #endif
1.183 brouard 2419: #ifdef POWELLORIGINAL
2420: if (t < 0.0) { /* Then we use it for new direction */
2421: #else
1.182 brouard 2422: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2423: 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 2424: 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 2425: 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 2426: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2427: }
1.181 brouard 2428: if (directest < 0.0) { /* Then we use it for new direction */
2429: #endif
1.191 brouard 2430: #ifdef DEBUGLINMIN
1.234 brouard 2431: printf("Before linmin in direction P%d-P0\n",n);
2432: for (j=1;j<=n;j++) {
2433: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2434: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2435: if(j % ncovmodel == 0){
2436: printf("\n");
2437: fprintf(ficlog,"\n");
2438: }
2439: }
1.224 brouard 2440: #endif
2441: #ifdef LINMINORIGINAL
1.234 brouard 2442: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2443: #else
1.234 brouard 2444: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2445: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2446: #endif
1.234 brouard 2447:
1.191 brouard 2448: #ifdef DEBUGLINMIN
1.234 brouard 2449: for (j=1;j<=n;j++) {
2450: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2451: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2452: if(j % ncovmodel == 0){
2453: printf("\n");
2454: fprintf(ficlog,"\n");
2455: }
2456: }
1.224 brouard 2457: #endif
1.234 brouard 2458: for (j=1;j<=n;j++) {
2459: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2460: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2461: }
1.224 brouard 2462: #ifdef LINMINORIGINAL
2463: #else
1.234 brouard 2464: for (j=1, flatd=0;j<=n;j++) {
2465: if(flatdir[j]>0)
2466: flatd++;
2467: }
2468: if(flatd >0){
1.255 brouard 2469: printf("%d flat directions: ",flatd);
2470: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2471: for (j=1;j<=n;j++) {
2472: if(flatdir[j]>0){
2473: printf("%d ",j);
2474: fprintf(ficlog,"%d ",j);
2475: }
2476: }
2477: printf("\n");
2478: fprintf(ficlog,"\n");
2479: }
1.191 brouard 2480: #endif
1.234 brouard 2481: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2482: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2483:
1.126 brouard 2484: #ifdef DEBUG
1.234 brouard 2485: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2486: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2487: for(j=1;j<=n;j++){
2488: printf(" %lf",xit[j]);
2489: fprintf(ficlog," %lf",xit[j]);
2490: }
2491: printf("\n");
2492: fprintf(ficlog,"\n");
1.126 brouard 2493: #endif
1.192 brouard 2494: } /* end of t or directest negative */
1.224 brouard 2495: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2496: #else
1.234 brouard 2497: } /* end if (fptt < fp) */
1.192 brouard 2498: #endif
1.225 brouard 2499: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2500: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2501: #else
1.224 brouard 2502: #endif
1.234 brouard 2503: } /* loop iteration */
1.126 brouard 2504: }
1.234 brouard 2505:
1.126 brouard 2506: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2507:
1.235 brouard 2508: 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 2509: {
1.235 brouard 2510: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2511: (and selected quantitative values in nres)
2512: by left multiplying the unit
1.234 brouard 2513: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2514: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2515: /* Wx is row vector: population in state 1, population in state 2, population dead */
2516: /* or prevalence in state 1, prevalence in state 2, 0 */
2517: /* newm is the matrix after multiplications, its rows are identical at a factor */
2518: /* Initial matrix pimij */
2519: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2520: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2521: /* 0, 0 , 1} */
2522: /*
2523: * and after some iteration: */
2524: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2525: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2526: /* 0, 0 , 1} */
2527: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2528: /* {0.51571254859325999, 0.4842874514067399, */
2529: /* 0.51326036147820708, 0.48673963852179264} */
2530: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2531:
1.126 brouard 2532: int i, ii,j,k;
1.209 brouard 2533: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2534: /* double **matprod2(); */ /* test */
1.218 brouard 2535: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2536: double **newm;
1.209 brouard 2537: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2538: int ncvloop=0;
1.169 brouard 2539:
1.209 brouard 2540: min=vector(1,nlstate);
2541: max=vector(1,nlstate);
2542: meandiff=vector(1,nlstate);
2543:
1.218 brouard 2544: /* Starting with matrix unity */
1.126 brouard 2545: for (ii=1;ii<=nlstate+ndeath;ii++)
2546: for (j=1;j<=nlstate+ndeath;j++){
2547: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2548: }
1.169 brouard 2549:
2550: cov[1]=1.;
2551:
2552: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2553: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2554: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2555: ncvloop++;
1.126 brouard 2556: newm=savm;
2557: /* Covariates have to be included here again */
1.138 brouard 2558: cov[2]=agefin;
1.187 brouard 2559: if(nagesqr==1)
2560: cov[3]= agefin*agefin;;
1.234 brouard 2561: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2562: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2563: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2564: /* 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 2565: }
2566: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2567: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2568: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2569: /* 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 2570: }
1.237 brouard 2571: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2572: if(Dummy[Tvar[Tage[k]]]){
2573: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2574: } else{
1.235 brouard 2575: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2576: }
1.235 brouard 2577: /* 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 2578: }
1.237 brouard 2579: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2580: /* 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 2581: if(Dummy[Tvard[k][1]==0]){
2582: if(Dummy[Tvard[k][2]==0]){
2583: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2584: }else{
2585: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2586: }
2587: }else{
2588: if(Dummy[Tvard[k][2]==0]){
2589: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2590: }else{
2591: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2592: }
2593: }
1.234 brouard 2594: }
1.138 brouard 2595: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2596: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2597: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2598: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2599: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2600: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2601: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2602:
1.126 brouard 2603: savm=oldm;
2604: oldm=newm;
1.209 brouard 2605:
2606: for(j=1; j<=nlstate; j++){
2607: max[j]=0.;
2608: min[j]=1.;
2609: }
2610: for(i=1;i<=nlstate;i++){
2611: sumnew=0;
2612: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2613: for(j=1; j<=nlstate; j++){
2614: prlim[i][j]= newm[i][j]/(1-sumnew);
2615: max[j]=FMAX(max[j],prlim[i][j]);
2616: min[j]=FMIN(min[j],prlim[i][j]);
2617: }
2618: }
2619:
1.126 brouard 2620: maxmax=0.;
1.209 brouard 2621: for(j=1; j<=nlstate; j++){
2622: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2623: maxmax=FMAX(maxmax,meandiff[j]);
2624: /* 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 2625: } /* j loop */
1.203 brouard 2626: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2627: /* 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 2628: if(maxmax < ftolpl){
1.209 brouard 2629: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2630: free_vector(min,1,nlstate);
2631: free_vector(max,1,nlstate);
2632: free_vector(meandiff,1,nlstate);
1.126 brouard 2633: return prlim;
2634: }
1.169 brouard 2635: } /* age loop */
1.208 brouard 2636: /* After some age loop it doesn't converge */
1.209 brouard 2637: 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 2638: 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 2639: /* 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); */
2640: free_vector(min,1,nlstate);
2641: free_vector(max,1,nlstate);
2642: free_vector(meandiff,1,nlstate);
1.208 brouard 2643:
1.169 brouard 2644: return prlim; /* should not reach here */
1.126 brouard 2645: }
2646:
1.217 brouard 2647:
2648: /**** Back Prevalence limit (stable or period prevalence) ****************/
2649:
1.218 brouard 2650: /* 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) */
2651: /* 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 2652: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2653: {
1.264 brouard 2654: /* 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 2655: matrix by transitions matrix until convergence is reached with precision ftolpl */
2656: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2657: /* Wx is row vector: population in state 1, population in state 2, population dead */
2658: /* or prevalence in state 1, prevalence in state 2, 0 */
2659: /* newm is the matrix after multiplications, its rows are identical at a factor */
2660: /* Initial matrix pimij */
2661: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2662: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2663: /* 0, 0 , 1} */
2664: /*
2665: * and after some iteration: */
2666: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2667: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2668: /* 0, 0 , 1} */
2669: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2670: /* {0.51571254859325999, 0.4842874514067399, */
2671: /* 0.51326036147820708, 0.48673963852179264} */
2672: /* If we start from prlim again, prlim tends to a constant matrix */
2673:
2674: int i, ii,j,k;
1.247 brouard 2675: int first=0;
1.217 brouard 2676: double *min, *max, *meandiff, maxmax,sumnew=0.;
2677: /* double **matprod2(); */ /* test */
2678: double **out, cov[NCOVMAX+1], **bmij();
2679: double **newm;
1.218 brouard 2680: double **dnewm, **doldm, **dsavm; /* for use */
2681: double **oldm, **savm; /* for use */
2682:
1.217 brouard 2683: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2684: int ncvloop=0;
2685:
2686: min=vector(1,nlstate);
2687: max=vector(1,nlstate);
2688: meandiff=vector(1,nlstate);
2689:
1.266 brouard 2690: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2691: oldm=oldms; savm=savms;
2692:
2693: /* Starting with matrix unity */
2694: for (ii=1;ii<=nlstate+ndeath;ii++)
2695: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2696: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2697: }
2698:
2699: cov[1]=1.;
2700:
2701: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2702: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2703: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2704: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2705: ncvloop++;
1.218 brouard 2706: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2707: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2708: /* Covariates have to be included here again */
2709: cov[2]=agefin;
2710: if(nagesqr==1)
2711: cov[3]= agefin*agefin;;
1.242 brouard 2712: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2713: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2714: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2715: /* 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 2716: }
2717: /* for (k=1; k<=cptcovn;k++) { */
2718: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2719: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2720: /* /\* 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])]); *\/ */
2721: /* } */
2722: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2723: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2724: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2725: /* 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]); */
2726: }
2727: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2728: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2729: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2730: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2731: for (k=1; k<=cptcovage;k++){ /* For product with age */
2732: if(Dummy[Tvar[Tage[k]]]){
2733: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2734: } else{
2735: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2736: }
2737: /* 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]); */
2738: }
2739: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2740: /* 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]); */
2741: if(Dummy[Tvard[k][1]==0]){
2742: if(Dummy[Tvard[k][2]==0]){
2743: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2744: }else{
2745: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2746: }
2747: }else{
2748: if(Dummy[Tvard[k][2]==0]){
2749: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2750: }else{
2751: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2752: }
2753: }
1.217 brouard 2754: }
2755:
2756: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2757: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2758: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2759: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2760: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2761: /* ij should be linked to the correct index of cov */
2762: /* age and covariate values ij are in 'cov', but we need to pass
2763: * ij for the observed prevalence at age and status and covariate
2764: * number: prevacurrent[(int)agefin][ii][ij]
2765: */
2766: /* 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 *\/ */
2767: /* 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 *\/ */
2768: 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 2769: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2770: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2771: /* for(i=1; i<=nlstate+ndeath; i++) { */
2772: /* printf("%d newm= ",i); */
2773: /* for(j=1;j<=nlstate+ndeath;j++) { */
2774: /* printf("%f ",newm[i][j]); */
2775: /* } */
2776: /* printf("oldm * "); */
2777: /* for(j=1;j<=nlstate+ndeath;j++) { */
2778: /* printf("%f ",oldm[i][j]); */
2779: /* } */
1.268 brouard 2780: /* printf(" bmmij "); */
1.266 brouard 2781: /* for(j=1;j<=nlstate+ndeath;j++) { */
2782: /* printf("%f ",pmmij[i][j]); */
2783: /* } */
2784: /* printf("\n"); */
2785: /* } */
2786: /* } */
1.217 brouard 2787: savm=oldm;
2788: oldm=newm;
1.266 brouard 2789:
1.217 brouard 2790: for(j=1; j<=nlstate; j++){
2791: max[j]=0.;
2792: min[j]=1.;
2793: }
2794: for(j=1; j<=nlstate; j++){
2795: for(i=1;i<=nlstate;i++){
1.234 brouard 2796: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2797: bprlim[i][j]= newm[i][j];
2798: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2799: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2800: }
2801: }
1.218 brouard 2802:
1.217 brouard 2803: maxmax=0.;
2804: for(i=1; i<=nlstate; i++){
2805: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2806: maxmax=FMAX(maxmax,meandiff[i]);
2807: /* 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 2808: } /* i loop */
1.217 brouard 2809: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2810: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2811: if(maxmax < ftolpl){
1.220 brouard 2812: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2813: free_vector(min,1,nlstate);
2814: free_vector(max,1,nlstate);
2815: free_vector(meandiff,1,nlstate);
2816: return bprlim;
2817: }
2818: } /* age loop */
2819: /* After some age loop it doesn't converge */
1.247 brouard 2820: if(first){
2821: first=1;
2822: 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\
2823: Oldest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
2824: }
2825: 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 2826: 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);
2827: /* 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); */
2828: free_vector(min,1,nlstate);
2829: free_vector(max,1,nlstate);
2830: free_vector(meandiff,1,nlstate);
2831:
2832: return bprlim; /* should not reach here */
2833: }
2834:
1.126 brouard 2835: /*************** transition probabilities ***************/
2836:
2837: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2838: {
1.138 brouard 2839: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2840: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2841: model to the ncovmodel covariates (including constant and age).
2842: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2843: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2844: ncth covariate in the global vector x is given by the formula:
2845: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2846: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2847: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2848: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2849: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2850: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2851: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2852: */
2853: double s1, lnpijopii;
1.126 brouard 2854: /*double t34;*/
1.164 brouard 2855: int i,j, nc, ii, jj;
1.126 brouard 2856:
1.223 brouard 2857: for(i=1; i<= nlstate; i++){
2858: for(j=1; j<i;j++){
2859: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2860: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2861: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2862: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2863: }
2864: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2865: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2866: }
2867: for(j=i+1; j<=nlstate+ndeath;j++){
2868: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2869: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2870: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2871: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2872: }
2873: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2874: }
2875: }
1.218 brouard 2876:
1.223 brouard 2877: for(i=1; i<= nlstate; i++){
2878: s1=0;
2879: for(j=1; j<i; j++){
2880: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2881: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2882: }
2883: for(j=i+1; j<=nlstate+ndeath; j++){
2884: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2885: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2886: }
2887: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2888: ps[i][i]=1./(s1+1.);
2889: /* Computing other pijs */
2890: for(j=1; j<i; j++)
2891: ps[i][j]= exp(ps[i][j])*ps[i][i];
2892: for(j=i+1; j<=nlstate+ndeath; j++)
2893: ps[i][j]= exp(ps[i][j])*ps[i][i];
2894: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2895: } /* end i */
1.218 brouard 2896:
1.223 brouard 2897: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2898: for(jj=1; jj<= nlstate+ndeath; jj++){
2899: ps[ii][jj]=0;
2900: ps[ii][ii]=1;
2901: }
2902: }
1.218 brouard 2903:
2904:
1.223 brouard 2905: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2906: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2907: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2908: /* } */
2909: /* printf("\n "); */
2910: /* } */
2911: /* printf("\n ");printf("%lf ",cov[2]);*/
2912: /*
2913: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2914: goto end;*/
1.266 brouard 2915: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2916: }
2917:
1.218 brouard 2918: /*************** backward transition probabilities ***************/
2919:
2920: /* 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 ) */
2921: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2922: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2923: {
1.266 brouard 2924: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2925: * 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 2926: */
1.218 brouard 2927: int i, ii, j,k;
1.222 brouard 2928:
2929: double **out, **pmij();
2930: double sumnew=0.;
1.218 brouard 2931: double agefin;
1.268 brouard 2932: 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 2933: double **dnewm, **dsavm, **doldm;
2934: double **bbmij;
2935:
1.218 brouard 2936: doldm=ddoldms; /* global pointers */
1.222 brouard 2937: dnewm=ddnewms;
2938: dsavm=ddsavms;
2939:
2940: agefin=cov[2];
1.268 brouard 2941: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2942: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2943: the observed prevalence (with this covariate ij) at beginning of transition */
2944: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2945:
2946: /* P_x */
1.266 brouard 2947: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2948: /* outputs pmmij which is a stochastic matrix in row */
2949:
2950: /* Diag(w_x) */
2951: /* Problem with prevacurrent which can be zero */
2952: sumnew=0.;
1.269 brouard 2953: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2954: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2955: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2956: sumnew+=prevacurrent[(int)agefin][ii][ij];
2957: }
2958: if(sumnew >0.01){ /* At least some value in the prevalence */
2959: for (ii=1;ii<=nlstate+ndeath;ii++){
2960: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2961: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2962: }
2963: }else{
2964: for (ii=1;ii<=nlstate+ndeath;ii++){
2965: for (j=1;j<=nlstate+ndeath;j++)
2966: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
2967: }
2968: /* if(sumnew <0.9){ */
2969: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2970: /* } */
2971: }
2972: k3=0.0; /* We put the last diagonal to 0 */
2973: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
2974: doldm[ii][ii]= k3;
2975: }
2976: /* End doldm, At the end doldm is diag[(w_i)] */
2977:
2978: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
2979: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
2980:
2981: /* Diag(Sum_i w^i_x p^ij_x */
2982: /* 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 2983: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 2984: sumnew=0.;
1.222 brouard 2985: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 2986: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 2987: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 2988: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 2989: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 2990: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 2991: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2992: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 2993: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2994: /* }else */
1.268 brouard 2995: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2996: } /*End ii */
2997: } /* 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 */
2998:
2999: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
3000: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3001: /* end bmij */
1.266 brouard 3002: return ps; /*pointer is unchanged */
1.218 brouard 3003: }
1.217 brouard 3004: /*************** transition probabilities ***************/
3005:
1.218 brouard 3006: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3007: {
3008: /* According to parameters values stored in x and the covariate's values stored in cov,
3009: computes the probability to be observed in state j being in state i by appying the
3010: model to the ncovmodel covariates (including constant and age).
3011: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3012: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3013: ncth covariate in the global vector x is given by the formula:
3014: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3015: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3016: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3017: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3018: Outputs ps[i][j] the probability to be observed in j being in j according to
3019: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3020: */
3021: double s1, lnpijopii;
3022: /*double t34;*/
3023: int i,j, nc, ii, jj;
3024:
1.234 brouard 3025: for(i=1; i<= nlstate; i++){
3026: for(j=1; j<i;j++){
3027: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3028: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3029: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3030: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3031: }
3032: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3033: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3034: }
3035: for(j=i+1; j<=nlstate+ndeath;j++){
3036: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3037: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3038: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3039: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3040: }
3041: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3042: }
3043: }
3044:
3045: for(i=1; i<= nlstate; i++){
3046: s1=0;
3047: for(j=1; j<i; j++){
3048: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3049: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3050: }
3051: for(j=i+1; j<=nlstate+ndeath; j++){
3052: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3053: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3054: }
3055: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3056: ps[i][i]=1./(s1+1.);
3057: /* Computing other pijs */
3058: for(j=1; j<i; j++)
3059: ps[i][j]= exp(ps[i][j])*ps[i][i];
3060: for(j=i+1; j<=nlstate+ndeath; j++)
3061: ps[i][j]= exp(ps[i][j])*ps[i][i];
3062: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3063: } /* end i */
3064:
3065: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3066: for(jj=1; jj<= nlstate+ndeath; jj++){
3067: ps[ii][jj]=0;
3068: ps[ii][ii]=1;
3069: }
3070: }
3071: /* Added for backcast */ /* Transposed matrix too */
3072: for(jj=1; jj<= nlstate+ndeath; jj++){
3073: s1=0.;
3074: for(ii=1; ii<= nlstate+ndeath; ii++){
3075: s1+=ps[ii][jj];
3076: }
3077: for(ii=1; ii<= nlstate; ii++){
3078: ps[ii][jj]=ps[ii][jj]/s1;
3079: }
3080: }
3081: /* Transposition */
3082: for(jj=1; jj<= nlstate+ndeath; jj++){
3083: for(ii=jj; ii<= nlstate+ndeath; ii++){
3084: s1=ps[ii][jj];
3085: ps[ii][jj]=ps[jj][ii];
3086: ps[jj][ii]=s1;
3087: }
3088: }
3089: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3090: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3091: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3092: /* } */
3093: /* printf("\n "); */
3094: /* } */
3095: /* printf("\n ");printf("%lf ",cov[2]);*/
3096: /*
3097: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3098: goto end;*/
3099: return ps;
1.217 brouard 3100: }
3101:
3102:
1.126 brouard 3103: /**************** Product of 2 matrices ******************/
3104:
1.145 brouard 3105: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3106: {
3107: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3108: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3109: /* in, b, out are matrice of pointers which should have been initialized
3110: before: only the contents of out is modified. The function returns
3111: a pointer to pointers identical to out */
1.145 brouard 3112: int i, j, k;
1.126 brouard 3113: for(i=nrl; i<= nrh; i++)
1.145 brouard 3114: for(k=ncolol; k<=ncoloh; k++){
3115: out[i][k]=0.;
3116: for(j=ncl; j<=nch; j++)
3117: out[i][k] +=in[i][j]*b[j][k];
3118: }
1.126 brouard 3119: return out;
3120: }
3121:
3122:
3123: /************* Higher Matrix Product ***************/
3124:
1.235 brouard 3125: 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 3126: {
1.218 brouard 3127: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3128: 'nhstepm*hstepm*stepm' months (i.e. until
3129: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3130: nhstepm*hstepm matrices.
3131: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3132: (typically every 2 years instead of every month which is too big
3133: for the memory).
3134: Model is determined by parameters x and covariates have to be
3135: included manually here.
3136:
3137: */
3138:
3139: int i, j, d, h, k;
1.131 brouard 3140: double **out, cov[NCOVMAX+1];
1.126 brouard 3141: double **newm;
1.187 brouard 3142: double agexact;
1.214 brouard 3143: double agebegin, ageend;
1.126 brouard 3144:
3145: /* Hstepm could be zero and should return the unit matrix */
3146: for (i=1;i<=nlstate+ndeath;i++)
3147: for (j=1;j<=nlstate+ndeath;j++){
3148: oldm[i][j]=(i==j ? 1.0 : 0.0);
3149: po[i][j][0]=(i==j ? 1.0 : 0.0);
3150: }
3151: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3152: for(h=1; h <=nhstepm; h++){
3153: for(d=1; d <=hstepm; d++){
3154: newm=savm;
3155: /* Covariates have to be included here again */
3156: cov[1]=1.;
1.214 brouard 3157: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3158: cov[2]=agexact;
3159: if(nagesqr==1)
1.227 brouard 3160: cov[3]= agexact*agexact;
1.235 brouard 3161: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3162: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3163: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3164: /* 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)); */
3165: }
3166: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3167: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3168: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3169: /* 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]); */
3170: }
3171: for (k=1; k<=cptcovage;k++){
3172: if(Dummy[Tvar[Tage[k]]]){
3173: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3174: } else{
3175: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3176: }
3177: /* 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]); */
3178: }
3179: for (k=1; k<=cptcovprod;k++){ /* */
3180: /* 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]); */
3181: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3182: }
3183: /* for (k=1; k<=cptcovn;k++) */
3184: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3185: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3186: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3187: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3188: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3189:
3190:
1.126 brouard 3191: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3192: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3193: /* right multiplication of oldm by the current matrix */
1.126 brouard 3194: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3195: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3196: /* if((int)age == 70){ */
3197: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3198: /* for(i=1; i<=nlstate+ndeath; i++) { */
3199: /* printf("%d pmmij ",i); */
3200: /* for(j=1;j<=nlstate+ndeath;j++) { */
3201: /* printf("%f ",pmmij[i][j]); */
3202: /* } */
3203: /* printf(" oldm "); */
3204: /* for(j=1;j<=nlstate+ndeath;j++) { */
3205: /* printf("%f ",oldm[i][j]); */
3206: /* } */
3207: /* printf("\n"); */
3208: /* } */
3209: /* } */
1.126 brouard 3210: savm=oldm;
3211: oldm=newm;
3212: }
3213: for(i=1; i<=nlstate+ndeath; i++)
3214: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3215: po[i][j][h]=newm[i][j];
3216: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3217: }
1.128 brouard 3218: /*printf("h=%d ",h);*/
1.126 brouard 3219: } /* end h */
1.267 brouard 3220: /* printf("\n H=%d \n",h); */
1.126 brouard 3221: return po;
3222: }
3223:
1.217 brouard 3224: /************* Higher Back Matrix Product ***************/
1.218 brouard 3225: /* 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 3226: 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 3227: {
1.266 brouard 3228: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3229: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3230: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3231: nhstepm*hstepm matrices.
3232: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3233: (typically every 2 years instead of every month which is too big
1.217 brouard 3234: for the memory).
1.218 brouard 3235: Model is determined by parameters x and covariates have to be
1.266 brouard 3236: included manually here. Then we use a call to bmij(x and cov)
3237: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3238: */
1.217 brouard 3239:
3240: int i, j, d, h, k;
1.266 brouard 3241: double **out, cov[NCOVMAX+1], **bmij();
3242: double **newm, ***newmm;
1.217 brouard 3243: double agexact;
3244: double agebegin, ageend;
1.222 brouard 3245: double **oldm, **savm;
1.217 brouard 3246:
1.266 brouard 3247: newmm=po; /* To be saved */
3248: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3249: /* Hstepm could be zero and should return the unit matrix */
3250: for (i=1;i<=nlstate+ndeath;i++)
3251: for (j=1;j<=nlstate+ndeath;j++){
3252: oldm[i][j]=(i==j ? 1.0 : 0.0);
3253: po[i][j][0]=(i==j ? 1.0 : 0.0);
3254: }
3255: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3256: for(h=1; h <=nhstepm; h++){
3257: for(d=1; d <=hstepm; d++){
3258: newm=savm;
3259: /* Covariates have to be included here again */
3260: cov[1]=1.;
1.271 brouard 3261: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3262: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3263: cov[2]=agexact;
3264: if(nagesqr==1)
1.222 brouard 3265: cov[3]= agexact*agexact;
1.266 brouard 3266: for (k=1; k<=cptcovn;k++){
3267: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3268: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3269: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3270: /* 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)); */
3271: }
1.267 brouard 3272: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3273: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3274: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3275: /* 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]); */
3276: }
3277: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3278: if(Dummy[Tvar[Tage[k]]]){
3279: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3280: } else{
3281: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3282: }
3283: /* 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]); */
3284: }
3285: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3286: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3287: }
1.217 brouard 3288: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3289: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3290:
1.218 brouard 3291: /* Careful transposed matrix */
1.266 brouard 3292: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3293: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3294: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3295: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3296: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3297: /* if((int)age == 70){ */
3298: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3299: /* for(i=1; i<=nlstate+ndeath; i++) { */
3300: /* printf("%d pmmij ",i); */
3301: /* for(j=1;j<=nlstate+ndeath;j++) { */
3302: /* printf("%f ",pmmij[i][j]); */
3303: /* } */
3304: /* printf(" oldm "); */
3305: /* for(j=1;j<=nlstate+ndeath;j++) { */
3306: /* printf("%f ",oldm[i][j]); */
3307: /* } */
3308: /* printf("\n"); */
3309: /* } */
3310: /* } */
3311: savm=oldm;
3312: oldm=newm;
3313: }
3314: for(i=1; i<=nlstate+ndeath; i++)
3315: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3316: po[i][j][h]=newm[i][j];
1.268 brouard 3317: /* if(h==nhstepm) */
3318: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3319: }
1.268 brouard 3320: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3321: } /* end h */
1.268 brouard 3322: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3323: return po;
3324: }
3325:
3326:
1.162 brouard 3327: #ifdef NLOPT
3328: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3329: double fret;
3330: double *xt;
3331: int j;
3332: myfunc_data *d2 = (myfunc_data *) pd;
3333: /* xt = (p1-1); */
3334: xt=vector(1,n);
3335: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3336:
3337: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3338: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3339: printf("Function = %.12lf ",fret);
3340: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3341: printf("\n");
3342: free_vector(xt,1,n);
3343: return fret;
3344: }
3345: #endif
1.126 brouard 3346:
3347: /*************** log-likelihood *************/
3348: double func( double *x)
3349: {
1.226 brouard 3350: int i, ii, j, k, mi, d, kk;
3351: int ioffset=0;
3352: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3353: double **out;
3354: double lli; /* Individual log likelihood */
3355: int s1, s2;
1.228 brouard 3356: 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 3357: double bbh, survp;
3358: long ipmx;
3359: double agexact;
3360: /*extern weight */
3361: /* We are differentiating ll according to initial status */
3362: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3363: /*for(i=1;i<imx;i++)
3364: printf(" %d\n",s[4][i]);
3365: */
1.162 brouard 3366:
1.226 brouard 3367: ++countcallfunc;
1.162 brouard 3368:
1.226 brouard 3369: cov[1]=1.;
1.126 brouard 3370:
1.226 brouard 3371: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3372: ioffset=0;
1.226 brouard 3373: if(mle==1){
3374: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3375: /* Computes the values of the ncovmodel covariates of the model
3376: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3377: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3378: to be observed in j being in i according to the model.
3379: */
1.243 brouard 3380: ioffset=2+nagesqr ;
1.233 brouard 3381: /* Fixed */
1.234 brouard 3382: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3383: 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)*/
3384: }
1.226 brouard 3385: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3386: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3387: has been calculated etc */
3388: /* For an individual i, wav[i] gives the number of effective waves */
3389: /* We compute the contribution to Likelihood of each effective transition
3390: mw[mi][i] is real wave of the mi th effectve wave */
3391: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3392: s2=s[mw[mi+1][i]][i];
3393: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3394: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3395: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3396: */
3397: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3398: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3399: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3400: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3401: }
3402: for (ii=1;ii<=nlstate+ndeath;ii++)
3403: for (j=1;j<=nlstate+ndeath;j++){
3404: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3405: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3406: }
3407: for(d=0; d<dh[mi][i]; d++){
3408: newm=savm;
3409: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3410: cov[2]=agexact;
3411: if(nagesqr==1)
3412: cov[3]= agexact*agexact; /* Should be changed here */
3413: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3414: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3415: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3416: else
3417: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3418: }
3419: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3420: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3421: savm=oldm;
3422: oldm=newm;
3423: } /* end mult */
3424:
3425: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3426: /* But now since version 0.9 we anticipate for bias at large stepm.
3427: * If stepm is larger than one month (smallest stepm) and if the exact delay
3428: * (in months) between two waves is not a multiple of stepm, we rounded to
3429: * the nearest (and in case of equal distance, to the lowest) interval but now
3430: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3431: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3432: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3433: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3434: * -stepm/2 to stepm/2 .
3435: * For stepm=1 the results are the same as for previous versions of Imach.
3436: * For stepm > 1 the results are less biased than in previous versions.
3437: */
1.234 brouard 3438: s1=s[mw[mi][i]][i];
3439: s2=s[mw[mi+1][i]][i];
3440: bbh=(double)bh[mi][i]/(double)stepm;
3441: /* bias bh is positive if real duration
3442: * is higher than the multiple of stepm and negative otherwise.
3443: */
3444: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3445: if( s2 > nlstate){
3446: /* i.e. if s2 is a death state and if the date of death is known
3447: then the contribution to the likelihood is the probability to
3448: die between last step unit time and current step unit time,
3449: which is also equal to probability to die before dh
3450: minus probability to die before dh-stepm .
3451: In version up to 0.92 likelihood was computed
3452: as if date of death was unknown. Death was treated as any other
3453: health state: the date of the interview describes the actual state
3454: and not the date of a change in health state. The former idea was
3455: to consider that at each interview the state was recorded
3456: (healthy, disable or death) and IMaCh was corrected; but when we
3457: introduced the exact date of death then we should have modified
3458: the contribution of an exact death to the likelihood. This new
3459: contribution is smaller and very dependent of the step unit
3460: stepm. It is no more the probability to die between last interview
3461: and month of death but the probability to survive from last
3462: interview up to one month before death multiplied by the
3463: probability to die within a month. Thanks to Chris
3464: Jackson for correcting this bug. Former versions increased
3465: mortality artificially. The bad side is that we add another loop
3466: which slows down the processing. The difference can be up to 10%
3467: lower mortality.
3468: */
3469: /* If, at the beginning of the maximization mostly, the
3470: cumulative probability or probability to be dead is
3471: constant (ie = 1) over time d, the difference is equal to
3472: 0. out[s1][3] = savm[s1][3]: probability, being at state
3473: s1 at precedent wave, to be dead a month before current
3474: wave is equal to probability, being at state s1 at
3475: precedent wave, to be dead at mont of the current
3476: wave. Then the observed probability (that this person died)
3477: is null according to current estimated parameter. In fact,
3478: it should be very low but not zero otherwise the log go to
3479: infinity.
3480: */
1.183 brouard 3481: /* #ifdef INFINITYORIGINAL */
3482: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3483: /* #else */
3484: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3485: /* lli=log(mytinydouble); */
3486: /* else */
3487: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3488: /* #endif */
1.226 brouard 3489: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3490:
1.226 brouard 3491: } else if ( s2==-1 ) { /* alive */
3492: for (j=1,survp=0. ; j<=nlstate; j++)
3493: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3494: /*survp += out[s1][j]; */
3495: lli= log(survp);
3496: }
3497: else if (s2==-4) {
3498: for (j=3,survp=0. ; j<=nlstate; j++)
3499: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3500: lli= log(survp);
3501: }
3502: else if (s2==-5) {
3503: for (j=1,survp=0. ; j<=2; j++)
3504: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3505: lli= log(survp);
3506: }
3507: else{
3508: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3509: /* 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 */
3510: }
3511: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3512: /*if(lli ==000.0)*/
3513: /*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); */
3514: ipmx +=1;
3515: sw += weight[i];
3516: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3517: /* if (lli < log(mytinydouble)){ */
3518: /* 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); */
3519: /* 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]); */
3520: /* } */
3521: } /* end of wave */
3522: } /* end of individual */
3523: } else if(mle==2){
3524: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3525: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3526: for(mi=1; mi<= wav[i]-1; mi++){
3527: for (ii=1;ii<=nlstate+ndeath;ii++)
3528: for (j=1;j<=nlstate+ndeath;j++){
3529: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3530: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3531: }
3532: for(d=0; d<=dh[mi][i]; d++){
3533: newm=savm;
3534: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3535: cov[2]=agexact;
3536: if(nagesqr==1)
3537: cov[3]= agexact*agexact;
3538: for (kk=1; kk<=cptcovage;kk++) {
3539: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3540: }
3541: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3542: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3543: savm=oldm;
3544: oldm=newm;
3545: } /* end mult */
3546:
3547: s1=s[mw[mi][i]][i];
3548: s2=s[mw[mi+1][i]][i];
3549: bbh=(double)bh[mi][i]/(double)stepm;
3550: 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 */
3551: ipmx +=1;
3552: sw += weight[i];
3553: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3554: } /* end of wave */
3555: } /* end of individual */
3556: } else if(mle==3){ /* exponential inter-extrapolation */
3557: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3558: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3559: for(mi=1; mi<= wav[i]-1; mi++){
3560: for (ii=1;ii<=nlstate+ndeath;ii++)
3561: for (j=1;j<=nlstate+ndeath;j++){
3562: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3563: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3564: }
3565: for(d=0; d<dh[mi][i]; d++){
3566: newm=savm;
3567: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3568: cov[2]=agexact;
3569: if(nagesqr==1)
3570: cov[3]= agexact*agexact;
3571: for (kk=1; kk<=cptcovage;kk++) {
3572: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3573: }
3574: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3575: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3576: savm=oldm;
3577: oldm=newm;
3578: } /* end mult */
3579:
3580: s1=s[mw[mi][i]][i];
3581: s2=s[mw[mi+1][i]][i];
3582: bbh=(double)bh[mi][i]/(double)stepm;
3583: 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 */
3584: ipmx +=1;
3585: sw += weight[i];
3586: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3587: } /* end of wave */
3588: } /* end of individual */
3589: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3590: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3591: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3592: for(mi=1; mi<= wav[i]-1; mi++){
3593: for (ii=1;ii<=nlstate+ndeath;ii++)
3594: for (j=1;j<=nlstate+ndeath;j++){
3595: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3596: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3597: }
3598: for(d=0; d<dh[mi][i]; d++){
3599: newm=savm;
3600: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3601: cov[2]=agexact;
3602: if(nagesqr==1)
3603: cov[3]= agexact*agexact;
3604: for (kk=1; kk<=cptcovage;kk++) {
3605: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3606: }
1.126 brouard 3607:
1.226 brouard 3608: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3609: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3610: savm=oldm;
3611: oldm=newm;
3612: } /* end mult */
3613:
3614: s1=s[mw[mi][i]][i];
3615: s2=s[mw[mi+1][i]][i];
3616: if( s2 > nlstate){
3617: lli=log(out[s1][s2] - savm[s1][s2]);
3618: } else if ( s2==-1 ) { /* alive */
3619: for (j=1,survp=0. ; j<=nlstate; j++)
3620: survp += out[s1][j];
3621: lli= log(survp);
3622: }else{
3623: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3624: }
3625: ipmx +=1;
3626: sw += weight[i];
3627: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3628: /* 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 3629: } /* end of wave */
3630: } /* end of individual */
3631: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3632: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3633: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3634: for(mi=1; mi<= wav[i]-1; mi++){
3635: for (ii=1;ii<=nlstate+ndeath;ii++)
3636: for (j=1;j<=nlstate+ndeath;j++){
3637: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3638: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3639: }
3640: for(d=0; d<dh[mi][i]; d++){
3641: newm=savm;
3642: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3643: cov[2]=agexact;
3644: if(nagesqr==1)
3645: cov[3]= agexact*agexact;
3646: for (kk=1; kk<=cptcovage;kk++) {
3647: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3648: }
1.126 brouard 3649:
1.226 brouard 3650: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3651: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3652: savm=oldm;
3653: oldm=newm;
3654: } /* end mult */
3655:
3656: s1=s[mw[mi][i]][i];
3657: s2=s[mw[mi+1][i]][i];
3658: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3659: ipmx +=1;
3660: sw += weight[i];
3661: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3662: /*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]);*/
3663: } /* end of wave */
3664: } /* end of individual */
3665: } /* End of if */
3666: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3667: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3668: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3669: return -l;
1.126 brouard 3670: }
3671:
3672: /*************** log-likelihood *************/
3673: double funcone( double *x)
3674: {
1.228 brouard 3675: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3676: int i, ii, j, k, mi, d, kk;
1.228 brouard 3677: int ioffset=0;
1.131 brouard 3678: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3679: double **out;
3680: double lli; /* Individual log likelihood */
3681: double llt;
3682: int s1, s2;
1.228 brouard 3683: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3684:
1.126 brouard 3685: double bbh, survp;
1.187 brouard 3686: double agexact;
1.214 brouard 3687: double agebegin, ageend;
1.126 brouard 3688: /*extern weight */
3689: /* We are differentiating ll according to initial status */
3690: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3691: /*for(i=1;i<imx;i++)
3692: printf(" %d\n",s[4][i]);
3693: */
3694: cov[1]=1.;
3695:
3696: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3697: ioffset=0;
3698: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3699: /* ioffset=2+nagesqr+cptcovage; */
3700: ioffset=2+nagesqr;
1.232 brouard 3701: /* Fixed */
1.224 brouard 3702: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3703: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3704: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3705: 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)*/
3706: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3707: /* cov[2+6]=covar[Tvar[6]][i]; */
3708: /* cov[2+6]=covar[2][i]; V2 */
3709: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3710: /* cov[2+7]=covar[Tvar[7]][i]; */
3711: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3712: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3713: /* cov[2+9]=covar[Tvar[9]][i]; */
3714: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3715: }
1.232 brouard 3716: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3717: /* 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?)*\/ */
3718: /* } */
1.231 brouard 3719: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3720: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3721: /* } */
1.225 brouard 3722:
1.233 brouard 3723:
3724: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3725: /* Wave varying (but not age varying) */
3726: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3727: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3728: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3729: }
1.232 brouard 3730: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3731: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3732: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3733: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3734: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3735: /* 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 3736: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3737: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3738: /* /\* 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]); *\/ */
3739: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3740: /* } */
1.126 brouard 3741: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3742: for (j=1;j<=nlstate+ndeath;j++){
3743: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3744: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3745: }
1.214 brouard 3746:
3747: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3748: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3749: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3750: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3751: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3752: and mw[mi+1][i]. dh depends on stepm.*/
3753: newm=savm;
1.247 brouard 3754: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3755: cov[2]=agexact;
3756: if(nagesqr==1)
3757: cov[3]= agexact*agexact;
3758: for (kk=1; kk<=cptcovage;kk++) {
3759: if(!FixedV[Tvar[Tage[kk]]])
3760: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3761: else
3762: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3763: }
3764: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3765: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3766: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3767: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3768: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3769: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3770: savm=oldm;
3771: oldm=newm;
1.126 brouard 3772: } /* end mult */
3773:
3774: s1=s[mw[mi][i]][i];
3775: s2=s[mw[mi+1][i]][i];
1.217 brouard 3776: /* if(s2==-1){ */
1.268 brouard 3777: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3778: /* /\* exit(1); *\/ */
3779: /* } */
1.126 brouard 3780: bbh=(double)bh[mi][i]/(double)stepm;
3781: /* bias is positive if real duration
3782: * is higher than the multiple of stepm and negative otherwise.
3783: */
3784: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3785: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3786: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3787: for (j=1,survp=0. ; j<=nlstate; j++)
3788: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3789: lli= log(survp);
1.126 brouard 3790: }else if (mle==1){
1.242 brouard 3791: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3792: } else if(mle==2){
1.242 brouard 3793: 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 3794: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3795: 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 3796: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3797: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3798: } else{ /* mle=0 back to 1 */
1.242 brouard 3799: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3800: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3801: } /* End of if */
3802: ipmx +=1;
3803: sw += weight[i];
3804: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3805: /*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 3806: if(globpr){
1.246 brouard 3807: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3808: %11.6f %11.6f %11.6f ", \
1.242 brouard 3809: 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 3810: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3811: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3812: llt +=ll[k]*gipmx/gsw;
3813: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3814: }
3815: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3816: }
1.232 brouard 3817: } /* end of wave */
3818: } /* end of individual */
3819: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3820: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3821: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3822: if(globpr==0){ /* First time we count the contributions and weights */
3823: gipmx=ipmx;
3824: gsw=sw;
3825: }
3826: return -l;
1.126 brouard 3827: }
3828:
3829:
3830: /*************** function likelione ***********/
3831: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3832: {
3833: /* This routine should help understanding what is done with
3834: the selection of individuals/waves and
3835: to check the exact contribution to the likelihood.
3836: Plotting could be done.
3837: */
3838: int k;
3839:
3840: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3841: strcpy(fileresilk,"ILK_");
1.202 brouard 3842: strcat(fileresilk,fileresu);
1.126 brouard 3843: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3844: printf("Problem with resultfile: %s\n", fileresilk);
3845: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3846: }
1.214 brouard 3847: 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");
3848: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3849: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3850: for(k=1; k<=nlstate; k++)
3851: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3852: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3853: }
3854:
3855: *fretone=(*funcone)(p);
3856: if(*globpri !=0){
3857: fclose(ficresilk);
1.205 brouard 3858: if (mle ==0)
3859: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3860: else if(mle >=1)
3861: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3862: 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 3863: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3864:
3865: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3866: 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 3867: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3868: }
1.207 brouard 3869: 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 3870: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3871: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3872: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3873: fflush(fichtm);
1.205 brouard 3874: }
1.126 brouard 3875: return;
3876: }
3877:
3878:
3879: /*********** Maximum Likelihood Estimation ***************/
3880:
3881: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3882: {
1.165 brouard 3883: int i,j, iter=0;
1.126 brouard 3884: double **xi;
3885: double fret;
3886: double fretone; /* Only one call to likelihood */
3887: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3888:
3889: #ifdef NLOPT
3890: int creturn;
3891: nlopt_opt opt;
3892: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3893: double *lb;
3894: double minf; /* the minimum objective value, upon return */
3895: double * p1; /* Shifted parameters from 0 instead of 1 */
3896: myfunc_data dinst, *d = &dinst;
3897: #endif
3898:
3899:
1.126 brouard 3900: xi=matrix(1,npar,1,npar);
3901: for (i=1;i<=npar;i++)
3902: for (j=1;j<=npar;j++)
3903: xi[i][j]=(i==j ? 1.0 : 0.0);
3904: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3905: strcpy(filerespow,"POW_");
1.126 brouard 3906: strcat(filerespow,fileres);
3907: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3908: printf("Problem with resultfile: %s\n", filerespow);
3909: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3910: }
3911: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3912: for (i=1;i<=nlstate;i++)
3913: for(j=1;j<=nlstate+ndeath;j++)
3914: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3915: fprintf(ficrespow,"\n");
1.162 brouard 3916: #ifdef POWELL
1.126 brouard 3917: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3918: #endif
1.126 brouard 3919:
1.162 brouard 3920: #ifdef NLOPT
3921: #ifdef NEWUOA
3922: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3923: #else
3924: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3925: #endif
3926: lb=vector(0,npar-1);
3927: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3928: nlopt_set_lower_bounds(opt, lb);
3929: nlopt_set_initial_step1(opt, 0.1);
3930:
3931: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3932: d->function = func;
3933: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3934: nlopt_set_min_objective(opt, myfunc, d);
3935: nlopt_set_xtol_rel(opt, ftol);
3936: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3937: printf("nlopt failed! %d\n",creturn);
3938: }
3939: else {
3940: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3941: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3942: iter=1; /* not equal */
3943: }
3944: nlopt_destroy(opt);
3945: #endif
1.126 brouard 3946: free_matrix(xi,1,npar,1,npar);
3947: fclose(ficrespow);
1.203 brouard 3948: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3949: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3950: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3951:
3952: }
3953:
3954: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3955: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3956: {
3957: double **a,**y,*x,pd;
1.203 brouard 3958: /* double **hess; */
1.164 brouard 3959: int i, j;
1.126 brouard 3960: int *indx;
3961:
3962: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3963: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3964: void lubksb(double **a, int npar, int *indx, double b[]) ;
3965: void ludcmp(double **a, int npar, int *indx, double *d) ;
3966: double gompertz(double p[]);
1.203 brouard 3967: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3968:
3969: printf("\nCalculation of the hessian matrix. Wait...\n");
3970: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3971: for (i=1;i<=npar;i++){
1.203 brouard 3972: printf("%d-",i);fflush(stdout);
3973: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3974:
3975: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3976:
3977: /* printf(" %f ",p[i]);
3978: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3979: }
3980:
3981: for (i=1;i<=npar;i++) {
3982: for (j=1;j<=npar;j++) {
3983: if (j>i) {
1.203 brouard 3984: printf(".%d-%d",i,j);fflush(stdout);
3985: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3986: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3987:
3988: hess[j][i]=hess[i][j];
3989: /*printf(" %lf ",hess[i][j]);*/
3990: }
3991: }
3992: }
3993: printf("\n");
3994: fprintf(ficlog,"\n");
3995:
3996: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3997: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3998:
3999: a=matrix(1,npar,1,npar);
4000: y=matrix(1,npar,1,npar);
4001: x=vector(1,npar);
4002: indx=ivector(1,npar);
4003: for (i=1;i<=npar;i++)
4004: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4005: ludcmp(a,npar,indx,&pd);
4006:
4007: for (j=1;j<=npar;j++) {
4008: for (i=1;i<=npar;i++) x[i]=0;
4009: x[j]=1;
4010: lubksb(a,npar,indx,x);
4011: for (i=1;i<=npar;i++){
4012: matcov[i][j]=x[i];
4013: }
4014: }
4015:
4016: printf("\n#Hessian matrix#\n");
4017: fprintf(ficlog,"\n#Hessian matrix#\n");
4018: for (i=1;i<=npar;i++) {
4019: for (j=1;j<=npar;j++) {
1.203 brouard 4020: printf("%.6e ",hess[i][j]);
4021: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4022: }
4023: printf("\n");
4024: fprintf(ficlog,"\n");
4025: }
4026:
1.203 brouard 4027: /* printf("\n#Covariance matrix#\n"); */
4028: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4029: /* for (i=1;i<=npar;i++) { */
4030: /* for (j=1;j<=npar;j++) { */
4031: /* printf("%.6e ",matcov[i][j]); */
4032: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4033: /* } */
4034: /* printf("\n"); */
4035: /* fprintf(ficlog,"\n"); */
4036: /* } */
4037:
1.126 brouard 4038: /* Recompute Inverse */
1.203 brouard 4039: /* for (i=1;i<=npar;i++) */
4040: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4041: /* ludcmp(a,npar,indx,&pd); */
4042:
4043: /* printf("\n#Hessian matrix recomputed#\n"); */
4044:
4045: /* for (j=1;j<=npar;j++) { */
4046: /* for (i=1;i<=npar;i++) x[i]=0; */
4047: /* x[j]=1; */
4048: /* lubksb(a,npar,indx,x); */
4049: /* for (i=1;i<=npar;i++){ */
4050: /* y[i][j]=x[i]; */
4051: /* printf("%.3e ",y[i][j]); */
4052: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4053: /* } */
4054: /* printf("\n"); */
4055: /* fprintf(ficlog,"\n"); */
4056: /* } */
4057:
4058: /* Verifying the inverse matrix */
4059: #ifdef DEBUGHESS
4060: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4061:
1.203 brouard 4062: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4063: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4064:
4065: for (j=1;j<=npar;j++) {
4066: for (i=1;i<=npar;i++){
1.203 brouard 4067: printf("%.2f ",y[i][j]);
4068: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4069: }
4070: printf("\n");
4071: fprintf(ficlog,"\n");
4072: }
1.203 brouard 4073: #endif
1.126 brouard 4074:
4075: free_matrix(a,1,npar,1,npar);
4076: free_matrix(y,1,npar,1,npar);
4077: free_vector(x,1,npar);
4078: free_ivector(indx,1,npar);
1.203 brouard 4079: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4080:
4081:
4082: }
4083:
4084: /*************** hessian matrix ****************/
4085: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4086: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4087: int i;
4088: int l=1, lmax=20;
1.203 brouard 4089: double k1,k2, res, fx;
1.132 brouard 4090: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4091: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4092: int k=0,kmax=10;
4093: double l1;
4094:
4095: fx=func(x);
4096: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4097: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4098: l1=pow(10,l);
4099: delts=delt;
4100: for(k=1 ; k <kmax; k=k+1){
4101: delt = delta*(l1*k);
4102: p2[theta]=x[theta] +delt;
1.145 brouard 4103: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4104: p2[theta]=x[theta]-delt;
4105: k2=func(p2)-fx;
4106: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4107: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4108:
1.203 brouard 4109: #ifdef DEBUGHESSII
1.126 brouard 4110: 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);
4111: 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);
4112: #endif
4113: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4114: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4115: k=kmax;
4116: }
4117: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4118: k=kmax; l=lmax*10;
1.126 brouard 4119: }
4120: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4121: delts=delt;
4122: }
1.203 brouard 4123: } /* End loop k */
1.126 brouard 4124: }
4125: delti[theta]=delts;
4126: return res;
4127:
4128: }
4129:
1.203 brouard 4130: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4131: {
4132: int i;
1.164 brouard 4133: int l=1, lmax=20;
1.126 brouard 4134: double k1,k2,k3,k4,res,fx;
1.132 brouard 4135: double p2[MAXPARM+1];
1.203 brouard 4136: int k, kmax=1;
4137: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4138:
4139: int firstime=0;
1.203 brouard 4140:
1.126 brouard 4141: fx=func(x);
1.203 brouard 4142: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4143: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4144: p2[thetai]=x[thetai]+delti[thetai]*k;
4145: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4146: k1=func(p2)-fx;
4147:
1.203 brouard 4148: p2[thetai]=x[thetai]+delti[thetai]*k;
4149: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4150: k2=func(p2)-fx;
4151:
1.203 brouard 4152: p2[thetai]=x[thetai]-delti[thetai]*k;
4153: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4154: k3=func(p2)-fx;
4155:
1.203 brouard 4156: p2[thetai]=x[thetai]-delti[thetai]*k;
4157: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4158: k4=func(p2)-fx;
1.203 brouard 4159: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4160: if(k1*k2*k3*k4 <0.){
1.208 brouard 4161: firstime=1;
1.203 brouard 4162: kmax=kmax+10;
1.208 brouard 4163: }
4164: if(kmax >=10 || firstime ==1){
1.246 brouard 4165: 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);
4166: 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 4167: 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);
4168: 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);
4169: }
4170: #ifdef DEBUGHESSIJ
4171: v1=hess[thetai][thetai];
4172: v2=hess[thetaj][thetaj];
4173: cv12=res;
4174: /* Computing eigen value of Hessian matrix */
4175: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4176: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4177: if ((lc2 <0) || (lc1 <0) ){
4178: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4179: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4180: 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);
4181: 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);
4182: }
1.126 brouard 4183: #endif
4184: }
4185: return res;
4186: }
4187:
1.203 brouard 4188: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4189: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4190: /* { */
4191: /* int i; */
4192: /* int l=1, lmax=20; */
4193: /* double k1,k2,k3,k4,res,fx; */
4194: /* double p2[MAXPARM+1]; */
4195: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4196: /* int k=0,kmax=10; */
4197: /* double l1; */
4198:
4199: /* fx=func(x); */
4200: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4201: /* l1=pow(10,l); */
4202: /* delts=delt; */
4203: /* for(k=1 ; k <kmax; k=k+1){ */
4204: /* delt = delti*(l1*k); */
4205: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4206: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4207: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4208: /* k1=func(p2)-fx; */
4209:
4210: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4211: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4212: /* k2=func(p2)-fx; */
4213:
4214: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4215: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4216: /* k3=func(p2)-fx; */
4217:
4218: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4219: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4220: /* k4=func(p2)-fx; */
4221: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4222: /* #ifdef DEBUGHESSIJ */
4223: /* 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); */
4224: /* 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); */
4225: /* #endif */
4226: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4227: /* k=kmax; */
4228: /* } */
4229: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4230: /* k=kmax; l=lmax*10; */
4231: /* } */
4232: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4233: /* delts=delt; */
4234: /* } */
4235: /* } /\* End loop k *\/ */
4236: /* } */
4237: /* delti[theta]=delts; */
4238: /* return res; */
4239: /* } */
4240:
4241:
1.126 brouard 4242: /************** Inverse of matrix **************/
4243: void ludcmp(double **a, int n, int *indx, double *d)
4244: {
4245: int i,imax,j,k;
4246: double big,dum,sum,temp;
4247: double *vv;
4248:
4249: vv=vector(1,n);
4250: *d=1.0;
4251: for (i=1;i<=n;i++) {
4252: big=0.0;
4253: for (j=1;j<=n;j++)
4254: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4255: if (big == 0.0){
4256: printf(" Singular Hessian matrix at row %d:\n",i);
4257: for (j=1;j<=n;j++) {
4258: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4259: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4260: }
4261: fflush(ficlog);
4262: fclose(ficlog);
4263: nrerror("Singular matrix in routine ludcmp");
4264: }
1.126 brouard 4265: vv[i]=1.0/big;
4266: }
4267: for (j=1;j<=n;j++) {
4268: for (i=1;i<j;i++) {
4269: sum=a[i][j];
4270: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4271: a[i][j]=sum;
4272: }
4273: big=0.0;
4274: for (i=j;i<=n;i++) {
4275: sum=a[i][j];
4276: for (k=1;k<j;k++)
4277: sum -= a[i][k]*a[k][j];
4278: a[i][j]=sum;
4279: if ( (dum=vv[i]*fabs(sum)) >= big) {
4280: big=dum;
4281: imax=i;
4282: }
4283: }
4284: if (j != imax) {
4285: for (k=1;k<=n;k++) {
4286: dum=a[imax][k];
4287: a[imax][k]=a[j][k];
4288: a[j][k]=dum;
4289: }
4290: *d = -(*d);
4291: vv[imax]=vv[j];
4292: }
4293: indx[j]=imax;
4294: if (a[j][j] == 0.0) a[j][j]=TINY;
4295: if (j != n) {
4296: dum=1.0/(a[j][j]);
4297: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4298: }
4299: }
4300: free_vector(vv,1,n); /* Doesn't work */
4301: ;
4302: }
4303:
4304: void lubksb(double **a, int n, int *indx, double b[])
4305: {
4306: int i,ii=0,ip,j;
4307: double sum;
4308:
4309: for (i=1;i<=n;i++) {
4310: ip=indx[i];
4311: sum=b[ip];
4312: b[ip]=b[i];
4313: if (ii)
4314: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4315: else if (sum) ii=i;
4316: b[i]=sum;
4317: }
4318: for (i=n;i>=1;i--) {
4319: sum=b[i];
4320: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4321: b[i]=sum/a[i][i];
4322: }
4323: }
4324:
4325: void pstamp(FILE *fichier)
4326: {
1.196 brouard 4327: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4328: }
4329:
1.253 brouard 4330:
4331:
1.126 brouard 4332: /************ Frequencies ********************/
1.251 brouard 4333: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4334: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4335: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4336: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4337:
1.265 brouard 4338: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4339: int iind=0, iage=0;
4340: int mi; /* Effective wave */
4341: int first;
4342: double ***freq; /* Frequencies */
1.268 brouard 4343: 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 */
4344: 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 4345: double *meanq;
4346: double **meanqt;
4347: double *pp, **prop, *posprop, *pospropt;
4348: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4349: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4350: double agebegin, ageend;
4351:
4352: pp=vector(1,nlstate);
1.251 brouard 4353: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4354: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4355: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4356: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4357: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4358: meanqt=matrix(1,lastpass,1,nqtveff);
4359: strcpy(fileresp,"P_");
4360: strcat(fileresp,fileresu);
4361: /*strcat(fileresphtm,fileresu);*/
4362: if((ficresp=fopen(fileresp,"w"))==NULL) {
4363: printf("Problem with prevalence resultfile: %s\n", fileresp);
4364: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4365: exit(0);
4366: }
1.240 brouard 4367:
1.226 brouard 4368: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4369: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4370: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4371: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4372: fflush(ficlog);
4373: exit(70);
4374: }
4375: else{
4376: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4377: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4378: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4379: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4380: }
1.237 brouard 4381: 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 4382:
1.226 brouard 4383: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4384: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4385: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4386: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4387: fflush(ficlog);
4388: exit(70);
1.240 brouard 4389: } else{
1.226 brouard 4390: 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 4391: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4392: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4393: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4394: }
1.240 brouard 4395: 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);
4396:
1.253 brouard 4397: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4398: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4399: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4400: j1=0;
1.126 brouard 4401:
1.227 brouard 4402: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4403: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4404: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4405:
4406:
1.226 brouard 4407: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4408: reference=low_education V1=0,V2=0
4409: med_educ V1=1 V2=0,
4410: high_educ V1=0 V2=1
4411: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4412: */
1.249 brouard 4413: dateintsum=0;
4414: k2cpt=0;
4415:
1.253 brouard 4416: if(cptcoveff == 0 )
1.265 brouard 4417: nl=1; /* Constant and age model only */
1.253 brouard 4418: else
4419: nl=2;
1.265 brouard 4420:
4421: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4422: /* Loop on nj=1 or 2 if dummy covariates j!=0
4423: * Loop on j1(1 to 2**cptcoveff) covariate combination
4424: * freq[s1][s2][iage] =0.
4425: * Loop on iind
4426: * ++freq[s1][s2][iage] weighted
4427: * end iind
4428: * if covariate and j!0
4429: * headers Variable on one line
4430: * endif cov j!=0
4431: * header of frequency table by age
4432: * Loop on age
4433: * pp[s1]+=freq[s1][s2][iage] weighted
4434: * pos+=freq[s1][s2][iage] weighted
4435: * Loop on s1 initial state
4436: * fprintf(ficresp
4437: * end s1
4438: * end age
4439: * if j!=0 computes starting values
4440: * end compute starting values
4441: * end j1
4442: * end nl
4443: */
1.253 brouard 4444: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4445: if(nj==1)
4446: j=0; /* First pass for the constant */
1.265 brouard 4447: else{
1.253 brouard 4448: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4449: }
1.251 brouard 4450: first=1;
1.265 brouard 4451: 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 4452: posproptt=0.;
4453: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4454: scanf("%d", i);*/
4455: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4456: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4457: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4458: freq[i][s2][m]=0;
1.251 brouard 4459:
4460: for (i=1; i<=nlstate; i++) {
1.240 brouard 4461: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4462: prop[i][m]=0;
4463: posprop[i]=0;
4464: pospropt[i]=0;
4465: }
4466: /* for (z1=1; z1<= nqfveff; z1++) { */
4467: /* meanq[z1]+=0.; */
4468: /* for(m=1;m<=lastpass;m++){ */
4469: /* meanqt[m][z1]=0.; */
4470: /* } */
4471: /* } */
4472:
4473: /* dateintsum=0; */
4474: /* k2cpt=0; */
4475:
1.265 brouard 4476: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4477: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4478: bool=1;
4479: if(j !=0){
4480: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4481: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4482: /* for (z1=1; z1<= nqfveff; z1++) { */
4483: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4484: /* } */
4485: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4486: /* if(Tvaraff[z1] ==-20){ */
4487: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4488: /* }else if(Tvaraff[z1] ==-10){ */
4489: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4490: /* }else */
4491: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4492: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4493: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4494: /* 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",
4495: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4496: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4497: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4498: } /* Onlyf fixed */
4499: } /* end z1 */
4500: } /* cptcovn > 0 */
4501: } /* end any */
4502: }/* end j==0 */
1.265 brouard 4503: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4504: /* for(m=firstpass; m<=lastpass; m++){ */
4505: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4506: m=mw[mi][iind];
4507: if(j!=0){
4508: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4509: for (z1=1; z1<=cptcoveff; z1++) {
4510: if( Fixed[Tmodelind[z1]]==1){
4511: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4512: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4513: value is -1, we don't select. It differs from the
4514: constant and age model which counts them. */
4515: bool=0; /* not selected */
4516: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4517: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4518: bool=0;
4519: }
4520: }
4521: }
4522: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4523: } /* end j==0 */
4524: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4525: if(bool==1){
4526: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4527: and mw[mi+1][iind]. dh depends on stepm. */
4528: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4529: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4530: if(m >=firstpass && m <=lastpass){
4531: k2=anint[m][iind]+(mint[m][iind]/12.);
4532: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4533: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4534: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4535: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4536: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4537: if (m<lastpass) {
4538: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4539: /* 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]); */
4540: if(s[m][iind]==-1)
4541: 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.));
4542: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4543: /* if((int)agev[m][iind] == 55) */
4544: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4545: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4546: 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 4547: }
1.251 brouard 4548: } /* end if between passes */
4549: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4550: dateintsum=dateintsum+k2; /* on all covariates ?*/
4551: k2cpt++;
4552: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4553: }
1.251 brouard 4554: }else{
4555: bool=1;
4556: }/* end bool 2 */
4557: } /* end m */
4558: } /* end bool */
4559: } /* end iind = 1 to imx */
4560: /* prop[s][age] is feeded for any initial and valid live state as well as
4561: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4562:
4563:
4564: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4565: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4566: pstamp(ficresp);
1.251 brouard 4567: if (cptcoveff>0 && j!=0){
1.265 brouard 4568: pstamp(ficresp);
1.251 brouard 4569: printf( "\n#********** Variable ");
4570: fprintf(ficresp, "\n#********** Variable ");
4571: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4572: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4573: fprintf(ficlog, "\n#********** Variable ");
4574: for (z1=1; z1<=cptcoveff; z1++){
4575: if(!FixedV[Tvaraff[z1]]){
4576: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4577: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4578: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4579: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4580: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4581: }else{
1.251 brouard 4582: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4583: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4584: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4585: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4586: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4587: }
4588: }
4589: printf( "**********\n#");
4590: fprintf(ficresp, "**********\n#");
4591: fprintf(ficresphtm, "**********</h3>\n");
4592: fprintf(ficresphtmfr, "**********</h3>\n");
4593: fprintf(ficlog, "**********\n");
4594: }
4595: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4596: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4597: fprintf(ficresp, " Age");
4598: 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 4599: for(i=1; i<=nlstate;i++) {
1.265 brouard 4600: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4601: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4602: }
1.265 brouard 4603: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4604: fprintf(ficresphtm, "\n");
4605:
4606: /* Header of frequency table by age */
4607: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4608: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4609: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4610: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4611: if(s2!=0 && m!=0)
4612: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4613: }
1.226 brouard 4614: }
1.251 brouard 4615: fprintf(ficresphtmfr, "\n");
4616:
4617: /* For each age */
4618: for(iage=iagemin; iage <= iagemax+3; iage++){
4619: fprintf(ficresphtm,"<tr>");
4620: if(iage==iagemax+1){
4621: fprintf(ficlog,"1");
4622: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4623: }else if(iage==iagemax+2){
4624: fprintf(ficlog,"0");
4625: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4626: }else if(iage==iagemax+3){
4627: fprintf(ficlog,"Total");
4628: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4629: }else{
1.240 brouard 4630: if(first==1){
1.251 brouard 4631: first=0;
4632: printf("See log file for details...\n");
4633: }
4634: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4635: fprintf(ficlog,"Age %d", iage);
4636: }
1.265 brouard 4637: for(s1=1; s1 <=nlstate ; s1++){
4638: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4639: pp[s1] += freq[s1][m][iage];
1.251 brouard 4640: }
1.265 brouard 4641: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4642: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4643: pos += freq[s1][m][iage];
4644: if(pp[s1]>=1.e-10){
1.251 brouard 4645: if(first==1){
1.265 brouard 4646: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4647: }
1.265 brouard 4648: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4649: }else{
4650: if(first==1)
1.265 brouard 4651: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4652: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4653: }
4654: }
4655:
1.265 brouard 4656: for(s1=1; s1 <=nlstate ; s1++){
4657: /* posprop[s1]=0; */
4658: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4659: pp[s1] += freq[s1][m][iage];
4660: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4661:
4662: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4663: pos += pp[s1]; /* pos is the total number of transitions until this age */
4664: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4665: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4666: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4667: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4668: }
4669:
4670: /* Writing ficresp */
4671: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4672: if( iage <= iagemax){
4673: fprintf(ficresp," %d",iage);
4674: }
4675: }else if( nj==2){
4676: if( iage <= iagemax){
4677: fprintf(ficresp," %d",iage);
4678: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4679: }
1.240 brouard 4680: }
1.265 brouard 4681: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4682: if(pos>=1.e-5){
1.251 brouard 4683: if(first==1)
1.265 brouard 4684: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4685: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4686: }else{
4687: if(first==1)
1.265 brouard 4688: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4689: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4690: }
4691: if( iage <= iagemax){
4692: if(pos>=1.e-5){
1.265 brouard 4693: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4694: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4695: }else if( nj==2){
4696: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4697: }
4698: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4699: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4700: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4701: } else{
4702: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4703: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4704: }
1.240 brouard 4705: }
1.265 brouard 4706: pospropt[s1] +=posprop[s1];
4707: } /* end loop s1 */
1.251 brouard 4708: /* pospropt=0.; */
1.265 brouard 4709: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4710: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4711: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4712: if(first==1){
1.265 brouard 4713: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4714: }
1.265 brouard 4715: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4716: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4717: }
1.265 brouard 4718: if(s1!=0 && m!=0)
4719: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4720: }
1.265 brouard 4721: } /* end loop s1 */
1.251 brouard 4722: posproptt=0.;
1.265 brouard 4723: for(s1=1; s1 <=nlstate; s1++){
4724: posproptt += pospropt[s1];
1.251 brouard 4725: }
4726: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4727: fprintf(ficresphtm,"</tr>\n");
4728: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4729: if(iage <= iagemax)
4730: fprintf(ficresp,"\n");
1.240 brouard 4731: }
1.251 brouard 4732: if(first==1)
4733: printf("Others in log...\n");
4734: fprintf(ficlog,"\n");
4735: } /* end loop age iage */
1.265 brouard 4736:
1.251 brouard 4737: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4738: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4739: if(posproptt < 1.e-5){
1.265 brouard 4740: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4741: }else{
1.265 brouard 4742: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4743: }
1.226 brouard 4744: }
1.251 brouard 4745: fprintf(ficresphtm,"</tr>\n");
4746: fprintf(ficresphtm,"</table>\n");
4747: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4748: if(posproptt < 1.e-5){
1.251 brouard 4749: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4750: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4751: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4752: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4753: invalidvarcomb[j1]=1;
1.226 brouard 4754: }else{
1.251 brouard 4755: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4756: invalidvarcomb[j1]=0;
1.226 brouard 4757: }
1.251 brouard 4758: fprintf(ficresphtmfr,"</table>\n");
4759: fprintf(ficlog,"\n");
4760: if(j!=0){
4761: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4762: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4763: for(k=1; k <=(nlstate+ndeath); k++){
4764: if (k != i) {
1.265 brouard 4765: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4766: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4767: if(j1==1){ /* All dummy covariates to zero */
4768: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4769: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4770: printf("%d%d ",i,k);
4771: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4772: 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]));
4773: 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]));
4774: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4775: }
1.253 brouard 4776: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4777: for(iage=iagemin; iage <= iagemax+3; iage++){
4778: x[iage]= (double)iage;
4779: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4780: /* 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 4781: }
1.268 brouard 4782: /* Some are not finite, but linreg will ignore these ages */
4783: no=0;
1.253 brouard 4784: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4785: pstart[s1]=b;
4786: pstart[s1-1]=a;
1.252 brouard 4787: }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 */
4788: 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]);
4789: 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 4790: 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 4791: printf("%d%d ",i,k);
4792: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4793: 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 4794: }else{ /* Other cases, like quantitative fixed or varying covariates */
4795: ;
4796: }
4797: /* printf("%12.7f )", param[i][jj][k]); */
4798: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4799: s1++;
1.251 brouard 4800: } /* end jj */
4801: } /* end k!= i */
4802: } /* end k */
1.265 brouard 4803: } /* end i, s1 */
1.251 brouard 4804: } /* end j !=0 */
4805: } /* end selected combination of covariate j1 */
4806: if(j==0){ /* We can estimate starting values from the occurences in each case */
4807: printf("#Freqsummary: Starting values for the constants:\n");
4808: fprintf(ficlog,"\n");
1.265 brouard 4809: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4810: for(k=1; k <=(nlstate+ndeath); k++){
4811: if (k != i) {
4812: printf("%d%d ",i,k);
4813: fprintf(ficlog,"%d%d ",i,k);
4814: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4815: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4816: if(jj==1){ /* Age has to be done */
1.265 brouard 4817: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4818: 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]));
4819: 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 4820: }
4821: /* printf("%12.7f )", param[i][jj][k]); */
4822: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4823: s1++;
1.250 brouard 4824: }
1.251 brouard 4825: printf("\n");
4826: fprintf(ficlog,"\n");
1.250 brouard 4827: }
4828: }
4829: }
1.251 brouard 4830: printf("#Freqsummary\n");
4831: fprintf(ficlog,"\n");
1.265 brouard 4832: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4833: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4834: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4835: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4836: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4837: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4838: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4839: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4840: /* } */
4841: }
1.265 brouard 4842: } /* end loop s1 */
1.251 brouard 4843:
4844: printf("\n");
4845: fprintf(ficlog,"\n");
4846: } /* end j=0 */
1.249 brouard 4847: } /* end j */
1.252 brouard 4848:
1.253 brouard 4849: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4850: for(i=1, jk=1; i <=nlstate; i++){
4851: for(j=1; j <=nlstate+ndeath; j++){
4852: if(j!=i){
4853: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4854: printf("%1d%1d",i,j);
4855: fprintf(ficparo,"%1d%1d",i,j);
4856: for(k=1; k<=ncovmodel;k++){
4857: /* printf(" %lf",param[i][j][k]); */
4858: /* fprintf(ficparo," %lf",param[i][j][k]); */
4859: p[jk]=pstart[jk];
4860: printf(" %f ",pstart[jk]);
4861: fprintf(ficparo," %f ",pstart[jk]);
4862: jk++;
4863: }
4864: printf("\n");
4865: fprintf(ficparo,"\n");
4866: }
4867: }
4868: }
4869: } /* end mle=-2 */
1.226 brouard 4870: dateintmean=dateintsum/k2cpt;
1.240 brouard 4871:
1.226 brouard 4872: fclose(ficresp);
4873: fclose(ficresphtm);
4874: fclose(ficresphtmfr);
4875: free_vector(meanq,1,nqfveff);
4876: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4877: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4878: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4879: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4880: free_vector(pospropt,1,nlstate);
4881: free_vector(posprop,1,nlstate);
1.251 brouard 4882: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4883: free_vector(pp,1,nlstate);
4884: /* End of freqsummary */
4885: }
1.126 brouard 4886:
1.268 brouard 4887: /* Simple linear regression */
4888: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4889:
4890: /* y=a+bx regression */
4891: double sumx = 0.0; /* sum of x */
4892: double sumx2 = 0.0; /* sum of x**2 */
4893: double sumxy = 0.0; /* sum of x * y */
4894: double sumy = 0.0; /* sum of y */
4895: double sumy2 = 0.0; /* sum of y**2 */
4896: double sume2 = 0.0; /* sum of square or residuals */
4897: double yhat;
4898:
4899: double denom=0;
4900: int i;
4901: int ne=*no;
4902:
4903: for ( i=ifi, ne=0;i<=ila;i++) {
4904: if(!isfinite(x[i]) || !isfinite(y[i])){
4905: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4906: continue;
4907: }
4908: ne=ne+1;
4909: sumx += x[i];
4910: sumx2 += x[i]*x[i];
4911: sumxy += x[i] * y[i];
4912: sumy += y[i];
4913: sumy2 += y[i]*y[i];
4914: denom = (ne * sumx2 - sumx*sumx);
4915: /* 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); */
4916: }
4917:
4918: denom = (ne * sumx2 - sumx*sumx);
4919: if (denom == 0) {
4920: // vertical, slope m is infinity
4921: *b = INFINITY;
4922: *a = 0;
4923: if (r) *r = 0;
4924: return 1;
4925: }
4926:
4927: *b = (ne * sumxy - sumx * sumy) / denom;
4928: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4929: if (r!=NULL) {
4930: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4931: sqrt((sumx2 - sumx*sumx/ne) *
4932: (sumy2 - sumy*sumy/ne));
4933: }
4934: *no=ne;
4935: for ( i=ifi, ne=0;i<=ila;i++) {
4936: if(!isfinite(x[i]) || !isfinite(y[i])){
4937: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4938: continue;
4939: }
4940: ne=ne+1;
4941: yhat = y[i] - *a -*b* x[i];
4942: sume2 += yhat * yhat ;
4943:
4944: denom = (ne * sumx2 - sumx*sumx);
4945: /* 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); */
4946: }
4947: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
4948: *sa= *sb * sqrt(sumx2/ne);
4949:
4950: return 0;
4951: }
4952:
1.126 brouard 4953: /************ Prevalence ********************/
1.227 brouard 4954: 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)
4955: {
4956: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4957: in each health status at the date of interview (if between dateprev1 and dateprev2).
4958: We still use firstpass and lastpass as another selection.
4959: */
1.126 brouard 4960:
1.227 brouard 4961: int i, m, jk, j1, bool, z1,j, iv;
4962: int mi; /* Effective wave */
4963: int iage;
4964: double agebegin, ageend;
4965:
4966: double **prop;
4967: double posprop;
4968: double y2; /* in fractional years */
4969: int iagemin, iagemax;
4970: int first; /** to stop verbosity which is redirected to log file */
4971:
4972: iagemin= (int) agemin;
4973: iagemax= (int) agemax;
4974: /*pp=vector(1,nlstate);*/
1.251 brouard 4975: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4976: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4977: j1=0;
1.222 brouard 4978:
1.227 brouard 4979: /*j=cptcoveff;*/
4980: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4981:
1.227 brouard 4982: first=1;
4983: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4984: for (i=1; i<=nlstate; i++)
1.251 brouard 4985: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4986: prop[i][iage]=0.0;
4987: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4988: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4989: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4990:
4991: for (i=1; i<=imx; i++) { /* Each individual */
4992: bool=1;
4993: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4994: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4995: m=mw[mi][i];
4996: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4997: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4998: for (z1=1; z1<=cptcoveff; z1++){
4999: if( Fixed[Tmodelind[z1]]==1){
5000: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5001: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5002: bool=0;
5003: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5004: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5005: bool=0;
5006: }
5007: }
5008: if(bool==1){ /* Otherwise we skip that wave/person */
5009: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5010: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5011: if(m >=firstpass && m <=lastpass){
5012: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5013: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5014: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5015: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5016: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5017: 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);
5018: exit(1);
5019: }
5020: if (s[m][i]>0 && s[m][i]<=nlstate) {
5021: /*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]]);*/
5022: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5023: prop[s[m][i]][iagemax+3] += weight[i];
5024: } /* end valid statuses */
5025: } /* end selection of dates */
5026: } /* end selection of waves */
5027: } /* end bool */
5028: } /* end wave */
5029: } /* end individual */
5030: for(i=iagemin; i <= iagemax+3; i++){
5031: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5032: posprop += prop[jk][i];
5033: }
5034:
5035: for(jk=1; jk <=nlstate ; jk++){
5036: if( i <= iagemax){
5037: if(posprop>=1.e-5){
5038: probs[i][jk][j1]= prop[jk][i]/posprop;
5039: } else{
5040: if(first==1){
5041: first=0;
1.266 brouard 5042: 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]);
5043: 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]);
5044: }else{
5045: 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 5046: }
5047: }
5048: }
5049: }/* end jk */
5050: }/* end i */
1.222 brouard 5051: /*} *//* end i1 */
1.227 brouard 5052: } /* end j1 */
1.222 brouard 5053:
1.227 brouard 5054: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5055: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5056: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5057: } /* End of prevalence */
1.126 brouard 5058:
5059: /************* Waves Concatenation ***************/
5060:
5061: 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)
5062: {
5063: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5064: Death is a valid wave (if date is known).
5065: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5066: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5067: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5068: */
1.126 brouard 5069:
1.224 brouard 5070: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5071: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5072: double sum=0., jmean=0.;*/
1.224 brouard 5073: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5074: int j, k=0,jk, ju, jl;
5075: double sum=0.;
5076: first=0;
1.214 brouard 5077: firstwo=0;
1.217 brouard 5078: firsthree=0;
1.218 brouard 5079: firstfour=0;
1.164 brouard 5080: jmin=100000;
1.126 brouard 5081: jmax=-1;
5082: jmean=0.;
1.224 brouard 5083:
5084: /* Treating live states */
1.214 brouard 5085: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5086: mi=0; /* First valid wave */
1.227 brouard 5087: mli=0; /* Last valid wave */
1.126 brouard 5088: m=firstpass;
1.214 brouard 5089: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5090: 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 */
5091: mli=m-1;/* mw[++mi][i]=m-1; */
5092: }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 */
5093: mw[++mi][i]=m;
5094: mli=m;
1.224 brouard 5095: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5096: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5097: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5098: }
1.227 brouard 5099: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5100: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5101: break;
1.224 brouard 5102: #else
1.227 brouard 5103: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5104: if(firsthree == 0){
1.262 brouard 5105: 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 5106: firsthree=1;
5107: }
1.262 brouard 5108: 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 5109: mw[++mi][i]=m;
5110: mli=m;
5111: }
5112: if(s[m][i]==-2){ /* Vital status is really unknown */
5113: nbwarn++;
5114: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5115: 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);
5116: 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);
5117: }
5118: break;
5119: }
5120: break;
1.224 brouard 5121: #endif
1.227 brouard 5122: }/* End m >= lastpass */
1.126 brouard 5123: }/* end while */
1.224 brouard 5124:
1.227 brouard 5125: /* 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 5126: /* After last pass */
1.224 brouard 5127: /* Treating death states */
1.214 brouard 5128: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5129: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5130: /* } */
1.126 brouard 5131: mi++; /* Death is another wave */
5132: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5133: /* Only death is a correct wave */
1.126 brouard 5134: mw[mi][i]=m;
1.257 brouard 5135: } /* else not in a death state */
1.224 brouard 5136: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5137: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5138: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5139: 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 */
5140: nbwarn++;
5141: if(firstfiv==0){
5142: 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 );
5143: firstfiv=1;
5144: }else{
5145: 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 );
5146: }
5147: }else{ /* Death occured afer last wave potential bias */
5148: nberr++;
5149: if(firstwo==0){
1.257 brouard 5150: 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 5151: firstwo=1;
5152: }
1.257 brouard 5153: 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 5154: }
1.257 brouard 5155: }else{ /* if date of interview is unknown */
1.227 brouard 5156: /* death is known but not confirmed by death status at any wave */
5157: if(firstfour==0){
5158: 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 );
5159: firstfour=1;
5160: }
5161: 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 5162: }
1.224 brouard 5163: } /* end if date of death is known */
5164: #endif
5165: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5166: /* wav[i]=mw[mi][i]; */
1.126 brouard 5167: if(mi==0){
5168: nbwarn++;
5169: if(first==0){
1.227 brouard 5170: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5171: first=1;
1.126 brouard 5172: }
5173: if(first==1){
1.227 brouard 5174: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5175: }
5176: } /* end mi==0 */
5177: } /* End individuals */
1.214 brouard 5178: /* wav and mw are no more changed */
1.223 brouard 5179:
1.214 brouard 5180:
1.126 brouard 5181: for(i=1; i<=imx; i++){
5182: for(mi=1; mi<wav[i];mi++){
5183: if (stepm <=0)
1.227 brouard 5184: dh[mi][i]=1;
1.126 brouard 5185: else{
1.260 brouard 5186: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5187: if (agedc[i] < 2*AGESUP) {
5188: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5189: if(j==0) j=1; /* Survives at least one month after exam */
5190: else if(j<0){
5191: nberr++;
5192: 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]);
5193: j=1; /* Temporary Dangerous patch */
5194: 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);
5195: 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]);
5196: 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);
5197: }
5198: k=k+1;
5199: if (j >= jmax){
5200: jmax=j;
5201: ijmax=i;
5202: }
5203: if (j <= jmin){
5204: jmin=j;
5205: ijmin=i;
5206: }
5207: sum=sum+j;
5208: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5209: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5210: }
5211: }
5212: else{
5213: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5214: /* 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 5215:
1.227 brouard 5216: k=k+1;
5217: if (j >= jmax) {
5218: jmax=j;
5219: ijmax=i;
5220: }
5221: else if (j <= jmin){
5222: jmin=j;
5223: ijmin=i;
5224: }
5225: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5226: /*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]);*/
5227: if(j<0){
5228: nberr++;
5229: 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]);
5230: 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]);
5231: }
5232: sum=sum+j;
5233: }
5234: jk= j/stepm;
5235: jl= j -jk*stepm;
5236: ju= j -(jk+1)*stepm;
5237: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5238: if(jl==0){
5239: dh[mi][i]=jk;
5240: bh[mi][i]=0;
5241: }else{ /* We want a negative bias in order to only have interpolation ie
5242: * to avoid the price of an extra matrix product in likelihood */
5243: dh[mi][i]=jk+1;
5244: bh[mi][i]=ju;
5245: }
5246: }else{
5247: if(jl <= -ju){
5248: dh[mi][i]=jk;
5249: bh[mi][i]=jl; /* bias is positive if real duration
5250: * is higher than the multiple of stepm and negative otherwise.
5251: */
5252: }
5253: else{
5254: dh[mi][i]=jk+1;
5255: bh[mi][i]=ju;
5256: }
5257: if(dh[mi][i]==0){
5258: dh[mi][i]=1; /* At least one step */
5259: bh[mi][i]=ju; /* At least one step */
5260: /* 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);*/
5261: }
5262: } /* end if mle */
1.126 brouard 5263: }
5264: } /* end wave */
5265: }
5266: jmean=sum/k;
5267: 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 5268: 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 5269: }
1.126 brouard 5270:
5271: /*********** Tricode ****************************/
1.220 brouard 5272: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5273: {
5274: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5275: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5276: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5277: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5278: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5279: */
1.130 brouard 5280:
1.242 brouard 5281: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5282: int modmaxcovj=0; /* Modality max of covariates j */
5283: int cptcode=0; /* Modality max of covariates j */
5284: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5285:
5286:
1.242 brouard 5287: /* cptcoveff=0; */
5288: /* *cptcov=0; */
1.126 brouard 5289:
1.242 brouard 5290: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5291:
1.242 brouard 5292: /* Loop on covariates without age and products and no quantitative variable */
5293: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5294: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5295: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5296: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5297: switch(Fixed[k]) {
5298: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5299: 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*/
5300: ij=(int)(covar[Tvar[k]][i]);
5301: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5302: * If product of Vn*Vm, still boolean *:
5303: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5304: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5305: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5306: modality of the nth covariate of individual i. */
5307: if (ij > modmaxcovj)
5308: modmaxcovj=ij;
5309: else if (ij < modmincovj)
5310: modmincovj=ij;
5311: if ((ij < -1) && (ij > NCOVMAX)){
5312: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5313: exit(1);
5314: }else
5315: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5316: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5317: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5318: /* getting the maximum value of the modality of the covariate
5319: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5320: female ies 1, then modmaxcovj=1.
5321: */
5322: } /* end for loop on individuals i */
5323: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5324: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5325: cptcode=modmaxcovj;
5326: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5327: /*for (i=0; i<=cptcode; i++) {*/
5328: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5329: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5330: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5331: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5332: if( j != -1){
5333: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5334: covariate for which somebody answered excluding
5335: undefined. Usually 2: 0 and 1. */
5336: }
5337: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5338: covariate for which somebody answered including
5339: undefined. Usually 3: -1, 0 and 1. */
5340: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5341: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5342: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5343:
1.242 brouard 5344: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5345: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5346: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5347: /* modmincovj=3; modmaxcovj = 7; */
5348: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5349: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5350: /* defining two dummy variables: variables V1_1 and V1_2.*/
5351: /* nbcode[Tvar[j]][ij]=k; */
5352: /* nbcode[Tvar[j]][1]=0; */
5353: /* nbcode[Tvar[j]][2]=1; */
5354: /* nbcode[Tvar[j]][3]=2; */
5355: /* To be continued (not working yet). */
5356: ij=0; /* ij is similar to i but can jump over null modalities */
5357: 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*/
5358: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5359: break;
5360: }
5361: ij++;
5362: 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*/
5363: cptcode = ij; /* New max modality for covar j */
5364: } /* end of loop on modality i=-1 to 1 or more */
5365: break;
5366: case 1: /* Testing on varying covariate, could be simple and
5367: * should look at waves or product of fixed *
5368: * varying. No time to test -1, assuming 0 and 1 only */
5369: ij=0;
5370: for(i=0; i<=1;i++){
5371: nbcode[Tvar[k]][++ij]=i;
5372: }
5373: break;
5374: default:
5375: break;
5376: } /* end switch */
5377: } /* end dummy test */
5378:
5379: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5380: /* /\*recode from 0 *\/ */
5381: /* k is a modality. If we have model=V1+V1*sex */
5382: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5383: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5384: /* } */
5385: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5386: /* if (ij > ncodemax[j]) { */
5387: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5388: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5389: /* break; */
5390: /* } */
5391: /* } /\* end of loop on modality k *\/ */
5392: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5393:
5394: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5395: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5396: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5397: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5398: 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 */
5399: 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 */
5400: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5401: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5402:
5403: ij=0;
5404: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5405: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5406: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5407: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5408: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5409: /* If product not in single variable we don't print results */
5410: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5411: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5412: 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*/
5413: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5414: 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 */
5415: if(Fixed[k]!=0)
5416: anyvaryingduminmodel=1;
5417: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5418: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5419: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5420: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5421: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5422: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5423: }
5424: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5425: /* ij--; */
5426: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5427: *cptcov=ij; /*Number of total real effective covariates: effective
5428: * because they can be excluded from the model and real
5429: * if in the model but excluded because missing values, but how to get k from ij?*/
5430: for(j=ij+1; j<= cptcovt; j++){
5431: Tvaraff[j]=0;
5432: Tmodelind[j]=0;
5433: }
5434: for(j=ntveff+1; j<= cptcovt; j++){
5435: TmodelInvind[j]=0;
5436: }
5437: /* To be sorted */
5438: ;
5439: }
1.126 brouard 5440:
1.145 brouard 5441:
1.126 brouard 5442: /*********** Health Expectancies ****************/
5443:
1.235 brouard 5444: 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 5445:
5446: {
5447: /* Health expectancies, no variances */
1.164 brouard 5448: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5449: int nhstepma, nstepma; /* Decreasing with age */
5450: double age, agelim, hf;
5451: double ***p3mat;
5452: double eip;
5453:
1.238 brouard 5454: /* pstamp(ficreseij); */
1.126 brouard 5455: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5456: fprintf(ficreseij,"# Age");
5457: for(i=1; i<=nlstate;i++){
5458: for(j=1; j<=nlstate;j++){
5459: fprintf(ficreseij," e%1d%1d ",i,j);
5460: }
5461: fprintf(ficreseij," e%1d. ",i);
5462: }
5463: fprintf(ficreseij,"\n");
5464:
5465:
5466: if(estepm < stepm){
5467: printf ("Problem %d lower than %d\n",estepm, stepm);
5468: }
5469: else hstepm=estepm;
5470: /* We compute the life expectancy from trapezoids spaced every estepm months
5471: * This is mainly to measure the difference between two models: for example
5472: * if stepm=24 months pijx are given only every 2 years and by summing them
5473: * we are calculating an estimate of the Life Expectancy assuming a linear
5474: * progression in between and thus overestimating or underestimating according
5475: * to the curvature of the survival function. If, for the same date, we
5476: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5477: * to compare the new estimate of Life expectancy with the same linear
5478: * hypothesis. A more precise result, taking into account a more precise
5479: * curvature will be obtained if estepm is as small as stepm. */
5480:
5481: /* For example we decided to compute the life expectancy with the smallest unit */
5482: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5483: nhstepm is the number of hstepm from age to agelim
5484: nstepm is the number of stepm from age to agelin.
1.270 brouard 5485: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5486: and note for a fixed period like estepm months */
5487: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5488: survival function given by stepm (the optimization length). Unfortunately it
5489: means that if the survival funtion is printed only each two years of age and if
5490: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5491: results. So we changed our mind and took the option of the best precision.
5492: */
5493: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5494:
5495: agelim=AGESUP;
5496: /* If stepm=6 months */
5497: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5498: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5499:
5500: /* nhstepm age range expressed in number of stepm */
5501: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5502: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5503: /* if (stepm >= YEARM) hstepm=1;*/
5504: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5505: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5506:
5507: for (age=bage; age<=fage; age ++){
5508: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5509: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5510: /* if (stepm >= YEARM) hstepm=1;*/
5511: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5512:
5513: /* If stepm=6 months */
5514: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5515: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5516:
1.235 brouard 5517: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5518:
5519: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5520:
5521: printf("%d|",(int)age);fflush(stdout);
5522: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5523:
5524: /* Computing expectancies */
5525: for(i=1; i<=nlstate;i++)
5526: for(j=1; j<=nlstate;j++)
5527: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5528: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5529:
5530: /* 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]);*/
5531:
5532: }
5533:
5534: fprintf(ficreseij,"%3.0f",age );
5535: for(i=1; i<=nlstate;i++){
5536: eip=0;
5537: for(j=1; j<=nlstate;j++){
5538: eip +=eij[i][j][(int)age];
5539: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5540: }
5541: fprintf(ficreseij,"%9.4f", eip );
5542: }
5543: fprintf(ficreseij,"\n");
5544:
5545: }
5546: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5547: printf("\n");
5548: fprintf(ficlog,"\n");
5549:
5550: }
5551:
1.235 brouard 5552: 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 5553:
5554: {
5555: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5556: to initial status i, ei. .
1.126 brouard 5557: */
5558: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5559: int nhstepma, nstepma; /* Decreasing with age */
5560: double age, agelim, hf;
5561: double ***p3matp, ***p3matm, ***varhe;
5562: double **dnewm,**doldm;
5563: double *xp, *xm;
5564: double **gp, **gm;
5565: double ***gradg, ***trgradg;
5566: int theta;
5567:
5568: double eip, vip;
5569:
5570: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5571: xp=vector(1,npar);
5572: xm=vector(1,npar);
5573: dnewm=matrix(1,nlstate*nlstate,1,npar);
5574: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5575:
5576: pstamp(ficresstdeij);
5577: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5578: fprintf(ficresstdeij,"# Age");
5579: for(i=1; i<=nlstate;i++){
5580: for(j=1; j<=nlstate;j++)
5581: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5582: fprintf(ficresstdeij," e%1d. ",i);
5583: }
5584: fprintf(ficresstdeij,"\n");
5585:
5586: pstamp(ficrescveij);
5587: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5588: fprintf(ficrescveij,"# Age");
5589: for(i=1; i<=nlstate;i++)
5590: for(j=1; j<=nlstate;j++){
5591: cptj= (j-1)*nlstate+i;
5592: for(i2=1; i2<=nlstate;i2++)
5593: for(j2=1; j2<=nlstate;j2++){
5594: cptj2= (j2-1)*nlstate+i2;
5595: if(cptj2 <= cptj)
5596: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5597: }
5598: }
5599: fprintf(ficrescveij,"\n");
5600:
5601: if(estepm < stepm){
5602: printf ("Problem %d lower than %d\n",estepm, stepm);
5603: }
5604: else hstepm=estepm;
5605: /* We compute the life expectancy from trapezoids spaced every estepm months
5606: * This is mainly to measure the difference between two models: for example
5607: * if stepm=24 months pijx are given only every 2 years and by summing them
5608: * we are calculating an estimate of the Life Expectancy assuming a linear
5609: * progression in between and thus overestimating or underestimating according
5610: * to the curvature of the survival function. If, for the same date, we
5611: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5612: * to compare the new estimate of Life expectancy with the same linear
5613: * hypothesis. A more precise result, taking into account a more precise
5614: * curvature will be obtained if estepm is as small as stepm. */
5615:
5616: /* For example we decided to compute the life expectancy with the smallest unit */
5617: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5618: nhstepm is the number of hstepm from age to agelim
5619: nstepm is the number of stepm from age to agelin.
5620: Look at hpijx to understand the reason of that which relies in memory size
5621: and note for a fixed period like estepm months */
5622: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5623: survival function given by stepm (the optimization length). Unfortunately it
5624: means that if the survival funtion is printed only each two years of age and if
5625: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5626: results. So we changed our mind and took the option of the best precision.
5627: */
5628: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5629:
5630: /* If stepm=6 months */
5631: /* nhstepm age range expressed in number of stepm */
5632: agelim=AGESUP;
5633: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5634: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5635: /* if (stepm >= YEARM) hstepm=1;*/
5636: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5637:
5638: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5639: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5640: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5641: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5642: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5643: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5644:
5645: for (age=bage; age<=fage; age ++){
5646: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5647: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5648: /* if (stepm >= YEARM) hstepm=1;*/
5649: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5650:
1.126 brouard 5651: /* If stepm=6 months */
5652: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5653: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5654:
5655: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5656:
1.126 brouard 5657: /* Computing Variances of health expectancies */
5658: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5659: decrease memory allocation */
5660: for(theta=1; theta <=npar; theta++){
5661: for(i=1; i<=npar; i++){
1.222 brouard 5662: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5663: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5664: }
1.235 brouard 5665: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5666: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5667:
1.126 brouard 5668: for(j=1; j<= nlstate; j++){
1.222 brouard 5669: for(i=1; i<=nlstate; i++){
5670: for(h=0; h<=nhstepm-1; h++){
5671: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5672: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5673: }
5674: }
1.126 brouard 5675: }
1.218 brouard 5676:
1.126 brouard 5677: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5678: for(h=0; h<=nhstepm-1; h++){
5679: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5680: }
1.126 brouard 5681: }/* End theta */
5682:
5683:
5684: for(h=0; h<=nhstepm-1; h++)
5685: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5686: for(theta=1; theta <=npar; theta++)
5687: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5688:
1.218 brouard 5689:
1.222 brouard 5690: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5691: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5692: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5693:
1.222 brouard 5694: printf("%d|",(int)age);fflush(stdout);
5695: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5696: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5697: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5698: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5699: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5700: for(ij=1;ij<=nlstate*nlstate;ij++)
5701: for(ji=1;ji<=nlstate*nlstate;ji++)
5702: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5703: }
5704: }
1.218 brouard 5705:
1.126 brouard 5706: /* Computing expectancies */
1.235 brouard 5707: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5708: for(i=1; i<=nlstate;i++)
5709: for(j=1; j<=nlstate;j++)
1.222 brouard 5710: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5711: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5712:
1.222 brouard 5713: /* 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 5714:
1.222 brouard 5715: }
1.269 brouard 5716:
5717: /* Standard deviation of expectancies ij */
1.126 brouard 5718: fprintf(ficresstdeij,"%3.0f",age );
5719: for(i=1; i<=nlstate;i++){
5720: eip=0.;
5721: vip=0.;
5722: for(j=1; j<=nlstate;j++){
1.222 brouard 5723: eip += eij[i][j][(int)age];
5724: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5725: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5726: 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 5727: }
5728: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5729: }
5730: fprintf(ficresstdeij,"\n");
1.218 brouard 5731:
1.269 brouard 5732: /* Variance of expectancies ij */
1.126 brouard 5733: fprintf(ficrescveij,"%3.0f",age );
5734: for(i=1; i<=nlstate;i++)
5735: for(j=1; j<=nlstate;j++){
1.222 brouard 5736: cptj= (j-1)*nlstate+i;
5737: for(i2=1; i2<=nlstate;i2++)
5738: for(j2=1; j2<=nlstate;j2++){
5739: cptj2= (j2-1)*nlstate+i2;
5740: if(cptj2 <= cptj)
5741: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5742: }
1.126 brouard 5743: }
5744: fprintf(ficrescveij,"\n");
1.218 brouard 5745:
1.126 brouard 5746: }
5747: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5748: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5749: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5750: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5751: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5752: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5753: printf("\n");
5754: fprintf(ficlog,"\n");
1.218 brouard 5755:
1.126 brouard 5756: free_vector(xm,1,npar);
5757: free_vector(xp,1,npar);
5758: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5759: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5760: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5761: }
1.218 brouard 5762:
1.126 brouard 5763: /************ Variance ******************/
1.235 brouard 5764: 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 5765: {
5766: /* Variance of health expectancies */
5767: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5768: /* double **newm;*/
5769: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5770:
5771: /* int movingaverage(); */
5772: double **dnewm,**doldm;
5773: double **dnewmp,**doldmp;
5774: int i, j, nhstepm, hstepm, h, nstepm ;
5775: int k;
5776: double *xp;
5777: double **gp, **gm; /* for var eij */
5778: double ***gradg, ***trgradg; /*for var eij */
5779: double **gradgp, **trgradgp; /* for var p point j */
5780: double *gpp, *gmp; /* for var p point j */
5781: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5782: double ***p3mat;
5783: double age,agelim, hf;
5784: /* double ***mobaverage; */
5785: int theta;
5786: char digit[4];
5787: char digitp[25];
5788:
5789: char fileresprobmorprev[FILENAMELENGTH];
5790:
5791: if(popbased==1){
5792: if(mobilav!=0)
5793: strcpy(digitp,"-POPULBASED-MOBILAV_");
5794: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5795: }
5796: else
5797: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5798:
1.218 brouard 5799: /* if (mobilav!=0) { */
5800: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5801: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5802: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5803: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5804: /* } */
5805: /* } */
5806:
5807: strcpy(fileresprobmorprev,"PRMORPREV-");
5808: sprintf(digit,"%-d",ij);
5809: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5810: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5811: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5812: strcat(fileresprobmorprev,fileresu);
5813: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5814: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5815: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5816: }
5817: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5818: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5819: pstamp(ficresprobmorprev);
5820: 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 5821: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5822: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5823: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5824: }
5825: for(j=1;j<=cptcoveff;j++)
5826: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5827: fprintf(ficresprobmorprev,"\n");
5828:
1.218 brouard 5829: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5830: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5831: fprintf(ficresprobmorprev," p.%-d SE",j);
5832: for(i=1; i<=nlstate;i++)
5833: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5834: }
5835: fprintf(ficresprobmorprev,"\n");
5836:
5837: fprintf(ficgp,"\n# Routine varevsij");
5838: fprintf(ficgp,"\nunset title \n");
5839: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5840: 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");
5841: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5842: /* } */
5843: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5844: pstamp(ficresvij);
5845: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5846: if(popbased==1)
5847: 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);
5848: else
5849: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5850: fprintf(ficresvij,"# Age");
5851: for(i=1; i<=nlstate;i++)
5852: for(j=1; j<=nlstate;j++)
5853: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5854: fprintf(ficresvij,"\n");
5855:
5856: xp=vector(1,npar);
5857: dnewm=matrix(1,nlstate,1,npar);
5858: doldm=matrix(1,nlstate,1,nlstate);
5859: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5860: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5861:
5862: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5863: gpp=vector(nlstate+1,nlstate+ndeath);
5864: gmp=vector(nlstate+1,nlstate+ndeath);
5865: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5866:
1.218 brouard 5867: if(estepm < stepm){
5868: printf ("Problem %d lower than %d\n",estepm, stepm);
5869: }
5870: else hstepm=estepm;
5871: /* For example we decided to compute the life expectancy with the smallest unit */
5872: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5873: nhstepm is the number of hstepm from age to agelim
5874: nstepm is the number of stepm from age to agelim.
5875: Look at function hpijx to understand why because of memory size limitations,
5876: we decided (b) to get a life expectancy respecting the most precise curvature of the
5877: survival function given by stepm (the optimization length). Unfortunately it
5878: means that if the survival funtion is printed every two years of age and if
5879: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5880: results. So we changed our mind and took the option of the best precision.
5881: */
5882: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5883: agelim = AGESUP;
5884: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5885: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5886: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5887: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5888: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5889: gp=matrix(0,nhstepm,1,nlstate);
5890: gm=matrix(0,nhstepm,1,nlstate);
5891:
5892:
5893: for(theta=1; theta <=npar; theta++){
5894: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5895: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5896: }
5897:
1.242 brouard 5898: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5899:
5900: if (popbased==1) {
5901: if(mobilav ==0){
5902: for(i=1; i<=nlstate;i++)
5903: prlim[i][i]=probs[(int)age][i][ij];
5904: }else{ /* mobilav */
5905: for(i=1; i<=nlstate;i++)
5906: prlim[i][i]=mobaverage[(int)age][i][ij];
5907: }
5908: }
5909:
1.235 brouard 5910: 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 5911: for(j=1; j<= nlstate; j++){
5912: for(h=0; h<=nhstepm; h++){
5913: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5914: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5915: }
5916: }
5917: /* Next for computing probability of death (h=1 means
5918: computed over hstepm matrices product = hstepm*stepm months)
5919: as a weighted average of prlim.
5920: */
5921: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5922: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5923: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5924: }
5925: /* end probability of death */
5926:
5927: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5928: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5929:
1.242 brouard 5930: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5931:
5932: if (popbased==1) {
5933: if(mobilav ==0){
5934: for(i=1; i<=nlstate;i++)
5935: prlim[i][i]=probs[(int)age][i][ij];
5936: }else{ /* mobilav */
5937: for(i=1; i<=nlstate;i++)
5938: prlim[i][i]=mobaverage[(int)age][i][ij];
5939: }
5940: }
5941:
1.235 brouard 5942: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5943:
5944: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5945: for(h=0; h<=nhstepm; h++){
5946: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5947: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5948: }
5949: }
5950: /* This for computing probability of death (h=1 means
5951: computed over hstepm matrices product = hstepm*stepm months)
5952: as a weighted average of prlim.
5953: */
5954: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5955: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5956: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5957: }
5958: /* end probability of death */
5959:
5960: for(j=1; j<= nlstate; j++) /* vareij */
5961: for(h=0; h<=nhstepm; h++){
5962: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5963: }
5964:
5965: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5966: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5967: }
5968:
5969: } /* End theta */
5970:
5971: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5972:
5973: for(h=0; h<=nhstepm; h++) /* veij */
5974: for(j=1; j<=nlstate;j++)
5975: for(theta=1; theta <=npar; theta++)
5976: trgradg[h][j][theta]=gradg[h][theta][j];
5977:
5978: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5979: for(theta=1; theta <=npar; theta++)
5980: trgradgp[j][theta]=gradgp[theta][j];
5981:
5982:
5983: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5984: for(i=1;i<=nlstate;i++)
5985: for(j=1;j<=nlstate;j++)
5986: vareij[i][j][(int)age] =0.;
5987:
5988: for(h=0;h<=nhstepm;h++){
5989: for(k=0;k<=nhstepm;k++){
5990: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5991: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5992: for(i=1;i<=nlstate;i++)
5993: for(j=1;j<=nlstate;j++)
5994: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5995: }
5996: }
5997:
5998: /* pptj */
5999: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6000: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6001: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6002: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6003: varppt[j][i]=doldmp[j][i];
6004: /* end ppptj */
6005: /* x centered again */
6006:
1.242 brouard 6007: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6008:
6009: if (popbased==1) {
6010: if(mobilav ==0){
6011: for(i=1; i<=nlstate;i++)
6012: prlim[i][i]=probs[(int)age][i][ij];
6013: }else{ /* mobilav */
6014: for(i=1; i<=nlstate;i++)
6015: prlim[i][i]=mobaverage[(int)age][i][ij];
6016: }
6017: }
6018:
6019: /* This for computing probability of death (h=1 means
6020: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6021: as a weighted average of prlim.
6022: */
1.235 brouard 6023: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6024: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6025: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6026: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6027: }
6028: /* end probability of death */
6029:
6030: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6031: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6032: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6033: for(i=1; i<=nlstate;i++){
6034: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6035: }
6036: }
6037: fprintf(ficresprobmorprev,"\n");
6038:
6039: fprintf(ficresvij,"%.0f ",age );
6040: for(i=1; i<=nlstate;i++)
6041: for(j=1; j<=nlstate;j++){
6042: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6043: }
6044: fprintf(ficresvij,"\n");
6045: free_matrix(gp,0,nhstepm,1,nlstate);
6046: free_matrix(gm,0,nhstepm,1,nlstate);
6047: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6048: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6049: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6050: } /* End age */
6051: free_vector(gpp,nlstate+1,nlstate+ndeath);
6052: free_vector(gmp,nlstate+1,nlstate+ndeath);
6053: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6054: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6055: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6056: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6057: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6058: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6059: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6060: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6061: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6062: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6063: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6064: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6065: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6066: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6067: 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);
6068: /* 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 6069: */
1.218 brouard 6070: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6071: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6072:
1.218 brouard 6073: free_vector(xp,1,npar);
6074: free_matrix(doldm,1,nlstate,1,nlstate);
6075: free_matrix(dnewm,1,nlstate,1,npar);
6076: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6077: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6078: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6079: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6080: fclose(ficresprobmorprev);
6081: fflush(ficgp);
6082: fflush(fichtm);
6083: } /* end varevsij */
1.126 brouard 6084:
6085: /************ Variance of prevlim ******************/
1.269 brouard 6086: 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 6087: {
1.205 brouard 6088: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6089: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6090:
1.268 brouard 6091: double **dnewmpar,**doldm;
1.126 brouard 6092: int i, j, nhstepm, hstepm;
6093: double *xp;
6094: double *gp, *gm;
6095: double **gradg, **trgradg;
1.208 brouard 6096: double **mgm, **mgp;
1.126 brouard 6097: double age,agelim;
6098: int theta;
6099:
6100: pstamp(ficresvpl);
6101: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6102: fprintf(ficresvpl,"# Age ");
6103: if(nresult >=1)
6104: fprintf(ficresvpl," Result# ");
1.126 brouard 6105: for(i=1; i<=nlstate;i++)
6106: fprintf(ficresvpl," %1d-%1d",i,i);
6107: fprintf(ficresvpl,"\n");
6108:
6109: xp=vector(1,npar);
1.268 brouard 6110: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6111: doldm=matrix(1,nlstate,1,nlstate);
6112:
6113: hstepm=1*YEARM; /* Every year of age */
6114: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6115: agelim = AGESUP;
6116: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6117: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6118: if (stepm >= YEARM) hstepm=1;
6119: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6120: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6121: mgp=matrix(1,npar,1,nlstate);
6122: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6123: gp=vector(1,nlstate);
6124: gm=vector(1,nlstate);
6125:
6126: for(theta=1; theta <=npar; theta++){
6127: for(i=1; i<=npar; i++){ /* Computes gradient */
6128: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6129: }
1.209 brouard 6130: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6131: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6132: else
1.235 brouard 6133: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6134: for(i=1;i<=nlstate;i++){
1.126 brouard 6135: gp[i] = prlim[i][i];
1.208 brouard 6136: mgp[theta][i] = prlim[i][i];
6137: }
1.126 brouard 6138: for(i=1; i<=npar; i++) /* Computes gradient */
6139: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6140: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6141: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6142: else
1.235 brouard 6143: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6144: for(i=1;i<=nlstate;i++){
1.126 brouard 6145: gm[i] = prlim[i][i];
1.208 brouard 6146: mgm[theta][i] = prlim[i][i];
6147: }
1.126 brouard 6148: for(i=1;i<=nlstate;i++)
6149: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6150: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6151: } /* End theta */
6152:
6153: trgradg =matrix(1,nlstate,1,npar);
6154:
6155: for(j=1; j<=nlstate;j++)
6156: for(theta=1; theta <=npar; theta++)
6157: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6158: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6159: /* printf("\nmgm mgp %d ",(int)age); */
6160: /* for(j=1; j<=nlstate;j++){ */
6161: /* printf(" %d ",j); */
6162: /* for(theta=1; theta <=npar; theta++) */
6163: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6164: /* printf("\n "); */
6165: /* } */
6166: /* } */
6167: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6168: /* printf("\n gradg %d ",(int)age); */
6169: /* for(j=1; j<=nlstate;j++){ */
6170: /* printf("%d ",j); */
6171: /* for(theta=1; theta <=npar; theta++) */
6172: /* printf("%d %lf ",theta,gradg[theta][j]); */
6173: /* printf("\n "); */
6174: /* } */
6175: /* } */
1.126 brouard 6176:
6177: for(i=1;i<=nlstate;i++)
6178: varpl[i][(int)age] =0.;
1.209 brouard 6179: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6180: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6181: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6182: }else{
1.268 brouard 6183: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6184: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6185: }
1.126 brouard 6186: for(i=1;i<=nlstate;i++)
6187: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6188:
6189: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6190: if(nresult >=1)
6191: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6192: for(i=1; i<=nlstate;i++)
6193: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6194: fprintf(ficresvpl,"\n");
6195: free_vector(gp,1,nlstate);
6196: free_vector(gm,1,nlstate);
1.208 brouard 6197: free_matrix(mgm,1,npar,1,nlstate);
6198: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6199: free_matrix(gradg,1,npar,1,nlstate);
6200: free_matrix(trgradg,1,nlstate,1,npar);
6201: } /* End age */
6202:
6203: free_vector(xp,1,npar);
6204: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6205: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6206:
6207: }
6208:
6209:
6210: /************ Variance of backprevalence limit ******************/
1.269 brouard 6211: 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 6212: {
6213: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6214: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6215:
6216: double **dnewmpar,**doldm;
6217: int i, j, nhstepm, hstepm;
6218: double *xp;
6219: double *gp, *gm;
6220: double **gradg, **trgradg;
6221: double **mgm, **mgp;
6222: double age,agelim;
6223: int theta;
6224:
6225: pstamp(ficresvbl);
6226: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6227: fprintf(ficresvbl,"# Age ");
6228: if(nresult >=1)
6229: fprintf(ficresvbl," Result# ");
6230: for(i=1; i<=nlstate;i++)
6231: fprintf(ficresvbl," %1d-%1d",i,i);
6232: fprintf(ficresvbl,"\n");
6233:
6234: xp=vector(1,npar);
6235: dnewmpar=matrix(1,nlstate,1,npar);
6236: doldm=matrix(1,nlstate,1,nlstate);
6237:
6238: hstepm=1*YEARM; /* Every year of age */
6239: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6240: agelim = AGEINF;
6241: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6242: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6243: if (stepm >= YEARM) hstepm=1;
6244: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6245: gradg=matrix(1,npar,1,nlstate);
6246: mgp=matrix(1,npar,1,nlstate);
6247: mgm=matrix(1,npar,1,nlstate);
6248: gp=vector(1,nlstate);
6249: gm=vector(1,nlstate);
6250:
6251: for(theta=1; theta <=npar; theta++){
6252: for(i=1; i<=npar; i++){ /* Computes gradient */
6253: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6254: }
6255: if(mobilavproj > 0 )
6256: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6257: else
6258: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6259: for(i=1;i<=nlstate;i++){
6260: gp[i] = bprlim[i][i];
6261: mgp[theta][i] = bprlim[i][i];
6262: }
6263: for(i=1; i<=npar; i++) /* Computes gradient */
6264: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6265: if(mobilavproj > 0 )
6266: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6267: else
6268: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6269: for(i=1;i<=nlstate;i++){
6270: gm[i] = bprlim[i][i];
6271: mgm[theta][i] = bprlim[i][i];
6272: }
6273: for(i=1;i<=nlstate;i++)
6274: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6275: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6276: } /* End theta */
6277:
6278: trgradg =matrix(1,nlstate,1,npar);
6279:
6280: for(j=1; j<=nlstate;j++)
6281: for(theta=1; theta <=npar; theta++)
6282: trgradg[j][theta]=gradg[theta][j];
6283: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6284: /* printf("\nmgm mgp %d ",(int)age); */
6285: /* for(j=1; j<=nlstate;j++){ */
6286: /* printf(" %d ",j); */
6287: /* for(theta=1; theta <=npar; theta++) */
6288: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6289: /* printf("\n "); */
6290: /* } */
6291: /* } */
6292: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6293: /* printf("\n gradg %d ",(int)age); */
6294: /* for(j=1; j<=nlstate;j++){ */
6295: /* printf("%d ",j); */
6296: /* for(theta=1; theta <=npar; theta++) */
6297: /* printf("%d %lf ",theta,gradg[theta][j]); */
6298: /* printf("\n "); */
6299: /* } */
6300: /* } */
6301:
6302: for(i=1;i<=nlstate;i++)
6303: varbpl[i][(int)age] =0.;
6304: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6305: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6306: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6307: }else{
6308: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6309: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6310: }
6311: for(i=1;i<=nlstate;i++)
6312: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6313:
6314: fprintf(ficresvbl,"%.0f ",age );
6315: if(nresult >=1)
6316: fprintf(ficresvbl,"%d ",nres );
6317: for(i=1; i<=nlstate;i++)
6318: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6319: fprintf(ficresvbl,"\n");
6320: free_vector(gp,1,nlstate);
6321: free_vector(gm,1,nlstate);
6322: free_matrix(mgm,1,npar,1,nlstate);
6323: free_matrix(mgp,1,npar,1,nlstate);
6324: free_matrix(gradg,1,npar,1,nlstate);
6325: free_matrix(trgradg,1,nlstate,1,npar);
6326: } /* End age */
6327:
6328: free_vector(xp,1,npar);
6329: free_matrix(doldm,1,nlstate,1,npar);
6330: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6331:
6332: }
6333:
6334: /************ Variance of one-step probabilities ******************/
6335: 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 6336: {
6337: int i, j=0, k1, l1, tj;
6338: int k2, l2, j1, z1;
6339: int k=0, l;
6340: int first=1, first1, first2;
6341: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6342: double **dnewm,**doldm;
6343: double *xp;
6344: double *gp, *gm;
6345: double **gradg, **trgradg;
6346: double **mu;
6347: double age, cov[NCOVMAX+1];
6348: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6349: int theta;
6350: char fileresprob[FILENAMELENGTH];
6351: char fileresprobcov[FILENAMELENGTH];
6352: char fileresprobcor[FILENAMELENGTH];
6353: double ***varpij;
6354:
6355: strcpy(fileresprob,"PROB_");
6356: strcat(fileresprob,fileres);
6357: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6358: printf("Problem with resultfile: %s\n", fileresprob);
6359: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6360: }
6361: strcpy(fileresprobcov,"PROBCOV_");
6362: strcat(fileresprobcov,fileresu);
6363: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6364: printf("Problem with resultfile: %s\n", fileresprobcov);
6365: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6366: }
6367: strcpy(fileresprobcor,"PROBCOR_");
6368: strcat(fileresprobcor,fileresu);
6369: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6370: printf("Problem with resultfile: %s\n", fileresprobcor);
6371: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6372: }
6373: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6374: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6375: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6376: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6377: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6378: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6379: pstamp(ficresprob);
6380: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6381: fprintf(ficresprob,"# Age");
6382: pstamp(ficresprobcov);
6383: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6384: fprintf(ficresprobcov,"# Age");
6385: pstamp(ficresprobcor);
6386: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6387: fprintf(ficresprobcor,"# Age");
1.126 brouard 6388:
6389:
1.222 brouard 6390: for(i=1; i<=nlstate;i++)
6391: for(j=1; j<=(nlstate+ndeath);j++){
6392: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6393: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6394: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6395: }
6396: /* fprintf(ficresprob,"\n");
6397: fprintf(ficresprobcov,"\n");
6398: fprintf(ficresprobcor,"\n");
6399: */
6400: xp=vector(1,npar);
6401: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6402: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6403: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6404: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6405: first=1;
6406: fprintf(ficgp,"\n# Routine varprob");
6407: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6408: fprintf(fichtm,"\n");
6409:
1.266 brouard 6410: 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 6411: 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);
6412: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6413: and drawn. It helps understanding how is the covariance between two incidences.\
6414: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6415: 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 6416: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6417: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6418: standard deviations wide on each axis. <br>\
6419: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6420: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6421: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6422:
1.222 brouard 6423: cov[1]=1;
6424: /* tj=cptcoveff; */
1.225 brouard 6425: tj = (int) pow(2,cptcoveff);
1.222 brouard 6426: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6427: j1=0;
1.224 brouard 6428: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6429: if (cptcovn>0) {
6430: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6431: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6432: fprintf(ficresprob, "**********\n#\n");
6433: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6434: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6435: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6436:
1.222 brouard 6437: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6438: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6439: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6440:
6441:
1.222 brouard 6442: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6443: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6444: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6445:
1.222 brouard 6446: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6447: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6448: fprintf(ficresprobcor, "**********\n#");
6449: if(invalidvarcomb[j1]){
6450: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6451: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6452: continue;
6453: }
6454: }
6455: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6456: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6457: gp=vector(1,(nlstate)*(nlstate+ndeath));
6458: gm=vector(1,(nlstate)*(nlstate+ndeath));
6459: for (age=bage; age<=fage; age ++){
6460: cov[2]=age;
6461: if(nagesqr==1)
6462: cov[3]= age*age;
6463: for (k=1; k<=cptcovn;k++) {
6464: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6465: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6466: * 1 1 1 1 1
6467: * 2 2 1 1 1
6468: * 3 1 2 1 1
6469: */
6470: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6471: }
6472: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6473: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6474: for (k=1; k<=cptcovprod;k++)
6475: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6476:
6477:
1.222 brouard 6478: for(theta=1; theta <=npar; theta++){
6479: for(i=1; i<=npar; i++)
6480: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6481:
1.222 brouard 6482: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6483:
1.222 brouard 6484: k=0;
6485: for(i=1; i<= (nlstate); i++){
6486: for(j=1; j<=(nlstate+ndeath);j++){
6487: k=k+1;
6488: gp[k]=pmmij[i][j];
6489: }
6490: }
1.220 brouard 6491:
1.222 brouard 6492: for(i=1; i<=npar; i++)
6493: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6494:
1.222 brouard 6495: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6496: k=0;
6497: for(i=1; i<=(nlstate); i++){
6498: for(j=1; j<=(nlstate+ndeath);j++){
6499: k=k+1;
6500: gm[k]=pmmij[i][j];
6501: }
6502: }
1.220 brouard 6503:
1.222 brouard 6504: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6505: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6506: }
1.126 brouard 6507:
1.222 brouard 6508: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6509: for(theta=1; theta <=npar; theta++)
6510: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6511:
1.222 brouard 6512: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6513: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6514:
1.222 brouard 6515: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6516:
1.222 brouard 6517: k=0;
6518: for(i=1; i<=(nlstate); i++){
6519: for(j=1; j<=(nlstate+ndeath);j++){
6520: k=k+1;
6521: mu[k][(int) age]=pmmij[i][j];
6522: }
6523: }
6524: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6525: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6526: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6527:
1.222 brouard 6528: /*printf("\n%d ",(int)age);
6529: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6530: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6531: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6532: }*/
1.220 brouard 6533:
1.222 brouard 6534: fprintf(ficresprob,"\n%d ",(int)age);
6535: fprintf(ficresprobcov,"\n%d ",(int)age);
6536: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6537:
1.222 brouard 6538: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6539: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6540: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6541: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6542: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6543: }
6544: i=0;
6545: for (k=1; k<=(nlstate);k++){
6546: for (l=1; l<=(nlstate+ndeath);l++){
6547: i++;
6548: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6549: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6550: for (j=1; j<=i;j++){
6551: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6552: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6553: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6554: }
6555: }
6556: }/* end of loop for state */
6557: } /* end of loop for age */
6558: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6559: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6560: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6561: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6562:
6563: /* Confidence intervalle of pij */
6564: /*
6565: fprintf(ficgp,"\nunset parametric;unset label");
6566: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6567: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6568: 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);
6569: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6570: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6571: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6572: */
6573:
6574: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6575: first1=1;first2=2;
6576: for (k2=1; k2<=(nlstate);k2++){
6577: for (l2=1; l2<=(nlstate+ndeath);l2++){
6578: if(l2==k2) continue;
6579: j=(k2-1)*(nlstate+ndeath)+l2;
6580: for (k1=1; k1<=(nlstate);k1++){
6581: for (l1=1; l1<=(nlstate+ndeath);l1++){
6582: if(l1==k1) continue;
6583: i=(k1-1)*(nlstate+ndeath)+l1;
6584: if(i<=j) continue;
6585: for (age=bage; age<=fage; age ++){
6586: if ((int)age %5==0){
6587: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6588: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6589: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6590: mu1=mu[i][(int) age]/stepm*YEARM ;
6591: mu2=mu[j][(int) age]/stepm*YEARM;
6592: c12=cv12/sqrt(v1*v2);
6593: /* Computing eigen value of matrix of covariance */
6594: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6595: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6596: if ((lc2 <0) || (lc1 <0) ){
6597: if(first2==1){
6598: first1=0;
6599: 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);
6600: }
6601: 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);
6602: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6603: /* lc2=fabs(lc2); */
6604: }
1.220 brouard 6605:
1.222 brouard 6606: /* Eigen vectors */
6607: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6608: /*v21=sqrt(1.-v11*v11); *//* error */
6609: v21=(lc1-v1)/cv12*v11;
6610: v12=-v21;
6611: v22=v11;
6612: tnalp=v21/v11;
6613: if(first1==1){
6614: first1=0;
6615: 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);
6616: }
6617: 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);
6618: /*printf(fignu*/
6619: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6620: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6621: if(first==1){
6622: first=0;
6623: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6624: fprintf(ficgp,"\nset parametric;unset label");
6625: 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);
6626: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6627: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6628: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6629: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6630: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6631: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6632: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6633: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6634: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6635: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6636: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6637: 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 6638: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6639: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6640: }else{
6641: first=0;
6642: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6643: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6644: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6645: 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 6646: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6647: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6648: }/* if first */
6649: } /* age mod 5 */
6650: } /* end loop age */
6651: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6652: first=1;
6653: } /*l12 */
6654: } /* k12 */
6655: } /*l1 */
6656: }/* k1 */
6657: } /* loop on combination of covariates j1 */
6658: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6659: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6660: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6661: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6662: free_vector(xp,1,npar);
6663: fclose(ficresprob);
6664: fclose(ficresprobcov);
6665: fclose(ficresprobcor);
6666: fflush(ficgp);
6667: fflush(fichtmcov);
6668: }
1.126 brouard 6669:
6670:
6671: /******************* Printing html file ***********/
1.201 brouard 6672: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6673: int lastpass, int stepm, int weightopt, char model[],\
6674: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6675: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6676: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6677: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6678: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6679:
6680: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6681: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6682: </ul>");
1.237 brouard 6683: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6684: </ul>", model);
1.214 brouard 6685: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6686: 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",
6687: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6688: 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 6689: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6690: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6691: fprintf(fichtm,"\
6692: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6693: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6694: fprintf(fichtm,"\
1.217 brouard 6695: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6696: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6697: fprintf(fichtm,"\
1.126 brouard 6698: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6699: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6700: fprintf(fichtm,"\
1.217 brouard 6701: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6702: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6703: fprintf(fichtm,"\
1.211 brouard 6704: - (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 6705: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6706: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6707: if(prevfcast==1){
6708: fprintf(fichtm,"\
6709: - Prevalence projections by age and states: \
1.201 brouard 6710: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6711: }
1.126 brouard 6712:
6713:
1.225 brouard 6714: m=pow(2,cptcoveff);
1.222 brouard 6715: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6716:
1.264 brouard 6717: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6718:
6719: jj1=0;
6720:
6721: fprintf(fichtm," \n<ul>");
6722: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6723: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6724: if(m != 1 && TKresult[nres]!= k1)
6725: continue;
6726: jj1++;
6727: if (cptcovn > 0) {
6728: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6729: for (cpt=1; cpt<=cptcoveff;cpt++){
6730: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6731: }
6732: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6733: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6734: }
6735: fprintf(fichtm,"\">");
6736:
6737: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6738: fprintf(fichtm,"************ Results for covariates");
6739: for (cpt=1; cpt<=cptcoveff;cpt++){
6740: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6741: }
6742: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6743: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6744: }
6745: if(invalidvarcomb[k1]){
6746: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6747: continue;
6748: }
6749: fprintf(fichtm,"</a></li>");
6750: } /* cptcovn >0 */
6751: }
6752: fprintf(fichtm," \n</ul>");
6753:
1.222 brouard 6754: jj1=0;
1.237 brouard 6755:
6756: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6757: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6758: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6759: continue;
1.220 brouard 6760:
1.222 brouard 6761: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6762: jj1++;
6763: if (cptcovn > 0) {
1.264 brouard 6764: fprintf(fichtm,"\n<p><a name=\"rescov");
6765: for (cpt=1; cpt<=cptcoveff;cpt++){
6766: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6767: }
6768: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6769: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6770: }
6771: fprintf(fichtm,"\"</a>");
6772:
1.222 brouard 6773: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6774: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6775: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6776: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6777: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6778: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6779: }
1.237 brouard 6780: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6781: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6782: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6783: }
6784:
1.230 brouard 6785: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6786: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6787: if(invalidvarcomb[k1]){
6788: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6789: printf("\nCombination (%d) ignored because no cases \n",k1);
6790: continue;
6791: }
6792: }
6793: /* aij, bij */
1.259 brouard 6794: 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 6795: <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 6796: /* Pij */
1.241 brouard 6797: 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> \
6798: <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 6799: /* Quasi-incidences */
6800: 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 6801: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6802: 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 6803: 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> \
6804: <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 6805: /* Survival functions (period) in state j */
6806: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6807: 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> \
6808: <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 6809: }
6810: /* State specific survival functions (period) */
6811: for(cpt=1; cpt<=nlstate;cpt++){
6812: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6813: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6814: <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 6815: }
6816: /* Period (stable) prevalence in each health state */
6817: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6818: 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> \
6819: <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 6820: }
6821: if(backcast==1){
6822: /* Period (stable) back prevalence in each health state */
6823: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6824: 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 6825: <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 6826: }
1.217 brouard 6827: }
1.222 brouard 6828: if(prevfcast==1){
6829: /* Projection of prevalence up to period (stable) prevalence in each health state */
6830: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6831: 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> \
6832: <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 6833: }
6834: }
1.268 brouard 6835: if(backcast==1){
6836: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6837: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6838: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6839: 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 \
6840: 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) \
6841: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6842: <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 6843: }
6844: }
1.220 brouard 6845:
1.222 brouard 6846: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6847: 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> \
6848: <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 6849: }
6850: /* } /\* end i1 *\/ */
6851: }/* End k1 */
6852: fprintf(fichtm,"</ul>");
1.126 brouard 6853:
1.222 brouard 6854: fprintf(fichtm,"\
1.126 brouard 6855: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6856: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6857: - 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 6858: But because parameters are usually highly correlated (a higher incidence of disability \
6859: and a higher incidence of recovery can give very close observed transition) it might \
6860: be very useful to look not only at linear confidence intervals estimated from the \
6861: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6862: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6863: covariance matrix of the one-step probabilities. \
6864: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6865:
1.222 brouard 6866: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6867: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6868: fprintf(fichtm,"\
1.126 brouard 6869: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6870: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6871:
1.222 brouard 6872: fprintf(fichtm,"\
1.126 brouard 6873: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6874: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6875: fprintf(fichtm,"\
1.126 brouard 6876: - 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): \
6877: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6878: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6879: fprintf(fichtm,"\
1.126 brouard 6880: - (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): \
6881: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6882: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6883: fprintf(fichtm,"\
1.128 brouard 6884: - 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 6885: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6886: fprintf(fichtm,"\
1.128 brouard 6887: - 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 6888: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6889: fprintf(fichtm,"\
1.126 brouard 6890: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6891: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6892:
6893: /* if(popforecast==1) fprintf(fichtm,"\n */
6894: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6895: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6896: /* <br>",fileres,fileres,fileres,fileres); */
6897: /* else */
6898: /* 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 6899: fflush(fichtm);
6900: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6901:
1.225 brouard 6902: m=pow(2,cptcoveff);
1.222 brouard 6903: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6904:
1.222 brouard 6905: jj1=0;
1.237 brouard 6906:
1.241 brouard 6907: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6908: for(k1=1; k1<=m;k1++){
1.253 brouard 6909: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6910: continue;
1.222 brouard 6911: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6912: jj1++;
1.126 brouard 6913: if (cptcovn > 0) {
6914: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6915: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6916: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6917: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6918: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6919: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6920: }
6921:
1.126 brouard 6922: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6923:
1.222 brouard 6924: if(invalidvarcomb[k1]){
6925: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6926: continue;
6927: }
1.126 brouard 6928: }
6929: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6930: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6931: 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 6932: <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 6933: }
6934: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6935: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6936: true period expectancies (those weighted with period prevalences are also\
6937: drawn in addition to the population based expectancies computed using\
1.241 brouard 6938: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6939: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6940: /* } /\* end i1 *\/ */
6941: }/* End k1 */
1.241 brouard 6942: }/* End nres */
1.222 brouard 6943: fprintf(fichtm,"</ul>");
6944: fflush(fichtm);
1.126 brouard 6945: }
6946:
6947: /******************* Gnuplot file **************/
1.270 brouard 6948: 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 6949:
6950: char dirfileres[132],optfileres[132];
1.264 brouard 6951: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6952: 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 6953: int lv=0, vlv=0, kl=0;
1.130 brouard 6954: int ng=0;
1.201 brouard 6955: int vpopbased;
1.223 brouard 6956: int ioffset; /* variable offset for columns */
1.270 brouard 6957: int iyearc=1; /* variable column for year of projection */
6958: int iagec=1; /* variable column for age of projection */
1.235 brouard 6959: int nres=0; /* Index of resultline */
1.266 brouard 6960: int istart=1; /* For starting graphs in projections */
1.219 brouard 6961:
1.126 brouard 6962: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6963: /* printf("Problem with file %s",optionfilegnuplot); */
6964: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6965: /* } */
6966:
6967: /*#ifdef windows */
6968: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6969: /*#endif */
1.225 brouard 6970: m=pow(2,cptcoveff);
1.126 brouard 6971:
1.274 brouard 6972: /* diagram of the model */
6973: fprintf(ficgp,"\n#Diagram of the model \n");
6974: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
6975: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
6976: 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);
6977:
6978: 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);
6979: fprintf(ficgp,"\n#show arrow\nunset label\n");
6980: 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);
6981: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
6982: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
6983: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
6984: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
6985:
1.202 brouard 6986: /* Contribution to likelihood */
6987: /* Plot the probability implied in the likelihood */
1.223 brouard 6988: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6989: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6990: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6991: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6992: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6993: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6994: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6995: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6996: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6997: 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));
6998: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6999: 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));
7000: for (i=1; i<= nlstate ; i ++) {
7001: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7002: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7003: 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);
7004: for (j=2; j<= nlstate+ndeath ; j ++) {
7005: 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);
7006: }
7007: fprintf(ficgp,";\nset out; unset ylabel;\n");
7008: }
7009: /* 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 */
7010: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7011: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7012: fprintf(ficgp,"\nset out;unset log\n");
7013: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7014:
1.126 brouard 7015: strcpy(dirfileres,optionfilefiname);
7016: strcpy(optfileres,"vpl");
1.223 brouard 7017: /* 1eme*/
1.238 brouard 7018: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7019: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7020: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7021: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7022: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7023: continue;
7024: /* We are interested in selected combination by the resultline */
1.246 brouard 7025: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7026: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7027: strcpy(gplotlabel,"(");
1.238 brouard 7028: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7029: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7030: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7031: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7032: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7033: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7034: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7035: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7036: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7037: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7038: }
7039: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7040: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7041: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7042: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7043: }
7044: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7045: /* printf("\n#\n"); */
1.238 brouard 7046: fprintf(ficgp,"\n#\n");
7047: if(invalidvarcomb[k1]){
1.260 brouard 7048: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7049: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7050: continue;
7051: }
1.235 brouard 7052:
1.241 brouard 7053: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7054: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 ! brouard 7055: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
! 7056: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7057: 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);
7058: /* 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); */
7059: /* k1-1 error should be nres-1*/
1.238 brouard 7060: for (i=1; i<= nlstate ; i ++) {
7061: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7062: else fprintf(ficgp," %%*lf (%%*lf)");
7063: }
1.260 brouard 7064: 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 7065: for (i=1; i<= nlstate ; i ++) {
7066: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7067: else fprintf(ficgp," %%*lf (%%*lf)");
7068: }
1.260 brouard 7069: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7070: for (i=1; i<= nlstate ; i ++) {
7071: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7072: else fprintf(ficgp," %%*lf (%%*lf)");
7073: }
1.265 brouard 7074: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7075:
7076: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7077: if(cptcoveff ==0){
1.271 brouard 7078: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7079: }else{
7080: kl=0;
7081: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7082: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7083: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7084: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7085: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7086: vlv= nbcode[Tvaraff[k]][lv];
7087: kl++;
7088: /* 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 *\/ */
7089: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7090: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7091: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7092: if(k==cptcoveff){
7093: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7094: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7095: }else{
7096: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7097: kl++;
7098: }
7099: } /* end covariate */
7100: } /* end if no covariate */
7101:
1.238 brouard 7102: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7103: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7104: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7105: if(cptcoveff ==0){
1.245 brouard 7106: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7107: }else{
7108: kl=0;
7109: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7110: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7111: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7112: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7113: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7114: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7115: kl++;
1.238 brouard 7116: /* 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 *\/ */
7117: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7118: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7119: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7120: if(k==cptcoveff){
1.245 brouard 7121: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7122: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7123: }else{
7124: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7125: kl++;
7126: }
7127: } /* end covariate */
7128: } /* end if no covariate */
1.268 brouard 7129: if(backcast == 1){
7130: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7131: /* k1-1 error should be nres-1*/
7132: for (i=1; i<= nlstate ; i ++) {
7133: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7134: else fprintf(ficgp," %%*lf (%%*lf)");
7135: }
1.271 brouard 7136: 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 7137: for (i=1; i<= nlstate ; i ++) {
7138: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7139: else fprintf(ficgp," %%*lf (%%*lf)");
7140: }
1.276 ! brouard 7141: 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 7142: for (i=1; i<= nlstate ; i ++) {
7143: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7144: else fprintf(ficgp," %%*lf (%%*lf)");
7145: }
1.274 brouard 7146: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7147: } /* end if backprojcast */
1.238 brouard 7148: } /* end if backcast */
1.276 ! brouard 7149: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
! 7150: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7151: } /* nres */
1.201 brouard 7152: } /* k1 */
7153: } /* cpt */
1.235 brouard 7154:
7155:
1.126 brouard 7156: /*2 eme*/
1.238 brouard 7157: for (k1=1; k1<= m ; k1 ++){
7158: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7159: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7160: continue;
7161: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7162: strcpy(gplotlabel,"(");
1.238 brouard 7163: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7164: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7165: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7166: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7167: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7168: vlv= nbcode[Tvaraff[k]][lv];
7169: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7170: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7171: }
1.237 brouard 7172: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7173: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7174: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7175: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7176: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7177: }
1.264 brouard 7178: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7179: fprintf(ficgp,"\n#\n");
1.223 brouard 7180: if(invalidvarcomb[k1]){
7181: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7182: continue;
7183: }
1.219 brouard 7184:
1.241 brouard 7185: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7186: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7187: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7188: if(vpopbased==0){
1.238 brouard 7189: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7190: }else
1.238 brouard 7191: fprintf(ficgp,"\nreplot ");
7192: for (i=1; i<= nlstate+1 ; i ++) {
7193: k=2*i;
1.261 brouard 7194: 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 7195: for (j=1; j<= nlstate+1 ; j ++) {
7196: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7197: else fprintf(ficgp," %%*lf (%%*lf)");
7198: }
7199: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7200: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7201: 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 7202: for (j=1; j<= nlstate+1 ; j ++) {
7203: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7204: else fprintf(ficgp," %%*lf (%%*lf)");
7205: }
7206: fprintf(ficgp,"\" t\"\" w l lt 0,");
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: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7213: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7214: } /* state */
7215: } /* vpopbased */
1.264 brouard 7216: 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 7217: } /* end nres */
7218: } /* k1 end 2 eme*/
7219:
7220:
7221: /*3eme*/
7222: for (k1=1; k1<= m ; k1 ++){
7223: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7224: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7225: continue;
7226:
7227: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7228: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7229: strcpy(gplotlabel,"(");
1.238 brouard 7230: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7231: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7232: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7233: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7234: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7235: vlv= nbcode[Tvaraff[k]][lv];
7236: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7237: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7238: }
7239: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7240: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7241: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7242: }
1.264 brouard 7243: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7244: fprintf(ficgp,"\n#\n");
7245: if(invalidvarcomb[k1]){
7246: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7247: continue;
7248: }
7249:
7250: /* k=2+nlstate*(2*cpt-2); */
7251: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7252: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7253: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7254: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7255: 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 7256: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7257: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7258: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7259: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7260: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7261: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7262:
1.238 brouard 7263: */
7264: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7265: 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 7266: /* 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 7267:
1.238 brouard 7268: }
1.261 brouard 7269: 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 7270: }
1.264 brouard 7271: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7272: } /* end nres */
7273: } /* end kl 3eme */
1.126 brouard 7274:
1.223 brouard 7275: /* 4eme */
1.201 brouard 7276: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7277: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7278: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7279: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7280: continue;
1.238 brouard 7281: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7282: strcpy(gplotlabel,"(");
1.238 brouard 7283: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7284: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7285: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7286: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7287: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7288: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7289: vlv= nbcode[Tvaraff[k]][lv];
7290: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7291: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7292: }
7293: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7294: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7295: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7296: }
1.264 brouard 7297: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7298: fprintf(ficgp,"\n#\n");
7299: if(invalidvarcomb[k1]){
7300: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7301: continue;
1.223 brouard 7302: }
1.238 brouard 7303:
1.241 brouard 7304: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7305: 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 7306: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7307: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7308: k=3;
7309: for (i=1; i<= nlstate ; i ++){
7310: if(i==1){
7311: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7312: }else{
7313: fprintf(ficgp,", '' ");
7314: }
7315: l=(nlstate+ndeath)*(i-1)+1;
7316: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7317: for (j=2; j<= nlstate+ndeath ; j ++)
7318: fprintf(ficgp,"+$%d",k+l+j-1);
7319: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7320: } /* nlstate */
1.264 brouard 7321: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7322: } /* end cpt state*/
7323: } /* end nres */
7324: } /* end covariate k1 */
7325:
1.220 brouard 7326: /* 5eme */
1.201 brouard 7327: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7328: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7329: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7330: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7331: continue;
1.238 brouard 7332: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7333: strcpy(gplotlabel,"(");
1.238 brouard 7334: 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);
7335: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7336: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7337: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7338: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7339: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7340: vlv= nbcode[Tvaraff[k]][lv];
7341: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7342: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7343: }
7344: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7345: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7346: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7347: }
1.264 brouard 7348: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7349: fprintf(ficgp,"\n#\n");
7350: if(invalidvarcomb[k1]){
7351: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7352: continue;
7353: }
1.227 brouard 7354:
1.241 brouard 7355: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7356: 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 7357: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7358: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7359: k=3;
7360: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7361: if(j==1)
7362: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7363: else
7364: fprintf(ficgp,", '' ");
7365: l=(nlstate+ndeath)*(cpt-1) +j;
7366: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7367: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7368: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7369: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7370: } /* nlstate */
7371: fprintf(ficgp,", '' ");
7372: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7373: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7374: l=(nlstate+ndeath)*(cpt-1) +j;
7375: if(j < nlstate)
7376: fprintf(ficgp,"$%d +",k+l);
7377: else
7378: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7379: }
1.264 brouard 7380: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7381: } /* end cpt state*/
7382: } /* end covariate */
7383: } /* end nres */
1.227 brouard 7384:
1.220 brouard 7385: /* 6eme */
1.202 brouard 7386: /* CV preval stable (period) for each covariate */
1.237 brouard 7387: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7388: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7389: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7390: continue;
1.255 brouard 7391: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7392: strcpy(gplotlabel,"(");
1.211 brouard 7393: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7394: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7395: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7396: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7397: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7398: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7399: vlv= nbcode[Tvaraff[k]][lv];
7400: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7401: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7402: }
1.237 brouard 7403: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7404: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7405: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7406: }
1.264 brouard 7407: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7408: fprintf(ficgp,"\n#\n");
1.223 brouard 7409: if(invalidvarcomb[k1]){
1.227 brouard 7410: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7411: continue;
1.223 brouard 7412: }
1.227 brouard 7413:
1.241 brouard 7414: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7415: 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 7416: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7417: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7418: k=3; /* Offset */
1.255 brouard 7419: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7420: if(i==1)
7421: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7422: else
7423: fprintf(ficgp,", '' ");
1.255 brouard 7424: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7425: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7426: for (j=2; j<= nlstate ; j ++)
7427: fprintf(ficgp,"+$%d",k+l+j-1);
7428: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7429: } /* nlstate */
1.264 brouard 7430: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7431: } /* end cpt state*/
7432: } /* end covariate */
1.227 brouard 7433:
7434:
1.220 brouard 7435: /* 7eme */
1.218 brouard 7436: if(backcast == 1){
1.217 brouard 7437: /* CV back preval stable (period) for each covariate */
1.237 brouard 7438: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7439: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7440: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7441: continue;
1.268 brouard 7442: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7443: strcpy(gplotlabel,"(");
7444: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7445: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7446: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7447: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7448: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7449: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7450: vlv= nbcode[Tvaraff[k]][lv];
7451: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7452: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7453: }
1.237 brouard 7454: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7455: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7456: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7457: }
1.264 brouard 7458: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7459: fprintf(ficgp,"\n#\n");
7460: if(invalidvarcomb[k1]){
7461: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7462: continue;
7463: }
7464:
1.241 brouard 7465: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7466: 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 7467: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7468: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7469: k=3; /* Offset */
1.268 brouard 7470: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7471: if(i==1)
7472: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7473: else
7474: fprintf(ficgp,", '' ");
7475: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7476: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7477: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7478: /* 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 7479: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7480: /* for (j=2; j<= nlstate ; j ++) */
7481: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7482: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7483: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7484: } /* nlstate */
1.264 brouard 7485: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7486: } /* end cpt state*/
7487: } /* end covariate */
7488: } /* End if backcast */
7489:
1.223 brouard 7490: /* 8eme */
1.218 brouard 7491: if(prevfcast==1){
7492: /* Projection from cross-sectional to stable (period) for each covariate */
7493:
1.237 brouard 7494: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7495: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7496: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7497: continue;
1.211 brouard 7498: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7499: strcpy(gplotlabel,"(");
1.227 brouard 7500: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7501: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7502: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7503: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7504: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7505: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7506: vlv= nbcode[Tvaraff[k]][lv];
7507: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7508: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7509: }
1.237 brouard 7510: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7511: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7512: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7513: }
1.264 brouard 7514: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7515: fprintf(ficgp,"\n#\n");
7516: if(invalidvarcomb[k1]){
7517: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7518: continue;
7519: }
7520:
7521: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7522: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7523: 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 7524: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7525: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7526:
7527: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7528: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7529: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7530: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7531: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7532: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7533: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7534: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7535: if(i==istart){
1.227 brouard 7536: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7537: }else{
7538: fprintf(ficgp,",\\\n '' ");
7539: }
7540: if(cptcoveff ==0){ /* No covariate */
7541: ioffset=2; /* Age is in 2 */
7542: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7543: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7544: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7545: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7546: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7547: if(i==nlstate+1){
1.270 brouard 7548: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7549: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7550: fprintf(ficgp,",\\\n '' ");
7551: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7552: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7553: offyear, \
1.268 brouard 7554: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7555: }else
1.227 brouard 7556: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7557: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7558: }else{ /* more than 2 covariates */
1.270 brouard 7559: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7560: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7561: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7562: iyearc=ioffset-1;
7563: iagec=ioffset;
1.227 brouard 7564: fprintf(ficgp," u %d:(",ioffset);
7565: kl=0;
7566: strcpy(gplotcondition,"(");
7567: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7568: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7569: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7570: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7571: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7572: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7573: kl++;
7574: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7575: kl++;
7576: if(k <cptcoveff && cptcoveff>1)
7577: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7578: }
7579: strcpy(gplotcondition+strlen(gplotcondition),")");
7580: /* 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 *\/ */
7581: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7582: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7583: /* '' 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*/
7584: if(i==nlstate+1){
1.270 brouard 7585: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7586: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7587: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7588: fprintf(ficgp," u %d:(",iagec);
7589: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7590: iyearc, iagec, offyear, \
7591: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7592: /* '' 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 7593: }else{
7594: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7595: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7596: }
7597: } /* end if covariate */
7598: } /* nlstate */
1.264 brouard 7599: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7600: } /* end cpt state*/
7601: } /* end covariate */
7602: } /* End if prevfcast */
1.227 brouard 7603:
1.268 brouard 7604: if(backcast==1){
7605: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7606:
7607: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7608: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7609: if(m != 1 && TKresult[nres]!= k1)
7610: continue;
7611: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7612: strcpy(gplotlabel,"(");
7613: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7614: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7615: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7616: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7617: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7618: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7619: vlv= nbcode[Tvaraff[k]][lv];
7620: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7621: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7622: }
7623: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7624: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7625: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7626: }
7627: strcpy(gplotlabel+strlen(gplotlabel),")");
7628: fprintf(ficgp,"\n#\n");
7629: if(invalidvarcomb[k1]){
7630: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7631: continue;
7632: }
7633:
7634: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7635: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7636: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7637: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7638: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7639:
7640: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7641: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7642: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7643: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7644: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7645: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7646: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7647: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7648: if(i==istart){
7649: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7650: }else{
7651: fprintf(ficgp,",\\\n '' ");
7652: }
7653: if(cptcoveff ==0){ /* No covariate */
7654: ioffset=2; /* Age is in 2 */
7655: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7656: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7657: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7658: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7659: fprintf(ficgp," u %d:(", ioffset);
7660: if(i==nlstate+1){
1.270 brouard 7661: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7662: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7663: fprintf(ficgp,",\\\n '' ");
7664: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7665: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7666: offbyear, \
7667: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7668: }else
7669: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7670: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7671: }else{ /* more than 2 covariates */
1.270 brouard 7672: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7673: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7674: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7675: iyearc=ioffset-1;
7676: iagec=ioffset;
1.268 brouard 7677: fprintf(ficgp," u %d:(",ioffset);
7678: kl=0;
7679: strcpy(gplotcondition,"(");
7680: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7681: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7682: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7683: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7684: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7685: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7686: kl++;
7687: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7688: kl++;
7689: if(k <cptcoveff && cptcoveff>1)
7690: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7691: }
7692: strcpy(gplotcondition+strlen(gplotcondition),")");
7693: /* 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 *\/ */
7694: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7695: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7696: /* '' 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*/
7697: if(i==nlstate+1){
1.270 brouard 7698: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7699: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7700: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7701: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7702: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7703: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7704: iyearc,iagec,offbyear, \
7705: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7706: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7707: }else{
7708: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7709: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7710: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7711: }
7712: } /* end if covariate */
7713: } /* nlstate */
7714: fprintf(ficgp,"\nset out; unset label;\n");
7715: } /* end cpt state*/
7716: } /* end covariate */
7717: } /* End if backcast */
7718:
1.227 brouard 7719:
1.238 brouard 7720: /* 9eme writing MLE parameters */
7721: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7722: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7723: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7724: for(k=1; k <=(nlstate+ndeath); k++){
7725: if (k != i) {
1.227 brouard 7726: fprintf(ficgp,"# current state %d\n",k);
7727: for(j=1; j <=ncovmodel; j++){
7728: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7729: jk++;
7730: }
7731: fprintf(ficgp,"\n");
1.126 brouard 7732: }
7733: }
1.223 brouard 7734: }
1.187 brouard 7735: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7736:
1.145 brouard 7737: /*goto avoid;*/
1.238 brouard 7738: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7739: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7740: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7741: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7742: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7743: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7744: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7745: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7746: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7747: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7748: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7749: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7750: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7751: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7752: fprintf(ficgp,"#\n");
1.223 brouard 7753: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7754: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7755: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7756: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7757: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7758: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7759: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7760: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7761: continue;
1.264 brouard 7762: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7763: strcpy(gplotlabel,"(");
1.276 ! brouard 7764: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7765: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7766: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7767: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7768: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7769: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7770: vlv= nbcode[Tvaraff[k]][lv];
7771: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7772: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7773: }
1.237 brouard 7774: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7775: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7776: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7777: }
1.264 brouard 7778: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7779: fprintf(ficgp,"\n#\n");
1.264 brouard 7780: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 ! brouard 7781: fprintf(ficgp,"\nset key outside ");
! 7782: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
! 7783: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7784: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7785: if (ng==1){
7786: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7787: fprintf(ficgp,"\nunset log y");
7788: }else if (ng==2){
7789: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7790: fprintf(ficgp,"\nset log y");
7791: }else if (ng==3){
7792: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7793: fprintf(ficgp,"\nset log y");
7794: }else
7795: fprintf(ficgp,"\nunset title ");
7796: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7797: i=1;
7798: for(k2=1; k2<=nlstate; k2++) {
7799: k3=i;
7800: for(k=1; k<=(nlstate+ndeath); k++) {
7801: if (k != k2){
7802: switch( ng) {
7803: case 1:
7804: if(nagesqr==0)
7805: fprintf(ficgp," p%d+p%d*x",i,i+1);
7806: else /* nagesqr =1 */
7807: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7808: break;
7809: case 2: /* ng=2 */
7810: if(nagesqr==0)
7811: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7812: else /* nagesqr =1 */
7813: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7814: break;
7815: case 3:
7816: if(nagesqr==0)
7817: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7818: else /* nagesqr =1 */
7819: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7820: break;
7821: }
7822: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7823: ijp=1; /* product no age */
7824: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7825: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7826: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7827: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7828: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7829: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7830: if(DummyV[j]==0){
7831: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7832: }else{ /* quantitative */
7833: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7834: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7835: }
7836: ij++;
1.237 brouard 7837: }
1.268 brouard 7838: }
7839: }else if(cptcovprod >0){
7840: if(j==Tprod[ijp]) { /* */
7841: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7842: if(ijp <=cptcovprod) { /* Product */
7843: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7844: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7845: /* 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)]); */
7846: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7847: }else{ /* Vn is dummy and Vm is quanti */
7848: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7849: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7850: }
7851: }else{ /* Vn*Vm Vn is quanti */
7852: if(DummyV[Tvard[ijp][2]]==0){
7853: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7854: }else{ /* Both quanti */
7855: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7856: }
1.237 brouard 7857: }
1.268 brouard 7858: ijp++;
1.237 brouard 7859: }
1.268 brouard 7860: } /* end Tprod */
1.237 brouard 7861: } else{ /* simple covariate */
1.264 brouard 7862: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7863: if(Dummy[j]==0){
7864: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7865: }else{ /* quantitative */
7866: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7867: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7868: }
1.237 brouard 7869: } /* end simple */
7870: } /* end j */
1.223 brouard 7871: }else{
7872: i=i-ncovmodel;
7873: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7874: fprintf(ficgp," (1.");
7875: }
1.227 brouard 7876:
1.223 brouard 7877: if(ng != 1){
7878: fprintf(ficgp,")/(1");
1.227 brouard 7879:
1.264 brouard 7880: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7881: if(nagesqr==0)
1.264 brouard 7882: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7883: else /* nagesqr =1 */
1.264 brouard 7884: 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 7885:
1.223 brouard 7886: ij=1;
7887: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7888: if(cptcovage >0){
7889: if((j-2)==Tage[ij]) { /* Bug valgrind */
7890: if(ij <=cptcovage) { /* Bug valgrind */
7891: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7892: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7893: ij++;
7894: }
7895: }
7896: }else
7897: 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 7898: }
7899: fprintf(ficgp,")");
7900: }
7901: fprintf(ficgp,")");
7902: if(ng ==2)
1.276 ! brouard 7903: 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 7904: else /* ng= 3 */
1.276 ! brouard 7905: 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 7906: }else{ /* end ng <> 1 */
7907: if( k !=k2) /* logit p11 is hard to draw */
1.276 ! brouard 7908: 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 7909: }
7910: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7911: fprintf(ficgp,",");
7912: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7913: fprintf(ficgp,",");
7914: i=i+ncovmodel;
7915: } /* end k */
7916: } /* end k2 */
1.276 ! brouard 7917: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
! 7918: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 7919: } /* end k1 */
1.223 brouard 7920: } /* end ng */
7921: /* avoid: */
7922: fflush(ficgp);
1.126 brouard 7923: } /* end gnuplot */
7924:
7925:
7926: /*************** Moving average **************/
1.219 brouard 7927: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7928: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7929:
1.222 brouard 7930: int i, cpt, cptcod;
7931: int modcovmax =1;
7932: int mobilavrange, mob;
7933: int iage=0;
7934:
1.266 brouard 7935: double sum=0., sumr=0.;
1.222 brouard 7936: double age;
1.266 brouard 7937: double *sumnewp, *sumnewm, *sumnewmr;
7938: double *agemingood, *agemaxgood;
7939: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7940:
7941:
1.225 brouard 7942: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7943: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7944:
7945: sumnewp = vector(1,ncovcombmax);
7946: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7947: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7948: agemingood = vector(1,ncovcombmax);
1.266 brouard 7949: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7950: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7951: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7952:
7953: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7954: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7955: sumnewp[cptcod]=0.;
1.266 brouard 7956: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
7957: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7958: }
7959: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7960:
1.266 brouard 7961: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7962: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7963: else mobilavrange=mobilav;
7964: for (age=bage; age<=fage; age++)
7965: for (i=1; i<=nlstate;i++)
7966: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7967: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7968: /* We keep the original values on the extreme ages bage, fage and for
7969: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7970: we use a 5 terms etc. until the borders are no more concerned.
7971: */
7972: for (mob=3;mob <=mobilavrange;mob=mob+2){
7973: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 7974: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7975: sumnewm[cptcod]=0.;
7976: for (i=1; i<=nlstate;i++){
1.222 brouard 7977: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7978: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7979: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7980: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7981: }
7982: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 7983: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7984: } /* end i */
7985: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
7986: } /* end cptcod */
1.222 brouard 7987: }/* end age */
7988: }/* end mob */
1.266 brouard 7989: }else{
7990: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 7991: return -1;
1.266 brouard 7992: }
7993:
7994: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 7995: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7996: if(invalidvarcomb[cptcod]){
7997: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7998: continue;
7999: }
1.219 brouard 8000:
1.266 brouard 8001: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8002: sumnewm[cptcod]=0.;
8003: sumnewmr[cptcod]=0.;
8004: for (i=1; i<=nlstate;i++){
8005: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8006: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8007: }
8008: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8009: agemingoodr[cptcod]=age;
8010: }
8011: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8012: agemingood[cptcod]=age;
8013: }
8014: } /* age */
8015: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8016: sumnewm[cptcod]=0.;
1.266 brouard 8017: sumnewmr[cptcod]=0.;
1.222 brouard 8018: for (i=1; i<=nlstate;i++){
8019: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8020: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8021: }
8022: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8023: agemaxgoodr[cptcod]=age;
1.222 brouard 8024: }
8025: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8026: agemaxgood[cptcod]=age;
8027: }
8028: } /* age */
8029: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8030: /* but they will change */
8031: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8032: sumnewm[cptcod]=0.;
8033: sumnewmr[cptcod]=0.;
8034: for (i=1; i<=nlstate;i++){
8035: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8036: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8037: }
8038: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8039: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8040: agemaxgoodr[cptcod]=age; /* age min */
8041: for (i=1; i<=nlstate;i++)
8042: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8043: }else{ /* bad we change the value with the values of good ages */
8044: for (i=1; i<=nlstate;i++){
8045: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8046: } /* i */
8047: } /* end bad */
8048: }else{
8049: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8050: agemaxgood[cptcod]=age;
8051: }else{ /* bad we change the value with the values of good ages */
8052: for (i=1; i<=nlstate;i++){
8053: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8054: } /* i */
8055: } /* end bad */
8056: }/* end else */
8057: sum=0.;sumr=0.;
8058: for (i=1; i<=nlstate;i++){
8059: sum+=mobaverage[(int)age][i][cptcod];
8060: sumr+=probs[(int)age][i][cptcod];
8061: }
8062: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8063: 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 8064: } /* end bad */
8065: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8066: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8067: 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 8068: } /* end bad */
8069: }/* age */
1.266 brouard 8070:
8071: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8072: sumnewm[cptcod]=0.;
1.266 brouard 8073: sumnewmr[cptcod]=0.;
1.222 brouard 8074: for (i=1; i<=nlstate;i++){
8075: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8076: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8077: }
8078: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8079: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8080: agemingoodr[cptcod]=age;
8081: for (i=1; i<=nlstate;i++)
8082: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8083: }else{ /* bad we change the value with the values of good ages */
8084: for (i=1; i<=nlstate;i++){
8085: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8086: } /* i */
8087: } /* end bad */
8088: }else{
8089: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8090: agemingood[cptcod]=age;
8091: }else{ /* bad */
8092: for (i=1; i<=nlstate;i++){
8093: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8094: } /* i */
8095: } /* end bad */
8096: }/* end else */
8097: sum=0.;sumr=0.;
8098: for (i=1; i<=nlstate;i++){
8099: sum+=mobaverage[(int)age][i][cptcod];
8100: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8101: }
1.266 brouard 8102: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8103: 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 8104: } /* end bad */
8105: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8106: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8107: 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 8108: } /* end bad */
8109: }/* age */
1.266 brouard 8110:
1.222 brouard 8111:
8112: for (age=bage; age<=fage; age++){
1.235 brouard 8113: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8114: sumnewp[cptcod]=0.;
8115: sumnewm[cptcod]=0.;
8116: for (i=1; i<=nlstate;i++){
8117: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8118: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8119: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8120: }
8121: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8122: }
8123: /* printf("\n"); */
8124: /* } */
1.266 brouard 8125:
1.222 brouard 8126: /* brutal averaging */
1.266 brouard 8127: /* for (i=1; i<=nlstate;i++){ */
8128: /* for (age=1; age<=bage; age++){ */
8129: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8130: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8131: /* } */
8132: /* for (age=fage; age<=AGESUP; age++){ */
8133: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8134: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8135: /* } */
8136: /* } /\* end i status *\/ */
8137: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8138: /* for (age=1; age<=AGESUP; age++){ */
8139: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8140: /* mobaverage[(int)age][i][cptcod]=0.; */
8141: /* } */
8142: /* } */
1.222 brouard 8143: }/* end cptcod */
1.266 brouard 8144: free_vector(agemaxgoodr,1, ncovcombmax);
8145: free_vector(agemaxgood,1, ncovcombmax);
8146: free_vector(agemingood,1, ncovcombmax);
8147: free_vector(agemingoodr,1, ncovcombmax);
8148: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8149: free_vector(sumnewm,1, ncovcombmax);
8150: free_vector(sumnewp,1, ncovcombmax);
8151: return 0;
8152: }/* End movingaverage */
1.218 brouard 8153:
1.126 brouard 8154:
8155: /************** Forecasting ******************/
1.269 brouard 8156: 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 8157: /* proj1, year, month, day of starting projection
8158: agemin, agemax range of age
8159: dateprev1 dateprev2 range of dates during which prevalence is computed
8160: anproj2 year of en of projection (same day and month as proj1).
8161: */
1.267 brouard 8162: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8163: double agec; /* generic age */
8164: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8165: double *popeffectif,*popcount;
8166: double ***p3mat;
1.218 brouard 8167: /* double ***mobaverage; */
1.126 brouard 8168: char fileresf[FILENAMELENGTH];
8169:
8170: agelim=AGESUP;
1.211 brouard 8171: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8172: in each health status at the date of interview (if between dateprev1 and dateprev2).
8173: We still use firstpass and lastpass as another selection.
8174: */
1.214 brouard 8175: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8176: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8177:
1.201 brouard 8178: strcpy(fileresf,"F_");
8179: strcat(fileresf,fileresu);
1.126 brouard 8180: if((ficresf=fopen(fileresf,"w"))==NULL) {
8181: printf("Problem with forecast resultfile: %s\n", fileresf);
8182: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8183: }
1.235 brouard 8184: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8185: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8186:
1.225 brouard 8187: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8188:
8189:
8190: stepsize=(int) (stepm+YEARM-1)/YEARM;
8191: if (stepm<=12) stepsize=1;
8192: if(estepm < stepm){
8193: printf ("Problem %d lower than %d\n",estepm, stepm);
8194: }
1.270 brouard 8195: else{
8196: hstepm=estepm;
8197: }
8198: if(estepm > stepm){ /* Yes every two year */
8199: stepsize=2;
8200: }
1.126 brouard 8201:
8202: hstepm=hstepm/stepm;
8203: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8204: fractional in yp1 */
8205: anprojmean=yp;
8206: yp2=modf((yp1*12),&yp);
8207: mprojmean=yp;
8208: yp1=modf((yp2*30.5),&yp);
8209: jprojmean=yp;
8210: if(jprojmean==0) jprojmean=1;
8211: if(mprojmean==0) jprojmean=1;
8212:
1.227 brouard 8213: i1=pow(2,cptcoveff);
1.126 brouard 8214: if (cptcovn < 1){i1=1;}
8215:
8216: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8217:
8218: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8219:
1.126 brouard 8220: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8221: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8222: for(k=1; k<=i1;k++){
1.253 brouard 8223: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8224: continue;
1.227 brouard 8225: if(invalidvarcomb[k]){
8226: printf("\nCombination (%d) projection ignored because no cases \n",k);
8227: continue;
8228: }
8229: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8230: for(j=1;j<=cptcoveff;j++) {
8231: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8232: }
1.235 brouard 8233: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8234: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8235: }
1.227 brouard 8236: fprintf(ficresf," yearproj age");
8237: for(j=1; j<=nlstate+ndeath;j++){
8238: for(i=1; i<=nlstate;i++)
8239: fprintf(ficresf," p%d%d",i,j);
8240: fprintf(ficresf," wp.%d",j);
8241: }
8242: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8243: fprintf(ficresf,"\n");
8244: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8245: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8246: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8247: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8248: nhstepm = nhstepm/hstepm;
8249: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8250: oldm=oldms;savm=savms;
1.268 brouard 8251: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8252: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8253: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8254: for (h=0; h<=nhstepm; h++){
8255: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8256: break;
8257: }
8258: }
8259: fprintf(ficresf,"\n");
8260: for(j=1;j<=cptcoveff;j++)
8261: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8262: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8263:
8264: for(j=1; j<=nlstate+ndeath;j++) {
8265: ppij=0.;
8266: for(i=1; i<=nlstate;i++) {
8267: /* if (mobilav>=1) */
1.269 brouard 8268: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
1.268 brouard 8269: /* else { */ /* even if mobilav==-1 we use mobaverage */
8270: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8271: /* } */
8272: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8273: } /* end i */
8274: fprintf(ficresf," %.3f", ppij);
8275: }/* end j */
1.227 brouard 8276: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8277: } /* end agec */
1.266 brouard 8278: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8279: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8280: } /* end yearp */
8281: } /* end k */
1.219 brouard 8282:
1.126 brouard 8283: fclose(ficresf);
1.215 brouard 8284: printf("End of Computing forecasting \n");
8285: fprintf(ficlog,"End of Computing forecasting\n");
8286:
1.126 brouard 8287: }
8288:
1.269 brouard 8289: /************** Back Forecasting ******************/
8290: 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 8291: /* back1, year, month, day of starting backection
8292: agemin, agemax range of age
8293: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8294: anback2 year of end of backprojection (same day and month as back1).
8295: prevacurrent and prev are prevalences.
1.267 brouard 8296: */
8297: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8298: double agec; /* generic age */
1.268 brouard 8299: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8300: double *popeffectif,*popcount;
8301: double ***p3mat;
8302: /* double ***mobaverage; */
8303: char fileresfb[FILENAMELENGTH];
8304:
1.268 brouard 8305: agelim=AGEINF;
1.267 brouard 8306: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8307: in each health status at the date of interview (if between dateprev1 and dateprev2).
8308: We still use firstpass and lastpass as another selection.
8309: */
8310: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8311: /* firstpass, lastpass, stepm, weightopt, model); */
8312:
8313: /*Do we need to compute prevalence again?*/
8314:
8315: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8316:
8317: strcpy(fileresfb,"FB_");
8318: strcat(fileresfb,fileresu);
8319: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8320: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8321: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8322: }
8323: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8324: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8325:
8326: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8327:
8328:
8329: stepsize=(int) (stepm+YEARM-1)/YEARM;
8330: if (stepm<=12) stepsize=1;
8331: if(estepm < stepm){
8332: printf ("Problem %d lower than %d\n",estepm, stepm);
8333: }
1.270 brouard 8334: else{
8335: hstepm=estepm;
8336: }
8337: if(estepm >= stepm){ /* Yes every two year */
8338: stepsize=2;
8339: }
1.267 brouard 8340:
8341: hstepm=hstepm/stepm;
8342: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8343: fractional in yp1 */
8344: anprojmean=yp;
8345: yp2=modf((yp1*12),&yp);
8346: mprojmean=yp;
8347: yp1=modf((yp2*30.5),&yp);
8348: jprojmean=yp;
8349: if(jprojmean==0) jprojmean=1;
8350: if(mprojmean==0) jprojmean=1;
8351:
8352: i1=pow(2,cptcoveff);
8353: if (cptcovn < 1){i1=1;}
8354:
8355: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8356: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8357:
8358: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8359:
8360: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8361: for(k=1; k<=i1;k++){
8362: if(i1 != 1 && TKresult[nres]!= k)
8363: continue;
8364: if(invalidvarcomb[k]){
8365: printf("\nCombination (%d) projection ignored because no cases \n",k);
8366: continue;
8367: }
1.268 brouard 8368: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8369: for(j=1;j<=cptcoveff;j++) {
8370: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8371: }
8372: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8373: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8374: }
8375: fprintf(ficresfb," yearbproj age");
8376: for(j=1; j<=nlstate+ndeath;j++){
8377: for(i=1; i<=nlstate;i++)
1.268 brouard 8378: fprintf(ficresfb," b%d%d",i,j);
8379: fprintf(ficresfb," b.%d",j);
1.267 brouard 8380: }
8381: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8382: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8383: fprintf(ficresfb,"\n");
8384: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8385: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8386: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8387: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8388: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8389: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8390: nhstepm = nhstepm/hstepm;
8391: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8392: oldm=oldms;savm=savms;
1.268 brouard 8393: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8394: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8395: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8396: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8397: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8398: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8399: for (h=0; h<=nhstepm; h++){
1.268 brouard 8400: if (h*hstepm/YEARM*stepm ==-yearp) {
8401: break;
8402: }
8403: }
8404: fprintf(ficresfb,"\n");
8405: for(j=1;j<=cptcoveff;j++)
8406: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8407: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8408: for(i=1; i<=nlstate+ndeath;i++) {
8409: ppij=0.;ppi=0.;
8410: for(j=1; j<=nlstate;j++) {
8411: /* if (mobilav==1) */
1.269 brouard 8412: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8413: ppi=ppi+prevacurrent[(int)agec][j][k];
8414: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8415: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8416: /* else { */
8417: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8418: /* } */
1.268 brouard 8419: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8420: } /* end j */
8421: if(ppi <0.99){
8422: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8423: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8424: }
8425: fprintf(ficresfb," %.3f", ppij);
8426: }/* end j */
1.267 brouard 8427: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8428: } /* end agec */
8429: } /* end yearp */
8430: } /* end k */
1.217 brouard 8431:
1.267 brouard 8432: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8433:
1.267 brouard 8434: fclose(ficresfb);
8435: printf("End of Computing Back forecasting \n");
8436: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8437:
1.267 brouard 8438: }
1.217 brouard 8439:
1.269 brouard 8440: /* Variance of prevalence limit: varprlim */
8441: 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){
8442: /*------- Variance of period (stable) prevalence------*/
8443:
8444: char fileresvpl[FILENAMELENGTH];
8445: FILE *ficresvpl;
8446: double **oldm, **savm;
8447: double **varpl; /* Variances of prevalence limits by age */
8448: int i1, k, nres, j ;
8449:
8450: strcpy(fileresvpl,"VPL_");
8451: strcat(fileresvpl,fileresu);
8452: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8453: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8454: exit(0);
8455: }
8456: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8457: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8458:
8459: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8460: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8461:
8462: i1=pow(2,cptcoveff);
8463: if (cptcovn < 1){i1=1;}
8464:
8465: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8466: for(k=1; k<=i1;k++){
8467: if(i1 != 1 && TKresult[nres]!= k)
8468: continue;
8469: fprintf(ficresvpl,"\n#****** ");
8470: printf("\n#****** ");
8471: fprintf(ficlog,"\n#****** ");
8472: for(j=1;j<=cptcoveff;j++) {
8473: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8474: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8475: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8476: }
8477: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8478: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8479: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8480: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8481: }
8482: fprintf(ficresvpl,"******\n");
8483: printf("******\n");
8484: fprintf(ficlog,"******\n");
8485:
8486: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8487: oldm=oldms;savm=savms;
8488: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8489: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8490: /*}*/
8491: }
8492:
8493: fclose(ficresvpl);
8494: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8495: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8496:
8497: }
8498: /* Variance of back prevalence: varbprlim */
8499: 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){
8500: /*------- Variance of back (stable) prevalence------*/
8501:
8502: char fileresvbl[FILENAMELENGTH];
8503: FILE *ficresvbl;
8504:
8505: double **oldm, **savm;
8506: double **varbpl; /* Variances of back prevalence limits by age */
8507: int i1, k, nres, j ;
8508:
8509: strcpy(fileresvbl,"VBL_");
8510: strcat(fileresvbl,fileresu);
8511: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8512: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8513: exit(0);
8514: }
8515: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8516: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8517:
8518:
8519: i1=pow(2,cptcoveff);
8520: if (cptcovn < 1){i1=1;}
8521:
8522: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8523: for(k=1; k<=i1;k++){
8524: if(i1 != 1 && TKresult[nres]!= k)
8525: continue;
8526: fprintf(ficresvbl,"\n#****** ");
8527: printf("\n#****** ");
8528: fprintf(ficlog,"\n#****** ");
8529: for(j=1;j<=cptcoveff;j++) {
8530: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8531: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8532: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8533: }
8534: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8535: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8536: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8537: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8538: }
8539: fprintf(ficresvbl,"******\n");
8540: printf("******\n");
8541: fprintf(ficlog,"******\n");
8542:
8543: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8544: oldm=oldms;savm=savms;
8545:
8546: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8547: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8548: /*}*/
8549: }
8550:
8551: fclose(ficresvbl);
8552: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8553: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8554:
8555: } /* End of varbprlim */
8556:
1.126 brouard 8557: /************** Forecasting *****not tested NB*************/
1.227 brouard 8558: /* 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 8559:
1.227 brouard 8560: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8561: /* int *popage; */
8562: /* double calagedatem, agelim, kk1, kk2; */
8563: /* double *popeffectif,*popcount; */
8564: /* double ***p3mat,***tabpop,***tabpopprev; */
8565: /* /\* double ***mobaverage; *\/ */
8566: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8567:
1.227 brouard 8568: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8569: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8570: /* agelim=AGESUP; */
8571: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8572:
1.227 brouard 8573: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8574:
8575:
1.227 brouard 8576: /* strcpy(filerespop,"POP_"); */
8577: /* strcat(filerespop,fileresu); */
8578: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8579: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8580: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8581: /* } */
8582: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8583: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8584:
1.227 brouard 8585: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8586:
1.227 brouard 8587: /* /\* if (mobilav!=0) { *\/ */
8588: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8589: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8590: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8591: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8592: /* /\* } *\/ */
8593: /* /\* } *\/ */
1.126 brouard 8594:
1.227 brouard 8595: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8596: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8597:
1.227 brouard 8598: /* agelim=AGESUP; */
1.126 brouard 8599:
1.227 brouard 8600: /* hstepm=1; */
8601: /* hstepm=hstepm/stepm; */
1.218 brouard 8602:
1.227 brouard 8603: /* if (popforecast==1) { */
8604: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8605: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8606: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8607: /* } */
8608: /* popage=ivector(0,AGESUP); */
8609: /* popeffectif=vector(0,AGESUP); */
8610: /* popcount=vector(0,AGESUP); */
1.126 brouard 8611:
1.227 brouard 8612: /* i=1; */
8613: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8614:
1.227 brouard 8615: /* imx=i; */
8616: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8617: /* } */
1.218 brouard 8618:
1.227 brouard 8619: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8620: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8621: /* k=k+1; */
8622: /* fprintf(ficrespop,"\n#******"); */
8623: /* for(j=1;j<=cptcoveff;j++) { */
8624: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8625: /* } */
8626: /* fprintf(ficrespop,"******\n"); */
8627: /* fprintf(ficrespop,"# Age"); */
8628: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8629: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8630:
1.227 brouard 8631: /* for (cpt=0; cpt<=0;cpt++) { */
8632: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8633:
1.227 brouard 8634: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8635: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8636: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8637:
1.227 brouard 8638: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8639: /* oldm=oldms;savm=savms; */
8640: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8641:
1.227 brouard 8642: /* for (h=0; h<=nhstepm; h++){ */
8643: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8644: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8645: /* } */
8646: /* for(j=1; j<=nlstate+ndeath;j++) { */
8647: /* kk1=0.;kk2=0; */
8648: /* for(i=1; i<=nlstate;i++) { */
8649: /* if (mobilav==1) */
8650: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8651: /* else { */
8652: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8653: /* } */
8654: /* } */
8655: /* if (h==(int)(calagedatem+12*cpt)){ */
8656: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8657: /* /\*fprintf(ficrespop," %.3f", kk1); */
8658: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8659: /* } */
8660: /* } */
8661: /* for(i=1; i<=nlstate;i++){ */
8662: /* kk1=0.; */
8663: /* for(j=1; j<=nlstate;j++){ */
8664: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8665: /* } */
8666: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8667: /* } */
1.218 brouard 8668:
1.227 brouard 8669: /* if (h==(int)(calagedatem+12*cpt)) */
8670: /* for(j=1; j<=nlstate;j++) */
8671: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8672: /* } */
8673: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8674: /* } */
8675: /* } */
1.218 brouard 8676:
1.227 brouard 8677: /* /\******\/ */
1.218 brouard 8678:
1.227 brouard 8679: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8680: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8681: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8682: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8683: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8684:
1.227 brouard 8685: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8686: /* oldm=oldms;savm=savms; */
8687: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8688: /* for (h=0; h<=nhstepm; h++){ */
8689: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8690: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8691: /* } */
8692: /* for(j=1; j<=nlstate+ndeath;j++) { */
8693: /* kk1=0.;kk2=0; */
8694: /* for(i=1; i<=nlstate;i++) { */
8695: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8696: /* } */
8697: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8698: /* } */
8699: /* } */
8700: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8701: /* } */
8702: /* } */
8703: /* } */
8704: /* } */
1.218 brouard 8705:
1.227 brouard 8706: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8707:
1.227 brouard 8708: /* if (popforecast==1) { */
8709: /* free_ivector(popage,0,AGESUP); */
8710: /* free_vector(popeffectif,0,AGESUP); */
8711: /* free_vector(popcount,0,AGESUP); */
8712: /* } */
8713: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8714: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8715: /* fclose(ficrespop); */
8716: /* } /\* End of popforecast *\/ */
1.218 brouard 8717:
1.126 brouard 8718: int fileappend(FILE *fichier, char *optionfich)
8719: {
8720: if((fichier=fopen(optionfich,"a"))==NULL) {
8721: printf("Problem with file: %s\n", optionfich);
8722: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8723: return (0);
8724: }
8725: fflush(fichier);
8726: return (1);
8727: }
8728:
8729:
8730: /**************** function prwizard **********************/
8731: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8732: {
8733:
8734: /* Wizard to print covariance matrix template */
8735:
1.164 brouard 8736: char ca[32], cb[32];
8737: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8738: int numlinepar;
8739:
8740: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8741: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8742: for(i=1; i <=nlstate; i++){
8743: jj=0;
8744: for(j=1; j <=nlstate+ndeath; j++){
8745: if(j==i) continue;
8746: jj++;
8747: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8748: printf("%1d%1d",i,j);
8749: fprintf(ficparo,"%1d%1d",i,j);
8750: for(k=1; k<=ncovmodel;k++){
8751: /* printf(" %lf",param[i][j][k]); */
8752: /* fprintf(ficparo," %lf",param[i][j][k]); */
8753: printf(" 0.");
8754: fprintf(ficparo," 0.");
8755: }
8756: printf("\n");
8757: fprintf(ficparo,"\n");
8758: }
8759: }
8760: printf("# Scales (for hessian or gradient estimation)\n");
8761: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8762: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8763: for(i=1; i <=nlstate; i++){
8764: jj=0;
8765: for(j=1; j <=nlstate+ndeath; j++){
8766: if(j==i) continue;
8767: jj++;
8768: fprintf(ficparo,"%1d%1d",i,j);
8769: printf("%1d%1d",i,j);
8770: fflush(stdout);
8771: for(k=1; k<=ncovmodel;k++){
8772: /* printf(" %le",delti3[i][j][k]); */
8773: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8774: printf(" 0.");
8775: fprintf(ficparo," 0.");
8776: }
8777: numlinepar++;
8778: printf("\n");
8779: fprintf(ficparo,"\n");
8780: }
8781: }
8782: printf("# Covariance matrix\n");
8783: /* # 121 Var(a12)\n\ */
8784: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8785: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8786: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8787: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8788: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8789: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8790: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8791: fflush(stdout);
8792: fprintf(ficparo,"# Covariance matrix\n");
8793: /* # 121 Var(a12)\n\ */
8794: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8795: /* # ...\n\ */
8796: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8797:
8798: for(itimes=1;itimes<=2;itimes++){
8799: jj=0;
8800: for(i=1; i <=nlstate; i++){
8801: for(j=1; j <=nlstate+ndeath; j++){
8802: if(j==i) continue;
8803: for(k=1; k<=ncovmodel;k++){
8804: jj++;
8805: ca[0]= k+'a'-1;ca[1]='\0';
8806: if(itimes==1){
8807: printf("#%1d%1d%d",i,j,k);
8808: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8809: }else{
8810: printf("%1d%1d%d",i,j,k);
8811: fprintf(ficparo,"%1d%1d%d",i,j,k);
8812: /* printf(" %.5le",matcov[i][j]); */
8813: }
8814: ll=0;
8815: for(li=1;li <=nlstate; li++){
8816: for(lj=1;lj <=nlstate+ndeath; lj++){
8817: if(lj==li) continue;
8818: for(lk=1;lk<=ncovmodel;lk++){
8819: ll++;
8820: if(ll<=jj){
8821: cb[0]= lk +'a'-1;cb[1]='\0';
8822: if(ll<jj){
8823: if(itimes==1){
8824: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8825: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8826: }else{
8827: printf(" 0.");
8828: fprintf(ficparo," 0.");
8829: }
8830: }else{
8831: if(itimes==1){
8832: printf(" Var(%s%1d%1d)",ca,i,j);
8833: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8834: }else{
8835: printf(" 0.");
8836: fprintf(ficparo," 0.");
8837: }
8838: }
8839: }
8840: } /* end lk */
8841: } /* end lj */
8842: } /* end li */
8843: printf("\n");
8844: fprintf(ficparo,"\n");
8845: numlinepar++;
8846: } /* end k*/
8847: } /*end j */
8848: } /* end i */
8849: } /* end itimes */
8850:
8851: } /* end of prwizard */
8852: /******************* Gompertz Likelihood ******************************/
8853: double gompertz(double x[])
8854: {
8855: double A,B,L=0.0,sump=0.,num=0.;
8856: int i,n=0; /* n is the size of the sample */
8857:
1.220 brouard 8858: for (i=1;i<=imx ; i++) {
1.126 brouard 8859: sump=sump+weight[i];
8860: /* sump=sump+1;*/
8861: num=num+1;
8862: }
8863:
8864:
8865: /* for (i=0; i<=imx; i++)
8866: 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]);*/
8867:
8868: for (i=1;i<=imx ; i++)
8869: {
8870: if (cens[i] == 1 && wav[i]>1)
8871: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8872:
8873: if (cens[i] == 0 && wav[i]>1)
8874: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8875: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8876:
8877: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8878: if (wav[i] > 1 ) { /* ??? */
8879: L=L+A*weight[i];
8880: /* 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]);*/
8881: }
8882: }
8883:
8884: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8885:
8886: return -2*L*num/sump;
8887: }
8888:
1.136 brouard 8889: #ifdef GSL
8890: /******************* Gompertz_f Likelihood ******************************/
8891: double gompertz_f(const gsl_vector *v, void *params)
8892: {
8893: double A,B,LL=0.0,sump=0.,num=0.;
8894: double *x= (double *) v->data;
8895: int i,n=0; /* n is the size of the sample */
8896:
8897: for (i=0;i<=imx-1 ; i++) {
8898: sump=sump+weight[i];
8899: /* sump=sump+1;*/
8900: num=num+1;
8901: }
8902:
8903:
8904: /* for (i=0; i<=imx; i++)
8905: 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]);*/
8906: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8907: for (i=1;i<=imx ; i++)
8908: {
8909: if (cens[i] == 1 && wav[i]>1)
8910: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8911:
8912: if (cens[i] == 0 && wav[i]>1)
8913: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8914: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8915:
8916: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8917: if (wav[i] > 1 ) { /* ??? */
8918: LL=LL+A*weight[i];
8919: /* 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]);*/
8920: }
8921: }
8922:
8923: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8924: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8925:
8926: return -2*LL*num/sump;
8927: }
8928: #endif
8929:
1.126 brouard 8930: /******************* Printing html file ***********/
1.201 brouard 8931: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8932: int lastpass, int stepm, int weightopt, char model[],\
8933: int imx, double p[],double **matcov,double agemortsup){
8934: int i,k;
8935:
8936: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8937: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8938: for (i=1;i<=2;i++)
8939: 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 8940: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8941: fprintf(fichtm,"</ul>");
8942:
8943: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8944:
8945: 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>");
8946:
8947: for (k=agegomp;k<(agemortsup-2);k++)
8948: 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]);
8949:
8950:
8951: fflush(fichtm);
8952: }
8953:
8954: /******************* Gnuplot file **************/
1.201 brouard 8955: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8956:
8957: char dirfileres[132],optfileres[132];
1.164 brouard 8958:
1.126 brouard 8959: int ng;
8960:
8961:
8962: /*#ifdef windows */
8963: fprintf(ficgp,"cd \"%s\" \n",pathc);
8964: /*#endif */
8965:
8966:
8967: strcpy(dirfileres,optionfilefiname);
8968: strcpy(optfileres,"vpl");
1.199 brouard 8969: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8970: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8971: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8972: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8973: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8974:
8975: }
8976:
1.136 brouard 8977: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8978: {
1.126 brouard 8979:
1.136 brouard 8980: /*-------- data file ----------*/
8981: FILE *fic;
8982: char dummy[]=" ";
1.240 brouard 8983: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8984: int lstra;
1.136 brouard 8985: int linei, month, year,iout;
8986: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8987: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8988: char *stratrunc;
1.223 brouard 8989:
1.240 brouard 8990: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8991: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8992:
1.240 brouard 8993: for(v=1; v <=ncovcol;v++){
8994: DummyV[v]=0;
8995: FixedV[v]=0;
8996: }
8997: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8998: DummyV[v]=1;
8999: FixedV[v]=0;
9000: }
9001: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9002: DummyV[v]=0;
9003: FixedV[v]=1;
9004: }
9005: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9006: DummyV[v]=1;
9007: FixedV[v]=1;
9008: }
9009: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9010: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9011: 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]);
9012: }
1.126 brouard 9013:
1.136 brouard 9014: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9015: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9016: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9017: }
1.126 brouard 9018:
1.136 brouard 9019: i=1;
9020: linei=0;
9021: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9022: linei=linei+1;
9023: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9024: if(line[j] == '\t')
9025: line[j] = ' ';
9026: }
9027: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9028: ;
9029: };
9030: line[j+1]=0; /* Trims blanks at end of line */
9031: if(line[0]=='#'){
9032: fprintf(ficlog,"Comment line\n%s\n",line);
9033: printf("Comment line\n%s\n",line);
9034: continue;
9035: }
9036: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9037: strcpy(line, linetmp);
1.223 brouard 9038:
9039: /* Loops on waves */
9040: for (j=maxwav;j>=1;j--){
9041: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9042: cutv(stra, strb, line, ' ');
9043: if(strb[0]=='.') { /* Missing value */
9044: lval=-1;
9045: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9046: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9047: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9048: 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);
9049: 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);
9050: return 1;
9051: }
9052: }else{
9053: errno=0;
9054: /* what_kind_of_number(strb); */
9055: dval=strtod(strb,&endptr);
9056: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9057: /* if(strb != endptr && *endptr == '\0') */
9058: /* dval=dlval; */
9059: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9060: if( strb[0]=='\0' || (*endptr != '\0')){
9061: 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);
9062: 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);
9063: return 1;
9064: }
9065: cotqvar[j][iv][i]=dval;
9066: cotvar[j][ntv+iv][i]=dval;
9067: }
9068: strcpy(line,stra);
1.223 brouard 9069: }/* end loop ntqv */
1.225 brouard 9070:
1.223 brouard 9071: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9072: cutv(stra, strb, line, ' ');
9073: if(strb[0]=='.') { /* Missing value */
9074: lval=-1;
9075: }else{
9076: errno=0;
9077: lval=strtol(strb,&endptr,10);
9078: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9079: if( strb[0]=='\0' || (*endptr != '\0')){
9080: 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);
9081: 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);
9082: return 1;
9083: }
9084: }
9085: if(lval <-1 || lval >1){
9086: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9087: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9088: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9089: For example, for multinomial values like 1, 2 and 3,\n \
9090: build V1=0 V2=0 for the reference value (1),\n \
9091: V1=1 V2=0 for (2) \n \
1.223 brouard 9092: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9093: output of IMaCh is often meaningless.\n \
1.223 brouard 9094: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9095: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9096: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9097: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9098: For example, for multinomial values like 1, 2 and 3,\n \
9099: build V1=0 V2=0 for the reference value (1),\n \
9100: V1=1 V2=0 for (2) \n \
1.223 brouard 9101: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9102: output of IMaCh is often meaningless.\n \
1.223 brouard 9103: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9104: return 1;
9105: }
9106: cotvar[j][iv][i]=(double)(lval);
9107: strcpy(line,stra);
1.223 brouard 9108: }/* end loop ntv */
1.225 brouard 9109:
1.223 brouard 9110: /* Statuses at wave */
1.137 brouard 9111: cutv(stra, strb, line, ' ');
1.223 brouard 9112: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9113: lval=-1;
1.136 brouard 9114: }else{
1.238 brouard 9115: errno=0;
9116: lval=strtol(strb,&endptr,10);
9117: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9118: if( strb[0]=='\0' || (*endptr != '\0')){
9119: 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);
9120: 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);
9121: return 1;
9122: }
1.136 brouard 9123: }
1.225 brouard 9124:
1.136 brouard 9125: s[j][i]=lval;
1.225 brouard 9126:
1.223 brouard 9127: /* Date of Interview */
1.136 brouard 9128: strcpy(line,stra);
9129: cutv(stra, strb,line,' ');
1.169 brouard 9130: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9131: }
1.169 brouard 9132: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9133: month=99;
9134: year=9999;
1.136 brouard 9135: }else{
1.225 brouard 9136: 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);
9137: 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);
9138: return 1;
1.136 brouard 9139: }
9140: anint[j][i]= (double) year;
9141: mint[j][i]= (double)month;
9142: strcpy(line,stra);
1.223 brouard 9143: } /* End loop on waves */
1.225 brouard 9144:
1.223 brouard 9145: /* Date of death */
1.136 brouard 9146: cutv(stra, strb,line,' ');
1.169 brouard 9147: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9148: }
1.169 brouard 9149: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9150: month=99;
9151: year=9999;
9152: }else{
1.141 brouard 9153: 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 9154: 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);
9155: return 1;
1.136 brouard 9156: }
9157: andc[i]=(double) year;
9158: moisdc[i]=(double) month;
9159: strcpy(line,stra);
9160:
1.223 brouard 9161: /* Date of birth */
1.136 brouard 9162: cutv(stra, strb,line,' ');
1.169 brouard 9163: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9164: }
1.169 brouard 9165: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9166: month=99;
9167: year=9999;
9168: }else{
1.141 brouard 9169: 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);
9170: 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 9171: return 1;
1.136 brouard 9172: }
9173: if (year==9999) {
1.141 brouard 9174: 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);
9175: 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 9176: return 1;
9177:
1.136 brouard 9178: }
9179: annais[i]=(double)(year);
9180: moisnais[i]=(double)(month);
9181: strcpy(line,stra);
1.225 brouard 9182:
1.223 brouard 9183: /* Sample weight */
1.136 brouard 9184: cutv(stra, strb,line,' ');
9185: errno=0;
9186: dval=strtod(strb,&endptr);
9187: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9188: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9189: 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 9190: fflush(ficlog);
9191: return 1;
9192: }
9193: weight[i]=dval;
9194: strcpy(line,stra);
1.225 brouard 9195:
1.223 brouard 9196: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9197: cutv(stra, strb, line, ' ');
9198: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9199: lval=-1;
1.223 brouard 9200: }else{
1.225 brouard 9201: errno=0;
9202: /* what_kind_of_number(strb); */
9203: dval=strtod(strb,&endptr);
9204: /* if(strb != endptr && *endptr == '\0') */
9205: /* dval=dlval; */
9206: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9207: if( strb[0]=='\0' || (*endptr != '\0')){
9208: 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);
9209: 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);
9210: return 1;
9211: }
9212: coqvar[iv][i]=dval;
1.226 brouard 9213: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9214: }
9215: strcpy(line,stra);
9216: }/* end loop nqv */
1.136 brouard 9217:
1.223 brouard 9218: /* Covariate values */
1.136 brouard 9219: for (j=ncovcol;j>=1;j--){
9220: cutv(stra, strb,line,' ');
1.223 brouard 9221: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9222: lval=-1;
1.136 brouard 9223: }else{
1.225 brouard 9224: errno=0;
9225: lval=strtol(strb,&endptr,10);
9226: if( strb[0]=='\0' || (*endptr != '\0')){
9227: 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);
9228: 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);
9229: return 1;
9230: }
1.136 brouard 9231: }
9232: if(lval <-1 || lval >1){
1.225 brouard 9233: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9234: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9235: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9236: For example, for multinomial values like 1, 2 and 3,\n \
9237: build V1=0 V2=0 for the reference value (1),\n \
9238: V1=1 V2=0 for (2) \n \
1.136 brouard 9239: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9240: output of IMaCh is often meaningless.\n \
1.136 brouard 9241: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9242: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9243: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9244: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9245: For example, for multinomial values like 1, 2 and 3,\n \
9246: build V1=0 V2=0 for the reference value (1),\n \
9247: V1=1 V2=0 for (2) \n \
1.136 brouard 9248: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9249: output of IMaCh is often meaningless.\n \
1.136 brouard 9250: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9251: return 1;
1.136 brouard 9252: }
9253: covar[j][i]=(double)(lval);
9254: strcpy(line,stra);
9255: }
9256: lstra=strlen(stra);
1.225 brouard 9257:
1.136 brouard 9258: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9259: stratrunc = &(stra[lstra-9]);
9260: num[i]=atol(stratrunc);
9261: }
9262: else
9263: num[i]=atol(stra);
9264: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9265: 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;}*/
9266:
9267: i=i+1;
9268: } /* End loop reading data */
1.225 brouard 9269:
1.136 brouard 9270: *imax=i-1; /* Number of individuals */
9271: fclose(fic);
1.225 brouard 9272:
1.136 brouard 9273: return (0);
1.164 brouard 9274: /* endread: */
1.225 brouard 9275: printf("Exiting readdata: ");
9276: fclose(fic);
9277: return (1);
1.223 brouard 9278: }
1.126 brouard 9279:
1.234 brouard 9280: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9281: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9282: while (*p2 == ' ')
1.234 brouard 9283: p2++;
9284: /* while ((*p1++ = *p2++) !=0) */
9285: /* ; */
9286: /* do */
9287: /* while (*p2 == ' ') */
9288: /* p2++; */
9289: /* while (*p1++ == *p2++); */
9290: *stri=p2;
1.145 brouard 9291: }
9292:
1.235 brouard 9293: int decoderesult ( char resultline[], int nres)
1.230 brouard 9294: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9295: {
1.235 brouard 9296: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9297: char resultsav[MAXLINE];
1.234 brouard 9298: int resultmodel[MAXLINE];
9299: int modelresult[MAXLINE];
1.230 brouard 9300: char stra[80], strb[80], strc[80], strd[80],stre[80];
9301:
1.234 brouard 9302: removefirstspace(&resultline);
1.233 brouard 9303: printf("decoderesult:%s\n",resultline);
1.230 brouard 9304:
9305: if (strstr(resultline,"v") !=0){
9306: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9307: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9308: return 1;
9309: }
9310: trimbb(resultsav, resultline);
9311: if (strlen(resultsav) >1){
9312: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9313: }
1.253 brouard 9314: if(j == 0){ /* Resultline but no = */
9315: TKresult[nres]=0; /* Combination for the nresult and the model */
9316: return (0);
9317: }
9318:
1.234 brouard 9319: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9320: 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);
9321: 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);
9322: }
9323: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9324: if(nbocc(resultsav,'=') >1){
9325: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9326: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9327: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9328: }else
9329: cutl(strc,strd,resultsav,'=');
1.230 brouard 9330: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9331:
1.230 brouard 9332: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9333: Tvarsel[k]=atoi(strc);
9334: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9335: /* cptcovsel++; */
9336: if (nbocc(stra,'=') >0)
9337: strcpy(resultsav,stra); /* and analyzes it */
9338: }
1.235 brouard 9339: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9340: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9341: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9342: match=0;
1.236 brouard 9343: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9344: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9345: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9346: match=1;
9347: break;
9348: }
9349: }
9350: if(match == 0){
9351: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9352: }
9353: }
9354: }
1.235 brouard 9355: /* Checking for missing or useless values in comparison of current model needs */
9356: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9357: match=0;
1.235 brouard 9358: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9359: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9360: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9361: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9362: ++match;
9363: }
9364: }
9365: }
9366: if(match == 0){
9367: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9368: }else if(match > 1){
9369: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9370: }
9371: }
1.235 brouard 9372:
1.234 brouard 9373: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9374: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9375: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9376: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9377: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9378: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9379: /* 1 0 0 0 */
9380: /* 2 1 0 0 */
9381: /* 3 0 1 0 */
9382: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9383: /* 5 0 0 1 */
9384: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9385: /* 7 0 1 1 */
9386: /* 8 1 1 1 */
1.237 brouard 9387: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9388: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9389: /* V5*age V5 known which value for nres? */
9390: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9391: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9392: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9393: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9394: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9395: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9396: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9397: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9398: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9399: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9400: k4++;;
9401: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9402: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9403: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9404: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9405: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9406: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9407: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9408: k4q++;;
9409: }
9410: }
1.234 brouard 9411:
1.235 brouard 9412: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9413: return (0);
9414: }
1.235 brouard 9415:
1.230 brouard 9416: int decodemodel( char model[], int lastobs)
9417: /**< This routine decodes the model and returns:
1.224 brouard 9418: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9419: * - nagesqr = 1 if age*age in the model, otherwise 0.
9420: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9421: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9422: * - cptcovage number of covariates with age*products =2
9423: * - cptcovs number of simple covariates
9424: * - 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
9425: * which is a new column after the 9 (ncovcol) variables.
9426: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9427: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9428: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9429: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9430: */
1.136 brouard 9431: {
1.238 brouard 9432: int i, j, k, ks, v;
1.227 brouard 9433: int j1, k1, k2, k3, k4;
1.136 brouard 9434: char modelsav[80];
1.145 brouard 9435: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9436: char *strpt;
1.136 brouard 9437:
1.145 brouard 9438: /*removespace(model);*/
1.136 brouard 9439: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9440: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9441: if (strstr(model,"AGE") !=0){
1.192 brouard 9442: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9443: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9444: return 1;
9445: }
1.141 brouard 9446: if (strstr(model,"v") !=0){
9447: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9448: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9449: return 1;
9450: }
1.187 brouard 9451: strcpy(modelsav,model);
9452: if ((strpt=strstr(model,"age*age")) !=0){
9453: printf(" strpt=%s, model=%s\n",strpt, model);
9454: if(strpt != model){
1.234 brouard 9455: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9456: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9457: corresponding column of parameters.\n",model);
1.234 brouard 9458: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9459: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9460: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9461: return 1;
1.225 brouard 9462: }
1.187 brouard 9463: nagesqr=1;
9464: if (strstr(model,"+age*age") !=0)
1.234 brouard 9465: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9466: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9467: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9468: else
1.234 brouard 9469: substrchaine(modelsav, model, "age*age");
1.187 brouard 9470: }else
9471: nagesqr=0;
9472: if (strlen(modelsav) >1){
9473: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9474: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9475: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9476: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9477: * cst, age and age*age
9478: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9479: /* including age products which are counted in cptcovage.
9480: * but the covariates which are products must be treated
9481: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9482: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9483: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9484:
9485:
1.187 brouard 9486: /* Design
9487: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9488: * < ncovcol=8 >
9489: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9490: * k= 1 2 3 4 5 6 7 8
9491: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9492: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9493: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9494: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9495: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9496: * Tage[++cptcovage]=k
9497: * if products, new covar are created after ncovcol with k1
9498: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9499: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9500: * 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
9501: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9502: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9503: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9504: * < ncovcol=8 >
9505: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9506: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9507: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9508: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9509: * p Tprod[1]@2={ 6, 5}
9510: *p Tvard[1][1]@4= {7, 8, 5, 6}
9511: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9512: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9513: *How to reorganize?
9514: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9515: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9516: * {2, 1, 4, 8, 5, 6, 3, 7}
9517: * Struct []
9518: */
1.225 brouard 9519:
1.187 brouard 9520: /* This loop fills the array Tvar from the string 'model'.*/
9521: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9522: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9523: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9524: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9525: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9526: /* k=1 Tvar[1]=2 (from V2) */
9527: /* k=5 Tvar[5] */
9528: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9529: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9530: /* } */
1.198 brouard 9531: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9532: /*
9533: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9534: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9535: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9536: }
1.187 brouard 9537: cptcovage=0;
9538: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9539: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9540: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9541: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9542: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9543: /*scanf("%d",i);*/
9544: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9545: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9546: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9547: /* covar is not filled and then is empty */
9548: cptcovprod--;
9549: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9550: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9551: Typevar[k]=1; /* 1 for age product */
9552: cptcovage++; /* Sums the number of covariates which include age as a product */
9553: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9554: /*printf("stre=%s ", stre);*/
9555: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9556: cptcovprod--;
9557: cutl(stre,strb,strc,'V');
9558: Tvar[k]=atoi(stre);
9559: Typevar[k]=1; /* 1 for age product */
9560: cptcovage++;
9561: Tage[cptcovage]=k;
9562: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9563: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9564: cptcovn++;
9565: cptcovprodnoage++;k1++;
9566: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9567: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9568: because this model-covariate is a construction we invent a new column
9569: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9570: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9571: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9572: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9573: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9574: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9575: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9576: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9577: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9578: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9579: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9580: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9581: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9582: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9583: for (i=1; i<=lastobs;i++){
9584: /* Computes the new covariate which is a product of
9585: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9586: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9587: }
9588: } /* End age is not in the model */
9589: } /* End if model includes a product */
9590: else { /* no more sum */
9591: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9592: /* scanf("%d",i);*/
9593: cutl(strd,strc,strb,'V');
9594: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9595: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9596: Tvar[k]=atoi(strd);
9597: Typevar[k]=0; /* 0 for simple covariates */
9598: }
9599: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9600: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9601: scanf("%d",i);*/
1.187 brouard 9602: } /* end of loop + on total covariates */
9603: } /* end if strlen(modelsave == 0) age*age might exist */
9604: } /* end if strlen(model == 0) */
1.136 brouard 9605:
9606: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9607: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9608:
1.136 brouard 9609: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9610: printf("cptcovprod=%d ", cptcovprod);
9611: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9612: scanf("%d ",i);*/
9613:
9614:
1.230 brouard 9615: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9616: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9617: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9618: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9619: k = 1 2 3 4 5 6 7 8 9
9620: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9621: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9622: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9623: Dummy[k] 1 0 0 0 3 1 1 2 3
9624: Tmodelind[combination of covar]=k;
1.225 brouard 9625: */
9626: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9627: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9628: /* 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 9629: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9630: printf("Model=%s\n\
9631: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9632: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9633: 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);
9634: fprintf(ficlog,"Model=%s\n\
9635: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9636: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9637: 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 9638: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9639: 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 */
9640: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9641: Fixed[k]= 0;
9642: Dummy[k]= 0;
1.225 brouard 9643: ncoveff++;
1.232 brouard 9644: ncovf++;
1.234 brouard 9645: nsd++;
9646: modell[k].maintype= FTYPE;
9647: TvarsD[nsd]=Tvar[k];
9648: TvarsDind[nsd]=k;
9649: TvarF[ncovf]=Tvar[k];
9650: TvarFind[ncovf]=k;
9651: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9652: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9653: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9654: Fixed[k]= 0;
9655: Dummy[k]= 0;
9656: ncoveff++;
9657: ncovf++;
9658: modell[k].maintype= FTYPE;
9659: TvarF[ncovf]=Tvar[k];
9660: TvarFind[ncovf]=k;
1.230 brouard 9661: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9662: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9663: }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 9664: Fixed[k]= 0;
9665: Dummy[k]= 1;
1.230 brouard 9666: nqfveff++;
1.234 brouard 9667: modell[k].maintype= FTYPE;
9668: modell[k].subtype= FQ;
9669: nsq++;
9670: TvarsQ[nsq]=Tvar[k];
9671: TvarsQind[nsq]=k;
1.232 brouard 9672: ncovf++;
1.234 brouard 9673: TvarF[ncovf]=Tvar[k];
9674: TvarFind[ncovf]=k;
1.231 brouard 9675: 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 9676: 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 9677: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9678: Fixed[k]= 1;
9679: Dummy[k]= 0;
1.225 brouard 9680: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9681: modell[k].maintype= VTYPE;
9682: modell[k].subtype= VD;
9683: nsd++;
9684: TvarsD[nsd]=Tvar[k];
9685: TvarsDind[nsd]=k;
9686: ncovv++; /* Only simple time varying variables */
9687: TvarV[ncovv]=Tvar[k];
1.242 brouard 9688: 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 9689: 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 */
9690: 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 9691: 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);
9692: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9693: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9694: Fixed[k]= 1;
9695: Dummy[k]= 1;
9696: nqtveff++;
9697: modell[k].maintype= VTYPE;
9698: modell[k].subtype= VQ;
9699: ncovv++; /* Only simple time varying variables */
9700: nsq++;
9701: TvarsQ[nsq]=Tvar[k];
9702: TvarsQind[nsq]=k;
9703: TvarV[ncovv]=Tvar[k];
1.242 brouard 9704: 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 9705: 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 */
9706: 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 9707: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9708: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9709: 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 9710: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9711: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9712: ncova++;
9713: TvarA[ncova]=Tvar[k];
9714: TvarAind[ncova]=k;
1.231 brouard 9715: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9716: Fixed[k]= 2;
9717: Dummy[k]= 2;
9718: modell[k].maintype= ATYPE;
9719: modell[k].subtype= APFD;
9720: /* ncoveff++; */
1.227 brouard 9721: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9722: Fixed[k]= 2;
9723: Dummy[k]= 3;
9724: modell[k].maintype= ATYPE;
9725: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9726: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9727: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9728: Fixed[k]= 3;
9729: Dummy[k]= 2;
9730: modell[k].maintype= ATYPE;
9731: modell[k].subtype= APVD; /* Product age * varying dummy */
9732: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9733: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9734: Fixed[k]= 3;
9735: Dummy[k]= 3;
9736: modell[k].maintype= ATYPE;
9737: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9738: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9739: }
9740: }else if (Typevar[k] == 2) { /* product without age */
9741: k1=Tposprod[k];
9742: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9743: if(Tvard[k1][2] <=ncovcol){
9744: Fixed[k]= 1;
9745: Dummy[k]= 0;
9746: modell[k].maintype= FTYPE;
9747: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9748: ncovf++; /* Fixed variables without age */
9749: TvarF[ncovf]=Tvar[k];
9750: TvarFind[ncovf]=k;
9751: }else if(Tvard[k1][2] <=ncovcol+nqv){
9752: Fixed[k]= 0; /* or 2 ?*/
9753: Dummy[k]= 1;
9754: modell[k].maintype= FTYPE;
9755: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9756: ncovf++; /* Varying variables without age */
9757: TvarF[ncovf]=Tvar[k];
9758: TvarFind[ncovf]=k;
9759: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9760: Fixed[k]= 1;
9761: Dummy[k]= 0;
9762: modell[k].maintype= VTYPE;
9763: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9764: ncovv++; /* Varying variables without age */
9765: TvarV[ncovv]=Tvar[k];
9766: TvarVind[ncovv]=k;
9767: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9768: Fixed[k]= 1;
9769: Dummy[k]= 1;
9770: modell[k].maintype= VTYPE;
9771: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9772: ncovv++; /* Varying variables without age */
9773: TvarV[ncovv]=Tvar[k];
9774: TvarVind[ncovv]=k;
9775: }
1.227 brouard 9776: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9777: if(Tvard[k1][2] <=ncovcol){
9778: Fixed[k]= 0; /* or 2 ?*/
9779: Dummy[k]= 1;
9780: modell[k].maintype= FTYPE;
9781: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9782: ncovf++; /* Fixed variables without age */
9783: TvarF[ncovf]=Tvar[k];
9784: TvarFind[ncovf]=k;
9785: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9786: Fixed[k]= 1;
9787: Dummy[k]= 1;
9788: modell[k].maintype= VTYPE;
9789: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9790: ncovv++; /* Varying variables without age */
9791: TvarV[ncovv]=Tvar[k];
9792: TvarVind[ncovv]=k;
9793: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9794: Fixed[k]= 1;
9795: Dummy[k]= 1;
9796: modell[k].maintype= VTYPE;
9797: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9798: ncovv++; /* Varying variables without age */
9799: TvarV[ncovv]=Tvar[k];
9800: TvarVind[ncovv]=k;
9801: ncovv++; /* Varying variables without age */
9802: TvarV[ncovv]=Tvar[k];
9803: TvarVind[ncovv]=k;
9804: }
1.227 brouard 9805: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9806: if(Tvard[k1][2] <=ncovcol){
9807: Fixed[k]= 1;
9808: Dummy[k]= 1;
9809: modell[k].maintype= VTYPE;
9810: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9811: ncovv++; /* Varying variables without age */
9812: TvarV[ncovv]=Tvar[k];
9813: TvarVind[ncovv]=k;
9814: }else if(Tvard[k1][2] <=ncovcol+nqv){
9815: Fixed[k]= 1;
9816: Dummy[k]= 1;
9817: modell[k].maintype= VTYPE;
9818: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9819: ncovv++; /* Varying variables without age */
9820: TvarV[ncovv]=Tvar[k];
9821: TvarVind[ncovv]=k;
9822: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9823: Fixed[k]= 1;
9824: Dummy[k]= 0;
9825: modell[k].maintype= VTYPE;
9826: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9827: ncovv++; /* Varying variables without age */
9828: TvarV[ncovv]=Tvar[k];
9829: TvarVind[ncovv]=k;
9830: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9831: Fixed[k]= 1;
9832: Dummy[k]= 1;
9833: modell[k].maintype= VTYPE;
9834: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9835: ncovv++; /* Varying variables without age */
9836: TvarV[ncovv]=Tvar[k];
9837: TvarVind[ncovv]=k;
9838: }
1.227 brouard 9839: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9840: if(Tvard[k1][2] <=ncovcol){
9841: Fixed[k]= 1;
9842: Dummy[k]= 1;
9843: modell[k].maintype= VTYPE;
9844: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9845: ncovv++; /* Varying variables without age */
9846: TvarV[ncovv]=Tvar[k];
9847: TvarVind[ncovv]=k;
9848: }else if(Tvard[k1][2] <=ncovcol+nqv){
9849: Fixed[k]= 1;
9850: Dummy[k]= 1;
9851: modell[k].maintype= VTYPE;
9852: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9853: ncovv++; /* Varying variables without age */
9854: TvarV[ncovv]=Tvar[k];
9855: TvarVind[ncovv]=k;
9856: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9857: Fixed[k]= 1;
9858: Dummy[k]= 1;
9859: modell[k].maintype= VTYPE;
9860: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9861: ncovv++; /* Varying variables without age */
9862: TvarV[ncovv]=Tvar[k];
9863: TvarVind[ncovv]=k;
9864: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9865: Fixed[k]= 1;
9866: Dummy[k]= 1;
9867: modell[k].maintype= VTYPE;
9868: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9869: ncovv++; /* Varying variables without age */
9870: TvarV[ncovv]=Tvar[k];
9871: TvarVind[ncovv]=k;
9872: }
1.227 brouard 9873: }else{
1.240 brouard 9874: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9875: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9876: } /*end k1*/
1.225 brouard 9877: }else{
1.226 brouard 9878: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9879: 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 9880: }
1.227 brouard 9881: 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 9882: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9883: 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]);
9884: }
9885: /* Searching for doublons in the model */
9886: for(k1=1; k1<= cptcovt;k1++){
9887: for(k2=1; k2 <k1;k2++){
9888: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9889: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9890: if(Tvar[k1]==Tvar[k2]){
9891: 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]]);
9892: 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);
9893: return(1);
9894: }
9895: }else if (Typevar[k1] ==2){
9896: k3=Tposprod[k1];
9897: k4=Tposprod[k2];
9898: 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])) ){
9899: 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]]);
9900: 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);
9901: return(1);
9902: }
9903: }
1.227 brouard 9904: }
9905: }
1.225 brouard 9906: }
9907: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9908: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9909: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9910: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9911: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9912: /*endread:*/
1.225 brouard 9913: printf("Exiting decodemodel: ");
9914: return (1);
1.136 brouard 9915: }
9916:
1.169 brouard 9917: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9918: {/* Check ages at death */
1.136 brouard 9919: int i, m;
1.218 brouard 9920: int firstone=0;
9921:
1.136 brouard 9922: for (i=1; i<=imx; i++) {
9923: for(m=2; (m<= maxwav); m++) {
9924: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9925: anint[m][i]=9999;
1.216 brouard 9926: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9927: s[m][i]=-1;
1.136 brouard 9928: }
9929: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9930: *nberr = *nberr + 1;
1.218 brouard 9931: if(firstone == 0){
9932: firstone=1;
1.260 brouard 9933: 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 9934: }
1.262 brouard 9935: 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 9936: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9937: }
9938: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9939: (*nberr)++;
1.259 brouard 9940: 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 9941: 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 9942: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9943: }
9944: }
9945: }
9946:
9947: for (i=1; i<=imx; i++) {
9948: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9949: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9950: 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 9951: if (s[m][i] >= nlstate+1) {
1.169 brouard 9952: if(agedc[i]>0){
9953: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9954: agev[m][i]=agedc[i];
1.214 brouard 9955: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9956: }else {
1.136 brouard 9957: if ((int)andc[i]!=9999){
9958: nbwarn++;
9959: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9960: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9961: agev[m][i]=-1;
9962: }
9963: }
1.169 brouard 9964: } /* agedc > 0 */
1.214 brouard 9965: } /* end if */
1.136 brouard 9966: else if(s[m][i] !=9){ /* Standard case, age in fractional
9967: years but with the precision of a month */
9968: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9969: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9970: agev[m][i]=1;
9971: else if(agev[m][i] < *agemin){
9972: *agemin=agev[m][i];
9973: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9974: }
9975: else if(agev[m][i] >*agemax){
9976: *agemax=agev[m][i];
1.156 brouard 9977: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9978: }
9979: /*agev[m][i]=anint[m][i]-annais[i];*/
9980: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9981: } /* en if 9*/
1.136 brouard 9982: else { /* =9 */
1.214 brouard 9983: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9984: agev[m][i]=1;
9985: s[m][i]=-1;
9986: }
9987: }
1.214 brouard 9988: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9989: agev[m][i]=1;
1.214 brouard 9990: else{
9991: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9992: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9993: agev[m][i]=0;
9994: }
9995: } /* End for lastpass */
9996: }
1.136 brouard 9997:
9998: for (i=1; i<=imx; i++) {
9999: for(m=firstpass; (m<=lastpass); m++){
10000: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10001: (*nberr)++;
1.136 brouard 10002: 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);
10003: 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);
10004: return 1;
10005: }
10006: }
10007: }
10008:
10009: /*for (i=1; i<=imx; i++){
10010: for (m=firstpass; (m<lastpass); m++){
10011: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10012: }
10013:
10014: }*/
10015:
10016:
1.139 brouard 10017: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10018: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10019:
10020: return (0);
1.164 brouard 10021: /* endread:*/
1.136 brouard 10022: printf("Exiting calandcheckages: ");
10023: return (1);
10024: }
10025:
1.172 brouard 10026: #if defined(_MSC_VER)
10027: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10028: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10029: //#include "stdafx.h"
10030: //#include <stdio.h>
10031: //#include <tchar.h>
10032: //#include <windows.h>
10033: //#include <iostream>
10034: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10035:
10036: LPFN_ISWOW64PROCESS fnIsWow64Process;
10037:
10038: BOOL IsWow64()
10039: {
10040: BOOL bIsWow64 = FALSE;
10041:
10042: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10043: // (HANDLE, PBOOL);
10044:
10045: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10046:
10047: HMODULE module = GetModuleHandle(_T("kernel32"));
10048: const char funcName[] = "IsWow64Process";
10049: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10050: GetProcAddress(module, funcName);
10051:
10052: if (NULL != fnIsWow64Process)
10053: {
10054: if (!fnIsWow64Process(GetCurrentProcess(),
10055: &bIsWow64))
10056: //throw std::exception("Unknown error");
10057: printf("Unknown error\n");
10058: }
10059: return bIsWow64 != FALSE;
10060: }
10061: #endif
1.177 brouard 10062:
1.191 brouard 10063: void syscompilerinfo(int logged)
1.167 brouard 10064: {
10065: /* #include "syscompilerinfo.h"*/
1.185 brouard 10066: /* command line Intel compiler 32bit windows, XP compatible:*/
10067: /* /GS /W3 /Gy
10068: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10069: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10070: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10071: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10072: */
10073: /* 64 bits */
1.185 brouard 10074: /*
10075: /GS /W3 /Gy
10076: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10077: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10078: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10079: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10080: /* Optimization are useless and O3 is slower than O2 */
10081: /*
10082: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10083: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10084: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10085: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10086: */
1.186 brouard 10087: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10088: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10089: /PDB:"visual studio
10090: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10091: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10092: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10093: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10094: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10095: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10096: uiAccess='false'"
10097: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10098: /NOLOGO /TLBID:1
10099: */
1.177 brouard 10100: #if defined __INTEL_COMPILER
1.178 brouard 10101: #if defined(__GNUC__)
10102: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10103: #endif
1.177 brouard 10104: #elif defined(__GNUC__)
1.179 brouard 10105: #ifndef __APPLE__
1.174 brouard 10106: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10107: #endif
1.177 brouard 10108: struct utsname sysInfo;
1.178 brouard 10109: int cross = CROSS;
10110: if (cross){
10111: printf("Cross-");
1.191 brouard 10112: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10113: }
1.174 brouard 10114: #endif
10115:
1.171 brouard 10116: #include <stdint.h>
1.178 brouard 10117:
1.191 brouard 10118: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10119: #if defined(__clang__)
1.191 brouard 10120: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10121: #endif
10122: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10123: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10124: #endif
10125: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10126: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10127: #endif
10128: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10129: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10130: #endif
10131: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10132: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10133: #endif
10134: #if defined(_MSC_VER)
1.191 brouard 10135: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10136: #endif
10137: #if defined(__PGI)
1.191 brouard 10138: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10139: #endif
10140: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10141: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10142: #endif
1.191 brouard 10143: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10144:
1.167 brouard 10145: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10146: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10147: // Windows (x64 and x86)
1.191 brouard 10148: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10149: #elif __unix__ // all unices, not all compilers
10150: // Unix
1.191 brouard 10151: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10152: #elif __linux__
10153: // linux
1.191 brouard 10154: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10155: #elif __APPLE__
1.174 brouard 10156: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10157: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10158: #endif
10159:
10160: /* __MINGW32__ */
10161: /* __CYGWIN__ */
10162: /* __MINGW64__ */
10163: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10164: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10165: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10166: /* _WIN64 // Defined for applications for Win64. */
10167: /* _M_X64 // Defined for compilations that target x64 processors. */
10168: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10169:
1.167 brouard 10170: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10171: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10172: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10173: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10174: #else
1.191 brouard 10175: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10176: #endif
10177:
1.169 brouard 10178: #if defined(__GNUC__)
10179: # if defined(__GNUC_PATCHLEVEL__)
10180: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10181: + __GNUC_MINOR__ * 100 \
10182: + __GNUC_PATCHLEVEL__)
10183: # else
10184: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10185: + __GNUC_MINOR__ * 100)
10186: # endif
1.174 brouard 10187: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10188: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10189:
10190: if (uname(&sysInfo) != -1) {
10191: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10192: 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 10193: }
10194: else
10195: perror("uname() error");
1.179 brouard 10196: //#ifndef __INTEL_COMPILER
10197: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10198: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10199: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10200: #endif
1.169 brouard 10201: #endif
1.172 brouard 10202:
10203: // void main()
10204: // {
1.169 brouard 10205: #if defined(_MSC_VER)
1.174 brouard 10206: if (IsWow64()){
1.191 brouard 10207: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10208: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10209: }
10210: else{
1.191 brouard 10211: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10212: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10213: }
1.172 brouard 10214: // printf("\nPress Enter to continue...");
10215: // getchar();
10216: // }
10217:
1.169 brouard 10218: #endif
10219:
1.167 brouard 10220:
1.219 brouard 10221: }
1.136 brouard 10222:
1.219 brouard 10223: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10224: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10225: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10226: /* double ftolpl = 1.e-10; */
1.180 brouard 10227: double age, agebase, agelim;
1.203 brouard 10228: double tot;
1.180 brouard 10229:
1.202 brouard 10230: strcpy(filerespl,"PL_");
10231: strcat(filerespl,fileresu);
10232: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10233: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10234: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10235: }
1.227 brouard 10236: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10237: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10238: pstamp(ficrespl);
1.203 brouard 10239: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10240: fprintf(ficrespl,"#Age ");
10241: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10242: fprintf(ficrespl,"\n");
1.180 brouard 10243:
1.219 brouard 10244: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10245:
1.219 brouard 10246: agebase=ageminpar;
10247: agelim=agemaxpar;
1.180 brouard 10248:
1.227 brouard 10249: /* i1=pow(2,ncoveff); */
1.234 brouard 10250: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10251: if (cptcovn < 1){i1=1;}
1.180 brouard 10252:
1.238 brouard 10253: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10254: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10255: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10256: continue;
1.235 brouard 10257:
1.238 brouard 10258: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10259: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10260: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10261: /* k=k+1; */
10262: /* to clean */
10263: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10264: fprintf(ficrespl,"#******");
10265: printf("#******");
10266: fprintf(ficlog,"#******");
10267: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10268: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10269: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10270: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10271: }
10272: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10273: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10274: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10275: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10276: }
10277: fprintf(ficrespl,"******\n");
10278: printf("******\n");
10279: fprintf(ficlog,"******\n");
10280: if(invalidvarcomb[k]){
10281: printf("\nCombination (%d) ignored because no case \n",k);
10282: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10283: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10284: continue;
10285: }
1.219 brouard 10286:
1.238 brouard 10287: fprintf(ficrespl,"#Age ");
10288: for(j=1;j<=cptcoveff;j++) {
10289: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10290: }
10291: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10292: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10293:
1.238 brouard 10294: for (age=agebase; age<=agelim; age++){
10295: /* for (age=agebase; age<=agebase; age++){ */
10296: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10297: fprintf(ficrespl,"%.0f ",age );
10298: for(j=1;j<=cptcoveff;j++)
10299: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10300: tot=0.;
10301: for(i=1; i<=nlstate;i++){
10302: tot += prlim[i][i];
10303: fprintf(ficrespl," %.5f", prlim[i][i]);
10304: }
10305: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10306: } /* Age */
10307: /* was end of cptcod */
10308: } /* cptcov */
10309: } /* nres */
1.219 brouard 10310: return 0;
1.180 brouard 10311: }
10312:
1.218 brouard 10313: 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){
10314: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10315:
10316: /* Computes the back prevalence limit for any combination of covariate values
10317: * at any age between ageminpar and agemaxpar
10318: */
1.235 brouard 10319: int i, j, k, i1, nres=0 ;
1.217 brouard 10320: /* double ftolpl = 1.e-10; */
10321: double age, agebase, agelim;
10322: double tot;
1.218 brouard 10323: /* double ***mobaverage; */
10324: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10325:
10326: strcpy(fileresplb,"PLB_");
10327: strcat(fileresplb,fileresu);
10328: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10329: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10330: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10331: }
10332: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10333: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10334: pstamp(ficresplb);
10335: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10336: fprintf(ficresplb,"#Age ");
10337: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10338: fprintf(ficresplb,"\n");
10339:
1.218 brouard 10340:
10341: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10342:
10343: agebase=ageminpar;
10344: agelim=agemaxpar;
10345:
10346:
1.227 brouard 10347: i1=pow(2,cptcoveff);
1.218 brouard 10348: if (cptcovn < 1){i1=1;}
1.227 brouard 10349:
1.238 brouard 10350: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10351: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10352: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10353: continue;
10354: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10355: fprintf(ficresplb,"#******");
10356: printf("#******");
10357: fprintf(ficlog,"#******");
10358: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10359: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10360: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10361: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10362: }
10363: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10364: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10365: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10366: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10367: }
10368: fprintf(ficresplb,"******\n");
10369: printf("******\n");
10370: fprintf(ficlog,"******\n");
10371: if(invalidvarcomb[k]){
10372: printf("\nCombination (%d) ignored because no cases \n",k);
10373: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10374: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10375: continue;
10376: }
1.218 brouard 10377:
1.238 brouard 10378: fprintf(ficresplb,"#Age ");
10379: for(j=1;j<=cptcoveff;j++) {
10380: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10381: }
10382: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10383: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10384:
10385:
1.238 brouard 10386: for (age=agebase; age<=agelim; age++){
10387: /* for (age=agebase; age<=agebase; age++){ */
10388: if(mobilavproj > 0){
10389: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10390: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10391: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10392: }else if (mobilavproj == 0){
10393: 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);
10394: 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);
10395: exit(1);
10396: }else{
10397: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10398: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10399: /* printf("TOTOT\n"); */
10400: /* exit(1); */
1.238 brouard 10401: }
10402: fprintf(ficresplb,"%.0f ",age );
10403: for(j=1;j<=cptcoveff;j++)
10404: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10405: tot=0.;
10406: for(i=1; i<=nlstate;i++){
10407: tot += bprlim[i][i];
10408: fprintf(ficresplb," %.5f", bprlim[i][i]);
10409: }
10410: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10411: } /* Age */
10412: /* was end of cptcod */
1.255 brouard 10413: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10414: } /* end of any combination */
10415: } /* end of nres */
1.218 brouard 10416: /* hBijx(p, bage, fage); */
10417: /* fclose(ficrespijb); */
10418:
10419: return 0;
1.217 brouard 10420: }
1.218 brouard 10421:
1.180 brouard 10422: int hPijx(double *p, int bage, int fage){
10423: /*------------- h Pij x at various ages ------------*/
10424:
10425: int stepsize;
10426: int agelim;
10427: int hstepm;
10428: int nhstepm;
1.235 brouard 10429: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10430:
10431: double agedeb;
10432: double ***p3mat;
10433:
1.201 brouard 10434: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10435: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10436: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10437: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10438: }
10439: printf("Computing pij: result on file '%s' \n", filerespij);
10440: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10441:
10442: stepsize=(int) (stepm+YEARM-1)/YEARM;
10443: /*if (stepm<=24) stepsize=2;*/
10444:
10445: agelim=AGESUP;
10446: hstepm=stepsize*YEARM; /* Every year of age */
10447: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10448:
1.180 brouard 10449: /* hstepm=1; aff par mois*/
10450: pstamp(ficrespij);
10451: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10452: i1= pow(2,cptcoveff);
1.218 brouard 10453: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10454: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10455: /* k=k+1; */
1.235 brouard 10456: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10457: for(k=1; k<=i1;k++){
1.253 brouard 10458: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10459: continue;
1.183 brouard 10460: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10461: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10462: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10463: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10464: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10465: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10466: }
1.183 brouard 10467: fprintf(ficrespij,"******\n");
10468:
10469: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10470: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10471: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10472:
10473: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10474:
1.183 brouard 10475: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10476: oldm=oldms;savm=savms;
1.235 brouard 10477: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10478: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10479: for(i=1; i<=nlstate;i++)
10480: for(j=1; j<=nlstate+ndeath;j++)
10481: fprintf(ficrespij," %1d-%1d",i,j);
10482: fprintf(ficrespij,"\n");
10483: for (h=0; h<=nhstepm; h++){
10484: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10485: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10486: for(i=1; i<=nlstate;i++)
10487: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10488: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10489: fprintf(ficrespij,"\n");
10490: }
1.183 brouard 10491: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10492: fprintf(ficrespij,"\n");
10493: }
1.180 brouard 10494: /*}*/
10495: }
1.218 brouard 10496: return 0;
1.180 brouard 10497: }
1.218 brouard 10498:
10499: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10500: /*------------- h Bij x at various ages ------------*/
10501:
10502: int stepsize;
1.218 brouard 10503: /* int agelim; */
10504: int ageminl;
1.217 brouard 10505: int hstepm;
10506: int nhstepm;
1.238 brouard 10507: int h, i, i1, j, k, nres;
1.218 brouard 10508:
1.217 brouard 10509: double agedeb;
10510: double ***p3mat;
1.218 brouard 10511:
10512: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10513: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10514: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10515: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10516: }
10517: printf("Computing pij back: result on file '%s' \n", filerespijb);
10518: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10519:
10520: stepsize=(int) (stepm+YEARM-1)/YEARM;
10521: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10522:
1.218 brouard 10523: /* agelim=AGESUP; */
10524: ageminl=30;
10525: hstepm=stepsize*YEARM; /* Every year of age */
10526: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10527:
10528: /* hstepm=1; aff par mois*/
10529: pstamp(ficrespijb);
1.255 brouard 10530: 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 10531: i1= pow(2,cptcoveff);
1.218 brouard 10532: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10533: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10534: /* k=k+1; */
1.238 brouard 10535: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10536: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10537: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10538: continue;
10539: fprintf(ficrespijb,"\n#****** ");
10540: for(j=1;j<=cptcoveff;j++)
10541: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10542: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10543: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10544: }
10545: fprintf(ficrespijb,"******\n");
1.264 brouard 10546: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10547: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10548: continue;
10549: }
10550:
10551: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10552: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10553: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10554: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10555: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10556:
10557: /* nhstepm=nhstepm*YEARM; aff par mois*/
10558:
1.266 brouard 10559: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10560: /* and memory limitations if stepm is small */
10561:
1.238 brouard 10562: /* oldm=oldms;savm=savms; */
10563: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10564: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10565: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10566: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10567: for(i=1; i<=nlstate;i++)
10568: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10569: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10570: fprintf(ficrespijb,"\n");
1.238 brouard 10571: for (h=0; h<=nhstepm; h++){
10572: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10573: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10574: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10575: for(i=1; i<=nlstate;i++)
10576: for(j=1; j<=nlstate+ndeath;j++)
10577: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10578: fprintf(ficrespijb,"\n");
10579: }
10580: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10581: fprintf(ficrespijb,"\n");
10582: } /* end age deb */
10583: } /* end combination */
10584: } /* end nres */
1.218 brouard 10585: return 0;
10586: } /* hBijx */
1.217 brouard 10587:
1.180 brouard 10588:
1.136 brouard 10589: /***********************************************/
10590: /**************** Main Program *****************/
10591: /***********************************************/
10592:
10593: int main(int argc, char *argv[])
10594: {
10595: #ifdef GSL
10596: const gsl_multimin_fminimizer_type *T;
10597: size_t iteri = 0, it;
10598: int rval = GSL_CONTINUE;
10599: int status = GSL_SUCCESS;
10600: double ssval;
10601: #endif
10602: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10603: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10604: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10605: int jj, ll, li, lj, lk;
1.136 brouard 10606: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10607: int num_filled;
1.136 brouard 10608: int itimes;
10609: int NDIM=2;
10610: int vpopbased=0;
1.235 brouard 10611: int nres=0;
1.258 brouard 10612: int endishere=0;
1.136 brouard 10613:
1.274 brouard 10614: int ncurrv=0; /* Temporary variable */
10615:
1.164 brouard 10616: char ca[32], cb[32];
1.136 brouard 10617: /* FILE *fichtm; *//* Html File */
10618: /* FILE *ficgp;*/ /*Gnuplot File */
10619: struct stat info;
1.191 brouard 10620: double agedeb=0.;
1.194 brouard 10621:
10622: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10623: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10624:
1.165 brouard 10625: double fret;
1.191 brouard 10626: double dum=0.; /* Dummy variable */
1.136 brouard 10627: double ***p3mat;
1.218 brouard 10628: /* double ***mobaverage; */
1.164 brouard 10629:
10630: char line[MAXLINE];
1.197 brouard 10631: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10632:
1.234 brouard 10633: char modeltemp[MAXLINE];
1.230 brouard 10634: char resultline[MAXLINE];
10635:
1.136 brouard 10636: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10637: char *tok, *val; /* pathtot */
1.136 brouard 10638: int firstobs=1, lastobs=10;
1.195 brouard 10639: int c, h , cpt, c2;
1.191 brouard 10640: int jl=0;
10641: int i1, j1, jk, stepsize=0;
1.194 brouard 10642: int count=0;
10643:
1.164 brouard 10644: int *tab;
1.136 brouard 10645: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10646: int backcast=0;
1.136 brouard 10647: int mobilav=0,popforecast=0;
1.191 brouard 10648: int hstepm=0, nhstepm=0;
1.136 brouard 10649: int agemortsup;
10650: float sumlpop=0.;
10651: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10652: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10653:
1.191 brouard 10654: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10655: double ftolpl=FTOL;
10656: double **prlim;
1.217 brouard 10657: double **bprlim;
1.136 brouard 10658: double ***param; /* Matrix of parameters */
1.251 brouard 10659: double ***paramstart; /* Matrix of starting parameter values */
10660: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10661: double **matcov; /* Matrix of covariance */
1.203 brouard 10662: double **hess; /* Hessian matrix */
1.136 brouard 10663: double ***delti3; /* Scale */
10664: double *delti; /* Scale */
10665: double ***eij, ***vareij;
10666: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10667:
1.136 brouard 10668: double *epj, vepp;
1.164 brouard 10669:
1.273 brouard 10670: double dateprev1, dateprev2;
10671: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10672: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10673:
1.136 brouard 10674: double **ximort;
1.145 brouard 10675: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10676: int *dcwave;
10677:
1.164 brouard 10678: char z[1]="c";
1.136 brouard 10679:
10680: /*char *strt;*/
10681: char strtend[80];
1.126 brouard 10682:
1.164 brouard 10683:
1.126 brouard 10684: /* setlocale (LC_ALL, ""); */
10685: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10686: /* textdomain (PACKAGE); */
10687: /* setlocale (LC_CTYPE, ""); */
10688: /* setlocale (LC_MESSAGES, ""); */
10689:
10690: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10691: rstart_time = time(NULL);
10692: /* (void) gettimeofday(&start_time,&tzp);*/
10693: start_time = *localtime(&rstart_time);
1.126 brouard 10694: curr_time=start_time;
1.157 brouard 10695: /*tml = *localtime(&start_time.tm_sec);*/
10696: /* strcpy(strstart,asctime(&tml)); */
10697: strcpy(strstart,asctime(&start_time));
1.126 brouard 10698:
10699: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10700: /* tp.tm_sec = tp.tm_sec +86400; */
10701: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10702: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10703: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10704: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10705: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10706: /* strt=asctime(&tmg); */
10707: /* printf("Time(after) =%s",strstart); */
10708: /* (void) time (&time_value);
10709: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10710: * tm = *localtime(&time_value);
10711: * strstart=asctime(&tm);
10712: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10713: */
10714:
10715: nberr=0; /* Number of errors and warnings */
10716: nbwarn=0;
1.184 brouard 10717: #ifdef WIN32
10718: _getcwd(pathcd, size);
10719: #else
1.126 brouard 10720: getcwd(pathcd, size);
1.184 brouard 10721: #endif
1.191 brouard 10722: syscompilerinfo(0);
1.196 brouard 10723: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10724: if(argc <=1){
10725: printf("\nEnter the parameter file name: ");
1.205 brouard 10726: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10727: printf("ERROR Empty parameter file name\n");
10728: goto end;
10729: }
1.126 brouard 10730: i=strlen(pathr);
10731: if(pathr[i-1]=='\n')
10732: pathr[i-1]='\0';
1.156 brouard 10733: i=strlen(pathr);
1.205 brouard 10734: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10735: pathr[i-1]='\0';
1.205 brouard 10736: }
10737: i=strlen(pathr);
10738: if( i==0 ){
10739: printf("ERROR Empty parameter file name\n");
10740: goto end;
10741: }
10742: for (tok = pathr; tok != NULL; ){
1.126 brouard 10743: printf("Pathr |%s|\n",pathr);
10744: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10745: printf("val= |%s| pathr=%s\n",val,pathr);
10746: strcpy (pathtot, val);
10747: if(pathr[0] == '\0') break; /* Dirty */
10748: }
10749: }
10750: else{
10751: strcpy(pathtot,argv[1]);
10752: }
10753: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10754: /*cygwin_split_path(pathtot,path,optionfile);
10755: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10756: /* cutv(path,optionfile,pathtot,'\\');*/
10757:
10758: /* Split argv[0], imach program to get pathimach */
10759: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10760: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10761: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10762: /* strcpy(pathimach,argv[0]); */
10763: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10764: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10765: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10766: #ifdef WIN32
10767: _chdir(path); /* Can be a relative path */
10768: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10769: #else
1.126 brouard 10770: chdir(path); /* Can be a relative path */
1.184 brouard 10771: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10772: #endif
10773: printf("Current directory %s!\n",pathcd);
1.126 brouard 10774: strcpy(command,"mkdir ");
10775: strcat(command,optionfilefiname);
10776: if((outcmd=system(command)) != 0){
1.169 brouard 10777: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10778: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10779: /* fclose(ficlog); */
10780: /* exit(1); */
10781: }
10782: /* if((imk=mkdir(optionfilefiname))<0){ */
10783: /* perror("mkdir"); */
10784: /* } */
10785:
10786: /*-------- arguments in the command line --------*/
10787:
1.186 brouard 10788: /* Main Log file */
1.126 brouard 10789: strcat(filelog, optionfilefiname);
10790: strcat(filelog,".log"); /* */
10791: if((ficlog=fopen(filelog,"w"))==NULL) {
10792: printf("Problem with logfile %s\n",filelog);
10793: goto end;
10794: }
10795: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10796: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10797: fprintf(ficlog,"\nEnter the parameter file name: \n");
10798: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10799: path=%s \n\
10800: optionfile=%s\n\
10801: optionfilext=%s\n\
1.156 brouard 10802: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10803:
1.197 brouard 10804: syscompilerinfo(1);
1.167 brouard 10805:
1.126 brouard 10806: printf("Local time (at start):%s",strstart);
10807: fprintf(ficlog,"Local time (at start): %s",strstart);
10808: fflush(ficlog);
10809: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10810: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10811:
10812: /* */
10813: strcpy(fileres,"r");
10814: strcat(fileres, optionfilefiname);
1.201 brouard 10815: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10816: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10817: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10818:
1.186 brouard 10819: /* Main ---------arguments file --------*/
1.126 brouard 10820:
10821: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10822: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10823: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10824: fflush(ficlog);
1.149 brouard 10825: /* goto end; */
10826: exit(70);
1.126 brouard 10827: }
10828:
10829:
10830:
10831: strcpy(filereso,"o");
1.201 brouard 10832: strcat(filereso,fileresu);
1.126 brouard 10833: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10834: printf("Problem with Output resultfile: %s\n", filereso);
10835: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10836: fflush(ficlog);
10837: goto end;
10838: }
10839:
10840: /* Reads comments: lines beginning with '#' */
10841: numlinepar=0;
1.197 brouard 10842:
10843: /* First parameter line */
10844: while(fgets(line, MAXLINE, ficpar)) {
10845: /* If line starts with a # it is a comment */
10846: if (line[0] == '#') {
10847: numlinepar++;
10848: fputs(line,stdout);
10849: fputs(line,ficparo);
10850: fputs(line,ficlog);
10851: continue;
10852: }else
10853: break;
10854: }
10855: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10856: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10857: if (num_filled != 5) {
10858: printf("Should be 5 parameters\n");
10859: }
1.126 brouard 10860: numlinepar++;
1.197 brouard 10861: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10862: }
10863: /* Second parameter line */
10864: while(fgets(line, MAXLINE, ficpar)) {
10865: /* If line starts with a # it is a comment */
10866: if (line[0] == '#') {
10867: numlinepar++;
10868: fputs(line,stdout);
10869: fputs(line,ficparo);
10870: fputs(line,ficlog);
10871: continue;
10872: }else
10873: break;
10874: }
1.223 brouard 10875: 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", \
10876: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10877: if (num_filled != 11) {
10878: 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 10879: printf("but line=%s\n",line);
1.197 brouard 10880: }
1.223 brouard 10881: 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 10882: }
1.203 brouard 10883: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10884: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10885: /* Third parameter line */
10886: while(fgets(line, MAXLINE, ficpar)) {
10887: /* If line starts with a # it is a comment */
10888: if (line[0] == '#') {
10889: numlinepar++;
10890: fputs(line,stdout);
10891: fputs(line,ficparo);
10892: fputs(line,ficlog);
10893: continue;
10894: }else
10895: break;
10896: }
1.201 brouard 10897: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 brouard 10898: if (num_filled == 0){
10899: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10900: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10901: model[0]='\0';
10902: goto end;
10903: } else if (num_filled != 1){
1.197 brouard 10904: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10905: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10906: model[0]='\0';
10907: goto end;
10908: }
10909: else{
10910: if (model[0]=='+'){
10911: for(i=1; i<=strlen(model);i++)
10912: modeltemp[i-1]=model[i];
1.201 brouard 10913: strcpy(model,modeltemp);
1.197 brouard 10914: }
10915: }
1.199 brouard 10916: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10917: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10918: }
10919: /* 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); */
10920: /* numlinepar=numlinepar+3; /\* In general *\/ */
10921: /* 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 10922: 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);
10923: 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 10924: fflush(ficlog);
1.190 brouard 10925: /* if(model[0]=='#'|| model[0]== '\0'){ */
10926: if(model[0]=='#'){
1.187 brouard 10927: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10928: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10929: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10930: if(mle != -1){
10931: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10932: exit(1);
10933: }
10934: }
1.126 brouard 10935: while((c=getc(ficpar))=='#' && c!= EOF){
10936: ungetc(c,ficpar);
10937: fgets(line, MAXLINE, ficpar);
10938: numlinepar++;
1.195 brouard 10939: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10940: z[0]=line[1];
10941: }
10942: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10943: fputs(line, stdout);
10944: //puts(line);
1.126 brouard 10945: fputs(line,ficparo);
10946: fputs(line,ficlog);
10947: }
10948: ungetc(c,ficpar);
10949:
10950:
1.145 brouard 10951: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 10952: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
10953: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
10954: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 10955: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10956: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10957: v1+v2*age+v2*v3 makes cptcovn = 3
10958: */
10959: if (strlen(model)>1)
1.187 brouard 10960: 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 10961: else
1.187 brouard 10962: ncovmodel=2; /* Constant and age */
1.133 brouard 10963: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10964: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10965: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10966: 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);
10967: 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);
10968: fflush(stdout);
10969: fclose (ficlog);
10970: goto end;
10971: }
1.126 brouard 10972: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10973: delti=delti3[1][1];
10974: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10975: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10976: /* We could also provide initial parameters values giving by simple logistic regression
10977: * only one way, that is without matrix product. We will have nlstate maximizations */
10978: /* for(i=1;i<nlstate;i++){ */
10979: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10980: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10981: /* } */
1.126 brouard 10982: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10983: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10984: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10985: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10986: fclose (ficparo);
10987: fclose (ficlog);
10988: goto end;
10989: exit(0);
1.220 brouard 10990: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10991: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10992: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10993: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10994: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10995: matcov=matrix(1,npar,1,npar);
1.203 brouard 10996: hess=matrix(1,npar,1,npar);
1.220 brouard 10997: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10998: /* Read guessed parameters */
1.126 brouard 10999: /* Reads comments: lines beginning with '#' */
11000: while((c=getc(ficpar))=='#' && c!= EOF){
11001: ungetc(c,ficpar);
11002: fgets(line, MAXLINE, ficpar);
11003: numlinepar++;
1.141 brouard 11004: fputs(line,stdout);
1.126 brouard 11005: fputs(line,ficparo);
11006: fputs(line,ficlog);
11007: }
11008: ungetc(c,ficpar);
11009:
11010: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11011: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11012: for(i=1; i <=nlstate; i++){
1.234 brouard 11013: j=0;
1.126 brouard 11014: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11015: if(jj==i) continue;
11016: j++;
11017: fscanf(ficpar,"%1d%1d",&i1,&j1);
11018: if ((i1 != i) || (j1 != jj)){
11019: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11020: It might be a problem of design; if ncovcol and the model are correct\n \
11021: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11022: exit(1);
11023: }
11024: fprintf(ficparo,"%1d%1d",i1,j1);
11025: if(mle==1)
11026: printf("%1d%1d",i,jj);
11027: fprintf(ficlog,"%1d%1d",i,jj);
11028: for(k=1; k<=ncovmodel;k++){
11029: fscanf(ficpar," %lf",¶m[i][j][k]);
11030: if(mle==1){
11031: printf(" %lf",param[i][j][k]);
11032: fprintf(ficlog," %lf",param[i][j][k]);
11033: }
11034: else
11035: fprintf(ficlog," %lf",param[i][j][k]);
11036: fprintf(ficparo," %lf",param[i][j][k]);
11037: }
11038: fscanf(ficpar,"\n");
11039: numlinepar++;
11040: if(mle==1)
11041: printf("\n");
11042: fprintf(ficlog,"\n");
11043: fprintf(ficparo,"\n");
1.126 brouard 11044: }
11045: }
11046: fflush(ficlog);
1.234 brouard 11047:
1.251 brouard 11048: /* Reads parameters values */
1.126 brouard 11049: p=param[1][1];
1.251 brouard 11050: pstart=paramstart[1][1];
1.126 brouard 11051:
11052: /* Reads comments: lines beginning with '#' */
11053: while((c=getc(ficpar))=='#' && c!= EOF){
11054: ungetc(c,ficpar);
11055: fgets(line, MAXLINE, ficpar);
11056: numlinepar++;
1.141 brouard 11057: fputs(line,stdout);
1.126 brouard 11058: fputs(line,ficparo);
11059: fputs(line,ficlog);
11060: }
11061: ungetc(c,ficpar);
11062:
11063: for(i=1; i <=nlstate; i++){
11064: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11065: fscanf(ficpar,"%1d%1d",&i1,&j1);
11066: if ( (i1-i) * (j1-j) != 0){
11067: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11068: exit(1);
11069: }
11070: printf("%1d%1d",i,j);
11071: fprintf(ficparo,"%1d%1d",i1,j1);
11072: fprintf(ficlog,"%1d%1d",i1,j1);
11073: for(k=1; k<=ncovmodel;k++){
11074: fscanf(ficpar,"%le",&delti3[i][j][k]);
11075: printf(" %le",delti3[i][j][k]);
11076: fprintf(ficparo," %le",delti3[i][j][k]);
11077: fprintf(ficlog," %le",delti3[i][j][k]);
11078: }
11079: fscanf(ficpar,"\n");
11080: numlinepar++;
11081: printf("\n");
11082: fprintf(ficparo,"\n");
11083: fprintf(ficlog,"\n");
1.126 brouard 11084: }
11085: }
11086: fflush(ficlog);
1.234 brouard 11087:
1.145 brouard 11088: /* Reads covariance matrix */
1.126 brouard 11089: delti=delti3[1][1];
1.220 brouard 11090:
11091:
1.126 brouard 11092: /* 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 11093:
1.126 brouard 11094: /* Reads comments: lines beginning with '#' */
11095: while((c=getc(ficpar))=='#' && c!= EOF){
11096: ungetc(c,ficpar);
11097: fgets(line, MAXLINE, ficpar);
11098: numlinepar++;
1.141 brouard 11099: fputs(line,stdout);
1.126 brouard 11100: fputs(line,ficparo);
11101: fputs(line,ficlog);
11102: }
11103: ungetc(c,ficpar);
1.220 brouard 11104:
1.126 brouard 11105: matcov=matrix(1,npar,1,npar);
1.203 brouard 11106: hess=matrix(1,npar,1,npar);
1.131 brouard 11107: for(i=1; i <=npar; i++)
11108: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11109:
1.194 brouard 11110: /* Scans npar lines */
1.126 brouard 11111: for(i=1; i <=npar; i++){
1.226 brouard 11112: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11113: if(count != 3){
1.226 brouard 11114: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11115: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11116: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11117: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11118: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11119: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11120: exit(1);
1.220 brouard 11121: }else{
1.226 brouard 11122: if(mle==1)
11123: printf("%1d%1d%d",i1,j1,jk);
11124: }
11125: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11126: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11127: for(j=1; j <=i; j++){
1.226 brouard 11128: fscanf(ficpar," %le",&matcov[i][j]);
11129: if(mle==1){
11130: printf(" %.5le",matcov[i][j]);
11131: }
11132: fprintf(ficlog," %.5le",matcov[i][j]);
11133: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11134: }
11135: fscanf(ficpar,"\n");
11136: numlinepar++;
11137: if(mle==1)
1.220 brouard 11138: printf("\n");
1.126 brouard 11139: fprintf(ficlog,"\n");
11140: fprintf(ficparo,"\n");
11141: }
1.194 brouard 11142: /* End of read covariance matrix npar lines */
1.126 brouard 11143: for(i=1; i <=npar; i++)
11144: for(j=i+1;j<=npar;j++)
1.226 brouard 11145: matcov[i][j]=matcov[j][i];
1.126 brouard 11146:
11147: if(mle==1)
11148: printf("\n");
11149: fprintf(ficlog,"\n");
11150:
11151: fflush(ficlog);
11152:
11153: /*-------- Rewriting parameter file ----------*/
11154: strcpy(rfileres,"r"); /* "Rparameterfile */
11155: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
11156: strcat(rfileres,"."); /* */
11157: strcat(rfileres,optionfilext); /* Other files have txt extension */
11158: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 11159: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11160: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 11161: }
11162: fprintf(ficres,"#%s\n",version);
11163: } /* End of mle != -3 */
1.218 brouard 11164:
1.186 brouard 11165: /* Main data
11166: */
1.126 brouard 11167: n= lastobs;
11168: num=lvector(1,n);
11169: moisnais=vector(1,n);
11170: annais=vector(1,n);
11171: moisdc=vector(1,n);
11172: andc=vector(1,n);
1.220 brouard 11173: weight=vector(1,n);
1.126 brouard 11174: agedc=vector(1,n);
11175: cod=ivector(1,n);
1.220 brouard 11176: for(i=1;i<=n;i++){
1.234 brouard 11177: num[i]=0;
11178: moisnais[i]=0;
11179: annais[i]=0;
11180: moisdc[i]=0;
11181: andc[i]=0;
11182: agedc[i]=0;
11183: cod[i]=0;
11184: weight[i]=1.0; /* Equal weights, 1 by default */
11185: }
1.126 brouard 11186: mint=matrix(1,maxwav,1,n);
11187: anint=matrix(1,maxwav,1,n);
1.131 brouard 11188: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11189: tab=ivector(1,NCOVMAX);
1.144 brouard 11190: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11191: 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 11192:
1.136 brouard 11193: /* Reads data from file datafile */
11194: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11195: goto end;
11196:
11197: /* Calculation of the number of parameters from char model */
1.234 brouard 11198: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11199: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11200: k=3 V4 Tvar[k=3]= 4 (from V4)
11201: k=2 V1 Tvar[k=2]= 1 (from V1)
11202: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11203: */
11204:
11205: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11206: TvarsDind=ivector(1,NCOVMAX); /* */
11207: TvarsD=ivector(1,NCOVMAX); /* */
11208: TvarsQind=ivector(1,NCOVMAX); /* */
11209: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11210: TvarF=ivector(1,NCOVMAX); /* */
11211: TvarFind=ivector(1,NCOVMAX); /* */
11212: TvarV=ivector(1,NCOVMAX); /* */
11213: TvarVind=ivector(1,NCOVMAX); /* */
11214: TvarA=ivector(1,NCOVMAX); /* */
11215: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11216: TvarFD=ivector(1,NCOVMAX); /* */
11217: TvarFDind=ivector(1,NCOVMAX); /* */
11218: TvarFQ=ivector(1,NCOVMAX); /* */
11219: TvarFQind=ivector(1,NCOVMAX); /* */
11220: TvarVD=ivector(1,NCOVMAX); /* */
11221: TvarVDind=ivector(1,NCOVMAX); /* */
11222: TvarVQ=ivector(1,NCOVMAX); /* */
11223: TvarVQind=ivector(1,NCOVMAX); /* */
11224:
1.230 brouard 11225: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11226: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11227: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11228: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11229: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11230: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11231: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11232: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11233: */
11234: /* For model-covariate k tells which data-covariate to use but
11235: because this model-covariate is a construction we invent a new column
11236: ncovcol + k1
11237: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11238: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11239: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11240: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11241: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11242: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11243: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11244: */
1.145 brouard 11245: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11246: 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 11247: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11248: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11249: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11250: 4 covariates (3 plus signs)
11251: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11252: */
1.230 brouard 11253: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11254: * individual dummy, fixed or varying:
11255: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11256: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11257: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11258: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11259: * Tmodelind[1]@9={9,0,3,2,}*/
11260: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11261: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11262: * individual quantitative, fixed or varying:
11263: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11264: * 3, 1, 0, 0, 0, 0, 0, 0},
11265: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11266: /* Main decodemodel */
11267:
1.187 brouard 11268:
1.223 brouard 11269: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11270: goto end;
11271:
1.137 brouard 11272: if((double)(lastobs-imx)/(double)imx > 1.10){
11273: nbwarn++;
11274: 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);
11275: 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);
11276: }
1.136 brouard 11277: /* if(mle==1){*/
1.137 brouard 11278: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11279: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11280: }
11281:
11282: /*-calculation of age at interview from date of interview and age at death -*/
11283: agev=matrix(1,maxwav,1,imx);
11284:
11285: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11286: goto end;
11287:
1.126 brouard 11288:
1.136 brouard 11289: agegomp=(int)agemin;
11290: free_vector(moisnais,1,n);
11291: free_vector(annais,1,n);
1.126 brouard 11292: /* free_matrix(mint,1,maxwav,1,n);
11293: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11294: /* free_vector(moisdc,1,n); */
11295: /* free_vector(andc,1,n); */
1.145 brouard 11296: /* */
11297:
1.126 brouard 11298: wav=ivector(1,imx);
1.214 brouard 11299: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11300: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11301: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11302: 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.*/
11303: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11304: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11305:
11306: /* Concatenates waves */
1.214 brouard 11307: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11308: Death is a valid wave (if date is known).
11309: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11310: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11311: and mw[mi+1][i]. dh depends on stepm.
11312: */
11313:
1.126 brouard 11314: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11315: /* Concatenates waves */
1.145 brouard 11316:
1.215 brouard 11317: free_vector(moisdc,1,n);
11318: free_vector(andc,1,n);
11319:
1.126 brouard 11320: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11321: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11322: ncodemax[1]=1;
1.145 brouard 11323: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11324: cptcoveff=0;
1.220 brouard 11325: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11326: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11327: }
11328:
11329: ncovcombmax=pow(2,cptcoveff);
11330: invalidvarcomb=ivector(1, ncovcombmax);
11331: for(i=1;i<ncovcombmax;i++)
11332: invalidvarcomb[i]=0;
11333:
1.211 brouard 11334: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11335: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11336: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11337:
1.200 brouard 11338: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11339: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11340: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11341: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11342: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11343: * (currently 0 or 1) in the data.
11344: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11345: * corresponding modality (h,j).
11346: */
11347:
1.145 brouard 11348: h=0;
11349: /*if (cptcovn > 0) */
1.126 brouard 11350: m=pow(2,cptcoveff);
11351:
1.144 brouard 11352: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11353: * For k=4 covariates, h goes from 1 to m=2**k
11354: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11355: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11356: * h\k 1 2 3 4
1.143 brouard 11357: *______________________________
11358: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11359: * 2 2 1 1 1
11360: * 3 i=2 1 2 1 1
11361: * 4 2 2 1 1
11362: * 5 i=3 1 i=2 1 2 1
11363: * 6 2 1 2 1
11364: * 7 i=4 1 2 2 1
11365: * 8 2 2 2 1
1.197 brouard 11366: * 9 i=5 1 i=3 1 i=2 1 2
11367: * 10 2 1 1 2
11368: * 11 i=6 1 2 1 2
11369: * 12 2 2 1 2
11370: * 13 i=7 1 i=4 1 2 2
11371: * 14 2 1 2 2
11372: * 15 i=8 1 2 2 2
11373: * 16 2 2 2 2
1.143 brouard 11374: */
1.212 brouard 11375: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11376: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11377: * and the value of each covariate?
11378: * V1=1, V2=1, V3=2, V4=1 ?
11379: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11380: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11381: * In order to get the real value in the data, we use nbcode
11382: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11383: * We are keeping this crazy system in order to be able (in the future?)
11384: * to have more than 2 values (0 or 1) for a covariate.
11385: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11386: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11387: * bbbbbbbb
11388: * 76543210
11389: * h-1 00000101 (6-1=5)
1.219 brouard 11390: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11391: * &
11392: * 1 00000001 (1)
1.219 brouard 11393: * 00000000 = 1 & ((h-1) >> (k-1))
11394: * +1= 00000001 =1
1.211 brouard 11395: *
11396: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11397: * h' 1101 =2^3+2^2+0x2^1+2^0
11398: * >>k' 11
11399: * & 00000001
11400: * = 00000001
11401: * +1 = 00000010=2 = codtabm(14,3)
11402: * Reverse h=6 and m=16?
11403: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11404: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11405: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11406: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11407: * V3=decodtabm(14,3,2**4)=2
11408: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11409: *(h-1) >> (j-1) 0011 =13 >> 2
11410: * &1 000000001
11411: * = 000000001
11412: * +1= 000000010 =2
11413: * 2211
11414: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11415: * V3=2
1.220 brouard 11416: * codtabm and decodtabm are identical
1.211 brouard 11417: */
11418:
1.145 brouard 11419:
11420: free_ivector(Ndum,-1,NCOVMAX);
11421:
11422:
1.126 brouard 11423:
1.186 brouard 11424: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11425: strcpy(optionfilegnuplot,optionfilefiname);
11426: if(mle==-3)
1.201 brouard 11427: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11428: strcat(optionfilegnuplot,".gp");
11429:
11430: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11431: printf("Problem with file %s",optionfilegnuplot);
11432: }
11433: else{
1.204 brouard 11434: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11435: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11436: //fprintf(ficgp,"set missing 'NaNq'\n");
11437: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11438: }
11439: /* fclose(ficgp);*/
1.186 brouard 11440:
11441:
11442: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11443:
11444: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11445: if(mle==-3)
1.201 brouard 11446: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11447: strcat(optionfilehtm,".htm");
11448: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11449: printf("Problem with %s \n",optionfilehtm);
11450: exit(0);
1.126 brouard 11451: }
11452:
11453: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11454: strcat(optionfilehtmcov,"-cov.htm");
11455: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11456: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11457: }
11458: else{
11459: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11460: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11461: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11462: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11463: }
11464:
1.213 brouard 11465: 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 11466: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11467: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11468: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11469: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11470: \n\
11471: <hr size=\"2\" color=\"#EC5E5E\">\
11472: <ul><li><h4>Parameter files</h4>\n\
11473: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11474: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11475: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11476: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11477: - Date and time at start: %s</ul>\n",\
11478: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11479: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11480: fileres,fileres,\
11481: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11482: fflush(fichtm);
11483:
11484: strcpy(pathr,path);
11485: strcat(pathr,optionfilefiname);
1.184 brouard 11486: #ifdef WIN32
11487: _chdir(optionfilefiname); /* Move to directory named optionfile */
11488: #else
1.126 brouard 11489: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11490: #endif
11491:
1.126 brouard 11492:
1.220 brouard 11493: /* Calculates basic frequencies. Computes observed prevalence at single age
11494: and for any valid combination of covariates
1.126 brouard 11495: and prints on file fileres'p'. */
1.251 brouard 11496: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11497: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11498:
11499: fprintf(fichtm,"\n");
1.274 brouard 11500: 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",\
11501: ftol, stepm);
11502: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11503: ncurrv=1;
11504: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11505: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11506: ncurrv=i;
11507: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
11508: fprintf(fichtm,"\n<li> Number of time varying (wave varying) covariates: ntv=%d ", ntv);
11509: ncurrv=i;
11510: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
11511: fprintf(fichtm,"\n<li>Number of quantitative time varying covariates: nqtv=%d ", nqtv);
11512: ncurrv=i;
11513: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11514: 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", \
11515: nlstate, ndeath, maxwav, mle, weightopt);
11516:
11517: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11518: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11519:
11520:
11521: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11522: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11523: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11524: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11525: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11526: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11527: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11528: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11529: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11530:
1.126 brouard 11531: /* For Powell, parameters are in a vector p[] starting at p[1]
11532: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11533: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11534:
11535: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11536: /* For mortality only */
1.126 brouard 11537: if (mle==-3){
1.136 brouard 11538: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11539: for(i=1;i<=NDIM;i++)
11540: for(j=1;j<=NDIM;j++)
11541: ximort[i][j]=0.;
1.186 brouard 11542: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11543: cens=ivector(1,n);
11544: ageexmed=vector(1,n);
11545: agecens=vector(1,n);
11546: dcwave=ivector(1,n);
1.223 brouard 11547:
1.126 brouard 11548: for (i=1; i<=imx; i++){
11549: dcwave[i]=-1;
11550: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11551: if (s[m][i]>nlstate) {
11552: dcwave[i]=m;
11553: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11554: break;
11555: }
1.126 brouard 11556: }
1.226 brouard 11557:
1.126 brouard 11558: for (i=1; i<=imx; i++) {
11559: if (wav[i]>0){
1.226 brouard 11560: ageexmed[i]=agev[mw[1][i]][i];
11561: j=wav[i];
11562: agecens[i]=1.;
11563:
11564: if (ageexmed[i]> 1 && wav[i] > 0){
11565: agecens[i]=agev[mw[j][i]][i];
11566: cens[i]= 1;
11567: }else if (ageexmed[i]< 1)
11568: cens[i]= -1;
11569: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11570: cens[i]=0 ;
1.126 brouard 11571: }
11572: else cens[i]=-1;
11573: }
11574:
11575: for (i=1;i<=NDIM;i++) {
11576: for (j=1;j<=NDIM;j++)
1.226 brouard 11577: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11578: }
11579:
1.145 brouard 11580: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11581: /*printf("%lf %lf", p[1], p[2]);*/
11582:
11583:
1.136 brouard 11584: #ifdef GSL
11585: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11586: #else
1.126 brouard 11587: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11588: #endif
1.201 brouard 11589: strcpy(filerespow,"POW-MORT_");
11590: strcat(filerespow,fileresu);
1.126 brouard 11591: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11592: printf("Problem with resultfile: %s\n", filerespow);
11593: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11594: }
1.136 brouard 11595: #ifdef GSL
11596: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11597: #else
1.126 brouard 11598: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11599: #endif
1.126 brouard 11600: /* for (i=1;i<=nlstate;i++)
11601: for(j=1;j<=nlstate+ndeath;j++)
11602: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11603: */
11604: fprintf(ficrespow,"\n");
1.136 brouard 11605: #ifdef GSL
11606: /* gsl starts here */
11607: T = gsl_multimin_fminimizer_nmsimplex;
11608: gsl_multimin_fminimizer *sfm = NULL;
11609: gsl_vector *ss, *x;
11610: gsl_multimin_function minex_func;
11611:
11612: /* Initial vertex size vector */
11613: ss = gsl_vector_alloc (NDIM);
11614:
11615: if (ss == NULL){
11616: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11617: }
11618: /* Set all step sizes to 1 */
11619: gsl_vector_set_all (ss, 0.001);
11620:
11621: /* Starting point */
1.126 brouard 11622:
1.136 brouard 11623: x = gsl_vector_alloc (NDIM);
11624:
11625: if (x == NULL){
11626: gsl_vector_free(ss);
11627: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11628: }
11629:
11630: /* Initialize method and iterate */
11631: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11632: /* gsl_vector_set(x, 0, 0.0268); */
11633: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11634: gsl_vector_set(x, 0, p[1]);
11635: gsl_vector_set(x, 1, p[2]);
11636:
11637: minex_func.f = &gompertz_f;
11638: minex_func.n = NDIM;
11639: minex_func.params = (void *)&p; /* ??? */
11640:
11641: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11642: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11643:
11644: printf("Iterations beginning .....\n\n");
11645: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11646:
11647: iteri=0;
11648: while (rval == GSL_CONTINUE){
11649: iteri++;
11650: status = gsl_multimin_fminimizer_iterate(sfm);
11651:
11652: if (status) printf("error: %s\n", gsl_strerror (status));
11653: fflush(0);
11654:
11655: if (status)
11656: break;
11657:
11658: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11659: ssval = gsl_multimin_fminimizer_size (sfm);
11660:
11661: if (rval == GSL_SUCCESS)
11662: printf ("converged to a local maximum at\n");
11663:
11664: printf("%5d ", iteri);
11665: for (it = 0; it < NDIM; it++){
11666: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11667: }
11668: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11669: }
11670:
11671: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11672:
11673: gsl_vector_free(x); /* initial values */
11674: gsl_vector_free(ss); /* inital step size */
11675: for (it=0; it<NDIM; it++){
11676: p[it+1]=gsl_vector_get(sfm->x,it);
11677: fprintf(ficrespow," %.12lf", p[it]);
11678: }
11679: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11680: #endif
11681: #ifdef POWELL
11682: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11683: #endif
1.126 brouard 11684: fclose(ficrespow);
11685:
1.203 brouard 11686: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11687:
11688: for(i=1; i <=NDIM; i++)
11689: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11690: matcov[i][j]=matcov[j][i];
1.126 brouard 11691:
11692: printf("\nCovariance matrix\n ");
1.203 brouard 11693: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11694: for(i=1; i <=NDIM; i++) {
11695: for(j=1;j<=NDIM;j++){
1.220 brouard 11696: printf("%f ",matcov[i][j]);
11697: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11698: }
1.203 brouard 11699: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11700: }
11701:
11702: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11703: for (i=1;i<=NDIM;i++) {
1.126 brouard 11704: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11705: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11706: }
1.126 brouard 11707: lsurv=vector(1,AGESUP);
11708: lpop=vector(1,AGESUP);
11709: tpop=vector(1,AGESUP);
11710: lsurv[agegomp]=100000;
11711:
11712: for (k=agegomp;k<=AGESUP;k++) {
11713: agemortsup=k;
11714: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11715: }
11716:
11717: for (k=agegomp;k<agemortsup;k++)
11718: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11719:
11720: for (k=agegomp;k<agemortsup;k++){
11721: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11722: sumlpop=sumlpop+lpop[k];
11723: }
11724:
11725: tpop[agegomp]=sumlpop;
11726: for (k=agegomp;k<(agemortsup-3);k++){
11727: /* tpop[k+1]=2;*/
11728: tpop[k+1]=tpop[k]-lpop[k];
11729: }
11730:
11731:
11732: printf("\nAge lx qx dx Lx Tx e(x)\n");
11733: for (k=agegomp;k<(agemortsup-2);k++)
11734: 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]);
11735:
11736:
11737: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11738: ageminpar=50;
11739: agemaxpar=100;
1.194 brouard 11740: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11741: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11742: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11743: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11744: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11745: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11746: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11747: }else{
11748: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11749: 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 11750: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11751: }
1.201 brouard 11752: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11753: stepm, weightopt,\
11754: model,imx,p,matcov,agemortsup);
11755:
11756: free_vector(lsurv,1,AGESUP);
11757: free_vector(lpop,1,AGESUP);
11758: free_vector(tpop,1,AGESUP);
1.220 brouard 11759: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11760: free_ivector(cens,1,n);
11761: free_vector(agecens,1,n);
11762: free_ivector(dcwave,1,n);
1.220 brouard 11763: #ifdef GSL
1.136 brouard 11764: #endif
1.186 brouard 11765: } /* Endof if mle==-3 mortality only */
1.205 brouard 11766: /* Standard */
11767: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11768: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11769: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11770: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11771: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11772: for (k=1; k<=npar;k++)
11773: printf(" %d %8.5f",k,p[k]);
11774: printf("\n");
1.205 brouard 11775: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11776: /* mlikeli uses func not funcone */
1.247 brouard 11777: /* for(i=1;i<nlstate;i++){ */
11778: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11779: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11780: /* } */
1.205 brouard 11781: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11782: }
11783: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11784: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11785: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11786: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11787: }
11788: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11789: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11790: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11791: for (k=1; k<=npar;k++)
11792: printf(" %d %8.5f",k,p[k]);
11793: printf("\n");
11794:
11795: /*--------- results files --------------*/
1.224 brouard 11796: 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 11797:
11798:
11799: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11800: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11801: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11802: for(i=1,jk=1; i <=nlstate; i++){
11803: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11804: if (k != i) {
11805: printf("%d%d ",i,k);
11806: fprintf(ficlog,"%d%d ",i,k);
11807: fprintf(ficres,"%1d%1d ",i,k);
11808: for(j=1; j <=ncovmodel; j++){
11809: printf("%12.7f ",p[jk]);
11810: fprintf(ficlog,"%12.7f ",p[jk]);
11811: fprintf(ficres,"%12.7f ",p[jk]);
11812: jk++;
11813: }
11814: printf("\n");
11815: fprintf(ficlog,"\n");
11816: fprintf(ficres,"\n");
11817: }
1.126 brouard 11818: }
11819: }
1.203 brouard 11820: if(mle != 0){
11821: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11822: ftolhess=ftol; /* Usually correct */
1.203 brouard 11823: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11824: 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");
11825: 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");
11826: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11827: for(k=1; k <=(nlstate+ndeath); k++){
11828: if (k != i) {
11829: printf("%d%d ",i,k);
11830: fprintf(ficlog,"%d%d ",i,k);
11831: for(j=1; j <=ncovmodel; j++){
11832: 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]));
11833: 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]));
11834: jk++;
11835: }
11836: printf("\n");
11837: fprintf(ficlog,"\n");
11838: }
11839: }
1.193 brouard 11840: }
1.203 brouard 11841: } /* end of hesscov and Wald tests */
1.225 brouard 11842:
1.203 brouard 11843: /* */
1.126 brouard 11844: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11845: printf("# Scales (for hessian or gradient estimation)\n");
11846: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11847: for(i=1,jk=1; i <=nlstate; i++){
11848: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11849: if (j!=i) {
11850: fprintf(ficres,"%1d%1d",i,j);
11851: printf("%1d%1d",i,j);
11852: fprintf(ficlog,"%1d%1d",i,j);
11853: for(k=1; k<=ncovmodel;k++){
11854: printf(" %.5e",delti[jk]);
11855: fprintf(ficlog," %.5e",delti[jk]);
11856: fprintf(ficres," %.5e",delti[jk]);
11857: jk++;
11858: }
11859: printf("\n");
11860: fprintf(ficlog,"\n");
11861: fprintf(ficres,"\n");
11862: }
1.126 brouard 11863: }
11864: }
11865:
11866: 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 11867: if(mle >= 1) /* To big for the screen */
1.126 brouard 11868: 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");
11869: 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");
11870: /* # 121 Var(a12)\n\ */
11871: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11872: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11873: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11874: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11875: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11876: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11877: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11878:
11879:
11880: /* Just to have a covariance matrix which will be more understandable
11881: even is we still don't want to manage dictionary of variables
11882: */
11883: for(itimes=1;itimes<=2;itimes++){
11884: jj=0;
11885: for(i=1; i <=nlstate; i++){
1.225 brouard 11886: for(j=1; j <=nlstate+ndeath; j++){
11887: if(j==i) continue;
11888: for(k=1; k<=ncovmodel;k++){
11889: jj++;
11890: ca[0]= k+'a'-1;ca[1]='\0';
11891: if(itimes==1){
11892: if(mle>=1)
11893: printf("#%1d%1d%d",i,j,k);
11894: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11895: fprintf(ficres,"#%1d%1d%d",i,j,k);
11896: }else{
11897: if(mle>=1)
11898: printf("%1d%1d%d",i,j,k);
11899: fprintf(ficlog,"%1d%1d%d",i,j,k);
11900: fprintf(ficres,"%1d%1d%d",i,j,k);
11901: }
11902: ll=0;
11903: for(li=1;li <=nlstate; li++){
11904: for(lj=1;lj <=nlstate+ndeath; lj++){
11905: if(lj==li) continue;
11906: for(lk=1;lk<=ncovmodel;lk++){
11907: ll++;
11908: if(ll<=jj){
11909: cb[0]= lk +'a'-1;cb[1]='\0';
11910: if(ll<jj){
11911: if(itimes==1){
11912: if(mle>=1)
11913: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11914: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11915: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11916: }else{
11917: if(mle>=1)
11918: printf(" %.5e",matcov[jj][ll]);
11919: fprintf(ficlog," %.5e",matcov[jj][ll]);
11920: fprintf(ficres," %.5e",matcov[jj][ll]);
11921: }
11922: }else{
11923: if(itimes==1){
11924: if(mle>=1)
11925: printf(" Var(%s%1d%1d)",ca,i,j);
11926: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11927: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11928: }else{
11929: if(mle>=1)
11930: printf(" %.7e",matcov[jj][ll]);
11931: fprintf(ficlog," %.7e",matcov[jj][ll]);
11932: fprintf(ficres," %.7e",matcov[jj][ll]);
11933: }
11934: }
11935: }
11936: } /* end lk */
11937: } /* end lj */
11938: } /* end li */
11939: if(mle>=1)
11940: printf("\n");
11941: fprintf(ficlog,"\n");
11942: fprintf(ficres,"\n");
11943: numlinepar++;
11944: } /* end k*/
11945: } /*end j */
1.126 brouard 11946: } /* end i */
11947: } /* end itimes */
11948:
11949: fflush(ficlog);
11950: fflush(ficres);
1.225 brouard 11951: while(fgets(line, MAXLINE, ficpar)) {
11952: /* If line starts with a # it is a comment */
11953: if (line[0] == '#') {
11954: numlinepar++;
11955: fputs(line,stdout);
11956: fputs(line,ficparo);
11957: fputs(line,ficlog);
11958: continue;
11959: }else
11960: break;
11961: }
11962:
1.209 brouard 11963: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11964: /* ungetc(c,ficpar); */
11965: /* fgets(line, MAXLINE, ficpar); */
11966: /* fputs(line,stdout); */
11967: /* fputs(line,ficparo); */
11968: /* } */
11969: /* ungetc(c,ficpar); */
1.126 brouard 11970:
11971: estepm=0;
1.209 brouard 11972: 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 11973:
11974: if (num_filled != 6) {
11975: 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);
11976: 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);
11977: goto end;
11978: }
11979: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11980: }
11981: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11982: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11983:
1.209 brouard 11984: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11985: if (estepm==0 || estepm < stepm) estepm=stepm;
11986: if (fage <= 2) {
11987: bage = ageminpar;
11988: fage = agemaxpar;
11989: }
11990:
11991: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11992: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11993: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11994:
1.186 brouard 11995: /* Other stuffs, more or less useful */
1.254 brouard 11996: while(fgets(line, MAXLINE, ficpar)) {
11997: /* If line starts with a # it is a comment */
11998: if (line[0] == '#') {
11999: numlinepar++;
12000: fputs(line,stdout);
12001: fputs(line,ficparo);
12002: fputs(line,ficlog);
12003: continue;
12004: }else
12005: break;
12006: }
12007:
12008: 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){
12009:
12010: if (num_filled != 7) {
12011: 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);
12012: 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);
12013: goto end;
12014: }
12015: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12016: 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);
12017: 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);
12018: 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 12019: }
1.254 brouard 12020:
12021: while(fgets(line, MAXLINE, ficpar)) {
12022: /* If line starts with a # it is a comment */
12023: if (line[0] == '#') {
12024: numlinepar++;
12025: fputs(line,stdout);
12026: fputs(line,ficparo);
12027: fputs(line,ficlog);
12028: continue;
12029: }else
12030: break;
1.126 brouard 12031: }
12032:
12033:
12034: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12035: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12036:
1.254 brouard 12037: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12038: if (num_filled != 1) {
12039: 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);
12040: 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);
12041: goto end;
12042: }
12043: printf("pop_based=%d\n",popbased);
12044: fprintf(ficlog,"pop_based=%d\n",popbased);
12045: fprintf(ficparo,"pop_based=%d\n",popbased);
12046: fprintf(ficres,"pop_based=%d\n",popbased);
12047: }
12048:
1.258 brouard 12049: /* Results */
12050: nresult=0;
12051: do{
12052: if(!fgets(line, MAXLINE, ficpar)){
12053: endishere=1;
12054: parameterline=14;
12055: }else if (line[0] == '#') {
12056: /* If line starts with a # it is a comment */
1.254 brouard 12057: numlinepar++;
12058: fputs(line,stdout);
12059: fputs(line,ficparo);
12060: fputs(line,ficlog);
12061: continue;
1.258 brouard 12062: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12063: parameterline=11;
12064: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12065: parameterline=12;
12066: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12067: parameterline=13;
12068: else{
12069: parameterline=14;
1.254 brouard 12070: }
1.258 brouard 12071: switch (parameterline){
12072: case 11:
12073: 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){
12074: if (num_filled != 8) {
12075: 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);
12076: 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);
12077: goto end;
12078: }
12079: 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);
12080: 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);
12081: 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);
12082: 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);
12083: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12084: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12085: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12086:
1.258 brouard 12087: }
1.254 brouard 12088: break;
1.258 brouard 12089: case 12:
12090: /*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);*/
12091: 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){
12092: if (num_filled != 8) {
1.262 brouard 12093: 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);
12094: 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 12095: goto end;
12096: }
12097: 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);
12098: 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);
12099: 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);
12100: 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);
12101: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12102: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12103: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12104: }
1.230 brouard 12105: break;
1.258 brouard 12106: case 13:
12107: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12108: if (num_filled == 0){
12109: resultline[0]='\0';
12110: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12111: 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);
12112: break;
12113: } else if (num_filled != 1){
12114: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12115: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12116: }
12117: nresult++; /* Sum of resultlines */
12118: printf("Result %d: result=%s\n",nresult, resultline);
12119: if(nresult > MAXRESULTLINES){
12120: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12121: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12122: goto end;
12123: }
12124: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12125: fprintf(ficparo,"result: %s\n",resultline);
12126: fprintf(ficres,"result: %s\n",resultline);
12127: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12128: break;
1.258 brouard 12129: case 14:
1.259 brouard 12130: if(ncovmodel >2 && nresult==0 ){
12131: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12132: goto end;
12133: }
1.259 brouard 12134: break;
1.258 brouard 12135: default:
12136: nresult=1;
12137: decoderesult(".",nresult ); /* No covariate */
12138: }
12139: } /* End switch parameterline */
12140: }while(endishere==0); /* End do */
1.126 brouard 12141:
1.230 brouard 12142: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12143: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12144:
12145: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12146: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12147: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12148: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12149: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12150: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12151: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12152: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12153: }else{
1.270 brouard 12154: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12155: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12156: }
12157: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12158: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12159: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12160:
1.225 brouard 12161: /*------------ free_vector -------------*/
12162: /* chdir(path); */
1.220 brouard 12163:
1.215 brouard 12164: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12165: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12166: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12167: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12168: free_lvector(num,1,n);
12169: free_vector(agedc,1,n);
12170: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12171: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12172: fclose(ficparo);
12173: fclose(ficres);
1.220 brouard 12174:
12175:
1.186 brouard 12176: /* Other results (useful)*/
1.220 brouard 12177:
12178:
1.126 brouard 12179: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12180: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12181: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12182: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12183: fclose(ficrespl);
12184:
12185: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12186: /*#include "hpijx.h"*/
12187: hPijx(p, bage, fage);
1.145 brouard 12188: fclose(ficrespij);
1.227 brouard 12189:
1.220 brouard 12190: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12191: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12192: k=1;
1.126 brouard 12193: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12194:
1.269 brouard 12195: /* Prevalence for each covariate combination in probs[age][status][cov] */
12196: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12197: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12198: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12199: for(k=1;k<=ncovcombmax;k++)
12200: probs[i][j][k]=0.;
1.269 brouard 12201: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12202: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12203: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12204: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12205: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12206: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12207: for(k=1;k<=ncovcombmax;k++)
12208: mobaverages[i][j][k]=0.;
1.219 brouard 12209: mobaverage=mobaverages;
12210: if (mobilav!=0) {
1.235 brouard 12211: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12212: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12213: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12214: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12215: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12216: }
1.269 brouard 12217: } else if (mobilavproj !=0) {
1.235 brouard 12218: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12219: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12220: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12221: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12222: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12223: }
1.269 brouard 12224: }else{
12225: printf("Internal error moving average\n");
12226: fflush(stdout);
12227: exit(1);
1.219 brouard 12228: }
12229: }/* end if moving average */
1.227 brouard 12230:
1.126 brouard 12231: /*---------- Forecasting ------------------*/
12232: if(prevfcast==1){
12233: /* if(stepm ==1){*/
1.269 brouard 12234: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12235: }
1.269 brouard 12236:
12237: /* Backcasting */
1.217 brouard 12238: if(backcast==1){
1.219 brouard 12239: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12240: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12241: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12242:
12243: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12244:
12245: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12246:
1.219 brouard 12247: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12248: fclose(ficresplb);
12249:
1.222 brouard 12250: hBijx(p, bage, fage, mobaverage);
12251: fclose(ficrespijb);
1.219 brouard 12252:
1.269 brouard 12253: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12254: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12255: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12256:
12257:
1.269 brouard 12258: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12259: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12260: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12261: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12262: } /* end Backcasting */
1.268 brouard 12263:
1.186 brouard 12264:
12265: /* ------ Other prevalence ratios------------ */
1.126 brouard 12266:
1.215 brouard 12267: free_ivector(wav,1,imx);
12268: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12269: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12270: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12271:
12272:
1.127 brouard 12273: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12274:
1.201 brouard 12275: strcpy(filerese,"E_");
12276: strcat(filerese,fileresu);
1.126 brouard 12277: if((ficreseij=fopen(filerese,"w"))==NULL) {
12278: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12279: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12280: }
1.208 brouard 12281: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12282: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12283:
12284: pstamp(ficreseij);
1.219 brouard 12285:
1.235 brouard 12286: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12287: if (cptcovn < 1){i1=1;}
12288:
12289: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12290: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12291: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12292: continue;
1.219 brouard 12293: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12294: printf("\n#****** ");
1.225 brouard 12295: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12296: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12297: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12298: }
12299: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12300: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12301: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12302: }
12303: fprintf(ficreseij,"******\n");
1.235 brouard 12304: printf("******\n");
1.219 brouard 12305:
12306: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12307: oldm=oldms;savm=savms;
1.235 brouard 12308: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12309:
1.219 brouard 12310: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12311: }
12312: fclose(ficreseij);
1.208 brouard 12313: printf("done evsij\n");fflush(stdout);
12314: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12315:
1.218 brouard 12316:
1.227 brouard 12317: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12318:
1.201 brouard 12319: strcpy(filerest,"T_");
12320: strcat(filerest,fileresu);
1.127 brouard 12321: if((ficrest=fopen(filerest,"w"))==NULL) {
12322: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12323: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12324: }
1.208 brouard 12325: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12326: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12327: strcpy(fileresstde,"STDE_");
12328: strcat(fileresstde,fileresu);
1.126 brouard 12329: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12330: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12331: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12332: }
1.227 brouard 12333: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12334: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12335:
1.201 brouard 12336: strcpy(filerescve,"CVE_");
12337: strcat(filerescve,fileresu);
1.126 brouard 12338: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12339: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12340: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12341: }
1.227 brouard 12342: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12343: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12344:
1.201 brouard 12345: strcpy(fileresv,"V_");
12346: strcat(fileresv,fileresu);
1.126 brouard 12347: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12348: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12349: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12350: }
1.227 brouard 12351: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12352: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12353:
1.235 brouard 12354: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12355: if (cptcovn < 1){i1=1;}
12356:
12357: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12358: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12359: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12360: continue;
1.242 brouard 12361: printf("\n#****** Result for:");
12362: fprintf(ficrest,"\n#****** Result for:");
12363: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12364: for(j=1;j<=cptcoveff;j++){
12365: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12366: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12367: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12368: }
1.235 brouard 12369: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12370: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12371: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12372: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12373: }
1.208 brouard 12374: fprintf(ficrest,"******\n");
1.227 brouard 12375: fprintf(ficlog,"******\n");
12376: printf("******\n");
1.208 brouard 12377:
12378: fprintf(ficresstdeij,"\n#****** ");
12379: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12380: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12381: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12382: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12383: }
1.235 brouard 12384: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12385: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12386: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12387: }
1.208 brouard 12388: fprintf(ficresstdeij,"******\n");
12389: fprintf(ficrescveij,"******\n");
12390:
12391: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12392: /* pstamp(ficresvij); */
1.225 brouard 12393: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12394: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12395: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12396: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12397: }
1.208 brouard 12398: fprintf(ficresvij,"******\n");
12399:
12400: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12401: oldm=oldms;savm=savms;
1.235 brouard 12402: printf(" cvevsij ");
12403: fprintf(ficlog, " cvevsij ");
12404: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12405: printf(" end cvevsij \n ");
12406: fprintf(ficlog, " end cvevsij \n ");
12407:
12408: /*
12409: */
12410: /* goto endfree; */
12411:
12412: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12413: pstamp(ficrest);
12414:
1.269 brouard 12415: epj=vector(1,nlstate+1);
1.208 brouard 12416: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12417: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12418: cptcod= 0; /* To be deleted */
12419: printf("varevsij vpopbased=%d \n",vpopbased);
12420: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12421: 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 12422: 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 ");
12423: if(vpopbased==1)
12424: 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);
12425: else
12426: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12427: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12428: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12429: fprintf(ficrest,"\n");
12430: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12431: printf("Computing age specific period (stable) prevalences in each health state \n");
12432: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12433: for(age=bage; age <=fage ;age++){
1.235 brouard 12434: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12435: if (vpopbased==1) {
12436: if(mobilav ==0){
12437: for(i=1; i<=nlstate;i++)
12438: prlim[i][i]=probs[(int)age][i][k];
12439: }else{ /* mobilav */
12440: for(i=1; i<=nlstate;i++)
12441: prlim[i][i]=mobaverage[(int)age][i][k];
12442: }
12443: }
1.219 brouard 12444:
1.227 brouard 12445: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12446: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12447: /* printf(" age %4.0f ",age); */
12448: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12449: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12450: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12451: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12452: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12453: }
12454: epj[nlstate+1] +=epj[j];
12455: }
12456: /* printf(" age %4.0f \n",age); */
1.219 brouard 12457:
1.227 brouard 12458: for(i=1, vepp=0.;i <=nlstate;i++)
12459: for(j=1;j <=nlstate;j++)
12460: vepp += vareij[i][j][(int)age];
12461: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12462: for(j=1;j <=nlstate;j++){
12463: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12464: }
12465: fprintf(ficrest,"\n");
12466: }
1.208 brouard 12467: } /* End vpopbased */
1.269 brouard 12468: free_vector(epj,1,nlstate+1);
1.208 brouard 12469: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12470: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12471: printf("done selection\n");fflush(stdout);
12472: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12473:
1.235 brouard 12474: } /* End k selection */
1.227 brouard 12475:
12476: printf("done State-specific expectancies\n");fflush(stdout);
12477: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12478:
1.269 brouard 12479: /* variance-covariance of period prevalence*/
12480: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12481:
1.227 brouard 12482:
12483: free_vector(weight,1,n);
12484: free_imatrix(Tvard,1,NCOVMAX,1,2);
12485: free_imatrix(s,1,maxwav+1,1,n);
12486: free_matrix(anint,1,maxwav,1,n);
12487: free_matrix(mint,1,maxwav,1,n);
12488: free_ivector(cod,1,n);
12489: free_ivector(tab,1,NCOVMAX);
12490: fclose(ficresstdeij);
12491: fclose(ficrescveij);
12492: fclose(ficresvij);
12493: fclose(ficrest);
12494: fclose(ficpar);
12495:
12496:
1.126 brouard 12497: /*---------- End : free ----------------*/
1.219 brouard 12498: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12499: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12500: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12501: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12502: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12503: } /* mle==-3 arrives here for freeing */
1.227 brouard 12504: /* endfree:*/
12505: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12506: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12507: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12508: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12509: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12510: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12511: free_matrix(covar,0,NCOVMAX,1,n);
12512: free_matrix(matcov,1,npar,1,npar);
12513: free_matrix(hess,1,npar,1,npar);
12514: /*free_vector(delti,1,npar);*/
12515: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12516: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12517: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12518: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12519:
12520: free_ivector(ncodemax,1,NCOVMAX);
12521: free_ivector(ncodemaxwundef,1,NCOVMAX);
12522: free_ivector(Dummy,-1,NCOVMAX);
12523: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12524: free_ivector(DummyV,1,NCOVMAX);
12525: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12526: free_ivector(Typevar,-1,NCOVMAX);
12527: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12528: free_ivector(TvarsQ,1,NCOVMAX);
12529: free_ivector(TvarsQind,1,NCOVMAX);
12530: free_ivector(TvarsD,1,NCOVMAX);
12531: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12532: free_ivector(TvarFD,1,NCOVMAX);
12533: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12534: free_ivector(TvarF,1,NCOVMAX);
12535: free_ivector(TvarFind,1,NCOVMAX);
12536: free_ivector(TvarV,1,NCOVMAX);
12537: free_ivector(TvarVind,1,NCOVMAX);
12538: free_ivector(TvarA,1,NCOVMAX);
12539: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12540: free_ivector(TvarFQ,1,NCOVMAX);
12541: free_ivector(TvarFQind,1,NCOVMAX);
12542: free_ivector(TvarVD,1,NCOVMAX);
12543: free_ivector(TvarVDind,1,NCOVMAX);
12544: free_ivector(TvarVQ,1,NCOVMAX);
12545: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12546: free_ivector(Tvarsel,1,NCOVMAX);
12547: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12548: free_ivector(Tposprod,1,NCOVMAX);
12549: free_ivector(Tprod,1,NCOVMAX);
12550: free_ivector(Tvaraff,1,NCOVMAX);
12551: free_ivector(invalidvarcomb,1,ncovcombmax);
12552: free_ivector(Tage,1,NCOVMAX);
12553: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12554: free_ivector(TmodelInvind,1,NCOVMAX);
12555: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12556:
12557: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12558: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12559: fflush(fichtm);
12560: fflush(ficgp);
12561:
1.227 brouard 12562:
1.126 brouard 12563: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12564: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12565: 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 12566: }else{
12567: printf("End of Imach\n");
12568: fprintf(ficlog,"End of Imach\n");
12569: }
12570: printf("See log file on %s\n",filelog);
12571: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12572: /*(void) gettimeofday(&end_time,&tzp);*/
12573: rend_time = time(NULL);
12574: end_time = *localtime(&rend_time);
12575: /* tml = *localtime(&end_time.tm_sec); */
12576: strcpy(strtend,asctime(&end_time));
1.126 brouard 12577: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12578: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12579: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12580:
1.157 brouard 12581: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12582: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12583: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12584: /* printf("Total time was %d uSec.\n", total_usecs);*/
12585: /* if(fileappend(fichtm,optionfilehtm)){ */
12586: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12587: fclose(fichtm);
12588: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12589: fclose(fichtmcov);
12590: fclose(ficgp);
12591: fclose(ficlog);
12592: /*------ End -----------*/
1.227 brouard 12593:
12594:
12595: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12596: #ifdef WIN32
1.227 brouard 12597: if (_chdir(pathcd) != 0)
12598: printf("Can't move to directory %s!\n",path);
12599: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12600: #else
1.227 brouard 12601: if(chdir(pathcd) != 0)
12602: printf("Can't move to directory %s!\n", path);
12603: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12604: #endif
1.126 brouard 12605: printf("Current directory %s!\n",pathcd);
12606: /*strcat(plotcmd,CHARSEPARATOR);*/
12607: sprintf(plotcmd,"gnuplot");
1.157 brouard 12608: #ifdef _WIN32
1.126 brouard 12609: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12610: #endif
12611: if(!stat(plotcmd,&info)){
1.158 brouard 12612: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12613: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12614: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12615: }else
12616: strcpy(pplotcmd,plotcmd);
1.157 brouard 12617: #ifdef __unix
1.126 brouard 12618: strcpy(plotcmd,GNUPLOTPROGRAM);
12619: if(!stat(plotcmd,&info)){
1.158 brouard 12620: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12621: }else
12622: strcpy(pplotcmd,plotcmd);
12623: #endif
12624: }else
12625: strcpy(pplotcmd,plotcmd);
12626:
12627: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12628: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12629:
1.126 brouard 12630: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12631: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12632: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12633: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12634: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12635: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12636: }
1.158 brouard 12637: printf(" Successful, please wait...");
1.126 brouard 12638: while (z[0] != 'q') {
12639: /* chdir(path); */
1.154 brouard 12640: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12641: scanf("%s",z);
12642: /* if (z[0] == 'c') system("./imach"); */
12643: if (z[0] == 'e') {
1.158 brouard 12644: #ifdef __APPLE__
1.152 brouard 12645: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12646: #elif __linux
12647: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12648: #else
1.152 brouard 12649: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12650: #endif
12651: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12652: system(pplotcmd);
1.126 brouard 12653: }
12654: else if (z[0] == 'g') system(plotcmd);
12655: else if (z[0] == 'q') exit(0);
12656: }
1.227 brouard 12657: end:
1.126 brouard 12658: while (z[0] != 'q') {
1.195 brouard 12659: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12660: scanf("%s",z);
12661: }
12662: }
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