Annotation of imach/src/imach.c, revision 1.277
1.277 ! brouard 1: /* $Id: imach.c,v 1.276 2017/06/30 15:48:31 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.277 ! brouard 4: Revision 1.276 2017/06/30 15:48:31 brouard
! 5: Summary: Graphs improvements
! 6:
1.276 brouard 7: Revision 1.275 2017/06/30 13:39:33 brouard
8: Summary: Saito's color
9:
1.275 brouard 10: Revision 1.274 2017/06/29 09:47:08 brouard
11: Summary: Version 0.99r14
12:
1.274 brouard 13: Revision 1.273 2017/06/27 11:06:02 brouard
14: Summary: More documentation on projections
15:
1.273 brouard 16: Revision 1.272 2017/06/27 10:22:40 brouard
17: Summary: Color of backprojection changed from 6 to 5(yellow)
18:
1.272 brouard 19: Revision 1.271 2017/06/27 10:17:50 brouard
20: Summary: Some bug with rint
21:
1.271 brouard 22: Revision 1.270 2017/05/24 05:45:29 brouard
23: *** empty log message ***
24:
1.270 brouard 25: Revision 1.269 2017/05/23 08:39:25 brouard
26: Summary: Code into subroutine, cleanings
27:
1.269 brouard 28: Revision 1.268 2017/05/18 20:09:32 brouard
29: Summary: backprojection and confidence intervals of backprevalence
30:
1.268 brouard 31: Revision 1.267 2017/05/13 10:25:05 brouard
32: Summary: temporary save for backprojection
33:
1.267 brouard 34: Revision 1.266 2017/05/13 07:26:12 brouard
35: Summary: Version 0.99r13 (improvements and bugs fixed)
36:
1.266 brouard 37: Revision 1.265 2017/04/26 16:22:11 brouard
38: Summary: imach 0.99r13 Some bugs fixed
39:
1.265 brouard 40: Revision 1.264 2017/04/26 06:01:29 brouard
41: Summary: Labels in graphs
42:
1.264 brouard 43: Revision 1.263 2017/04/24 15:23:15 brouard
44: Summary: to save
45:
1.263 brouard 46: Revision 1.262 2017/04/18 16:48:12 brouard
47: *** empty log message ***
48:
1.262 brouard 49: Revision 1.261 2017/04/05 10:14:09 brouard
50: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
51:
1.261 brouard 52: Revision 1.260 2017/04/04 17:46:59 brouard
53: Summary: Gnuplot indexations fixed (humm)
54:
1.260 brouard 55: Revision 1.259 2017/04/04 13:01:16 brouard
56: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
57:
1.259 brouard 58: Revision 1.258 2017/04/03 10:17:47 brouard
59: Summary: Version 0.99r12
60:
61: Some cleanings, conformed with updated documentation.
62:
1.258 brouard 63: Revision 1.257 2017/03/29 16:53:30 brouard
64: Summary: Temp
65:
1.257 brouard 66: Revision 1.256 2017/03/27 05:50:23 brouard
67: Summary: Temporary
68:
1.256 brouard 69: Revision 1.255 2017/03/08 16:02:28 brouard
70: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
71:
1.255 brouard 72: Revision 1.254 2017/03/08 07:13:00 brouard
73: Summary: Fixing data parameter line
74:
1.254 brouard 75: Revision 1.253 2016/12/15 11:59:41 brouard
76: Summary: 0.99 in progress
77:
1.253 brouard 78: Revision 1.252 2016/09/15 21:15:37 brouard
79: *** empty log message ***
80:
1.252 brouard 81: Revision 1.251 2016/09/15 15:01:13 brouard
82: Summary: not working
83:
1.251 brouard 84: Revision 1.250 2016/09/08 16:07:27 brouard
85: Summary: continue
86:
1.250 brouard 87: Revision 1.249 2016/09/07 17:14:18 brouard
88: Summary: Starting values from frequencies
89:
1.249 brouard 90: Revision 1.248 2016/09/07 14:10:18 brouard
91: *** empty log message ***
92:
1.248 brouard 93: Revision 1.247 2016/09/02 11:11:21 brouard
94: *** empty log message ***
95:
1.247 brouard 96: Revision 1.246 2016/09/02 08:49:22 brouard
97: *** empty log message ***
98:
1.246 brouard 99: Revision 1.245 2016/09/02 07:25:01 brouard
100: *** empty log message ***
101:
1.245 brouard 102: Revision 1.244 2016/09/02 07:17:34 brouard
103: *** empty log message ***
104:
1.244 brouard 105: Revision 1.243 2016/09/02 06:45:35 brouard
106: *** empty log message ***
107:
1.243 brouard 108: Revision 1.242 2016/08/30 15:01:20 brouard
109: Summary: Fixing a lots
110:
1.242 brouard 111: Revision 1.241 2016/08/29 17:17:25 brouard
112: Summary: gnuplot problem in Back projection to fix
113:
1.241 brouard 114: Revision 1.240 2016/08/29 07:53:18 brouard
115: Summary: Better
116:
1.240 brouard 117: Revision 1.239 2016/08/26 15:51:03 brouard
118: Summary: Improvement in Powell output in order to copy and paste
119:
120: Author:
121:
1.239 brouard 122: Revision 1.238 2016/08/26 14:23:35 brouard
123: Summary: Starting tests of 0.99
124:
1.238 brouard 125: Revision 1.237 2016/08/26 09:20:19 brouard
126: Summary: to valgrind
127:
1.237 brouard 128: Revision 1.236 2016/08/25 10:50:18 brouard
129: *** empty log message ***
130:
1.236 brouard 131: Revision 1.235 2016/08/25 06:59:23 brouard
132: *** empty log message ***
133:
1.235 brouard 134: Revision 1.234 2016/08/23 16:51:20 brouard
135: *** empty log message ***
136:
1.234 brouard 137: Revision 1.233 2016/08/23 07:40:50 brouard
138: Summary: not working
139:
1.233 brouard 140: Revision 1.232 2016/08/22 14:20:21 brouard
141: Summary: not working
142:
1.232 brouard 143: Revision 1.231 2016/08/22 07:17:15 brouard
144: Summary: not working
145:
1.231 brouard 146: Revision 1.230 2016/08/22 06:55:53 brouard
147: Summary: Not working
148:
1.230 brouard 149: Revision 1.229 2016/07/23 09:45:53 brouard
150: Summary: Completing for func too
151:
1.229 brouard 152: Revision 1.228 2016/07/22 17:45:30 brouard
153: Summary: Fixing some arrays, still debugging
154:
1.227 brouard 155: Revision 1.226 2016/07/12 18:42:34 brouard
156: Summary: temp
157:
1.226 brouard 158: Revision 1.225 2016/07/12 08:40:03 brouard
159: Summary: saving but not running
160:
1.225 brouard 161: Revision 1.224 2016/07/01 13:16:01 brouard
162: Summary: Fixes
163:
1.224 brouard 164: Revision 1.223 2016/02/19 09:23:35 brouard
165: Summary: temporary
166:
1.223 brouard 167: Revision 1.222 2016/02/17 08:14:50 brouard
168: Summary: Probably last 0.98 stable version 0.98r6
169:
1.222 brouard 170: Revision 1.221 2016/02/15 23:35:36 brouard
171: Summary: minor bug
172:
1.220 brouard 173: Revision 1.219 2016/02/15 00:48:12 brouard
174: *** empty log message ***
175:
1.219 brouard 176: Revision 1.218 2016/02/12 11:29:23 brouard
177: Summary: 0.99 Back projections
178:
1.218 brouard 179: Revision 1.217 2015/12/23 17:18:31 brouard
180: Summary: Experimental backcast
181:
1.217 brouard 182: Revision 1.216 2015/12/18 17:32:11 brouard
183: Summary: 0.98r4 Warning and status=-2
184:
185: Version 0.98r4 is now:
186: - displaying an error when status is -1, date of interview unknown and date of death known;
187: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
188: Older changes concerning s=-2, dating from 2005 have been supersed.
189:
1.216 brouard 190: Revision 1.215 2015/12/16 08:52:24 brouard
191: Summary: 0.98r4 working
192:
1.215 brouard 193: Revision 1.214 2015/12/16 06:57:54 brouard
194: Summary: temporary not working
195:
1.214 brouard 196: Revision 1.213 2015/12/11 18:22:17 brouard
197: Summary: 0.98r4
198:
1.213 brouard 199: Revision 1.212 2015/11/21 12:47:24 brouard
200: Summary: minor typo
201:
1.212 brouard 202: Revision 1.211 2015/11/21 12:41:11 brouard
203: Summary: 0.98r3 with some graph of projected cross-sectional
204:
205: Author: Nicolas Brouard
206:
1.211 brouard 207: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 208: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 209: Summary: Adding ftolpl parameter
210: Author: N Brouard
211:
212: We had difficulties to get smoothed confidence intervals. It was due
213: to the period prevalence which wasn't computed accurately. The inner
214: parameter ftolpl is now an outer parameter of the .imach parameter
215: file after estepm. If ftolpl is small 1.e-4 and estepm too,
216: computation are long.
217:
1.209 brouard 218: Revision 1.208 2015/11/17 14:31:57 brouard
219: Summary: temporary
220:
1.208 brouard 221: Revision 1.207 2015/10/27 17:36:57 brouard
222: *** empty log message ***
223:
1.207 brouard 224: Revision 1.206 2015/10/24 07:14:11 brouard
225: *** empty log message ***
226:
1.206 brouard 227: Revision 1.205 2015/10/23 15:50:53 brouard
228: Summary: 0.98r3 some clarification for graphs on likelihood contributions
229:
1.205 brouard 230: Revision 1.204 2015/10/01 16:20:26 brouard
231: Summary: Some new graphs of contribution to likelihood
232:
1.204 brouard 233: Revision 1.203 2015/09/30 17:45:14 brouard
234: Summary: looking at better estimation of the hessian
235:
236: Also a better criteria for convergence to the period prevalence And
237: therefore adding the number of years needed to converge. (The
238: prevalence in any alive state shold sum to one
239:
1.203 brouard 240: Revision 1.202 2015/09/22 19:45:16 brouard
241: Summary: Adding some overall graph on contribution to likelihood. Might change
242:
1.202 brouard 243: Revision 1.201 2015/09/15 17:34:58 brouard
244: Summary: 0.98r0
245:
246: - Some new graphs like suvival functions
247: - Some bugs fixed like model=1+age+V2.
248:
1.201 brouard 249: Revision 1.200 2015/09/09 16:53:55 brouard
250: Summary: Big bug thanks to Flavia
251:
252: Even model=1+age+V2. did not work anymore
253:
1.200 brouard 254: Revision 1.199 2015/09/07 14:09:23 brouard
255: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
256:
1.199 brouard 257: Revision 1.198 2015/09/03 07:14:39 brouard
258: Summary: 0.98q5 Flavia
259:
1.198 brouard 260: Revision 1.197 2015/09/01 18:24:39 brouard
261: *** empty log message ***
262:
1.197 brouard 263: Revision 1.196 2015/08/18 23:17:52 brouard
264: Summary: 0.98q5
265:
1.196 brouard 266: Revision 1.195 2015/08/18 16:28:39 brouard
267: Summary: Adding a hack for testing purpose
268:
269: After reading the title, ftol and model lines, if the comment line has
270: a q, starting with #q, the answer at the end of the run is quit. It
271: permits to run test files in batch with ctest. The former workaround was
272: $ echo q | imach foo.imach
273:
1.195 brouard 274: Revision 1.194 2015/08/18 13:32:00 brouard
275: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
276:
1.194 brouard 277: Revision 1.193 2015/08/04 07:17:42 brouard
278: Summary: 0.98q4
279:
1.193 brouard 280: Revision 1.192 2015/07/16 16:49:02 brouard
281: Summary: Fixing some outputs
282:
1.192 brouard 283: Revision 1.191 2015/07/14 10:00:33 brouard
284: Summary: Some fixes
285:
1.191 brouard 286: Revision 1.190 2015/05/05 08:51:13 brouard
287: Summary: Adding digits in output parameters (7 digits instead of 6)
288:
289: Fix 1+age+.
290:
1.190 brouard 291: Revision 1.189 2015/04/30 14:45:16 brouard
292: Summary: 0.98q2
293:
1.189 brouard 294: Revision 1.188 2015/04/30 08:27:53 brouard
295: *** empty log message ***
296:
1.188 brouard 297: Revision 1.187 2015/04/29 09:11:15 brouard
298: *** empty log message ***
299:
1.187 brouard 300: Revision 1.186 2015/04/23 12:01:52 brouard
301: Summary: V1*age is working now, version 0.98q1
302:
303: Some codes had been disabled in order to simplify and Vn*age was
304: working in the optimization phase, ie, giving correct MLE parameters,
305: but, as usual, outputs were not correct and program core dumped.
306:
1.186 brouard 307: Revision 1.185 2015/03/11 13:26:42 brouard
308: Summary: Inclusion of compile and links command line for Intel Compiler
309:
1.185 brouard 310: Revision 1.184 2015/03/11 11:52:39 brouard
311: Summary: Back from Windows 8. Intel Compiler
312:
1.184 brouard 313: Revision 1.183 2015/03/10 20:34:32 brouard
314: Summary: 0.98q0, trying with directest, mnbrak fixed
315:
316: We use directest instead of original Powell test; probably no
317: incidence on the results, but better justifications;
318: We fixed Numerical Recipes mnbrak routine which was wrong and gave
319: wrong results.
320:
1.183 brouard 321: Revision 1.182 2015/02/12 08:19:57 brouard
322: Summary: Trying to keep directest which seems simpler and more general
323: Author: Nicolas Brouard
324:
1.182 brouard 325: Revision 1.181 2015/02/11 23:22:24 brouard
326: Summary: Comments on Powell added
327:
328: Author:
329:
1.181 brouard 330: Revision 1.180 2015/02/11 17:33:45 brouard
331: Summary: Finishing move from main to function (hpijx and prevalence_limit)
332:
1.180 brouard 333: Revision 1.179 2015/01/04 09:57:06 brouard
334: Summary: back to OS/X
335:
1.179 brouard 336: Revision 1.178 2015/01/04 09:35:48 brouard
337: *** empty log message ***
338:
1.178 brouard 339: Revision 1.177 2015/01/03 18:40:56 brouard
340: Summary: Still testing ilc32 on OSX
341:
1.177 brouard 342: Revision 1.176 2015/01/03 16:45:04 brouard
343: *** empty log message ***
344:
1.176 brouard 345: Revision 1.175 2015/01/03 16:33:42 brouard
346: *** empty log message ***
347:
1.175 brouard 348: Revision 1.174 2015/01/03 16:15:49 brouard
349: Summary: Still in cross-compilation
350:
1.174 brouard 351: Revision 1.173 2015/01/03 12:06:26 brouard
352: Summary: trying to detect cross-compilation
353:
1.173 brouard 354: Revision 1.172 2014/12/27 12:07:47 brouard
355: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
356:
1.172 brouard 357: Revision 1.171 2014/12/23 13:26:59 brouard
358: Summary: Back from Visual C
359:
360: Still problem with utsname.h on Windows
361:
1.171 brouard 362: Revision 1.170 2014/12/23 11:17:12 brouard
363: Summary: Cleaning some \%% back to %%
364:
365: The escape was mandatory for a specific compiler (which one?), but too many warnings.
366:
1.170 brouard 367: Revision 1.169 2014/12/22 23:08:31 brouard
368: Summary: 0.98p
369:
370: Outputs some informations on compiler used, OS etc. Testing on different platforms.
371:
1.169 brouard 372: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 373: Summary: update
1.169 brouard 374:
1.168 brouard 375: Revision 1.167 2014/12/22 13:50:56 brouard
376: Summary: Testing uname and compiler version and if compiled 32 or 64
377:
378: Testing on Linux 64
379:
1.167 brouard 380: Revision 1.166 2014/12/22 11:40:47 brouard
381: *** empty log message ***
382:
1.166 brouard 383: Revision 1.165 2014/12/16 11:20:36 brouard
384: Summary: After compiling on Visual C
385:
386: * imach.c (Module): Merging 1.61 to 1.162
387:
1.165 brouard 388: Revision 1.164 2014/12/16 10:52:11 brouard
389: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
390:
391: * imach.c (Module): Merging 1.61 to 1.162
392:
1.164 brouard 393: Revision 1.163 2014/12/16 10:30:11 brouard
394: * imach.c (Module): Merging 1.61 to 1.162
395:
1.163 brouard 396: Revision 1.162 2014/09/25 11:43:39 brouard
397: Summary: temporary backup 0.99!
398:
1.162 brouard 399: Revision 1.1 2014/09/16 11:06:58 brouard
400: Summary: With some code (wrong) for nlopt
401:
402: Author:
403:
404: Revision 1.161 2014/09/15 20:41:41 brouard
405: Summary: Problem with macro SQR on Intel compiler
406:
1.161 brouard 407: Revision 1.160 2014/09/02 09:24:05 brouard
408: *** empty log message ***
409:
1.160 brouard 410: Revision 1.159 2014/09/01 10:34:10 brouard
411: Summary: WIN32
412: Author: Brouard
413:
1.159 brouard 414: Revision 1.158 2014/08/27 17:11:51 brouard
415: *** empty log message ***
416:
1.158 brouard 417: Revision 1.157 2014/08/27 16:26:55 brouard
418: Summary: Preparing windows Visual studio version
419: Author: Brouard
420:
421: In order to compile on Visual studio, time.h is now correct and time_t
422: and tm struct should be used. difftime should be used but sometimes I
423: just make the differences in raw time format (time(&now).
424: Trying to suppress #ifdef LINUX
425: Add xdg-open for __linux in order to open default browser.
426:
1.157 brouard 427: Revision 1.156 2014/08/25 20:10:10 brouard
428: *** empty log message ***
429:
1.156 brouard 430: Revision 1.155 2014/08/25 18:32:34 brouard
431: Summary: New compile, minor changes
432: Author: Brouard
433:
1.155 brouard 434: Revision 1.154 2014/06/20 17:32:08 brouard
435: Summary: Outputs now all graphs of convergence to period prevalence
436:
1.154 brouard 437: Revision 1.153 2014/06/20 16:45:46 brouard
438: Summary: If 3 live state, convergence to period prevalence on same graph
439: Author: Brouard
440:
1.153 brouard 441: Revision 1.152 2014/06/18 17:54:09 brouard
442: Summary: open browser, use gnuplot on same dir than imach if not found in the path
443:
1.152 brouard 444: Revision 1.151 2014/06/18 16:43:30 brouard
445: *** empty log message ***
446:
1.151 brouard 447: Revision 1.150 2014/06/18 16:42:35 brouard
448: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
449: Author: brouard
450:
1.150 brouard 451: Revision 1.149 2014/06/18 15:51:14 brouard
452: Summary: Some fixes in parameter files errors
453: Author: Nicolas Brouard
454:
1.149 brouard 455: Revision 1.148 2014/06/17 17:38:48 brouard
456: Summary: Nothing new
457: Author: Brouard
458:
459: Just a new packaging for OS/X version 0.98nS
460:
1.148 brouard 461: Revision 1.147 2014/06/16 10:33:11 brouard
462: *** empty log message ***
463:
1.147 brouard 464: Revision 1.146 2014/06/16 10:20:28 brouard
465: Summary: Merge
466: Author: Brouard
467:
468: Merge, before building revised version.
469:
1.146 brouard 470: Revision 1.145 2014/06/10 21:23:15 brouard
471: Summary: Debugging with valgrind
472: Author: Nicolas Brouard
473:
474: Lot of changes in order to output the results with some covariates
475: After the Edimburgh REVES conference 2014, it seems mandatory to
476: improve the code.
477: No more memory valgrind error but a lot has to be done in order to
478: continue the work of splitting the code into subroutines.
479: Also, decodemodel has been improved. Tricode is still not
480: optimal. nbcode should be improved. Documentation has been added in
481: the source code.
482:
1.144 brouard 483: Revision 1.143 2014/01/26 09:45:38 brouard
484: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
485:
486: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
487: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
488:
1.143 brouard 489: Revision 1.142 2014/01/26 03:57:36 brouard
490: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
491:
492: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
493:
1.142 brouard 494: Revision 1.141 2014/01/26 02:42:01 brouard
495: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
496:
1.141 brouard 497: Revision 1.140 2011/09/02 10:37:54 brouard
498: Summary: times.h is ok with mingw32 now.
499:
1.140 brouard 500: Revision 1.139 2010/06/14 07:50:17 brouard
501: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
502: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
503:
1.139 brouard 504: Revision 1.138 2010/04/30 18:19:40 brouard
505: *** empty log message ***
506:
1.138 brouard 507: Revision 1.137 2010/04/29 18:11:38 brouard
508: (Module): Checking covariates for more complex models
509: than V1+V2. A lot of change to be done. Unstable.
510:
1.137 brouard 511: Revision 1.136 2010/04/26 20:30:53 brouard
512: (Module): merging some libgsl code. Fixing computation
513: of likelione (using inter/intrapolation if mle = 0) in order to
514: get same likelihood as if mle=1.
515: Some cleaning of code and comments added.
516:
1.136 brouard 517: Revision 1.135 2009/10/29 15:33:14 brouard
518: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
519:
1.135 brouard 520: Revision 1.134 2009/10/29 13:18:53 brouard
521: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
522:
1.134 brouard 523: Revision 1.133 2009/07/06 10:21:25 brouard
524: just nforces
525:
1.133 brouard 526: Revision 1.132 2009/07/06 08:22:05 brouard
527: Many tings
528:
1.132 brouard 529: Revision 1.131 2009/06/20 16:22:47 brouard
530: Some dimensions resccaled
531:
1.131 brouard 532: Revision 1.130 2009/05/26 06:44:34 brouard
533: (Module): Max Covariate is now set to 20 instead of 8. A
534: lot of cleaning with variables initialized to 0. Trying to make
535: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
536:
1.130 brouard 537: Revision 1.129 2007/08/31 13:49:27 lievre
538: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
539:
1.129 lievre 540: Revision 1.128 2006/06/30 13:02:05 brouard
541: (Module): Clarifications on computing e.j
542:
1.128 brouard 543: Revision 1.127 2006/04/28 18:11:50 brouard
544: (Module): Yes the sum of survivors was wrong since
545: imach-114 because nhstepm was no more computed in the age
546: loop. Now we define nhstepma in the age loop.
547: (Module): In order to speed up (in case of numerous covariates) we
548: compute health expectancies (without variances) in a first step
549: and then all the health expectancies with variances or standard
550: deviation (needs data from the Hessian matrices) which slows the
551: computation.
552: In the future we should be able to stop the program is only health
553: expectancies and graph are needed without standard deviations.
554:
1.127 brouard 555: Revision 1.126 2006/04/28 17:23:28 brouard
556: (Module): Yes the sum of survivors was wrong since
557: imach-114 because nhstepm was no more computed in the age
558: loop. Now we define nhstepma in the age loop.
559: Version 0.98h
560:
1.126 brouard 561: Revision 1.125 2006/04/04 15:20:31 lievre
562: Errors in calculation of health expectancies. Age was not initialized.
563: Forecasting file added.
564:
565: Revision 1.124 2006/03/22 17:13:53 lievre
566: Parameters are printed with %lf instead of %f (more numbers after the comma).
567: The log-likelihood is printed in the log file
568:
569: Revision 1.123 2006/03/20 10:52:43 brouard
570: * imach.c (Module): <title> changed, corresponds to .htm file
571: name. <head> headers where missing.
572:
573: * imach.c (Module): Weights can have a decimal point as for
574: English (a comma might work with a correct LC_NUMERIC environment,
575: otherwise the weight is truncated).
576: Modification of warning when the covariates values are not 0 or
577: 1.
578: Version 0.98g
579:
580: Revision 1.122 2006/03/20 09:45:41 brouard
581: (Module): Weights can have a decimal point as for
582: English (a comma might work with a correct LC_NUMERIC environment,
583: otherwise the weight is truncated).
584: Modification of warning when the covariates values are not 0 or
585: 1.
586: Version 0.98g
587:
588: Revision 1.121 2006/03/16 17:45:01 lievre
589: * imach.c (Module): Comments concerning covariates added
590:
591: * imach.c (Module): refinements in the computation of lli if
592: status=-2 in order to have more reliable computation if stepm is
593: not 1 month. Version 0.98f
594:
595: Revision 1.120 2006/03/16 15:10:38 lievre
596: (Module): refinements in the computation of lli if
597: status=-2 in order to have more reliable computation if stepm is
598: not 1 month. Version 0.98f
599:
600: Revision 1.119 2006/03/15 17:42:26 brouard
601: (Module): Bug if status = -2, the loglikelihood was
602: computed as likelihood omitting the logarithm. Version O.98e
603:
604: Revision 1.118 2006/03/14 18:20:07 brouard
605: (Module): varevsij Comments added explaining the second
606: table of variances if popbased=1 .
607: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
608: (Module): Function pstamp added
609: (Module): Version 0.98d
610:
611: Revision 1.117 2006/03/14 17:16:22 brouard
612: (Module): varevsij Comments added explaining the second
613: table of variances if popbased=1 .
614: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
615: (Module): Function pstamp added
616: (Module): Version 0.98d
617:
618: Revision 1.116 2006/03/06 10:29:27 brouard
619: (Module): Variance-covariance wrong links and
620: varian-covariance of ej. is needed (Saito).
621:
622: Revision 1.115 2006/02/27 12:17:45 brouard
623: (Module): One freematrix added in mlikeli! 0.98c
624:
625: Revision 1.114 2006/02/26 12:57:58 brouard
626: (Module): Some improvements in processing parameter
627: filename with strsep.
628:
629: Revision 1.113 2006/02/24 14:20:24 brouard
630: (Module): Memory leaks checks with valgrind and:
631: datafile was not closed, some imatrix were not freed and on matrix
632: allocation too.
633:
634: Revision 1.112 2006/01/30 09:55:26 brouard
635: (Module): Back to gnuplot.exe instead of wgnuplot.exe
636:
637: Revision 1.111 2006/01/25 20:38:18 brouard
638: (Module): Lots of cleaning and bugs added (Gompertz)
639: (Module): Comments can be added in data file. Missing date values
640: can be a simple dot '.'.
641:
642: Revision 1.110 2006/01/25 00:51:50 brouard
643: (Module): Lots of cleaning and bugs added (Gompertz)
644:
645: Revision 1.109 2006/01/24 19:37:15 brouard
646: (Module): Comments (lines starting with a #) are allowed in data.
647:
648: Revision 1.108 2006/01/19 18:05:42 lievre
649: Gnuplot problem appeared...
650: To be fixed
651:
652: Revision 1.107 2006/01/19 16:20:37 brouard
653: Test existence of gnuplot in imach path
654:
655: Revision 1.106 2006/01/19 13:24:36 brouard
656: Some cleaning and links added in html output
657:
658: Revision 1.105 2006/01/05 20:23:19 lievre
659: *** empty log message ***
660:
661: Revision 1.104 2005/09/30 16:11:43 lievre
662: (Module): sump fixed, loop imx fixed, and simplifications.
663: (Module): If the status is missing at the last wave but we know
664: that the person is alive, then we can code his/her status as -2
665: (instead of missing=-1 in earlier versions) and his/her
666: contributions to the likelihood is 1 - Prob of dying from last
667: health status (= 1-p13= p11+p12 in the easiest case of somebody in
668: the healthy state at last known wave). Version is 0.98
669:
670: Revision 1.103 2005/09/30 15:54:49 lievre
671: (Module): sump fixed, loop imx fixed, and simplifications.
672:
673: Revision 1.102 2004/09/15 17:31:30 brouard
674: Add the possibility to read data file including tab characters.
675:
676: Revision 1.101 2004/09/15 10:38:38 brouard
677: Fix on curr_time
678:
679: Revision 1.100 2004/07/12 18:29:06 brouard
680: Add version for Mac OS X. Just define UNIX in Makefile
681:
682: Revision 1.99 2004/06/05 08:57:40 brouard
683: *** empty log message ***
684:
685: Revision 1.98 2004/05/16 15:05:56 brouard
686: New version 0.97 . First attempt to estimate force of mortality
687: directly from the data i.e. without the need of knowing the health
688: state at each age, but using a Gompertz model: log u =a + b*age .
689: This is the basic analysis of mortality and should be done before any
690: other analysis, in order to test if the mortality estimated from the
691: cross-longitudinal survey is different from the mortality estimated
692: from other sources like vital statistic data.
693:
694: The same imach parameter file can be used but the option for mle should be -3.
695:
1.133 brouard 696: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 697: former routines in order to include the new code within the former code.
698:
699: The output is very simple: only an estimate of the intercept and of
700: the slope with 95% confident intervals.
701:
702: Current limitations:
703: A) Even if you enter covariates, i.e. with the
704: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
705: B) There is no computation of Life Expectancy nor Life Table.
706:
707: Revision 1.97 2004/02/20 13:25:42 lievre
708: Version 0.96d. Population forecasting command line is (temporarily)
709: suppressed.
710:
711: Revision 1.96 2003/07/15 15:38:55 brouard
712: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
713: rewritten within the same printf. Workaround: many printfs.
714:
715: Revision 1.95 2003/07/08 07:54:34 brouard
716: * imach.c (Repository):
717: (Repository): Using imachwizard code to output a more meaningful covariance
718: matrix (cov(a12,c31) instead of numbers.
719:
720: Revision 1.94 2003/06/27 13:00:02 brouard
721: Just cleaning
722:
723: Revision 1.93 2003/06/25 16:33:55 brouard
724: (Module): On windows (cygwin) function asctime_r doesn't
725: exist so I changed back to asctime which exists.
726: (Module): Version 0.96b
727:
728: Revision 1.92 2003/06/25 16:30:45 brouard
729: (Module): On windows (cygwin) function asctime_r doesn't
730: exist so I changed back to asctime which exists.
731:
732: Revision 1.91 2003/06/25 15:30:29 brouard
733: * imach.c (Repository): Duplicated warning errors corrected.
734: (Repository): Elapsed time after each iteration is now output. It
735: helps to forecast when convergence will be reached. Elapsed time
736: is stamped in powell. We created a new html file for the graphs
737: concerning matrix of covariance. It has extension -cov.htm.
738:
739: Revision 1.90 2003/06/24 12:34:15 brouard
740: (Module): Some bugs corrected for windows. Also, when
741: mle=-1 a template is output in file "or"mypar.txt with the design
742: of the covariance matrix to be input.
743:
744: Revision 1.89 2003/06/24 12:30:52 brouard
745: (Module): Some bugs corrected for windows. Also, when
746: mle=-1 a template is output in file "or"mypar.txt with the design
747: of the covariance matrix to be input.
748:
749: Revision 1.88 2003/06/23 17:54:56 brouard
750: * 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.
751:
752: Revision 1.87 2003/06/18 12:26:01 brouard
753: Version 0.96
754:
755: Revision 1.86 2003/06/17 20:04:08 brouard
756: (Module): Change position of html and gnuplot routines and added
757: routine fileappend.
758:
759: Revision 1.85 2003/06/17 13:12:43 brouard
760: * imach.c (Repository): Check when date of death was earlier that
761: current date of interview. It may happen when the death was just
762: prior to the death. In this case, dh was negative and likelihood
763: was wrong (infinity). We still send an "Error" but patch by
764: assuming that the date of death was just one stepm after the
765: interview.
766: (Repository): Because some people have very long ID (first column)
767: we changed int to long in num[] and we added a new lvector for
768: memory allocation. But we also truncated to 8 characters (left
769: truncation)
770: (Repository): No more line truncation errors.
771:
772: Revision 1.84 2003/06/13 21:44:43 brouard
773: * imach.c (Repository): Replace "freqsummary" at a correct
774: place. It differs from routine "prevalence" which may be called
775: many times. Probs is memory consuming and must be used with
776: parcimony.
777: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
778:
779: Revision 1.83 2003/06/10 13:39:11 lievre
780: *** empty log message ***
781:
782: Revision 1.82 2003/06/05 15:57:20 brouard
783: Add log in imach.c and fullversion number is now printed.
784:
785: */
786: /*
787: Interpolated Markov Chain
788:
789: Short summary of the programme:
790:
1.227 brouard 791: This program computes Healthy Life Expectancies or State-specific
792: (if states aren't health statuses) Expectancies from
793: cross-longitudinal data. Cross-longitudinal data consist in:
794:
795: -1- a first survey ("cross") where individuals from different ages
796: are interviewed on their health status or degree of disability (in
797: the case of a health survey which is our main interest)
798:
799: -2- at least a second wave of interviews ("longitudinal") which
800: measure each change (if any) in individual health status. Health
801: expectancies are computed from the time spent in each health state
802: according to a model. More health states you consider, more time is
803: necessary to reach the Maximum Likelihood of the parameters involved
804: in the model. The simplest model is the multinomial logistic model
805: where pij is the probability to be observed in state j at the second
806: wave conditional to be observed in state i at the first
807: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
808: etc , where 'age' is age and 'sex' is a covariate. If you want to
809: have a more complex model than "constant and age", you should modify
810: the program where the markup *Covariates have to be included here
811: again* invites you to do it. More covariates you add, slower the
1.126 brouard 812: convergence.
813:
814: The advantage of this computer programme, compared to a simple
815: multinomial logistic model, is clear when the delay between waves is not
816: identical for each individual. Also, if a individual missed an
817: intermediate interview, the information is lost, but taken into
818: account using an interpolation or extrapolation.
819:
820: hPijx is the probability to be observed in state i at age x+h
821: conditional to the observed state i at age x. The delay 'h' can be
822: split into an exact number (nh*stepm) of unobserved intermediate
823: states. This elementary transition (by month, quarter,
824: semester or year) is modelled as a multinomial logistic. The hPx
825: matrix is simply the matrix product of nh*stepm elementary matrices
826: and the contribution of each individual to the likelihood is simply
827: hPijx.
828:
829: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 830: of the life expectancies. It also computes the period (stable) prevalence.
831:
832: Back prevalence and projections:
1.227 brouard 833:
834: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
835: double agemaxpar, double ftolpl, int *ncvyearp, double
836: dateprev1,double dateprev2, int firstpass, int lastpass, int
837: mobilavproj)
838:
839: Computes the back prevalence limit for any combination of
840: covariate values k at any age between ageminpar and agemaxpar and
841: returns it in **bprlim. In the loops,
842:
843: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
844: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
845:
846: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 847: Computes for any combination of covariates k and any age between bage and fage
848: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
849: oldm=oldms;savm=savms;
1.227 brouard 850:
1.267 brouard 851: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 852: Computes the transition matrix starting at age 'age' over
853: 'nhstepm*hstepm*stepm' months (i.e. until
854: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 855: nhstepm*hstepm matrices.
856:
857: Returns p3mat[i][j][h] after calling
858: p3mat[i][j][h]=matprod2(newm,
859: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
860: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
861: oldm);
1.226 brouard 862:
863: Important routines
864:
865: - func (or funcone), computes logit (pij) distinguishing
866: o fixed variables (single or product dummies or quantitative);
867: o varying variables by:
868: (1) wave (single, product dummies, quantitative),
869: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
870: % fixed dummy (treated) or quantitative (not done because time-consuming);
871: % varying dummy (not done) or quantitative (not done);
872: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
873: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
874: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
875: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
876: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 877:
1.226 brouard 878:
879:
1.133 brouard 880: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
881: Institut national d'études démographiques, Paris.
1.126 brouard 882: This software have been partly granted by Euro-REVES, a concerted action
883: from the European Union.
884: It is copyrighted identically to a GNU software product, ie programme and
885: software can be distributed freely for non commercial use. Latest version
886: can be accessed at http://euroreves.ined.fr/imach .
887:
888: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
889: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
890:
891: **********************************************************************/
892: /*
893: main
894: read parameterfile
895: read datafile
896: concatwav
897: freqsummary
898: if (mle >= 1)
899: mlikeli
900: print results files
901: if mle==1
902: computes hessian
903: read end of parameter file: agemin, agemax, bage, fage, estepm
904: begin-prev-date,...
905: open gnuplot file
906: open html file
1.145 brouard 907: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
908: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
909: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
910: freexexit2 possible for memory heap.
911:
912: h Pij x | pij_nom ficrestpij
913: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
914: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
915: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
916:
917: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
918: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
919: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
920: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
921: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
922:
1.126 brouard 923: forecasting if prevfcast==1 prevforecast call prevalence()
924: health expectancies
925: Variance-covariance of DFLE
926: prevalence()
927: movingaverage()
928: varevsij()
929: if popbased==1 varevsij(,popbased)
930: total life expectancies
931: Variance of period (stable) prevalence
932: end
933: */
934:
1.187 brouard 935: /* #define DEBUG */
936: /* #define DEBUGBRENT */
1.203 brouard 937: /* #define DEBUGLINMIN */
938: /* #define DEBUGHESS */
939: #define DEBUGHESSIJ
1.224 brouard 940: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 941: #define POWELL /* Instead of NLOPT */
1.224 brouard 942: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 943: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
944: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 945:
946: #include <math.h>
947: #include <stdio.h>
948: #include <stdlib.h>
949: #include <string.h>
1.226 brouard 950: #include <ctype.h>
1.159 brouard 951:
952: #ifdef _WIN32
953: #include <io.h>
1.172 brouard 954: #include <windows.h>
955: #include <tchar.h>
1.159 brouard 956: #else
1.126 brouard 957: #include <unistd.h>
1.159 brouard 958: #endif
1.126 brouard 959:
960: #include <limits.h>
961: #include <sys/types.h>
1.171 brouard 962:
963: #if defined(__GNUC__)
964: #include <sys/utsname.h> /* Doesn't work on Windows */
965: #endif
966:
1.126 brouard 967: #include <sys/stat.h>
968: #include <errno.h>
1.159 brouard 969: /* extern int errno; */
1.126 brouard 970:
1.157 brouard 971: /* #ifdef LINUX */
972: /* #include <time.h> */
973: /* #include "timeval.h" */
974: /* #else */
975: /* #include <sys/time.h> */
976: /* #endif */
977:
1.126 brouard 978: #include <time.h>
979:
1.136 brouard 980: #ifdef GSL
981: #include <gsl/gsl_errno.h>
982: #include <gsl/gsl_multimin.h>
983: #endif
984:
1.167 brouard 985:
1.162 brouard 986: #ifdef NLOPT
987: #include <nlopt.h>
988: typedef struct {
989: double (* function)(double [] );
990: } myfunc_data ;
991: #endif
992:
1.126 brouard 993: /* #include <libintl.h> */
994: /* #define _(String) gettext (String) */
995:
1.251 brouard 996: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 997:
998: #define GNUPLOTPROGRAM "gnuplot"
999: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1000: #define FILENAMELENGTH 132
1001:
1002: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1003: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1004:
1.144 brouard 1005: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1006: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1007:
1008: #define NINTERVMAX 8
1.144 brouard 1009: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1010: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1011: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1012: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1013: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1014: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1015: #define MAXN 20000
1.144 brouard 1016: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1017: /* #define AGESUP 130 */
1018: #define AGESUP 150
1.268 brouard 1019: #define AGEINF 0
1.218 brouard 1020: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1021: #define AGEBASE 40
1.194 brouard 1022: #define AGEOVERFLOW 1.e20
1.164 brouard 1023: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1024: #ifdef _WIN32
1025: #define DIRSEPARATOR '\\'
1026: #define CHARSEPARATOR "\\"
1027: #define ODIRSEPARATOR '/'
1028: #else
1.126 brouard 1029: #define DIRSEPARATOR '/'
1030: #define CHARSEPARATOR "/"
1031: #define ODIRSEPARATOR '\\'
1032: #endif
1033:
1.277 ! brouard 1034: /* $Id: imach.c,v 1.276 2017/06/30 15:48:31 brouard Exp $ */
1.126 brouard 1035: /* $State: Exp $ */
1.196 brouard 1036: #include "version.h"
1037: char version[]=__IMACH_VERSION__;
1.224 brouard 1038: 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";
1.277 ! brouard 1039: char fullversion[]="$Revision: 1.276 $ $Date: 2017/06/30 15:48:31 $";
1.126 brouard 1040: char strstart[80];
1041: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1042: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1043: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1044: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1045: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1046: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1047: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1048: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1049: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1050: int cptcovprodnoage=0; /**< Number of covariate products without age */
1051: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1052: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1053: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1054: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1055: int nsd=0; /**< Total number of single dummy variables (output) */
1056: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1057: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1058: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1059: int ntveff=0; /**< ntveff number of effective time varying variables */
1060: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1061: int cptcov=0; /* Working variable */
1.218 brouard 1062: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1063: int npar=NPARMAX;
1064: int nlstate=2; /* Number of live states */
1065: int ndeath=1; /* Number of dead states */
1.130 brouard 1066: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1067: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1068: int popbased=0;
1069:
1070: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1071: int maxwav=0; /* Maxim number of waves */
1072: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1073: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1074: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1075: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1076: int mle=1, weightopt=0;
1.126 brouard 1077: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1078: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1079: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1080: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1081: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1082: int selected(int kvar); /* Is covariate kvar selected for printing results */
1083:
1.130 brouard 1084: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1085: double **matprod2(); /* test */
1.126 brouard 1086: double **oldm, **newm, **savm; /* Working pointers to matrices */
1087: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1088: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1089:
1.136 brouard 1090: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1091: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1092: FILE *ficlog, *ficrespow;
1.130 brouard 1093: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1094: double fretone; /* Only one call to likelihood */
1.130 brouard 1095: long ipmx=0; /* Number of contributions */
1.126 brouard 1096: double sw; /* Sum of weights */
1097: char filerespow[FILENAMELENGTH];
1098: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1099: FILE *ficresilk;
1100: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1101: FILE *ficresprobmorprev;
1102: FILE *fichtm, *fichtmcov; /* Html File */
1103: FILE *ficreseij;
1104: char filerese[FILENAMELENGTH];
1105: FILE *ficresstdeij;
1106: char fileresstde[FILENAMELENGTH];
1107: FILE *ficrescveij;
1108: char filerescve[FILENAMELENGTH];
1109: FILE *ficresvij;
1110: char fileresv[FILENAMELENGTH];
1.269 brouard 1111:
1.126 brouard 1112: char title[MAXLINE];
1.234 brouard 1113: char model[MAXLINE]; /**< The model line */
1.217 brouard 1114: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1115: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1116: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1117: char command[FILENAMELENGTH];
1118: int outcmd=0;
1119:
1.217 brouard 1120: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1121: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1122: char filelog[FILENAMELENGTH]; /* Log file */
1123: char filerest[FILENAMELENGTH];
1124: char fileregp[FILENAMELENGTH];
1125: char popfile[FILENAMELENGTH];
1126:
1127: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1128:
1.157 brouard 1129: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1130: /* struct timezone tzp; */
1131: /* extern int gettimeofday(); */
1132: struct tm tml, *gmtime(), *localtime();
1133:
1134: extern time_t time();
1135:
1136: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1137: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1138: struct tm tm;
1139:
1.126 brouard 1140: char strcurr[80], strfor[80];
1141:
1142: char *endptr;
1143: long lval;
1144: double dval;
1145:
1146: #define NR_END 1
1147: #define FREE_ARG char*
1148: #define FTOL 1.0e-10
1149:
1150: #define NRANSI
1.240 brouard 1151: #define ITMAX 200
1152: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1153:
1154: #define TOL 2.0e-4
1155:
1156: #define CGOLD 0.3819660
1157: #define ZEPS 1.0e-10
1158: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1159:
1160: #define GOLD 1.618034
1161: #define GLIMIT 100.0
1162: #define TINY 1.0e-20
1163:
1164: static double maxarg1,maxarg2;
1165: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1166: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1167:
1168: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1169: #define rint(a) floor(a+0.5)
1.166 brouard 1170: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1171: #define mytinydouble 1.0e-16
1.166 brouard 1172: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1173: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1174: /* static double dsqrarg; */
1175: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1176: static double sqrarg;
1177: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1178: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1179: int agegomp= AGEGOMP;
1180:
1181: int imx;
1182: int stepm=1;
1183: /* Stepm, step in month: minimum step interpolation*/
1184:
1185: int estepm;
1186: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1187:
1188: int m,nb;
1189: long *num;
1.197 brouard 1190: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1191: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1192: covariate for which somebody answered excluding
1193: undefined. Usually 2: 0 and 1. */
1194: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1195: covariate for which somebody answered including
1196: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1197: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1198: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1199: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1200: double *ageexmed,*agecens;
1201: double dateintmean=0;
1202:
1203: double *weight;
1204: int **s; /* Status */
1.141 brouard 1205: double *agedc;
1.145 brouard 1206: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1207: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1208: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1209: double **coqvar; /* Fixed quantitative covariate nqv */
1210: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1211: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1212: double idx;
1213: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1214: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1215: /*k 1 2 3 4 5 6 7 8 9 */
1216: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1217: /* Tndvar[k] 1 2 3 4 5 */
1218: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1219: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1220: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1221: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1222: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1223: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1224: /* Tprod[i]=k 4 7 */
1225: /* Tage[i]=k 5 8 */
1226: /* */
1227: /* Type */
1228: /* V 1 2 3 4 5 */
1229: /* F F V V V */
1230: /* D Q D D Q */
1231: /* */
1232: int *TvarsD;
1233: int *TvarsDind;
1234: int *TvarsQ;
1235: int *TvarsQind;
1236:
1.235 brouard 1237: #define MAXRESULTLINES 10
1238: int nresult=0;
1.258 brouard 1239: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1240: int TKresult[MAXRESULTLINES];
1.237 brouard 1241: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1242: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1243: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1244: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1245: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1246: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1247:
1.234 brouard 1248: /* 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 1249: 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 */
1250: 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 */
1251: 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 */
1252: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1253: 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 */
1254: 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 1255: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1256: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1257: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1258: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1259: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1260: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1261: 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 */
1262: 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 */
1263:
1.230 brouard 1264: int *Tvarsel; /**< Selected covariates for output */
1265: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1266: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1267: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1268: 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 1269: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1270: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1271: int *Tage;
1.227 brouard 1272: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1273: 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 1274: 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*/
1275: 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 1276: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1277: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1278: int **Tvard;
1279: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1280: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1281: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1282: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1283: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1284: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1285: double *lsurv, *lpop, *tpop;
1286:
1.231 brouard 1287: #define FD 1; /* Fixed dummy covariate */
1288: #define FQ 2; /* Fixed quantitative covariate */
1289: #define FP 3; /* Fixed product covariate */
1290: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1291: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1292: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1293: #define VD 10; /* Varying dummy covariate */
1294: #define VQ 11; /* Varying quantitative covariate */
1295: #define VP 12; /* Varying product covariate */
1296: #define VPDD 13; /* Varying product dummy*dummy covariate */
1297: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1298: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1299: #define APFD 16; /* Age product * fixed dummy covariate */
1300: #define APFQ 17; /* Age product * fixed quantitative covariate */
1301: #define APVD 18; /* Age product * varying dummy covariate */
1302: #define APVQ 19; /* Age product * varying quantitative covariate */
1303:
1304: #define FTYPE 1; /* Fixed covariate */
1305: #define VTYPE 2; /* Varying covariate (loop in wave) */
1306: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1307:
1308: struct kmodel{
1309: int maintype; /* main type */
1310: int subtype; /* subtype */
1311: };
1312: struct kmodel modell[NCOVMAX];
1313:
1.143 brouard 1314: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1315: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1316:
1317: /**************** split *************************/
1318: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1319: {
1320: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1321: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1322: */
1323: char *ss; /* pointer */
1.186 brouard 1324: int l1=0, l2=0; /* length counters */
1.126 brouard 1325:
1326: l1 = strlen(path ); /* length of path */
1327: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1328: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1329: if ( ss == NULL ) { /* no directory, so determine current directory */
1330: strcpy( name, path ); /* we got the fullname name because no directory */
1331: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1332: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1333: /* get current working directory */
1334: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1335: #ifdef WIN32
1336: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1337: #else
1338: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1339: #endif
1.126 brouard 1340: return( GLOCK_ERROR_GETCWD );
1341: }
1342: /* got dirc from getcwd*/
1343: printf(" DIRC = %s \n",dirc);
1.205 brouard 1344: } else { /* strip directory from path */
1.126 brouard 1345: ss++; /* after this, the filename */
1346: l2 = strlen( ss ); /* length of filename */
1347: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1348: strcpy( name, ss ); /* save file name */
1349: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1350: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1351: printf(" DIRC2 = %s \n",dirc);
1352: }
1353: /* We add a separator at the end of dirc if not exists */
1354: l1 = strlen( dirc ); /* length of directory */
1355: if( dirc[l1-1] != DIRSEPARATOR ){
1356: dirc[l1] = DIRSEPARATOR;
1357: dirc[l1+1] = 0;
1358: printf(" DIRC3 = %s \n",dirc);
1359: }
1360: ss = strrchr( name, '.' ); /* find last / */
1361: if (ss >0){
1362: ss++;
1363: strcpy(ext,ss); /* save extension */
1364: l1= strlen( name);
1365: l2= strlen(ss)+1;
1366: strncpy( finame, name, l1-l2);
1367: finame[l1-l2]= 0;
1368: }
1369:
1370: return( 0 ); /* we're done */
1371: }
1372:
1373:
1374: /******************************************/
1375:
1376: void replace_back_to_slash(char *s, char*t)
1377: {
1378: int i;
1379: int lg=0;
1380: i=0;
1381: lg=strlen(t);
1382: for(i=0; i<= lg; i++) {
1383: (s[i] = t[i]);
1384: if (t[i]== '\\') s[i]='/';
1385: }
1386: }
1387:
1.132 brouard 1388: char *trimbb(char *out, char *in)
1.137 brouard 1389: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1390: char *s;
1391: s=out;
1392: while (*in != '\0'){
1.137 brouard 1393: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1394: in++;
1395: }
1396: *out++ = *in++;
1397: }
1398: *out='\0';
1399: return s;
1400: }
1401:
1.187 brouard 1402: /* char *substrchaine(char *out, char *in, char *chain) */
1403: /* { */
1404: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1405: /* char *s, *t; */
1406: /* t=in;s=out; */
1407: /* while ((*in != *chain) && (*in != '\0')){ */
1408: /* *out++ = *in++; */
1409: /* } */
1410:
1411: /* /\* *in matches *chain *\/ */
1412: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1413: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1414: /* } */
1415: /* in--; chain--; */
1416: /* while ( (*in != '\0')){ */
1417: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1418: /* *out++ = *in++; */
1419: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1420: /* } */
1421: /* *out='\0'; */
1422: /* out=s; */
1423: /* return out; */
1424: /* } */
1425: char *substrchaine(char *out, char *in, char *chain)
1426: {
1427: /* Substract chain 'chain' from 'in', return and output 'out' */
1428: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1429:
1430: char *strloc;
1431:
1432: strcpy (out, in);
1433: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1434: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1435: if(strloc != NULL){
1436: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1437: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1438: /* strcpy (strloc, strloc +strlen(chain));*/
1439: }
1440: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1441: return out;
1442: }
1443:
1444:
1.145 brouard 1445: char *cutl(char *blocc, char *alocc, char *in, char occ)
1446: {
1.187 brouard 1447: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1448: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1449: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1450: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1451: */
1.160 brouard 1452: char *s, *t;
1.145 brouard 1453: t=in;s=in;
1454: while ((*in != occ) && (*in != '\0')){
1455: *alocc++ = *in++;
1456: }
1457: if( *in == occ){
1458: *(alocc)='\0';
1459: s=++in;
1460: }
1461:
1462: if (s == t) {/* occ not found */
1463: *(alocc-(in-s))='\0';
1464: in=s;
1465: }
1466: while ( *in != '\0'){
1467: *blocc++ = *in++;
1468: }
1469:
1470: *blocc='\0';
1471: return t;
1472: }
1.137 brouard 1473: char *cutv(char *blocc, char *alocc, char *in, char occ)
1474: {
1.187 brouard 1475: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1476: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1477: gives blocc="abcdef2ghi" and alocc="j".
1478: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1479: */
1480: char *s, *t;
1481: t=in;s=in;
1482: while (*in != '\0'){
1483: while( *in == occ){
1484: *blocc++ = *in++;
1485: s=in;
1486: }
1487: *blocc++ = *in++;
1488: }
1489: if (s == t) /* occ not found */
1490: *(blocc-(in-s))='\0';
1491: else
1492: *(blocc-(in-s)-1)='\0';
1493: in=s;
1494: while ( *in != '\0'){
1495: *alocc++ = *in++;
1496: }
1497:
1498: *alocc='\0';
1499: return s;
1500: }
1501:
1.126 brouard 1502: int nbocc(char *s, char occ)
1503: {
1504: int i,j=0;
1505: int lg=20;
1506: i=0;
1507: lg=strlen(s);
1508: for(i=0; i<= lg; i++) {
1.234 brouard 1509: if (s[i] == occ ) j++;
1.126 brouard 1510: }
1511: return j;
1512: }
1513:
1.137 brouard 1514: /* void cutv(char *u,char *v, char*t, char occ) */
1515: /* { */
1516: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1517: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1518: /* gives u="abcdef2ghi" and v="j" *\/ */
1519: /* int i,lg,j,p=0; */
1520: /* i=0; */
1521: /* lg=strlen(t); */
1522: /* for(j=0; j<=lg-1; j++) { */
1523: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1524: /* } */
1.126 brouard 1525:
1.137 brouard 1526: /* for(j=0; j<p; j++) { */
1527: /* (u[j] = t[j]); */
1528: /* } */
1529: /* u[p]='\0'; */
1.126 brouard 1530:
1.137 brouard 1531: /* for(j=0; j<= lg; j++) { */
1532: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1533: /* } */
1534: /* } */
1.126 brouard 1535:
1.160 brouard 1536: #ifdef _WIN32
1537: char * strsep(char **pp, const char *delim)
1538: {
1539: char *p, *q;
1540:
1541: if ((p = *pp) == NULL)
1542: return 0;
1543: if ((q = strpbrk (p, delim)) != NULL)
1544: {
1545: *pp = q + 1;
1546: *q = '\0';
1547: }
1548: else
1549: *pp = 0;
1550: return p;
1551: }
1552: #endif
1553:
1.126 brouard 1554: /********************** nrerror ********************/
1555:
1556: void nrerror(char error_text[])
1557: {
1558: fprintf(stderr,"ERREUR ...\n");
1559: fprintf(stderr,"%s\n",error_text);
1560: exit(EXIT_FAILURE);
1561: }
1562: /*********************** vector *******************/
1563: double *vector(int nl, int nh)
1564: {
1565: double *v;
1566: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1567: if (!v) nrerror("allocation failure in vector");
1568: return v-nl+NR_END;
1569: }
1570:
1571: /************************ free vector ******************/
1572: void free_vector(double*v, int nl, int nh)
1573: {
1574: free((FREE_ARG)(v+nl-NR_END));
1575: }
1576:
1577: /************************ivector *******************************/
1578: int *ivector(long nl,long nh)
1579: {
1580: int *v;
1581: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1582: if (!v) nrerror("allocation failure in ivector");
1583: return v-nl+NR_END;
1584: }
1585:
1586: /******************free ivector **************************/
1587: void free_ivector(int *v, long nl, long nh)
1588: {
1589: free((FREE_ARG)(v+nl-NR_END));
1590: }
1591:
1592: /************************lvector *******************************/
1593: long *lvector(long nl,long nh)
1594: {
1595: long *v;
1596: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1597: if (!v) nrerror("allocation failure in ivector");
1598: return v-nl+NR_END;
1599: }
1600:
1601: /******************free lvector **************************/
1602: void free_lvector(long *v, long nl, long nh)
1603: {
1604: free((FREE_ARG)(v+nl-NR_END));
1605: }
1606:
1607: /******************* imatrix *******************************/
1608: int **imatrix(long nrl, long nrh, long ncl, long nch)
1609: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1610: {
1611: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1612: int **m;
1613:
1614: /* allocate pointers to rows */
1615: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1616: if (!m) nrerror("allocation failure 1 in matrix()");
1617: m += NR_END;
1618: m -= nrl;
1619:
1620:
1621: /* allocate rows and set pointers to them */
1622: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1623: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1624: m[nrl] += NR_END;
1625: m[nrl] -= ncl;
1626:
1627: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1628:
1629: /* return pointer to array of pointers to rows */
1630: return m;
1631: }
1632:
1633: /****************** free_imatrix *************************/
1634: void free_imatrix(m,nrl,nrh,ncl,nch)
1635: int **m;
1636: long nch,ncl,nrh,nrl;
1637: /* free an int matrix allocated by imatrix() */
1638: {
1639: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1640: free((FREE_ARG) (m+nrl-NR_END));
1641: }
1642:
1643: /******************* matrix *******************************/
1644: double **matrix(long nrl, long nrh, long ncl, long nch)
1645: {
1646: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1647: double **m;
1648:
1649: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1650: if (!m) nrerror("allocation failure 1 in matrix()");
1651: m += NR_END;
1652: m -= nrl;
1653:
1654: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1655: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1656: m[nrl] += NR_END;
1657: m[nrl] -= ncl;
1658:
1659: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1660: return m;
1.145 brouard 1661: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1662: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1663: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1664: */
1665: }
1666:
1667: /*************************free matrix ************************/
1668: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1669: {
1670: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1671: free((FREE_ARG)(m+nrl-NR_END));
1672: }
1673:
1674: /******************* ma3x *******************************/
1675: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1676: {
1677: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1678: double ***m;
1679:
1680: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1681: if (!m) nrerror("allocation failure 1 in matrix()");
1682: m += NR_END;
1683: m -= nrl;
1684:
1685: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1686: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1687: m[nrl] += NR_END;
1688: m[nrl] -= ncl;
1689:
1690: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1691:
1692: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1693: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1694: m[nrl][ncl] += NR_END;
1695: m[nrl][ncl] -= nll;
1696: for (j=ncl+1; j<=nch; j++)
1697: m[nrl][j]=m[nrl][j-1]+nlay;
1698:
1699: for (i=nrl+1; i<=nrh; i++) {
1700: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1701: for (j=ncl+1; j<=nch; j++)
1702: m[i][j]=m[i][j-1]+nlay;
1703: }
1704: return m;
1705: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1706: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1707: */
1708: }
1709:
1710: /*************************free ma3x ************************/
1711: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1712: {
1713: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1714: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1715: free((FREE_ARG)(m+nrl-NR_END));
1716: }
1717:
1718: /*************** function subdirf ***********/
1719: char *subdirf(char fileres[])
1720: {
1721: /* Caution optionfilefiname is hidden */
1722: strcpy(tmpout,optionfilefiname);
1723: strcat(tmpout,"/"); /* Add to the right */
1724: strcat(tmpout,fileres);
1725: return tmpout;
1726: }
1727:
1728: /*************** function subdirf2 ***********/
1729: char *subdirf2(char fileres[], char *preop)
1730: {
1731:
1732: /* Caution optionfilefiname is hidden */
1733: strcpy(tmpout,optionfilefiname);
1734: strcat(tmpout,"/");
1735: strcat(tmpout,preop);
1736: strcat(tmpout,fileres);
1737: return tmpout;
1738: }
1739:
1740: /*************** function subdirf3 ***********/
1741: char *subdirf3(char fileres[], char *preop, char *preop2)
1742: {
1743:
1744: /* Caution optionfilefiname is hidden */
1745: strcpy(tmpout,optionfilefiname);
1746: strcat(tmpout,"/");
1747: strcat(tmpout,preop);
1748: strcat(tmpout,preop2);
1749: strcat(tmpout,fileres);
1750: return tmpout;
1751: }
1.213 brouard 1752:
1753: /*************** function subdirfext ***********/
1754: char *subdirfext(char fileres[], char *preop, char *postop)
1755: {
1756:
1757: strcpy(tmpout,preop);
1758: strcat(tmpout,fileres);
1759: strcat(tmpout,postop);
1760: return tmpout;
1761: }
1.126 brouard 1762:
1.213 brouard 1763: /*************** function subdirfext3 ***********/
1764: char *subdirfext3(char fileres[], char *preop, char *postop)
1765: {
1766:
1767: /* Caution optionfilefiname is hidden */
1768: strcpy(tmpout,optionfilefiname);
1769: strcat(tmpout,"/");
1770: strcat(tmpout,preop);
1771: strcat(tmpout,fileres);
1772: strcat(tmpout,postop);
1773: return tmpout;
1774: }
1775:
1.162 brouard 1776: char *asc_diff_time(long time_sec, char ascdiff[])
1777: {
1778: long sec_left, days, hours, minutes;
1779: days = (time_sec) / (60*60*24);
1780: sec_left = (time_sec) % (60*60*24);
1781: hours = (sec_left) / (60*60) ;
1782: sec_left = (sec_left) %(60*60);
1783: minutes = (sec_left) /60;
1784: sec_left = (sec_left) % (60);
1785: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1786: return ascdiff;
1787: }
1788:
1.126 brouard 1789: /***************** f1dim *************************/
1790: extern int ncom;
1791: extern double *pcom,*xicom;
1792: extern double (*nrfunc)(double []);
1793:
1794: double f1dim(double x)
1795: {
1796: int j;
1797: double f;
1798: double *xt;
1799:
1800: xt=vector(1,ncom);
1801: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1802: f=(*nrfunc)(xt);
1803: free_vector(xt,1,ncom);
1804: return f;
1805: }
1806:
1807: /*****************brent *************************/
1808: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1809: {
1810: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1811: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1812: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1813: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1814: * returned function value.
1815: */
1.126 brouard 1816: int iter;
1817: double a,b,d,etemp;
1.159 brouard 1818: double fu=0,fv,fw,fx;
1.164 brouard 1819: double ftemp=0.;
1.126 brouard 1820: double p,q,r,tol1,tol2,u,v,w,x,xm;
1821: double e=0.0;
1822:
1823: a=(ax < cx ? ax : cx);
1824: b=(ax > cx ? ax : cx);
1825: x=w=v=bx;
1826: fw=fv=fx=(*f)(x);
1827: for (iter=1;iter<=ITMAX;iter++) {
1828: xm=0.5*(a+b);
1829: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1830: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1831: printf(".");fflush(stdout);
1832: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1833: #ifdef DEBUGBRENT
1.126 brouard 1834: 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);
1835: 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);
1836: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1837: #endif
1838: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1839: *xmin=x;
1840: return fx;
1841: }
1842: ftemp=fu;
1843: if (fabs(e) > tol1) {
1844: r=(x-w)*(fx-fv);
1845: q=(x-v)*(fx-fw);
1846: p=(x-v)*q-(x-w)*r;
1847: q=2.0*(q-r);
1848: if (q > 0.0) p = -p;
1849: q=fabs(q);
1850: etemp=e;
1851: e=d;
1852: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1853: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1854: else {
1.224 brouard 1855: d=p/q;
1856: u=x+d;
1857: if (u-a < tol2 || b-u < tol2)
1858: d=SIGN(tol1,xm-x);
1.126 brouard 1859: }
1860: } else {
1861: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1862: }
1863: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1864: fu=(*f)(u);
1865: if (fu <= fx) {
1866: if (u >= x) a=x; else b=x;
1867: SHFT(v,w,x,u)
1.183 brouard 1868: SHFT(fv,fw,fx,fu)
1869: } else {
1870: if (u < x) a=u; else b=u;
1871: if (fu <= fw || w == x) {
1.224 brouard 1872: v=w;
1873: w=u;
1874: fv=fw;
1875: fw=fu;
1.183 brouard 1876: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1877: v=u;
1878: fv=fu;
1.183 brouard 1879: }
1880: }
1.126 brouard 1881: }
1882: nrerror("Too many iterations in brent");
1883: *xmin=x;
1884: return fx;
1885: }
1886:
1887: /****************** mnbrak ***********************/
1888:
1889: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1890: double (*func)(double))
1.183 brouard 1891: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1892: the downhill direction (defined by the function as evaluated at the initial points) and returns
1893: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1894: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1895: */
1.126 brouard 1896: double ulim,u,r,q, dum;
1897: double fu;
1.187 brouard 1898:
1899: double scale=10.;
1900: int iterscale=0;
1901:
1902: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1903: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1904:
1905:
1906: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1907: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1908: /* *bx = *ax - (*ax - *bx)/scale; */
1909: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1910: /* } */
1911:
1.126 brouard 1912: if (*fb > *fa) {
1913: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1914: SHFT(dum,*fb,*fa,dum)
1915: }
1.126 brouard 1916: *cx=(*bx)+GOLD*(*bx-*ax);
1917: *fc=(*func)(*cx);
1.183 brouard 1918: #ifdef DEBUG
1.224 brouard 1919: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1920: 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 1921: #endif
1.224 brouard 1922: 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 1923: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1924: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1925: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1926: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1927: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1928: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1929: fu=(*func)(u);
1.163 brouard 1930: #ifdef DEBUG
1931: /* f(x)=A(x-u)**2+f(u) */
1932: double A, fparabu;
1933: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1934: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1935: 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);
1936: 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 1937: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1938: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1939: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1940: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1941: #endif
1.184 brouard 1942: #ifdef MNBRAKORIGINAL
1.183 brouard 1943: #else
1.191 brouard 1944: /* if (fu > *fc) { */
1945: /* #ifdef DEBUG */
1946: /* printf("mnbrak4 fu > fc \n"); */
1947: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1948: /* #endif */
1949: /* /\* 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 *\\/ *\/ */
1950: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1951: /* dum=u; /\* Shifting c and u *\/ */
1952: /* u = *cx; */
1953: /* *cx = dum; */
1954: /* dum = fu; */
1955: /* fu = *fc; */
1956: /* *fc =dum; */
1957: /* } else { /\* end *\/ */
1958: /* #ifdef DEBUG */
1959: /* printf("mnbrak3 fu < fc \n"); */
1960: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1961: /* #endif */
1962: /* dum=u; /\* Shifting c and u *\/ */
1963: /* u = *cx; */
1964: /* *cx = dum; */
1965: /* dum = fu; */
1966: /* fu = *fc; */
1967: /* *fc =dum; */
1968: /* } */
1.224 brouard 1969: #ifdef DEBUGMNBRAK
1970: double A, fparabu;
1971: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1972: fparabu= *fa - A*(*ax-u)*(*ax-u);
1973: 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);
1974: 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 1975: #endif
1.191 brouard 1976: dum=u; /* Shifting c and u */
1977: u = *cx;
1978: *cx = dum;
1979: dum = fu;
1980: fu = *fc;
1981: *fc =dum;
1.183 brouard 1982: #endif
1.162 brouard 1983: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1984: #ifdef DEBUG
1.224 brouard 1985: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1986: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1987: #endif
1.126 brouard 1988: fu=(*func)(u);
1989: if (fu < *fc) {
1.183 brouard 1990: #ifdef DEBUG
1.224 brouard 1991: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1992: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1993: #endif
1994: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1995: SHFT(*fb,*fc,fu,(*func)(u))
1996: #ifdef DEBUG
1997: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1998: #endif
1999: }
1.162 brouard 2000: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2001: #ifdef DEBUG
1.224 brouard 2002: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2003: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2004: #endif
1.126 brouard 2005: u=ulim;
2006: fu=(*func)(u);
1.183 brouard 2007: } else { /* u could be left to b (if r > q parabola has a maximum) */
2008: #ifdef DEBUG
1.224 brouard 2009: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2010: 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 2011: #endif
1.126 brouard 2012: u=(*cx)+GOLD*(*cx-*bx);
2013: fu=(*func)(u);
1.224 brouard 2014: #ifdef DEBUG
2015: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2016: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2017: #endif
1.183 brouard 2018: } /* end tests */
1.126 brouard 2019: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2020: SHFT(*fa,*fb,*fc,fu)
2021: #ifdef DEBUG
1.224 brouard 2022: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2023: 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 2024: #endif
2025: } /* 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 2026: }
2027:
2028: /*************** linmin ************************/
1.162 brouard 2029: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2030: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2031: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2032: the value of func at the returned location p . This is actually all accomplished by calling the
2033: routines mnbrak and brent .*/
1.126 brouard 2034: int ncom;
2035: double *pcom,*xicom;
2036: double (*nrfunc)(double []);
2037:
1.224 brouard 2038: #ifdef LINMINORIGINAL
1.126 brouard 2039: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2040: #else
2041: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2042: #endif
1.126 brouard 2043: {
2044: double brent(double ax, double bx, double cx,
2045: double (*f)(double), double tol, double *xmin);
2046: double f1dim(double x);
2047: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2048: double *fc, double (*func)(double));
2049: int j;
2050: double xx,xmin,bx,ax;
2051: double fx,fb,fa;
1.187 brouard 2052:
1.203 brouard 2053: #ifdef LINMINORIGINAL
2054: #else
2055: double scale=10., axs, xxs; /* Scale added for infinity */
2056: #endif
2057:
1.126 brouard 2058: ncom=n;
2059: pcom=vector(1,n);
2060: xicom=vector(1,n);
2061: nrfunc=func;
2062: for (j=1;j<=n;j++) {
2063: pcom[j]=p[j];
1.202 brouard 2064: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2065: }
1.187 brouard 2066:
1.203 brouard 2067: #ifdef LINMINORIGINAL
2068: xx=1.;
2069: #else
2070: axs=0.0;
2071: xxs=1.;
2072: do{
2073: xx= xxs;
2074: #endif
1.187 brouard 2075: ax=0.;
2076: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2077: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2078: /* 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)) */
2079: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2080: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2081: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2082: /* 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 2083: #ifdef LINMINORIGINAL
2084: #else
2085: if (fx != fx){
1.224 brouard 2086: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2087: printf("|");
2088: fprintf(ficlog,"|");
1.203 brouard 2089: #ifdef DEBUGLINMIN
1.224 brouard 2090: 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 2091: #endif
2092: }
1.224 brouard 2093: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2094: #endif
2095:
1.191 brouard 2096: #ifdef DEBUGLINMIN
2097: 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 2098: 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 2099: #endif
1.224 brouard 2100: #ifdef LINMINORIGINAL
2101: #else
2102: if(fb == fx){ /* Flat function in the direction */
2103: xmin=xx;
2104: *flat=1;
2105: }else{
2106: *flat=0;
2107: #endif
2108: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2109: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2110: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2111: /* fmin = f(p[j] + xmin * xi[j]) */
2112: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2113: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2114: #ifdef DEBUG
1.224 brouard 2115: 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);
2116: 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);
2117: #endif
2118: #ifdef LINMINORIGINAL
2119: #else
2120: }
1.126 brouard 2121: #endif
1.191 brouard 2122: #ifdef DEBUGLINMIN
2123: printf("linmin end ");
1.202 brouard 2124: fprintf(ficlog,"linmin end ");
1.191 brouard 2125: #endif
1.126 brouard 2126: for (j=1;j<=n;j++) {
1.203 brouard 2127: #ifdef LINMINORIGINAL
2128: xi[j] *= xmin;
2129: #else
2130: #ifdef DEBUGLINMIN
2131: if(xxs <1.0)
2132: printf(" before xi[%d]=%12.8f", j,xi[j]);
2133: #endif
2134: 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) */
2135: #ifdef DEBUGLINMIN
2136: if(xxs <1.0)
2137: 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 );
2138: #endif
2139: #endif
1.187 brouard 2140: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2141: }
1.191 brouard 2142: #ifdef DEBUGLINMIN
1.203 brouard 2143: printf("\n");
1.191 brouard 2144: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2145: 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 2146: for (j=1;j<=n;j++) {
1.202 brouard 2147: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2148: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2149: if(j % ncovmodel == 0){
1.191 brouard 2150: printf("\n");
1.202 brouard 2151: fprintf(ficlog,"\n");
2152: }
1.191 brouard 2153: }
1.203 brouard 2154: #else
1.191 brouard 2155: #endif
1.126 brouard 2156: free_vector(xicom,1,n);
2157: free_vector(pcom,1,n);
2158: }
2159:
2160:
2161: /*************** powell ************************/
1.162 brouard 2162: /*
2163: Minimization of a function func of n variables. Input consists of an initial starting point
2164: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2165: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2166: such that failure to decrease by more than this amount on one iteration signals doneness. On
2167: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2168: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2169: */
1.224 brouard 2170: #ifdef LINMINORIGINAL
2171: #else
2172: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2173: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2174: #endif
1.126 brouard 2175: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2176: double (*func)(double []))
2177: {
1.224 brouard 2178: #ifdef LINMINORIGINAL
2179: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2180: double (*func)(double []));
1.224 brouard 2181: #else
1.241 brouard 2182: void linmin(double p[], double xi[], int n, double *fret,
2183: double (*func)(double []),int *flat);
1.224 brouard 2184: #endif
1.239 brouard 2185: int i,ibig,j,jk,k;
1.126 brouard 2186: double del,t,*pt,*ptt,*xit;
1.181 brouard 2187: double directest;
1.126 brouard 2188: double fp,fptt;
2189: double *xits;
2190: int niterf, itmp;
1.224 brouard 2191: #ifdef LINMINORIGINAL
2192: #else
2193:
2194: flatdir=ivector(1,n);
2195: for (j=1;j<=n;j++) flatdir[j]=0;
2196: #endif
1.126 brouard 2197:
2198: pt=vector(1,n);
2199: ptt=vector(1,n);
2200: xit=vector(1,n);
2201: xits=vector(1,n);
2202: *fret=(*func)(p);
2203: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2204: rcurr_time = time(NULL);
1.126 brouard 2205: for (*iter=1;;++(*iter)) {
1.187 brouard 2206: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2207: ibig=0;
2208: del=0.0;
1.157 brouard 2209: rlast_time=rcurr_time;
2210: /* (void) gettimeofday(&curr_time,&tzp); */
2211: rcurr_time = time(NULL);
2212: curr_time = *localtime(&rcurr_time);
2213: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2214: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2215: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2216: for (i=1;i<=n;i++) {
1.126 brouard 2217: fprintf(ficrespow," %.12lf", p[i]);
2218: }
1.239 brouard 2219: fprintf(ficrespow,"\n");fflush(ficrespow);
2220: printf("\n#model= 1 + age ");
2221: fprintf(ficlog,"\n#model= 1 + age ");
2222: if(nagesqr==1){
1.241 brouard 2223: printf(" + age*age ");
2224: fprintf(ficlog," + age*age ");
1.239 brouard 2225: }
2226: for(j=1;j <=ncovmodel-2;j++){
2227: if(Typevar[j]==0) {
2228: printf(" + V%d ",Tvar[j]);
2229: fprintf(ficlog," + V%d ",Tvar[j]);
2230: }else if(Typevar[j]==1) {
2231: printf(" + V%d*age ",Tvar[j]);
2232: fprintf(ficlog," + V%d*age ",Tvar[j]);
2233: }else if(Typevar[j]==2) {
2234: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2235: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2236: }
2237: }
1.126 brouard 2238: printf("\n");
1.239 brouard 2239: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2240: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2241: fprintf(ficlog,"\n");
1.239 brouard 2242: for(i=1,jk=1; i <=nlstate; i++){
2243: for(k=1; k <=(nlstate+ndeath); k++){
2244: if (k != i) {
2245: printf("%d%d ",i,k);
2246: fprintf(ficlog,"%d%d ",i,k);
2247: for(j=1; j <=ncovmodel; j++){
2248: printf("%12.7f ",p[jk]);
2249: fprintf(ficlog,"%12.7f ",p[jk]);
2250: jk++;
2251: }
2252: printf("\n");
2253: fprintf(ficlog,"\n");
2254: }
2255: }
2256: }
1.241 brouard 2257: if(*iter <=3 && *iter >1){
1.157 brouard 2258: tml = *localtime(&rcurr_time);
2259: strcpy(strcurr,asctime(&tml));
2260: rforecast_time=rcurr_time;
1.126 brouard 2261: itmp = strlen(strcurr);
2262: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2263: strcurr[itmp-1]='\0';
1.162 brouard 2264: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2265: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2266: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2267: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2268: forecast_time = *localtime(&rforecast_time);
2269: strcpy(strfor,asctime(&forecast_time));
2270: itmp = strlen(strfor);
2271: if(strfor[itmp-1]=='\n')
2272: strfor[itmp-1]='\0';
2273: 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);
2274: 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 2275: }
2276: }
1.187 brouard 2277: for (i=1;i<=n;i++) { /* For each direction i */
2278: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2279: fptt=(*fret);
2280: #ifdef DEBUG
1.203 brouard 2281: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2282: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2283: #endif
1.203 brouard 2284: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2285: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2286: #ifdef LINMINORIGINAL
1.188 brouard 2287: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2288: #else
2289: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2290: flatdir[i]=flat; /* Function is vanishing in that direction i */
2291: #endif
2292: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2293: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2294: /* because that direction will be replaced unless the gain del is small */
2295: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2296: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2297: /* with the new direction. */
2298: del=fabs(fptt-(*fret));
2299: ibig=i;
1.126 brouard 2300: }
2301: #ifdef DEBUG
2302: printf("%d %.12e",i,(*fret));
2303: fprintf(ficlog,"%d %.12e",i,(*fret));
2304: for (j=1;j<=n;j++) {
1.224 brouard 2305: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2306: printf(" x(%d)=%.12e",j,xit[j]);
2307: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2308: }
2309: for(j=1;j<=n;j++) {
1.225 brouard 2310: printf(" p(%d)=%.12e",j,p[j]);
2311: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2312: }
2313: printf("\n");
2314: fprintf(ficlog,"\n");
2315: #endif
1.187 brouard 2316: } /* end loop on each direction i */
2317: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2318: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2319: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2320: for(j=1;j<=n;j++) {
1.225 brouard 2321: if(flatdir[j] >0){
2322: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2323: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2324: }
2325: /* printf("\n"); */
2326: /* fprintf(ficlog,"\n"); */
2327: }
1.243 brouard 2328: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2329: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2330: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2331: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2332: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2333: /* decreased of more than 3.84 */
2334: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2335: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2336: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2337:
1.188 brouard 2338: /* Starting the program with initial values given by a former maximization will simply change */
2339: /* the scales of the directions and the directions, because the are reset to canonical directions */
2340: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2341: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2342: #ifdef DEBUG
2343: int k[2],l;
2344: k[0]=1;
2345: k[1]=-1;
2346: printf("Max: %.12e",(*func)(p));
2347: fprintf(ficlog,"Max: %.12e",(*func)(p));
2348: for (j=1;j<=n;j++) {
2349: printf(" %.12e",p[j]);
2350: fprintf(ficlog," %.12e",p[j]);
2351: }
2352: printf("\n");
2353: fprintf(ficlog,"\n");
2354: for(l=0;l<=1;l++) {
2355: for (j=1;j<=n;j++) {
2356: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2357: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2358: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2359: }
2360: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2361: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2362: }
2363: #endif
2364:
1.224 brouard 2365: #ifdef LINMINORIGINAL
2366: #else
2367: free_ivector(flatdir,1,n);
2368: #endif
1.126 brouard 2369: free_vector(xit,1,n);
2370: free_vector(xits,1,n);
2371: free_vector(ptt,1,n);
2372: free_vector(pt,1,n);
2373: return;
1.192 brouard 2374: } /* enough precision */
1.240 brouard 2375: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2376: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2377: ptt[j]=2.0*p[j]-pt[j];
2378: xit[j]=p[j]-pt[j];
2379: pt[j]=p[j];
2380: }
1.181 brouard 2381: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2382: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2383: if (*iter <=4) {
1.225 brouard 2384: #else
2385: #endif
1.224 brouard 2386: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2387: #else
1.161 brouard 2388: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2389: #endif
1.162 brouard 2390: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2391: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2392: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2393: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2394: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2395: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2396: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2397: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2398: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2399: /* Even if f3 <f1, directest can be negative and t >0 */
2400: /* mu² and del² are equal when f3=f1 */
2401: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2402: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2403: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2404: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2405: #ifdef NRCORIGINAL
2406: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2407: #else
2408: 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 2409: t= t- del*SQR(fp-fptt);
1.183 brouard 2410: #endif
1.202 brouard 2411: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2412: #ifdef DEBUG
1.181 brouard 2413: 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);
2414: 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 2415: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2416: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2417: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2418: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2419: 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);
2420: 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);
2421: #endif
1.183 brouard 2422: #ifdef POWELLORIGINAL
2423: if (t < 0.0) { /* Then we use it for new direction */
2424: #else
1.182 brouard 2425: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2426: 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 2427: 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 2428: 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 2429: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2430: }
1.181 brouard 2431: if (directest < 0.0) { /* Then we use it for new direction */
2432: #endif
1.191 brouard 2433: #ifdef DEBUGLINMIN
1.234 brouard 2434: printf("Before linmin in direction P%d-P0\n",n);
2435: for (j=1;j<=n;j++) {
2436: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2437: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2438: if(j % ncovmodel == 0){
2439: printf("\n");
2440: fprintf(ficlog,"\n");
2441: }
2442: }
1.224 brouard 2443: #endif
2444: #ifdef LINMINORIGINAL
1.234 brouard 2445: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2446: #else
1.234 brouard 2447: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2448: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2449: #endif
1.234 brouard 2450:
1.191 brouard 2451: #ifdef DEBUGLINMIN
1.234 brouard 2452: for (j=1;j<=n;j++) {
2453: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2454: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2455: if(j % ncovmodel == 0){
2456: printf("\n");
2457: fprintf(ficlog,"\n");
2458: }
2459: }
1.224 brouard 2460: #endif
1.234 brouard 2461: for (j=1;j<=n;j++) {
2462: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2463: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2464: }
1.224 brouard 2465: #ifdef LINMINORIGINAL
2466: #else
1.234 brouard 2467: for (j=1, flatd=0;j<=n;j++) {
2468: if(flatdir[j]>0)
2469: flatd++;
2470: }
2471: if(flatd >0){
1.255 brouard 2472: printf("%d flat directions: ",flatd);
2473: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2474: for (j=1;j<=n;j++) {
2475: if(flatdir[j]>0){
2476: printf("%d ",j);
2477: fprintf(ficlog,"%d ",j);
2478: }
2479: }
2480: printf("\n");
2481: fprintf(ficlog,"\n");
2482: }
1.191 brouard 2483: #endif
1.234 brouard 2484: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2485: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2486:
1.126 brouard 2487: #ifdef DEBUG
1.234 brouard 2488: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2489: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2490: for(j=1;j<=n;j++){
2491: printf(" %lf",xit[j]);
2492: fprintf(ficlog," %lf",xit[j]);
2493: }
2494: printf("\n");
2495: fprintf(ficlog,"\n");
1.126 brouard 2496: #endif
1.192 brouard 2497: } /* end of t or directest negative */
1.224 brouard 2498: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2499: #else
1.234 brouard 2500: } /* end if (fptt < fp) */
1.192 brouard 2501: #endif
1.225 brouard 2502: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2503: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2504: #else
1.224 brouard 2505: #endif
1.234 brouard 2506: } /* loop iteration */
1.126 brouard 2507: }
1.234 brouard 2508:
1.126 brouard 2509: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2510:
1.235 brouard 2511: 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 2512: {
1.235 brouard 2513: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2514: (and selected quantitative values in nres)
2515: by left multiplying the unit
1.234 brouard 2516: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2517: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2518: /* Wx is row vector: population in state 1, population in state 2, population dead */
2519: /* or prevalence in state 1, prevalence in state 2, 0 */
2520: /* newm is the matrix after multiplications, its rows are identical at a factor */
2521: /* Initial matrix pimij */
2522: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2523: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2524: /* 0, 0 , 1} */
2525: /*
2526: * and after some iteration: */
2527: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2528: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2529: /* 0, 0 , 1} */
2530: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2531: /* {0.51571254859325999, 0.4842874514067399, */
2532: /* 0.51326036147820708, 0.48673963852179264} */
2533: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2534:
1.126 brouard 2535: int i, ii,j,k;
1.209 brouard 2536: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2537: /* double **matprod2(); */ /* test */
1.218 brouard 2538: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2539: double **newm;
1.209 brouard 2540: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2541: int ncvloop=0;
1.169 brouard 2542:
1.209 brouard 2543: min=vector(1,nlstate);
2544: max=vector(1,nlstate);
2545: meandiff=vector(1,nlstate);
2546:
1.218 brouard 2547: /* Starting with matrix unity */
1.126 brouard 2548: for (ii=1;ii<=nlstate+ndeath;ii++)
2549: for (j=1;j<=nlstate+ndeath;j++){
2550: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2551: }
1.169 brouard 2552:
2553: cov[1]=1.;
2554:
2555: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2556: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2557: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2558: ncvloop++;
1.126 brouard 2559: newm=savm;
2560: /* Covariates have to be included here again */
1.138 brouard 2561: cov[2]=agefin;
1.187 brouard 2562: if(nagesqr==1)
2563: cov[3]= agefin*agefin;;
1.234 brouard 2564: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2565: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2566: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2567: /* 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 2568: }
2569: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2570: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2571: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2572: /* 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 2573: }
1.237 brouard 2574: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2575: if(Dummy[Tvar[Tage[k]]]){
2576: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2577: } else{
1.235 brouard 2578: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2579: }
1.235 brouard 2580: /* 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 2581: }
1.237 brouard 2582: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2583: /* 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 2584: if(Dummy[Tvard[k][1]==0]){
2585: if(Dummy[Tvard[k][2]==0]){
2586: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2587: }else{
2588: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2589: }
2590: }else{
2591: if(Dummy[Tvard[k][2]==0]){
2592: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2593: }else{
2594: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2595: }
2596: }
1.234 brouard 2597: }
1.138 brouard 2598: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2599: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2600: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2601: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2602: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2603: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2604: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2605:
1.126 brouard 2606: savm=oldm;
2607: oldm=newm;
1.209 brouard 2608:
2609: for(j=1; j<=nlstate; j++){
2610: max[j]=0.;
2611: min[j]=1.;
2612: }
2613: for(i=1;i<=nlstate;i++){
2614: sumnew=0;
2615: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2616: for(j=1; j<=nlstate; j++){
2617: prlim[i][j]= newm[i][j]/(1-sumnew);
2618: max[j]=FMAX(max[j],prlim[i][j]);
2619: min[j]=FMIN(min[j],prlim[i][j]);
2620: }
2621: }
2622:
1.126 brouard 2623: maxmax=0.;
1.209 brouard 2624: for(j=1; j<=nlstate; j++){
2625: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2626: maxmax=FMAX(maxmax,meandiff[j]);
2627: /* 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 2628: } /* j loop */
1.203 brouard 2629: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2630: /* 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 2631: if(maxmax < ftolpl){
1.209 brouard 2632: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2633: free_vector(min,1,nlstate);
2634: free_vector(max,1,nlstate);
2635: free_vector(meandiff,1,nlstate);
1.126 brouard 2636: return prlim;
2637: }
1.169 brouard 2638: } /* age loop */
1.208 brouard 2639: /* After some age loop it doesn't converge */
1.209 brouard 2640: 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 2641: 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 2642: /* 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); */
2643: free_vector(min,1,nlstate);
2644: free_vector(max,1,nlstate);
2645: free_vector(meandiff,1,nlstate);
1.208 brouard 2646:
1.169 brouard 2647: return prlim; /* should not reach here */
1.126 brouard 2648: }
2649:
1.217 brouard 2650:
2651: /**** Back Prevalence limit (stable or period prevalence) ****************/
2652:
1.218 brouard 2653: /* 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) */
2654: /* 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 2655: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2656: {
1.264 brouard 2657: /* 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 2658: matrix by transitions matrix until convergence is reached with precision ftolpl */
2659: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2660: /* Wx is row vector: population in state 1, population in state 2, population dead */
2661: /* or prevalence in state 1, prevalence in state 2, 0 */
2662: /* newm is the matrix after multiplications, its rows are identical at a factor */
2663: /* Initial matrix pimij */
2664: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2665: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2666: /* 0, 0 , 1} */
2667: /*
2668: * and after some iteration: */
2669: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2670: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2671: /* 0, 0 , 1} */
2672: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2673: /* {0.51571254859325999, 0.4842874514067399, */
2674: /* 0.51326036147820708, 0.48673963852179264} */
2675: /* If we start from prlim again, prlim tends to a constant matrix */
2676:
2677: int i, ii,j,k;
1.247 brouard 2678: int first=0;
1.217 brouard 2679: double *min, *max, *meandiff, maxmax,sumnew=0.;
2680: /* double **matprod2(); */ /* test */
2681: double **out, cov[NCOVMAX+1], **bmij();
2682: double **newm;
1.218 brouard 2683: double **dnewm, **doldm, **dsavm; /* for use */
2684: double **oldm, **savm; /* for use */
2685:
1.217 brouard 2686: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2687: int ncvloop=0;
2688:
2689: min=vector(1,nlstate);
2690: max=vector(1,nlstate);
2691: meandiff=vector(1,nlstate);
2692:
1.266 brouard 2693: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2694: oldm=oldms; savm=savms;
2695:
2696: /* Starting with matrix unity */
2697: for (ii=1;ii<=nlstate+ndeath;ii++)
2698: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2699: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2700: }
2701:
2702: cov[1]=1.;
2703:
2704: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2705: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2706: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2707: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2708: ncvloop++;
1.218 brouard 2709: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2710: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2711: /* Covariates have to be included here again */
2712: cov[2]=agefin;
2713: if(nagesqr==1)
2714: cov[3]= agefin*agefin;;
1.242 brouard 2715: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2716: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2717: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2718: /* 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 2719: }
2720: /* for (k=1; k<=cptcovn;k++) { */
2721: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2722: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2723: /* /\* 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])]); *\/ */
2724: /* } */
2725: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2726: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2727: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2728: /* 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]); */
2729: }
2730: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2731: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2732: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2733: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2734: for (k=1; k<=cptcovage;k++){ /* For product with age */
2735: if(Dummy[Tvar[Tage[k]]]){
2736: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2737: } else{
2738: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2739: }
2740: /* 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]); */
2741: }
2742: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2743: /* 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]); */
2744: if(Dummy[Tvard[k][1]==0]){
2745: if(Dummy[Tvard[k][2]==0]){
2746: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2747: }else{
2748: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2749: }
2750: }else{
2751: if(Dummy[Tvard[k][2]==0]){
2752: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2753: }else{
2754: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2755: }
2756: }
1.217 brouard 2757: }
2758:
2759: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2760: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2761: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2762: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2763: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2764: /* ij should be linked to the correct index of cov */
2765: /* age and covariate values ij are in 'cov', but we need to pass
2766: * ij for the observed prevalence at age and status and covariate
2767: * number: prevacurrent[(int)agefin][ii][ij]
2768: */
2769: /* 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 *\/ */
2770: /* 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 *\/ */
2771: 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 2772: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2773: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2774: /* for(i=1; i<=nlstate+ndeath; i++) { */
2775: /* printf("%d newm= ",i); */
2776: /* for(j=1;j<=nlstate+ndeath;j++) { */
2777: /* printf("%f ",newm[i][j]); */
2778: /* } */
2779: /* printf("oldm * "); */
2780: /* for(j=1;j<=nlstate+ndeath;j++) { */
2781: /* printf("%f ",oldm[i][j]); */
2782: /* } */
1.268 brouard 2783: /* printf(" bmmij "); */
1.266 brouard 2784: /* for(j=1;j<=nlstate+ndeath;j++) { */
2785: /* printf("%f ",pmmij[i][j]); */
2786: /* } */
2787: /* printf("\n"); */
2788: /* } */
2789: /* } */
1.217 brouard 2790: savm=oldm;
2791: oldm=newm;
1.266 brouard 2792:
1.217 brouard 2793: for(j=1; j<=nlstate; j++){
2794: max[j]=0.;
2795: min[j]=1.;
2796: }
2797: for(j=1; j<=nlstate; j++){
2798: for(i=1;i<=nlstate;i++){
1.234 brouard 2799: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2800: bprlim[i][j]= newm[i][j];
2801: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2802: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2803: }
2804: }
1.218 brouard 2805:
1.217 brouard 2806: maxmax=0.;
2807: for(i=1; i<=nlstate; i++){
2808: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2809: maxmax=FMAX(maxmax,meandiff[i]);
2810: /* 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 2811: } /* i loop */
1.217 brouard 2812: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2813: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2814: if(maxmax < ftolpl){
1.220 brouard 2815: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2816: free_vector(min,1,nlstate);
2817: free_vector(max,1,nlstate);
2818: free_vector(meandiff,1,nlstate);
2819: return bprlim;
2820: }
2821: } /* age loop */
2822: /* After some age loop it doesn't converge */
1.247 brouard 2823: if(first){
2824: first=1;
2825: 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\
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: }
2828: 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 2829: Oldest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
2830: /* 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); */
2831: free_vector(min,1,nlstate);
2832: free_vector(max,1,nlstate);
2833: free_vector(meandiff,1,nlstate);
2834:
2835: return bprlim; /* should not reach here */
2836: }
2837:
1.126 brouard 2838: /*************** transition probabilities ***************/
2839:
2840: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2841: {
1.138 brouard 2842: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2843: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2844: model to the ncovmodel covariates (including constant and age).
2845: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2846: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2847: ncth covariate in the global vector x is given by the formula:
2848: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2849: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2850: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2851: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2852: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2853: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2854: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2855: */
2856: double s1, lnpijopii;
1.126 brouard 2857: /*double t34;*/
1.164 brouard 2858: int i,j, nc, ii, jj;
1.126 brouard 2859:
1.223 brouard 2860: for(i=1; i<= nlstate; i++){
2861: for(j=1; j<i;j++){
2862: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2863: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2864: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2865: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2866: }
2867: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2868: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2869: }
2870: for(j=i+1; j<=nlstate+ndeath;j++){
2871: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2872: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2873: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2874: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2875: }
2876: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2877: }
2878: }
1.218 brouard 2879:
1.223 brouard 2880: for(i=1; i<= nlstate; i++){
2881: s1=0;
2882: for(j=1; j<i; j++){
2883: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2884: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2885: }
2886: for(j=i+1; j<=nlstate+ndeath; j++){
2887: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2888: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2889: }
2890: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2891: ps[i][i]=1./(s1+1.);
2892: /* Computing other pijs */
2893: for(j=1; j<i; j++)
2894: ps[i][j]= exp(ps[i][j])*ps[i][i];
2895: for(j=i+1; j<=nlstate+ndeath; j++)
2896: ps[i][j]= exp(ps[i][j])*ps[i][i];
2897: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2898: } /* end i */
1.218 brouard 2899:
1.223 brouard 2900: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2901: for(jj=1; jj<= nlstate+ndeath; jj++){
2902: ps[ii][jj]=0;
2903: ps[ii][ii]=1;
2904: }
2905: }
1.218 brouard 2906:
2907:
1.223 brouard 2908: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2909: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2910: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2911: /* } */
2912: /* printf("\n "); */
2913: /* } */
2914: /* printf("\n ");printf("%lf ",cov[2]);*/
2915: /*
2916: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2917: goto end;*/
1.266 brouard 2918: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2919: }
2920:
1.218 brouard 2921: /*************** backward transition probabilities ***************/
2922:
2923: /* 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 ) */
2924: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2925: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2926: {
1.266 brouard 2927: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2928: * 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 2929: */
1.218 brouard 2930: int i, ii, j,k;
1.222 brouard 2931:
2932: double **out, **pmij();
2933: double sumnew=0.;
1.218 brouard 2934: double agefin;
1.268 brouard 2935: 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 2936: double **dnewm, **dsavm, **doldm;
2937: double **bbmij;
2938:
1.218 brouard 2939: doldm=ddoldms; /* global pointers */
1.222 brouard 2940: dnewm=ddnewms;
2941: dsavm=ddsavms;
2942:
2943: agefin=cov[2];
1.268 brouard 2944: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2945: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2946: the observed prevalence (with this covariate ij) at beginning of transition */
2947: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2948:
2949: /* P_x */
1.266 brouard 2950: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2951: /* outputs pmmij which is a stochastic matrix in row */
2952:
2953: /* Diag(w_x) */
2954: /* Problem with prevacurrent which can be zero */
2955: sumnew=0.;
1.269 brouard 2956: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2957: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2958: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2959: sumnew+=prevacurrent[(int)agefin][ii][ij];
2960: }
2961: if(sumnew >0.01){ /* At least some value in the prevalence */
2962: for (ii=1;ii<=nlstate+ndeath;ii++){
2963: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2964: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2965: }
2966: }else{
2967: for (ii=1;ii<=nlstate+ndeath;ii++){
2968: for (j=1;j<=nlstate+ndeath;j++)
2969: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
2970: }
2971: /* if(sumnew <0.9){ */
2972: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2973: /* } */
2974: }
2975: k3=0.0; /* We put the last diagonal to 0 */
2976: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
2977: doldm[ii][ii]= k3;
2978: }
2979: /* End doldm, At the end doldm is diag[(w_i)] */
2980:
2981: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
2982: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
2983:
2984: /* Diag(Sum_i w^i_x p^ij_x */
2985: /* 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 2986: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 2987: sumnew=0.;
1.222 brouard 2988: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 2989: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 2990: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 2991: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 2992: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 2993: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 2994: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2995: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 2996: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2997: /* }else */
1.268 brouard 2998: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2999: } /*End ii */
3000: } /* 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 */
3001:
3002: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
3003: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3004: /* end bmij */
1.266 brouard 3005: return ps; /*pointer is unchanged */
1.218 brouard 3006: }
1.217 brouard 3007: /*************** transition probabilities ***************/
3008:
1.218 brouard 3009: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3010: {
3011: /* According to parameters values stored in x and the covariate's values stored in cov,
3012: computes the probability to be observed in state j being in state i by appying the
3013: model to the ncovmodel covariates (including constant and age).
3014: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3015: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3016: ncth covariate in the global vector x is given by the formula:
3017: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3018: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3019: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3020: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3021: Outputs ps[i][j] the probability to be observed in j being in j according to
3022: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3023: */
3024: double s1, lnpijopii;
3025: /*double t34;*/
3026: int i,j, nc, ii, jj;
3027:
1.234 brouard 3028: for(i=1; i<= nlstate; i++){
3029: for(j=1; j<i;j++){
3030: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3031: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3032: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3033: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3034: }
3035: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3036: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3037: }
3038: for(j=i+1; j<=nlstate+ndeath;j++){
3039: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3040: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3041: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3042: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3043: }
3044: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3045: }
3046: }
3047:
3048: for(i=1; i<= nlstate; i++){
3049: s1=0;
3050: for(j=1; j<i; j++){
3051: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3052: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3053: }
3054: for(j=i+1; j<=nlstate+ndeath; j++){
3055: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3056: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3057: }
3058: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3059: ps[i][i]=1./(s1+1.);
3060: /* Computing other pijs */
3061: for(j=1; j<i; j++)
3062: ps[i][j]= exp(ps[i][j])*ps[i][i];
3063: for(j=i+1; j<=nlstate+ndeath; j++)
3064: ps[i][j]= exp(ps[i][j])*ps[i][i];
3065: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3066: } /* end i */
3067:
3068: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3069: for(jj=1; jj<= nlstate+ndeath; jj++){
3070: ps[ii][jj]=0;
3071: ps[ii][ii]=1;
3072: }
3073: }
3074: /* Added for backcast */ /* Transposed matrix too */
3075: for(jj=1; jj<= nlstate+ndeath; jj++){
3076: s1=0.;
3077: for(ii=1; ii<= nlstate+ndeath; ii++){
3078: s1+=ps[ii][jj];
3079: }
3080: for(ii=1; ii<= nlstate; ii++){
3081: ps[ii][jj]=ps[ii][jj]/s1;
3082: }
3083: }
3084: /* Transposition */
3085: for(jj=1; jj<= nlstate+ndeath; jj++){
3086: for(ii=jj; ii<= nlstate+ndeath; ii++){
3087: s1=ps[ii][jj];
3088: ps[ii][jj]=ps[jj][ii];
3089: ps[jj][ii]=s1;
3090: }
3091: }
3092: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3093: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3094: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3095: /* } */
3096: /* printf("\n "); */
3097: /* } */
3098: /* printf("\n ");printf("%lf ",cov[2]);*/
3099: /*
3100: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3101: goto end;*/
3102: return ps;
1.217 brouard 3103: }
3104:
3105:
1.126 brouard 3106: /**************** Product of 2 matrices ******************/
3107:
1.145 brouard 3108: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3109: {
3110: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3111: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3112: /* in, b, out are matrice of pointers which should have been initialized
3113: before: only the contents of out is modified. The function returns
3114: a pointer to pointers identical to out */
1.145 brouard 3115: int i, j, k;
1.126 brouard 3116: for(i=nrl; i<= nrh; i++)
1.145 brouard 3117: for(k=ncolol; k<=ncoloh; k++){
3118: out[i][k]=0.;
3119: for(j=ncl; j<=nch; j++)
3120: out[i][k] +=in[i][j]*b[j][k];
3121: }
1.126 brouard 3122: return out;
3123: }
3124:
3125:
3126: /************* Higher Matrix Product ***************/
3127:
1.235 brouard 3128: 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 3129: {
1.218 brouard 3130: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3131: 'nhstepm*hstepm*stepm' months (i.e. until
3132: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3133: nhstepm*hstepm matrices.
3134: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3135: (typically every 2 years instead of every month which is too big
3136: for the memory).
3137: Model is determined by parameters x and covariates have to be
3138: included manually here.
3139:
3140: */
3141:
3142: int i, j, d, h, k;
1.131 brouard 3143: double **out, cov[NCOVMAX+1];
1.126 brouard 3144: double **newm;
1.187 brouard 3145: double agexact;
1.214 brouard 3146: double agebegin, ageend;
1.126 brouard 3147:
3148: /* Hstepm could be zero and should return the unit matrix */
3149: for (i=1;i<=nlstate+ndeath;i++)
3150: for (j=1;j<=nlstate+ndeath;j++){
3151: oldm[i][j]=(i==j ? 1.0 : 0.0);
3152: po[i][j][0]=(i==j ? 1.0 : 0.0);
3153: }
3154: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3155: for(h=1; h <=nhstepm; h++){
3156: for(d=1; d <=hstepm; d++){
3157: newm=savm;
3158: /* Covariates have to be included here again */
3159: cov[1]=1.;
1.214 brouard 3160: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3161: cov[2]=agexact;
3162: if(nagesqr==1)
1.227 brouard 3163: cov[3]= agexact*agexact;
1.235 brouard 3164: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3165: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3166: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3167: /* 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)); */
3168: }
3169: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3170: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3171: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3172: /* 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]); */
3173: }
3174: for (k=1; k<=cptcovage;k++){
3175: if(Dummy[Tvar[Tage[k]]]){
3176: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3177: } else{
3178: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3179: }
3180: /* 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]); */
3181: }
3182: for (k=1; k<=cptcovprod;k++){ /* */
3183: /* 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]); */
3184: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3185: }
3186: /* for (k=1; k<=cptcovn;k++) */
3187: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3188: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3189: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3190: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3191: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3192:
3193:
1.126 brouard 3194: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3195: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3196: /* right multiplication of oldm by the current matrix */
1.126 brouard 3197: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3198: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3199: /* if((int)age == 70){ */
3200: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3201: /* for(i=1; i<=nlstate+ndeath; i++) { */
3202: /* printf("%d pmmij ",i); */
3203: /* for(j=1;j<=nlstate+ndeath;j++) { */
3204: /* printf("%f ",pmmij[i][j]); */
3205: /* } */
3206: /* printf(" oldm "); */
3207: /* for(j=1;j<=nlstate+ndeath;j++) { */
3208: /* printf("%f ",oldm[i][j]); */
3209: /* } */
3210: /* printf("\n"); */
3211: /* } */
3212: /* } */
1.126 brouard 3213: savm=oldm;
3214: oldm=newm;
3215: }
3216: for(i=1; i<=nlstate+ndeath; i++)
3217: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3218: po[i][j][h]=newm[i][j];
3219: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3220: }
1.128 brouard 3221: /*printf("h=%d ",h);*/
1.126 brouard 3222: } /* end h */
1.267 brouard 3223: /* printf("\n H=%d \n",h); */
1.126 brouard 3224: return po;
3225: }
3226:
1.217 brouard 3227: /************* Higher Back Matrix Product ***************/
1.218 brouard 3228: /* 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 3229: 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 3230: {
1.266 brouard 3231: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3232: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3233: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3234: nhstepm*hstepm matrices.
3235: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3236: (typically every 2 years instead of every month which is too big
1.217 brouard 3237: for the memory).
1.218 brouard 3238: Model is determined by parameters x and covariates have to be
1.266 brouard 3239: included manually here. Then we use a call to bmij(x and cov)
3240: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3241: */
1.217 brouard 3242:
3243: int i, j, d, h, k;
1.266 brouard 3244: double **out, cov[NCOVMAX+1], **bmij();
3245: double **newm, ***newmm;
1.217 brouard 3246: double agexact;
3247: double agebegin, ageend;
1.222 brouard 3248: double **oldm, **savm;
1.217 brouard 3249:
1.266 brouard 3250: newmm=po; /* To be saved */
3251: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3252: /* Hstepm could be zero and should return the unit matrix */
3253: for (i=1;i<=nlstate+ndeath;i++)
3254: for (j=1;j<=nlstate+ndeath;j++){
3255: oldm[i][j]=(i==j ? 1.0 : 0.0);
3256: po[i][j][0]=(i==j ? 1.0 : 0.0);
3257: }
3258: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3259: for(h=1; h <=nhstepm; h++){
3260: for(d=1; d <=hstepm; d++){
3261: newm=savm;
3262: /* Covariates have to be included here again */
3263: cov[1]=1.;
1.271 brouard 3264: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3265: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3266: cov[2]=agexact;
3267: if(nagesqr==1)
1.222 brouard 3268: cov[3]= agexact*agexact;
1.266 brouard 3269: for (k=1; k<=cptcovn;k++){
3270: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3271: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3272: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3273: /* 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)); */
3274: }
1.267 brouard 3275: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3276: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3277: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3278: /* 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]); */
3279: }
3280: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3281: if(Dummy[Tvar[Tage[k]]]){
3282: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3283: } else{
3284: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3285: }
3286: /* 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]); */
3287: }
3288: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3289: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3290: }
1.217 brouard 3291: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3292: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3293:
1.218 brouard 3294: /* Careful transposed matrix */
1.266 brouard 3295: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3296: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3297: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3298: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3299: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3300: /* if((int)age == 70){ */
3301: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3302: /* for(i=1; i<=nlstate+ndeath; i++) { */
3303: /* printf("%d pmmij ",i); */
3304: /* for(j=1;j<=nlstate+ndeath;j++) { */
3305: /* printf("%f ",pmmij[i][j]); */
3306: /* } */
3307: /* printf(" oldm "); */
3308: /* for(j=1;j<=nlstate+ndeath;j++) { */
3309: /* printf("%f ",oldm[i][j]); */
3310: /* } */
3311: /* printf("\n"); */
3312: /* } */
3313: /* } */
3314: savm=oldm;
3315: oldm=newm;
3316: }
3317: for(i=1; i<=nlstate+ndeath; i++)
3318: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3319: po[i][j][h]=newm[i][j];
1.268 brouard 3320: /* if(h==nhstepm) */
3321: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3322: }
1.268 brouard 3323: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3324: } /* end h */
1.268 brouard 3325: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3326: return po;
3327: }
3328:
3329:
1.162 brouard 3330: #ifdef NLOPT
3331: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3332: double fret;
3333: double *xt;
3334: int j;
3335: myfunc_data *d2 = (myfunc_data *) pd;
3336: /* xt = (p1-1); */
3337: xt=vector(1,n);
3338: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3339:
3340: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3341: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3342: printf("Function = %.12lf ",fret);
3343: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3344: printf("\n");
3345: free_vector(xt,1,n);
3346: return fret;
3347: }
3348: #endif
1.126 brouard 3349:
3350: /*************** log-likelihood *************/
3351: double func( double *x)
3352: {
1.226 brouard 3353: int i, ii, j, k, mi, d, kk;
3354: int ioffset=0;
3355: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3356: double **out;
3357: double lli; /* Individual log likelihood */
3358: int s1, s2;
1.228 brouard 3359: 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 3360: double bbh, survp;
3361: long ipmx;
3362: double agexact;
3363: /*extern weight */
3364: /* We are differentiating ll according to initial status */
3365: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3366: /*for(i=1;i<imx;i++)
3367: printf(" %d\n",s[4][i]);
3368: */
1.162 brouard 3369:
1.226 brouard 3370: ++countcallfunc;
1.162 brouard 3371:
1.226 brouard 3372: cov[1]=1.;
1.126 brouard 3373:
1.226 brouard 3374: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3375: ioffset=0;
1.226 brouard 3376: if(mle==1){
3377: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3378: /* Computes the values of the ncovmodel covariates of the model
3379: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3380: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3381: to be observed in j being in i according to the model.
3382: */
1.243 brouard 3383: ioffset=2+nagesqr ;
1.233 brouard 3384: /* Fixed */
1.234 brouard 3385: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3386: 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)*/
3387: }
1.226 brouard 3388: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3389: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3390: has been calculated etc */
3391: /* For an individual i, wav[i] gives the number of effective waves */
3392: /* We compute the contribution to Likelihood of each effective transition
3393: mw[mi][i] is real wave of the mi th effectve wave */
3394: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3395: s2=s[mw[mi+1][i]][i];
3396: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3397: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3398: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3399: */
3400: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3401: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3402: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3403: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3404: }
3405: for (ii=1;ii<=nlstate+ndeath;ii++)
3406: for (j=1;j<=nlstate+ndeath;j++){
3407: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3408: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3409: }
3410: for(d=0; d<dh[mi][i]; d++){
3411: newm=savm;
3412: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3413: cov[2]=agexact;
3414: if(nagesqr==1)
3415: cov[3]= agexact*agexact; /* Should be changed here */
3416: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3417: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3418: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3419: else
3420: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3421: }
3422: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3423: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3424: savm=oldm;
3425: oldm=newm;
3426: } /* end mult */
3427:
3428: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3429: /* But now since version 0.9 we anticipate for bias at large stepm.
3430: * If stepm is larger than one month (smallest stepm) and if the exact delay
3431: * (in months) between two waves is not a multiple of stepm, we rounded to
3432: * the nearest (and in case of equal distance, to the lowest) interval but now
3433: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3434: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3435: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3436: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3437: * -stepm/2 to stepm/2 .
3438: * For stepm=1 the results are the same as for previous versions of Imach.
3439: * For stepm > 1 the results are less biased than in previous versions.
3440: */
1.234 brouard 3441: s1=s[mw[mi][i]][i];
3442: s2=s[mw[mi+1][i]][i];
3443: bbh=(double)bh[mi][i]/(double)stepm;
3444: /* bias bh is positive if real duration
3445: * is higher than the multiple of stepm and negative otherwise.
3446: */
3447: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3448: if( s2 > nlstate){
3449: /* i.e. if s2 is a death state and if the date of death is known
3450: then the contribution to the likelihood is the probability to
3451: die between last step unit time and current step unit time,
3452: which is also equal to probability to die before dh
3453: minus probability to die before dh-stepm .
3454: In version up to 0.92 likelihood was computed
3455: as if date of death was unknown. Death was treated as any other
3456: health state: the date of the interview describes the actual state
3457: and not the date of a change in health state. The former idea was
3458: to consider that at each interview the state was recorded
3459: (healthy, disable or death) and IMaCh was corrected; but when we
3460: introduced the exact date of death then we should have modified
3461: the contribution of an exact death to the likelihood. This new
3462: contribution is smaller and very dependent of the step unit
3463: stepm. It is no more the probability to die between last interview
3464: and month of death but the probability to survive from last
3465: interview up to one month before death multiplied by the
3466: probability to die within a month. Thanks to Chris
3467: Jackson for correcting this bug. Former versions increased
3468: mortality artificially. The bad side is that we add another loop
3469: which slows down the processing. The difference can be up to 10%
3470: lower mortality.
3471: */
3472: /* If, at the beginning of the maximization mostly, the
3473: cumulative probability or probability to be dead is
3474: constant (ie = 1) over time d, the difference is equal to
3475: 0. out[s1][3] = savm[s1][3]: probability, being at state
3476: s1 at precedent wave, to be dead a month before current
3477: wave is equal to probability, being at state s1 at
3478: precedent wave, to be dead at mont of the current
3479: wave. Then the observed probability (that this person died)
3480: is null according to current estimated parameter. In fact,
3481: it should be very low but not zero otherwise the log go to
3482: infinity.
3483: */
1.183 brouard 3484: /* #ifdef INFINITYORIGINAL */
3485: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3486: /* #else */
3487: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3488: /* lli=log(mytinydouble); */
3489: /* else */
3490: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3491: /* #endif */
1.226 brouard 3492: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3493:
1.226 brouard 3494: } else if ( s2==-1 ) { /* alive */
3495: for (j=1,survp=0. ; j<=nlstate; j++)
3496: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3497: /*survp += out[s1][j]; */
3498: lli= log(survp);
3499: }
3500: else if (s2==-4) {
3501: for (j=3,survp=0. ; j<=nlstate; j++)
3502: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3503: lli= log(survp);
3504: }
3505: else if (s2==-5) {
3506: for (j=1,survp=0. ; j<=2; j++)
3507: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3508: lli= log(survp);
3509: }
3510: else{
3511: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3512: /* 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 */
3513: }
3514: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3515: /*if(lli ==000.0)*/
3516: /*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); */
3517: ipmx +=1;
3518: sw += weight[i];
3519: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3520: /* if (lli < log(mytinydouble)){ */
3521: /* 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); */
3522: /* 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]); */
3523: /* } */
3524: } /* end of wave */
3525: } /* end of individual */
3526: } else if(mle==2){
3527: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3528: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3529: for(mi=1; mi<= wav[i]-1; mi++){
3530: for (ii=1;ii<=nlstate+ndeath;ii++)
3531: for (j=1;j<=nlstate+ndeath;j++){
3532: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3533: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3534: }
3535: for(d=0; d<=dh[mi][i]; d++){
3536: newm=savm;
3537: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3538: cov[2]=agexact;
3539: if(nagesqr==1)
3540: cov[3]= agexact*agexact;
3541: for (kk=1; kk<=cptcovage;kk++) {
3542: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3543: }
3544: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3545: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3546: savm=oldm;
3547: oldm=newm;
3548: } /* end mult */
3549:
3550: s1=s[mw[mi][i]][i];
3551: s2=s[mw[mi+1][i]][i];
3552: bbh=(double)bh[mi][i]/(double)stepm;
3553: 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 */
3554: ipmx +=1;
3555: sw += weight[i];
3556: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3557: } /* end of wave */
3558: } /* end of individual */
3559: } else if(mle==3){ /* exponential inter-extrapolation */
3560: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3561: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3562: for(mi=1; mi<= wav[i]-1; mi++){
3563: for (ii=1;ii<=nlstate+ndeath;ii++)
3564: for (j=1;j<=nlstate+ndeath;j++){
3565: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3566: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3567: }
3568: for(d=0; d<dh[mi][i]; d++){
3569: newm=savm;
3570: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3571: cov[2]=agexact;
3572: if(nagesqr==1)
3573: cov[3]= agexact*agexact;
3574: for (kk=1; kk<=cptcovage;kk++) {
3575: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3576: }
3577: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3578: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3579: savm=oldm;
3580: oldm=newm;
3581: } /* end mult */
3582:
3583: s1=s[mw[mi][i]][i];
3584: s2=s[mw[mi+1][i]][i];
3585: bbh=(double)bh[mi][i]/(double)stepm;
3586: 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 */
3587: ipmx +=1;
3588: sw += weight[i];
3589: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3590: } /* end of wave */
3591: } /* end of individual */
3592: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3593: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3594: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3595: for(mi=1; mi<= wav[i]-1; mi++){
3596: for (ii=1;ii<=nlstate+ndeath;ii++)
3597: for (j=1;j<=nlstate+ndeath;j++){
3598: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3599: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3600: }
3601: for(d=0; d<dh[mi][i]; d++){
3602: newm=savm;
3603: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3604: cov[2]=agexact;
3605: if(nagesqr==1)
3606: cov[3]= agexact*agexact;
3607: for (kk=1; kk<=cptcovage;kk++) {
3608: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3609: }
1.126 brouard 3610:
1.226 brouard 3611: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3612: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3613: savm=oldm;
3614: oldm=newm;
3615: } /* end mult */
3616:
3617: s1=s[mw[mi][i]][i];
3618: s2=s[mw[mi+1][i]][i];
3619: if( s2 > nlstate){
3620: lli=log(out[s1][s2] - savm[s1][s2]);
3621: } else if ( s2==-1 ) { /* alive */
3622: for (j=1,survp=0. ; j<=nlstate; j++)
3623: survp += out[s1][j];
3624: lli= log(survp);
3625: }else{
3626: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3627: }
3628: ipmx +=1;
3629: sw += weight[i];
3630: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3631: /* 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 3632: } /* end of wave */
3633: } /* end of individual */
3634: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3635: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3636: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3637: for(mi=1; mi<= wav[i]-1; mi++){
3638: for (ii=1;ii<=nlstate+ndeath;ii++)
3639: for (j=1;j<=nlstate+ndeath;j++){
3640: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3641: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3642: }
3643: for(d=0; d<dh[mi][i]; d++){
3644: newm=savm;
3645: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3646: cov[2]=agexact;
3647: if(nagesqr==1)
3648: cov[3]= agexact*agexact;
3649: for (kk=1; kk<=cptcovage;kk++) {
3650: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3651: }
1.126 brouard 3652:
1.226 brouard 3653: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3654: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3655: savm=oldm;
3656: oldm=newm;
3657: } /* end mult */
3658:
3659: s1=s[mw[mi][i]][i];
3660: s2=s[mw[mi+1][i]][i];
3661: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3662: ipmx +=1;
3663: sw += weight[i];
3664: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3665: /*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]);*/
3666: } /* end of wave */
3667: } /* end of individual */
3668: } /* End of if */
3669: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3670: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3671: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3672: return -l;
1.126 brouard 3673: }
3674:
3675: /*************** log-likelihood *************/
3676: double funcone( double *x)
3677: {
1.228 brouard 3678: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3679: int i, ii, j, k, mi, d, kk;
1.228 brouard 3680: int ioffset=0;
1.131 brouard 3681: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3682: double **out;
3683: double lli; /* Individual log likelihood */
3684: double llt;
3685: int s1, s2;
1.228 brouard 3686: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3687:
1.126 brouard 3688: double bbh, survp;
1.187 brouard 3689: double agexact;
1.214 brouard 3690: double agebegin, ageend;
1.126 brouard 3691: /*extern weight */
3692: /* We are differentiating ll according to initial status */
3693: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3694: /*for(i=1;i<imx;i++)
3695: printf(" %d\n",s[4][i]);
3696: */
3697: cov[1]=1.;
3698:
3699: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3700: ioffset=0;
3701: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3702: /* ioffset=2+nagesqr+cptcovage; */
3703: ioffset=2+nagesqr;
1.232 brouard 3704: /* Fixed */
1.224 brouard 3705: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3706: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3707: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3708: 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)*/
3709: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3710: /* cov[2+6]=covar[Tvar[6]][i]; */
3711: /* cov[2+6]=covar[2][i]; V2 */
3712: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3713: /* cov[2+7]=covar[Tvar[7]][i]; */
3714: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3715: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3716: /* cov[2+9]=covar[Tvar[9]][i]; */
3717: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3718: }
1.232 brouard 3719: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3720: /* 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?)*\/ */
3721: /* } */
1.231 brouard 3722: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3723: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3724: /* } */
1.225 brouard 3725:
1.233 brouard 3726:
3727: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3728: /* Wave varying (but not age varying) */
3729: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3730: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3731: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3732: }
1.232 brouard 3733: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3734: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3735: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3736: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3737: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3738: /* 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 3739: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3740: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3741: /* /\* 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]); *\/ */
3742: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3743: /* } */
1.126 brouard 3744: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3745: for (j=1;j<=nlstate+ndeath;j++){
3746: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3747: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3748: }
1.214 brouard 3749:
3750: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3751: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3752: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3753: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3754: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3755: and mw[mi+1][i]. dh depends on stepm.*/
3756: newm=savm;
1.247 brouard 3757: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3758: cov[2]=agexact;
3759: if(nagesqr==1)
3760: cov[3]= agexact*agexact;
3761: for (kk=1; kk<=cptcovage;kk++) {
3762: if(!FixedV[Tvar[Tage[kk]]])
3763: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3764: else
3765: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3766: }
3767: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3768: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3769: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3770: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3771: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3772: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3773: savm=oldm;
3774: oldm=newm;
1.126 brouard 3775: } /* end mult */
3776:
3777: s1=s[mw[mi][i]][i];
3778: s2=s[mw[mi+1][i]][i];
1.217 brouard 3779: /* if(s2==-1){ */
1.268 brouard 3780: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3781: /* /\* exit(1); *\/ */
3782: /* } */
1.126 brouard 3783: bbh=(double)bh[mi][i]/(double)stepm;
3784: /* bias is positive if real duration
3785: * is higher than the multiple of stepm and negative otherwise.
3786: */
3787: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3788: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3789: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3790: for (j=1,survp=0. ; j<=nlstate; j++)
3791: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3792: lli= log(survp);
1.126 brouard 3793: }else if (mle==1){
1.242 brouard 3794: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3795: } else if(mle==2){
1.242 brouard 3796: 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 3797: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3798: 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 3799: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3800: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3801: } else{ /* mle=0 back to 1 */
1.242 brouard 3802: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3803: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3804: } /* End of if */
3805: ipmx +=1;
3806: sw += weight[i];
3807: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3808: /*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 3809: if(globpr){
1.246 brouard 3810: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3811: %11.6f %11.6f %11.6f ", \
1.242 brouard 3812: 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 3813: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3814: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3815: llt +=ll[k]*gipmx/gsw;
3816: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3817: }
3818: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3819: }
1.232 brouard 3820: } /* end of wave */
3821: } /* end of individual */
3822: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3823: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3824: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3825: if(globpr==0){ /* First time we count the contributions and weights */
3826: gipmx=ipmx;
3827: gsw=sw;
3828: }
3829: return -l;
1.126 brouard 3830: }
3831:
3832:
3833: /*************** function likelione ***********/
3834: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3835: {
3836: /* This routine should help understanding what is done with
3837: the selection of individuals/waves and
3838: to check the exact contribution to the likelihood.
3839: Plotting could be done.
3840: */
3841: int k;
3842:
3843: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3844: strcpy(fileresilk,"ILK_");
1.202 brouard 3845: strcat(fileresilk,fileresu);
1.126 brouard 3846: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3847: printf("Problem with resultfile: %s\n", fileresilk);
3848: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3849: }
1.214 brouard 3850: 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");
3851: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3852: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3853: for(k=1; k<=nlstate; k++)
3854: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3855: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3856: }
3857:
3858: *fretone=(*funcone)(p);
3859: if(*globpri !=0){
3860: fclose(ficresilk);
1.205 brouard 3861: if (mle ==0)
3862: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3863: else if(mle >=1)
3864: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3865: 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 3866: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3867:
3868: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3869: 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 3870: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3871: }
1.207 brouard 3872: 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 3873: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3874: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3875: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3876: fflush(fichtm);
1.205 brouard 3877: }
1.126 brouard 3878: return;
3879: }
3880:
3881:
3882: /*********** Maximum Likelihood Estimation ***************/
3883:
3884: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3885: {
1.165 brouard 3886: int i,j, iter=0;
1.126 brouard 3887: double **xi;
3888: double fret;
3889: double fretone; /* Only one call to likelihood */
3890: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3891:
3892: #ifdef NLOPT
3893: int creturn;
3894: nlopt_opt opt;
3895: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3896: double *lb;
3897: double minf; /* the minimum objective value, upon return */
3898: double * p1; /* Shifted parameters from 0 instead of 1 */
3899: myfunc_data dinst, *d = &dinst;
3900: #endif
3901:
3902:
1.126 brouard 3903: xi=matrix(1,npar,1,npar);
3904: for (i=1;i<=npar;i++)
3905: for (j=1;j<=npar;j++)
3906: xi[i][j]=(i==j ? 1.0 : 0.0);
3907: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3908: strcpy(filerespow,"POW_");
1.126 brouard 3909: strcat(filerespow,fileres);
3910: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3911: printf("Problem with resultfile: %s\n", filerespow);
3912: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3913: }
3914: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3915: for (i=1;i<=nlstate;i++)
3916: for(j=1;j<=nlstate+ndeath;j++)
3917: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3918: fprintf(ficrespow,"\n");
1.162 brouard 3919: #ifdef POWELL
1.126 brouard 3920: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3921: #endif
1.126 brouard 3922:
1.162 brouard 3923: #ifdef NLOPT
3924: #ifdef NEWUOA
3925: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3926: #else
3927: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3928: #endif
3929: lb=vector(0,npar-1);
3930: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3931: nlopt_set_lower_bounds(opt, lb);
3932: nlopt_set_initial_step1(opt, 0.1);
3933:
3934: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3935: d->function = func;
3936: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3937: nlopt_set_min_objective(opt, myfunc, d);
3938: nlopt_set_xtol_rel(opt, ftol);
3939: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3940: printf("nlopt failed! %d\n",creturn);
3941: }
3942: else {
3943: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3944: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3945: iter=1; /* not equal */
3946: }
3947: nlopt_destroy(opt);
3948: #endif
1.126 brouard 3949: free_matrix(xi,1,npar,1,npar);
3950: fclose(ficrespow);
1.203 brouard 3951: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3952: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3953: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3954:
3955: }
3956:
3957: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3958: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3959: {
3960: double **a,**y,*x,pd;
1.203 brouard 3961: /* double **hess; */
1.164 brouard 3962: int i, j;
1.126 brouard 3963: int *indx;
3964:
3965: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3966: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3967: void lubksb(double **a, int npar, int *indx, double b[]) ;
3968: void ludcmp(double **a, int npar, int *indx, double *d) ;
3969: double gompertz(double p[]);
1.203 brouard 3970: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3971:
3972: printf("\nCalculation of the hessian matrix. Wait...\n");
3973: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3974: for (i=1;i<=npar;i++){
1.203 brouard 3975: printf("%d-",i);fflush(stdout);
3976: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3977:
3978: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3979:
3980: /* printf(" %f ",p[i]);
3981: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3982: }
3983:
3984: for (i=1;i<=npar;i++) {
3985: for (j=1;j<=npar;j++) {
3986: if (j>i) {
1.203 brouard 3987: printf(".%d-%d",i,j);fflush(stdout);
3988: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3989: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3990:
3991: hess[j][i]=hess[i][j];
3992: /*printf(" %lf ",hess[i][j]);*/
3993: }
3994: }
3995: }
3996: printf("\n");
3997: fprintf(ficlog,"\n");
3998:
3999: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4000: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4001:
4002: a=matrix(1,npar,1,npar);
4003: y=matrix(1,npar,1,npar);
4004: x=vector(1,npar);
4005: indx=ivector(1,npar);
4006: for (i=1;i<=npar;i++)
4007: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4008: ludcmp(a,npar,indx,&pd);
4009:
4010: for (j=1;j<=npar;j++) {
4011: for (i=1;i<=npar;i++) x[i]=0;
4012: x[j]=1;
4013: lubksb(a,npar,indx,x);
4014: for (i=1;i<=npar;i++){
4015: matcov[i][j]=x[i];
4016: }
4017: }
4018:
4019: printf("\n#Hessian matrix#\n");
4020: fprintf(ficlog,"\n#Hessian matrix#\n");
4021: for (i=1;i<=npar;i++) {
4022: for (j=1;j<=npar;j++) {
1.203 brouard 4023: printf("%.6e ",hess[i][j]);
4024: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4025: }
4026: printf("\n");
4027: fprintf(ficlog,"\n");
4028: }
4029:
1.203 brouard 4030: /* printf("\n#Covariance matrix#\n"); */
4031: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4032: /* for (i=1;i<=npar;i++) { */
4033: /* for (j=1;j<=npar;j++) { */
4034: /* printf("%.6e ",matcov[i][j]); */
4035: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4036: /* } */
4037: /* printf("\n"); */
4038: /* fprintf(ficlog,"\n"); */
4039: /* } */
4040:
1.126 brouard 4041: /* Recompute Inverse */
1.203 brouard 4042: /* for (i=1;i<=npar;i++) */
4043: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4044: /* ludcmp(a,npar,indx,&pd); */
4045:
4046: /* printf("\n#Hessian matrix recomputed#\n"); */
4047:
4048: /* for (j=1;j<=npar;j++) { */
4049: /* for (i=1;i<=npar;i++) x[i]=0; */
4050: /* x[j]=1; */
4051: /* lubksb(a,npar,indx,x); */
4052: /* for (i=1;i<=npar;i++){ */
4053: /* y[i][j]=x[i]; */
4054: /* printf("%.3e ",y[i][j]); */
4055: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4056: /* } */
4057: /* printf("\n"); */
4058: /* fprintf(ficlog,"\n"); */
4059: /* } */
4060:
4061: /* Verifying the inverse matrix */
4062: #ifdef DEBUGHESS
4063: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4064:
1.203 brouard 4065: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4066: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4067:
4068: for (j=1;j<=npar;j++) {
4069: for (i=1;i<=npar;i++){
1.203 brouard 4070: printf("%.2f ",y[i][j]);
4071: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4072: }
4073: printf("\n");
4074: fprintf(ficlog,"\n");
4075: }
1.203 brouard 4076: #endif
1.126 brouard 4077:
4078: free_matrix(a,1,npar,1,npar);
4079: free_matrix(y,1,npar,1,npar);
4080: free_vector(x,1,npar);
4081: free_ivector(indx,1,npar);
1.203 brouard 4082: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4083:
4084:
4085: }
4086:
4087: /*************** hessian matrix ****************/
4088: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4089: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4090: int i;
4091: int l=1, lmax=20;
1.203 brouard 4092: double k1,k2, res, fx;
1.132 brouard 4093: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4094: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4095: int k=0,kmax=10;
4096: double l1;
4097:
4098: fx=func(x);
4099: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4100: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4101: l1=pow(10,l);
4102: delts=delt;
4103: for(k=1 ; k <kmax; k=k+1){
4104: delt = delta*(l1*k);
4105: p2[theta]=x[theta] +delt;
1.145 brouard 4106: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4107: p2[theta]=x[theta]-delt;
4108: k2=func(p2)-fx;
4109: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4110: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4111:
1.203 brouard 4112: #ifdef DEBUGHESSII
1.126 brouard 4113: 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);
4114: 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);
4115: #endif
4116: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4117: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4118: k=kmax;
4119: }
4120: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4121: k=kmax; l=lmax*10;
1.126 brouard 4122: }
4123: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4124: delts=delt;
4125: }
1.203 brouard 4126: } /* End loop k */
1.126 brouard 4127: }
4128: delti[theta]=delts;
4129: return res;
4130:
4131: }
4132:
1.203 brouard 4133: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4134: {
4135: int i;
1.164 brouard 4136: int l=1, lmax=20;
1.126 brouard 4137: double k1,k2,k3,k4,res,fx;
1.132 brouard 4138: double p2[MAXPARM+1];
1.203 brouard 4139: int k, kmax=1;
4140: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4141:
4142: int firstime=0;
1.203 brouard 4143:
1.126 brouard 4144: fx=func(x);
1.203 brouard 4145: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4146: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4147: p2[thetai]=x[thetai]+delti[thetai]*k;
4148: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4149: k1=func(p2)-fx;
4150:
1.203 brouard 4151: p2[thetai]=x[thetai]+delti[thetai]*k;
4152: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4153: k2=func(p2)-fx;
4154:
1.203 brouard 4155: p2[thetai]=x[thetai]-delti[thetai]*k;
4156: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4157: k3=func(p2)-fx;
4158:
1.203 brouard 4159: p2[thetai]=x[thetai]-delti[thetai]*k;
4160: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4161: k4=func(p2)-fx;
1.203 brouard 4162: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4163: if(k1*k2*k3*k4 <0.){
1.208 brouard 4164: firstime=1;
1.203 brouard 4165: kmax=kmax+10;
1.208 brouard 4166: }
4167: if(kmax >=10 || firstime ==1){
1.246 brouard 4168: 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);
4169: 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 4170: 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);
4171: 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);
4172: }
4173: #ifdef DEBUGHESSIJ
4174: v1=hess[thetai][thetai];
4175: v2=hess[thetaj][thetaj];
4176: cv12=res;
4177: /* Computing eigen value of Hessian matrix */
4178: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4179: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4180: if ((lc2 <0) || (lc1 <0) ){
4181: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4182: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4183: 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);
4184: 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);
4185: }
1.126 brouard 4186: #endif
4187: }
4188: return res;
4189: }
4190:
1.203 brouard 4191: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4192: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4193: /* { */
4194: /* int i; */
4195: /* int l=1, lmax=20; */
4196: /* double k1,k2,k3,k4,res,fx; */
4197: /* double p2[MAXPARM+1]; */
4198: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4199: /* int k=0,kmax=10; */
4200: /* double l1; */
4201:
4202: /* fx=func(x); */
4203: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4204: /* l1=pow(10,l); */
4205: /* delts=delt; */
4206: /* for(k=1 ; k <kmax; k=k+1){ */
4207: /* delt = delti*(l1*k); */
4208: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4209: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4210: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4211: /* k1=func(p2)-fx; */
4212:
4213: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4214: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4215: /* k2=func(p2)-fx; */
4216:
4217: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4218: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4219: /* k3=func(p2)-fx; */
4220:
4221: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4222: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4223: /* k4=func(p2)-fx; */
4224: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4225: /* #ifdef DEBUGHESSIJ */
4226: /* 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); */
4227: /* 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); */
4228: /* #endif */
4229: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4230: /* k=kmax; */
4231: /* } */
4232: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4233: /* k=kmax; l=lmax*10; */
4234: /* } */
4235: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4236: /* delts=delt; */
4237: /* } */
4238: /* } /\* End loop k *\/ */
4239: /* } */
4240: /* delti[theta]=delts; */
4241: /* return res; */
4242: /* } */
4243:
4244:
1.126 brouard 4245: /************** Inverse of matrix **************/
4246: void ludcmp(double **a, int n, int *indx, double *d)
4247: {
4248: int i,imax,j,k;
4249: double big,dum,sum,temp;
4250: double *vv;
4251:
4252: vv=vector(1,n);
4253: *d=1.0;
4254: for (i=1;i<=n;i++) {
4255: big=0.0;
4256: for (j=1;j<=n;j++)
4257: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4258: if (big == 0.0){
4259: printf(" Singular Hessian matrix at row %d:\n",i);
4260: for (j=1;j<=n;j++) {
4261: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4262: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4263: }
4264: fflush(ficlog);
4265: fclose(ficlog);
4266: nrerror("Singular matrix in routine ludcmp");
4267: }
1.126 brouard 4268: vv[i]=1.0/big;
4269: }
4270: for (j=1;j<=n;j++) {
4271: for (i=1;i<j;i++) {
4272: sum=a[i][j];
4273: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4274: a[i][j]=sum;
4275: }
4276: big=0.0;
4277: for (i=j;i<=n;i++) {
4278: sum=a[i][j];
4279: for (k=1;k<j;k++)
4280: sum -= a[i][k]*a[k][j];
4281: a[i][j]=sum;
4282: if ( (dum=vv[i]*fabs(sum)) >= big) {
4283: big=dum;
4284: imax=i;
4285: }
4286: }
4287: if (j != imax) {
4288: for (k=1;k<=n;k++) {
4289: dum=a[imax][k];
4290: a[imax][k]=a[j][k];
4291: a[j][k]=dum;
4292: }
4293: *d = -(*d);
4294: vv[imax]=vv[j];
4295: }
4296: indx[j]=imax;
4297: if (a[j][j] == 0.0) a[j][j]=TINY;
4298: if (j != n) {
4299: dum=1.0/(a[j][j]);
4300: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4301: }
4302: }
4303: free_vector(vv,1,n); /* Doesn't work */
4304: ;
4305: }
4306:
4307: void lubksb(double **a, int n, int *indx, double b[])
4308: {
4309: int i,ii=0,ip,j;
4310: double sum;
4311:
4312: for (i=1;i<=n;i++) {
4313: ip=indx[i];
4314: sum=b[ip];
4315: b[ip]=b[i];
4316: if (ii)
4317: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4318: else if (sum) ii=i;
4319: b[i]=sum;
4320: }
4321: for (i=n;i>=1;i--) {
4322: sum=b[i];
4323: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4324: b[i]=sum/a[i][i];
4325: }
4326: }
4327:
4328: void pstamp(FILE *fichier)
4329: {
1.196 brouard 4330: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4331: }
4332:
1.253 brouard 4333:
4334:
1.126 brouard 4335: /************ Frequencies ********************/
1.251 brouard 4336: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4337: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4338: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4339: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4340:
1.265 brouard 4341: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4342: int iind=0, iage=0;
4343: int mi; /* Effective wave */
4344: int first;
4345: double ***freq; /* Frequencies */
1.268 brouard 4346: 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 */
4347: 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 4348: double *meanq;
4349: double **meanqt;
4350: double *pp, **prop, *posprop, *pospropt;
4351: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4352: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4353: double agebegin, ageend;
4354:
4355: pp=vector(1,nlstate);
1.251 brouard 4356: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4357: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4358: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4359: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4360: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4361: meanqt=matrix(1,lastpass,1,nqtveff);
4362: strcpy(fileresp,"P_");
4363: strcat(fileresp,fileresu);
4364: /*strcat(fileresphtm,fileresu);*/
4365: if((ficresp=fopen(fileresp,"w"))==NULL) {
4366: printf("Problem with prevalence resultfile: %s\n", fileresp);
4367: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4368: exit(0);
4369: }
1.240 brouard 4370:
1.226 brouard 4371: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4372: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4373: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4374: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4375: fflush(ficlog);
4376: exit(70);
4377: }
4378: else{
4379: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4380: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4381: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4382: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4383: }
1.237 brouard 4384: 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 4385:
1.226 brouard 4386: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4387: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4388: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4389: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4390: fflush(ficlog);
4391: exit(70);
1.240 brouard 4392: } else{
1.226 brouard 4393: 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 4394: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4395: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4396: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4397: }
1.240 brouard 4398: 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);
4399:
1.253 brouard 4400: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4401: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4402: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4403: j1=0;
1.126 brouard 4404:
1.227 brouard 4405: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4406: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4407: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4408:
4409:
1.226 brouard 4410: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4411: reference=low_education V1=0,V2=0
4412: med_educ V1=1 V2=0,
4413: high_educ V1=0 V2=1
4414: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4415: */
1.249 brouard 4416: dateintsum=0;
4417: k2cpt=0;
4418:
1.253 brouard 4419: if(cptcoveff == 0 )
1.265 brouard 4420: nl=1; /* Constant and age model only */
1.253 brouard 4421: else
4422: nl=2;
1.265 brouard 4423:
4424: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4425: /* Loop on nj=1 or 2 if dummy covariates j!=0
4426: * Loop on j1(1 to 2**cptcoveff) covariate combination
4427: * freq[s1][s2][iage] =0.
4428: * Loop on iind
4429: * ++freq[s1][s2][iage] weighted
4430: * end iind
4431: * if covariate and j!0
4432: * headers Variable on one line
4433: * endif cov j!=0
4434: * header of frequency table by age
4435: * Loop on age
4436: * pp[s1]+=freq[s1][s2][iage] weighted
4437: * pos+=freq[s1][s2][iage] weighted
4438: * Loop on s1 initial state
4439: * fprintf(ficresp
4440: * end s1
4441: * end age
4442: * if j!=0 computes starting values
4443: * end compute starting values
4444: * end j1
4445: * end nl
4446: */
1.253 brouard 4447: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4448: if(nj==1)
4449: j=0; /* First pass for the constant */
1.265 brouard 4450: else{
1.253 brouard 4451: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4452: }
1.251 brouard 4453: first=1;
1.265 brouard 4454: 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 4455: posproptt=0.;
4456: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4457: scanf("%d", i);*/
4458: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4459: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4460: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4461: freq[i][s2][m]=0;
1.251 brouard 4462:
4463: for (i=1; i<=nlstate; i++) {
1.240 brouard 4464: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4465: prop[i][m]=0;
4466: posprop[i]=0;
4467: pospropt[i]=0;
4468: }
4469: /* for (z1=1; z1<= nqfveff; z1++) { */
4470: /* meanq[z1]+=0.; */
4471: /* for(m=1;m<=lastpass;m++){ */
4472: /* meanqt[m][z1]=0.; */
4473: /* } */
4474: /* } */
4475:
4476: /* dateintsum=0; */
4477: /* k2cpt=0; */
4478:
1.265 brouard 4479: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4480: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4481: bool=1;
4482: if(j !=0){
4483: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4484: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4485: /* for (z1=1; z1<= nqfveff; z1++) { */
4486: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4487: /* } */
4488: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4489: /* if(Tvaraff[z1] ==-20){ */
4490: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4491: /* }else if(Tvaraff[z1] ==-10){ */
4492: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4493: /* }else */
4494: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4495: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4496: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4497: /* 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",
4498: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4499: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4500: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4501: } /* Onlyf fixed */
4502: } /* end z1 */
4503: } /* cptcovn > 0 */
4504: } /* end any */
4505: }/* end j==0 */
1.265 brouard 4506: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4507: /* for(m=firstpass; m<=lastpass; m++){ */
4508: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4509: m=mw[mi][iind];
4510: if(j!=0){
4511: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4512: for (z1=1; z1<=cptcoveff; z1++) {
4513: if( Fixed[Tmodelind[z1]]==1){
4514: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4515: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4516: value is -1, we don't select. It differs from the
4517: constant and age model which counts them. */
4518: bool=0; /* not selected */
4519: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4520: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4521: bool=0;
4522: }
4523: }
4524: }
4525: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4526: } /* end j==0 */
4527: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4528: if(bool==1){
4529: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4530: and mw[mi+1][iind]. dh depends on stepm. */
4531: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4532: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4533: if(m >=firstpass && m <=lastpass){
4534: k2=anint[m][iind]+(mint[m][iind]/12.);
4535: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4536: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4537: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4538: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4539: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4540: if (m<lastpass) {
4541: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4542: /* 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]); */
4543: if(s[m][iind]==-1)
4544: 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.));
4545: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4546: /* if((int)agev[m][iind] == 55) */
4547: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4548: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4549: 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 4550: }
1.251 brouard 4551: } /* end if between passes */
4552: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4553: dateintsum=dateintsum+k2; /* on all covariates ?*/
4554: k2cpt++;
4555: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4556: }
1.251 brouard 4557: }else{
4558: bool=1;
4559: }/* end bool 2 */
4560: } /* end m */
4561: } /* end bool */
4562: } /* end iind = 1 to imx */
4563: /* prop[s][age] is feeded for any initial and valid live state as well as
4564: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4565:
4566:
4567: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4568: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4569: pstamp(ficresp);
1.251 brouard 4570: if (cptcoveff>0 && j!=0){
1.265 brouard 4571: pstamp(ficresp);
1.251 brouard 4572: printf( "\n#********** Variable ");
4573: fprintf(ficresp, "\n#********** Variable ");
4574: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4575: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4576: fprintf(ficlog, "\n#********** Variable ");
4577: for (z1=1; z1<=cptcoveff; z1++){
4578: if(!FixedV[Tvaraff[z1]]){
4579: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4580: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4581: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4582: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4583: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4584: }else{
1.251 brouard 4585: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4586: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4587: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4588: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4589: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4590: }
4591: }
4592: printf( "**********\n#");
4593: fprintf(ficresp, "**********\n#");
4594: fprintf(ficresphtm, "**********</h3>\n");
4595: fprintf(ficresphtmfr, "**********</h3>\n");
4596: fprintf(ficlog, "**********\n");
4597: }
4598: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4599: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4600: fprintf(ficresp, " Age");
4601: 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 4602: for(i=1; i<=nlstate;i++) {
1.265 brouard 4603: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4604: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4605: }
1.265 brouard 4606: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4607: fprintf(ficresphtm, "\n");
4608:
4609: /* Header of frequency table by age */
4610: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4611: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4612: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4613: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4614: if(s2!=0 && m!=0)
4615: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4616: }
1.226 brouard 4617: }
1.251 brouard 4618: fprintf(ficresphtmfr, "\n");
4619:
4620: /* For each age */
4621: for(iage=iagemin; iage <= iagemax+3; iage++){
4622: fprintf(ficresphtm,"<tr>");
4623: if(iage==iagemax+1){
4624: fprintf(ficlog,"1");
4625: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4626: }else if(iage==iagemax+2){
4627: fprintf(ficlog,"0");
4628: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4629: }else if(iage==iagemax+3){
4630: fprintf(ficlog,"Total");
4631: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4632: }else{
1.240 brouard 4633: if(first==1){
1.251 brouard 4634: first=0;
4635: printf("See log file for details...\n");
4636: }
4637: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4638: fprintf(ficlog,"Age %d", iage);
4639: }
1.265 brouard 4640: for(s1=1; s1 <=nlstate ; s1++){
4641: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4642: pp[s1] += freq[s1][m][iage];
1.251 brouard 4643: }
1.265 brouard 4644: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4645: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4646: pos += freq[s1][m][iage];
4647: if(pp[s1]>=1.e-10){
1.251 brouard 4648: if(first==1){
1.265 brouard 4649: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4650: }
1.265 brouard 4651: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4652: }else{
4653: if(first==1)
1.265 brouard 4654: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4655: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4656: }
4657: }
4658:
1.265 brouard 4659: for(s1=1; s1 <=nlstate ; s1++){
4660: /* posprop[s1]=0; */
4661: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4662: pp[s1] += freq[s1][m][iage];
4663: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4664:
4665: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4666: pos += pp[s1]; /* pos is the total number of transitions until this age */
4667: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4668: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4669: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4670: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4671: }
4672:
4673: /* Writing ficresp */
4674: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4675: if( iage <= iagemax){
4676: fprintf(ficresp," %d",iage);
4677: }
4678: }else if( nj==2){
4679: if( iage <= iagemax){
4680: fprintf(ficresp," %d",iage);
4681: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4682: }
1.240 brouard 4683: }
1.265 brouard 4684: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4685: if(pos>=1.e-5){
1.251 brouard 4686: if(first==1)
1.265 brouard 4687: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4688: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4689: }else{
4690: if(first==1)
1.265 brouard 4691: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4692: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4693: }
4694: if( iage <= iagemax){
4695: if(pos>=1.e-5){
1.265 brouard 4696: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4697: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4698: }else if( nj==2){
4699: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4700: }
4701: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4702: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4703: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4704: } else{
4705: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4706: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4707: }
1.240 brouard 4708: }
1.265 brouard 4709: pospropt[s1] +=posprop[s1];
4710: } /* end loop s1 */
1.251 brouard 4711: /* pospropt=0.; */
1.265 brouard 4712: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4713: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4714: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4715: if(first==1){
1.265 brouard 4716: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4717: }
1.265 brouard 4718: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4719: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4720: }
1.265 brouard 4721: if(s1!=0 && m!=0)
4722: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4723: }
1.265 brouard 4724: } /* end loop s1 */
1.251 brouard 4725: posproptt=0.;
1.265 brouard 4726: for(s1=1; s1 <=nlstate; s1++){
4727: posproptt += pospropt[s1];
1.251 brouard 4728: }
4729: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4730: fprintf(ficresphtm,"</tr>\n");
4731: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4732: if(iage <= iagemax)
4733: fprintf(ficresp,"\n");
1.240 brouard 4734: }
1.251 brouard 4735: if(first==1)
4736: printf("Others in log...\n");
4737: fprintf(ficlog,"\n");
4738: } /* end loop age iage */
1.265 brouard 4739:
1.251 brouard 4740: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4741: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4742: if(posproptt < 1.e-5){
1.265 brouard 4743: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4744: }else{
1.265 brouard 4745: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4746: }
1.226 brouard 4747: }
1.251 brouard 4748: fprintf(ficresphtm,"</tr>\n");
4749: fprintf(ficresphtm,"</table>\n");
4750: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4751: if(posproptt < 1.e-5){
1.251 brouard 4752: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4753: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4754: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4755: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4756: invalidvarcomb[j1]=1;
1.226 brouard 4757: }else{
1.251 brouard 4758: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4759: invalidvarcomb[j1]=0;
1.226 brouard 4760: }
1.251 brouard 4761: fprintf(ficresphtmfr,"</table>\n");
4762: fprintf(ficlog,"\n");
4763: if(j!=0){
4764: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4765: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4766: for(k=1; k <=(nlstate+ndeath); k++){
4767: if (k != i) {
1.265 brouard 4768: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4769: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4770: if(j1==1){ /* All dummy covariates to zero */
4771: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4772: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4773: printf("%d%d ",i,k);
4774: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4775: 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]));
4776: 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]));
4777: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4778: }
1.253 brouard 4779: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4780: for(iage=iagemin; iage <= iagemax+3; iage++){
4781: x[iage]= (double)iage;
4782: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4783: /* 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 4784: }
1.268 brouard 4785: /* Some are not finite, but linreg will ignore these ages */
4786: no=0;
1.253 brouard 4787: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4788: pstart[s1]=b;
4789: pstart[s1-1]=a;
1.252 brouard 4790: }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 */
4791: 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]);
4792: 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 4793: 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 4794: printf("%d%d ",i,k);
4795: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4796: 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 4797: }else{ /* Other cases, like quantitative fixed or varying covariates */
4798: ;
4799: }
4800: /* printf("%12.7f )", param[i][jj][k]); */
4801: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4802: s1++;
1.251 brouard 4803: } /* end jj */
4804: } /* end k!= i */
4805: } /* end k */
1.265 brouard 4806: } /* end i, s1 */
1.251 brouard 4807: } /* end j !=0 */
4808: } /* end selected combination of covariate j1 */
4809: if(j==0){ /* We can estimate starting values from the occurences in each case */
4810: printf("#Freqsummary: Starting values for the constants:\n");
4811: fprintf(ficlog,"\n");
1.265 brouard 4812: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4813: for(k=1; k <=(nlstate+ndeath); k++){
4814: if (k != i) {
4815: printf("%d%d ",i,k);
4816: fprintf(ficlog,"%d%d ",i,k);
4817: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4818: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4819: if(jj==1){ /* Age has to be done */
1.265 brouard 4820: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4821: 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]));
4822: 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 4823: }
4824: /* printf("%12.7f )", param[i][jj][k]); */
4825: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4826: s1++;
1.250 brouard 4827: }
1.251 brouard 4828: printf("\n");
4829: fprintf(ficlog,"\n");
1.250 brouard 4830: }
4831: }
4832: }
1.251 brouard 4833: printf("#Freqsummary\n");
4834: fprintf(ficlog,"\n");
1.265 brouard 4835: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4836: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4837: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
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]);
4840: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4841: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4842: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4843: /* } */
4844: }
1.265 brouard 4845: } /* end loop s1 */
1.251 brouard 4846:
4847: printf("\n");
4848: fprintf(ficlog,"\n");
4849: } /* end j=0 */
1.249 brouard 4850: } /* end j */
1.252 brouard 4851:
1.253 brouard 4852: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4853: for(i=1, jk=1; i <=nlstate; i++){
4854: for(j=1; j <=nlstate+ndeath; j++){
4855: if(j!=i){
4856: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4857: printf("%1d%1d",i,j);
4858: fprintf(ficparo,"%1d%1d",i,j);
4859: for(k=1; k<=ncovmodel;k++){
4860: /* printf(" %lf",param[i][j][k]); */
4861: /* fprintf(ficparo," %lf",param[i][j][k]); */
4862: p[jk]=pstart[jk];
4863: printf(" %f ",pstart[jk]);
4864: fprintf(ficparo," %f ",pstart[jk]);
4865: jk++;
4866: }
4867: printf("\n");
4868: fprintf(ficparo,"\n");
4869: }
4870: }
4871: }
4872: } /* end mle=-2 */
1.226 brouard 4873: dateintmean=dateintsum/k2cpt;
1.240 brouard 4874:
1.226 brouard 4875: fclose(ficresp);
4876: fclose(ficresphtm);
4877: fclose(ficresphtmfr);
4878: free_vector(meanq,1,nqfveff);
4879: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4880: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4881: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4882: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4883: free_vector(pospropt,1,nlstate);
4884: free_vector(posprop,1,nlstate);
1.251 brouard 4885: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4886: free_vector(pp,1,nlstate);
4887: /* End of freqsummary */
4888: }
1.126 brouard 4889:
1.268 brouard 4890: /* Simple linear regression */
4891: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4892:
4893: /* y=a+bx regression */
4894: double sumx = 0.0; /* sum of x */
4895: double sumx2 = 0.0; /* sum of x**2 */
4896: double sumxy = 0.0; /* sum of x * y */
4897: double sumy = 0.0; /* sum of y */
4898: double sumy2 = 0.0; /* sum of y**2 */
4899: double sume2 = 0.0; /* sum of square or residuals */
4900: double yhat;
4901:
4902: double denom=0;
4903: int i;
4904: int ne=*no;
4905:
4906: for ( i=ifi, ne=0;i<=ila;i++) {
4907: if(!isfinite(x[i]) || !isfinite(y[i])){
4908: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4909: continue;
4910: }
4911: ne=ne+1;
4912: sumx += x[i];
4913: sumx2 += x[i]*x[i];
4914: sumxy += x[i] * y[i];
4915: sumy += y[i];
4916: sumy2 += y[i]*y[i];
4917: denom = (ne * sumx2 - sumx*sumx);
4918: /* 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); */
4919: }
4920:
4921: denom = (ne * sumx2 - sumx*sumx);
4922: if (denom == 0) {
4923: // vertical, slope m is infinity
4924: *b = INFINITY;
4925: *a = 0;
4926: if (r) *r = 0;
4927: return 1;
4928: }
4929:
4930: *b = (ne * sumxy - sumx * sumy) / denom;
4931: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4932: if (r!=NULL) {
4933: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4934: sqrt((sumx2 - sumx*sumx/ne) *
4935: (sumy2 - sumy*sumy/ne));
4936: }
4937: *no=ne;
4938: for ( i=ifi, ne=0;i<=ila;i++) {
4939: if(!isfinite(x[i]) || !isfinite(y[i])){
4940: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4941: continue;
4942: }
4943: ne=ne+1;
4944: yhat = y[i] - *a -*b* x[i];
4945: sume2 += yhat * yhat ;
4946:
4947: denom = (ne * sumx2 - sumx*sumx);
4948: /* 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); */
4949: }
4950: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
4951: *sa= *sb * sqrt(sumx2/ne);
4952:
4953: return 0;
4954: }
4955:
1.126 brouard 4956: /************ Prevalence ********************/
1.227 brouard 4957: 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)
4958: {
4959: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4960: in each health status at the date of interview (if between dateprev1 and dateprev2).
4961: We still use firstpass and lastpass as another selection.
4962: */
1.126 brouard 4963:
1.227 brouard 4964: int i, m, jk, j1, bool, z1,j, iv;
4965: int mi; /* Effective wave */
4966: int iage;
4967: double agebegin, ageend;
4968:
4969: double **prop;
4970: double posprop;
4971: double y2; /* in fractional years */
4972: int iagemin, iagemax;
4973: int first; /** to stop verbosity which is redirected to log file */
4974:
4975: iagemin= (int) agemin;
4976: iagemax= (int) agemax;
4977: /*pp=vector(1,nlstate);*/
1.251 brouard 4978: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4979: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4980: j1=0;
1.222 brouard 4981:
1.227 brouard 4982: /*j=cptcoveff;*/
4983: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4984:
1.227 brouard 4985: first=1;
4986: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4987: for (i=1; i<=nlstate; i++)
1.251 brouard 4988: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4989: prop[i][iage]=0.0;
4990: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4991: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4992: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4993:
4994: for (i=1; i<=imx; i++) { /* Each individual */
4995: bool=1;
4996: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4997: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4998: m=mw[mi][i];
4999: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5000: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5001: for (z1=1; z1<=cptcoveff; z1++){
5002: if( Fixed[Tmodelind[z1]]==1){
5003: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5004: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5005: bool=0;
5006: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5007: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5008: bool=0;
5009: }
5010: }
5011: if(bool==1){ /* Otherwise we skip that wave/person */
5012: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5013: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5014: if(m >=firstpass && m <=lastpass){
5015: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5016: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5017: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5018: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5019: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5020: 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);
5021: exit(1);
5022: }
5023: if (s[m][i]>0 && s[m][i]<=nlstate) {
5024: /*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]]);*/
5025: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5026: prop[s[m][i]][iagemax+3] += weight[i];
5027: } /* end valid statuses */
5028: } /* end selection of dates */
5029: } /* end selection of waves */
5030: } /* end bool */
5031: } /* end wave */
5032: } /* end individual */
5033: for(i=iagemin; i <= iagemax+3; i++){
5034: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5035: posprop += prop[jk][i];
5036: }
5037:
5038: for(jk=1; jk <=nlstate ; jk++){
5039: if( i <= iagemax){
5040: if(posprop>=1.e-5){
5041: probs[i][jk][j1]= prop[jk][i]/posprop;
5042: } else{
5043: if(first==1){
5044: first=0;
1.266 brouard 5045: 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]);
5046: 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]);
5047: }else{
5048: 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 5049: }
5050: }
5051: }
5052: }/* end jk */
5053: }/* end i */
1.222 brouard 5054: /*} *//* end i1 */
1.227 brouard 5055: } /* end j1 */
1.222 brouard 5056:
1.227 brouard 5057: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5058: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5059: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5060: } /* End of prevalence */
1.126 brouard 5061:
5062: /************* Waves Concatenation ***************/
5063:
5064: 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)
5065: {
5066: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5067: Death is a valid wave (if date is known).
5068: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5069: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5070: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5071: */
1.126 brouard 5072:
1.224 brouard 5073: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5074: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5075: double sum=0., jmean=0.;*/
1.224 brouard 5076: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5077: int j, k=0,jk, ju, jl;
5078: double sum=0.;
5079: first=0;
1.214 brouard 5080: firstwo=0;
1.217 brouard 5081: firsthree=0;
1.218 brouard 5082: firstfour=0;
1.164 brouard 5083: jmin=100000;
1.126 brouard 5084: jmax=-1;
5085: jmean=0.;
1.224 brouard 5086:
5087: /* Treating live states */
1.214 brouard 5088: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5089: mi=0; /* First valid wave */
1.227 brouard 5090: mli=0; /* Last valid wave */
1.126 brouard 5091: m=firstpass;
1.214 brouard 5092: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5093: 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 */
5094: mli=m-1;/* mw[++mi][i]=m-1; */
5095: }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 */
5096: mw[++mi][i]=m;
5097: mli=m;
1.224 brouard 5098: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5099: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5100: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5101: }
1.227 brouard 5102: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5103: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5104: break;
1.224 brouard 5105: #else
1.227 brouard 5106: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5107: if(firsthree == 0){
1.262 brouard 5108: 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 5109: firsthree=1;
5110: }
1.262 brouard 5111: 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 5112: mw[++mi][i]=m;
5113: mli=m;
5114: }
5115: if(s[m][i]==-2){ /* Vital status is really unknown */
5116: nbwarn++;
5117: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5118: 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);
5119: 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);
5120: }
5121: break;
5122: }
5123: break;
1.224 brouard 5124: #endif
1.227 brouard 5125: }/* End m >= lastpass */
1.126 brouard 5126: }/* end while */
1.224 brouard 5127:
1.227 brouard 5128: /* 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 5129: /* After last pass */
1.224 brouard 5130: /* Treating death states */
1.214 brouard 5131: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5132: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5133: /* } */
1.126 brouard 5134: mi++; /* Death is another wave */
5135: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5136: /* Only death is a correct wave */
1.126 brouard 5137: mw[mi][i]=m;
1.257 brouard 5138: } /* else not in a death state */
1.224 brouard 5139: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5140: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5141: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5142: 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 */
5143: nbwarn++;
5144: if(firstfiv==0){
5145: 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 );
5146: firstfiv=1;
5147: }else{
5148: 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 );
5149: }
5150: }else{ /* Death occured afer last wave potential bias */
5151: nberr++;
5152: if(firstwo==0){
1.257 brouard 5153: 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 5154: firstwo=1;
5155: }
1.257 brouard 5156: 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 5157: }
1.257 brouard 5158: }else{ /* if date of interview is unknown */
1.227 brouard 5159: /* death is known but not confirmed by death status at any wave */
5160: if(firstfour==0){
5161: 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 );
5162: firstfour=1;
5163: }
5164: 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 5165: }
1.224 brouard 5166: } /* end if date of death is known */
5167: #endif
5168: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5169: /* wav[i]=mw[mi][i]; */
1.126 brouard 5170: if(mi==0){
5171: nbwarn++;
5172: if(first==0){
1.227 brouard 5173: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5174: first=1;
1.126 brouard 5175: }
5176: if(first==1){
1.227 brouard 5177: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5178: }
5179: } /* end mi==0 */
5180: } /* End individuals */
1.214 brouard 5181: /* wav and mw are no more changed */
1.223 brouard 5182:
1.214 brouard 5183:
1.126 brouard 5184: for(i=1; i<=imx; i++){
5185: for(mi=1; mi<wav[i];mi++){
5186: if (stepm <=0)
1.227 brouard 5187: dh[mi][i]=1;
1.126 brouard 5188: else{
1.260 brouard 5189: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5190: if (agedc[i] < 2*AGESUP) {
5191: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5192: if(j==0) j=1; /* Survives at least one month after exam */
5193: else if(j<0){
5194: nberr++;
5195: 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]);
5196: j=1; /* Temporary Dangerous patch */
5197: 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);
5198: 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]);
5199: 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);
5200: }
5201: k=k+1;
5202: if (j >= jmax){
5203: jmax=j;
5204: ijmax=i;
5205: }
5206: if (j <= jmin){
5207: jmin=j;
5208: ijmin=i;
5209: }
5210: sum=sum+j;
5211: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5212: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5213: }
5214: }
5215: else{
5216: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5217: /* 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 5218:
1.227 brouard 5219: k=k+1;
5220: if (j >= jmax) {
5221: jmax=j;
5222: ijmax=i;
5223: }
5224: else if (j <= jmin){
5225: jmin=j;
5226: ijmin=i;
5227: }
5228: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5229: /*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]);*/
5230: if(j<0){
5231: nberr++;
5232: 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]);
5233: 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]);
5234: }
5235: sum=sum+j;
5236: }
5237: jk= j/stepm;
5238: jl= j -jk*stepm;
5239: ju= j -(jk+1)*stepm;
5240: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5241: if(jl==0){
5242: dh[mi][i]=jk;
5243: bh[mi][i]=0;
5244: }else{ /* We want a negative bias in order to only have interpolation ie
5245: * to avoid the price of an extra matrix product in likelihood */
5246: dh[mi][i]=jk+1;
5247: bh[mi][i]=ju;
5248: }
5249: }else{
5250: if(jl <= -ju){
5251: dh[mi][i]=jk;
5252: bh[mi][i]=jl; /* bias is positive if real duration
5253: * is higher than the multiple of stepm and negative otherwise.
5254: */
5255: }
5256: else{
5257: dh[mi][i]=jk+1;
5258: bh[mi][i]=ju;
5259: }
5260: if(dh[mi][i]==0){
5261: dh[mi][i]=1; /* At least one step */
5262: bh[mi][i]=ju; /* At least one step */
5263: /* 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);*/
5264: }
5265: } /* end if mle */
1.126 brouard 5266: }
5267: } /* end wave */
5268: }
5269: jmean=sum/k;
5270: 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 5271: 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 5272: }
1.126 brouard 5273:
5274: /*********** Tricode ****************************/
1.220 brouard 5275: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5276: {
5277: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5278: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5279: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5280: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5281: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5282: */
1.130 brouard 5283:
1.242 brouard 5284: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5285: int modmaxcovj=0; /* Modality max of covariates j */
5286: int cptcode=0; /* Modality max of covariates j */
5287: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5288:
5289:
1.242 brouard 5290: /* cptcoveff=0; */
5291: /* *cptcov=0; */
1.126 brouard 5292:
1.242 brouard 5293: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5294:
1.242 brouard 5295: /* Loop on covariates without age and products and no quantitative variable */
5296: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5297: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5298: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5299: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5300: switch(Fixed[k]) {
5301: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5302: 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*/
5303: ij=(int)(covar[Tvar[k]][i]);
5304: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5305: * If product of Vn*Vm, still boolean *:
5306: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5307: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5308: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5309: modality of the nth covariate of individual i. */
5310: if (ij > modmaxcovj)
5311: modmaxcovj=ij;
5312: else if (ij < modmincovj)
5313: modmincovj=ij;
5314: if ((ij < -1) && (ij > NCOVMAX)){
5315: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5316: exit(1);
5317: }else
5318: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5319: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5320: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5321: /* getting the maximum value of the modality of the covariate
5322: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5323: female ies 1, then modmaxcovj=1.
5324: */
5325: } /* end for loop on individuals i */
5326: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5327: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5328: cptcode=modmaxcovj;
5329: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5330: /*for (i=0; i<=cptcode; i++) {*/
5331: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5332: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5333: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5334: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5335: if( j != -1){
5336: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5337: covariate for which somebody answered excluding
5338: undefined. Usually 2: 0 and 1. */
5339: }
5340: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5341: covariate for which somebody answered including
5342: undefined. Usually 3: -1, 0 and 1. */
5343: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5344: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5345: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5346:
1.242 brouard 5347: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5348: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5349: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5350: /* modmincovj=3; modmaxcovj = 7; */
5351: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5352: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5353: /* defining two dummy variables: variables V1_1 and V1_2.*/
5354: /* nbcode[Tvar[j]][ij]=k; */
5355: /* nbcode[Tvar[j]][1]=0; */
5356: /* nbcode[Tvar[j]][2]=1; */
5357: /* nbcode[Tvar[j]][3]=2; */
5358: /* To be continued (not working yet). */
5359: ij=0; /* ij is similar to i but can jump over null modalities */
5360: 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*/
5361: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5362: break;
5363: }
5364: ij++;
5365: 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*/
5366: cptcode = ij; /* New max modality for covar j */
5367: } /* end of loop on modality i=-1 to 1 or more */
5368: break;
5369: case 1: /* Testing on varying covariate, could be simple and
5370: * should look at waves or product of fixed *
5371: * varying. No time to test -1, assuming 0 and 1 only */
5372: ij=0;
5373: for(i=0; i<=1;i++){
5374: nbcode[Tvar[k]][++ij]=i;
5375: }
5376: break;
5377: default:
5378: break;
5379: } /* end switch */
5380: } /* end dummy test */
5381:
5382: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5383: /* /\*recode from 0 *\/ */
5384: /* k is a modality. If we have model=V1+V1*sex */
5385: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5386: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5387: /* } */
5388: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5389: /* if (ij > ncodemax[j]) { */
5390: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5391: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5392: /* break; */
5393: /* } */
5394: /* } /\* end of loop on modality k *\/ */
5395: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5396:
5397: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5398: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5399: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5400: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5401: 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 */
5402: 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 */
5403: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5404: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5405:
5406: ij=0;
5407: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5408: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5409: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5410: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5411: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5412: /* If product not in single variable we don't print results */
5413: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5414: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5415: 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*/
5416: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5417: 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 */
5418: if(Fixed[k]!=0)
5419: anyvaryingduminmodel=1;
5420: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5421: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5422: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5423: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5424: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5425: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5426: }
5427: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5428: /* ij--; */
5429: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5430: *cptcov=ij; /*Number of total real effective covariates: effective
5431: * because they can be excluded from the model and real
5432: * if in the model but excluded because missing values, but how to get k from ij?*/
5433: for(j=ij+1; j<= cptcovt; j++){
5434: Tvaraff[j]=0;
5435: Tmodelind[j]=0;
5436: }
5437: for(j=ntveff+1; j<= cptcovt; j++){
5438: TmodelInvind[j]=0;
5439: }
5440: /* To be sorted */
5441: ;
5442: }
1.126 brouard 5443:
1.145 brouard 5444:
1.126 brouard 5445: /*********** Health Expectancies ****************/
5446:
1.235 brouard 5447: 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 5448:
5449: {
5450: /* Health expectancies, no variances */
1.164 brouard 5451: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5452: int nhstepma, nstepma; /* Decreasing with age */
5453: double age, agelim, hf;
5454: double ***p3mat;
5455: double eip;
5456:
1.238 brouard 5457: /* pstamp(ficreseij); */
1.126 brouard 5458: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5459: fprintf(ficreseij,"# Age");
5460: for(i=1; i<=nlstate;i++){
5461: for(j=1; j<=nlstate;j++){
5462: fprintf(ficreseij," e%1d%1d ",i,j);
5463: }
5464: fprintf(ficreseij," e%1d. ",i);
5465: }
5466: fprintf(ficreseij,"\n");
5467:
5468:
5469: if(estepm < stepm){
5470: printf ("Problem %d lower than %d\n",estepm, stepm);
5471: }
5472: else hstepm=estepm;
5473: /* We compute the life expectancy from trapezoids spaced every estepm months
5474: * This is mainly to measure the difference between two models: for example
5475: * if stepm=24 months pijx are given only every 2 years and by summing them
5476: * we are calculating an estimate of the Life Expectancy assuming a linear
5477: * progression in between and thus overestimating or underestimating according
5478: * to the curvature of the survival function. If, for the same date, we
5479: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5480: * to compare the new estimate of Life expectancy with the same linear
5481: * hypothesis. A more precise result, taking into account a more precise
5482: * curvature will be obtained if estepm is as small as stepm. */
5483:
5484: /* For example we decided to compute the life expectancy with the smallest unit */
5485: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5486: nhstepm is the number of hstepm from age to agelim
5487: nstepm is the number of stepm from age to agelin.
1.270 brouard 5488: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5489: and note for a fixed period like estepm months */
5490: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5491: survival function given by stepm (the optimization length). Unfortunately it
5492: means that if the survival funtion is printed only each two years of age and if
5493: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5494: results. So we changed our mind and took the option of the best precision.
5495: */
5496: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5497:
5498: agelim=AGESUP;
5499: /* If stepm=6 months */
5500: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5501: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5502:
5503: /* nhstepm age range expressed in number of stepm */
5504: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5505: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5506: /* if (stepm >= YEARM) hstepm=1;*/
5507: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5508: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5509:
5510: for (age=bage; age<=fage; age ++){
5511: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5512: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5513: /* if (stepm >= YEARM) hstepm=1;*/
5514: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5515:
5516: /* If stepm=6 months */
5517: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5518: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5519:
1.235 brouard 5520: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5521:
5522: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5523:
5524: printf("%d|",(int)age);fflush(stdout);
5525: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5526:
5527: /* Computing expectancies */
5528: for(i=1; i<=nlstate;i++)
5529: for(j=1; j<=nlstate;j++)
5530: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5531: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5532:
5533: /* 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]);*/
5534:
5535: }
5536:
5537: fprintf(ficreseij,"%3.0f",age );
5538: for(i=1; i<=nlstate;i++){
5539: eip=0;
5540: for(j=1; j<=nlstate;j++){
5541: eip +=eij[i][j][(int)age];
5542: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5543: }
5544: fprintf(ficreseij,"%9.4f", eip );
5545: }
5546: fprintf(ficreseij,"\n");
5547:
5548: }
5549: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5550: printf("\n");
5551: fprintf(ficlog,"\n");
5552:
5553: }
5554:
1.235 brouard 5555: 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 5556:
5557: {
5558: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5559: to initial status i, ei. .
1.126 brouard 5560: */
5561: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5562: int nhstepma, nstepma; /* Decreasing with age */
5563: double age, agelim, hf;
5564: double ***p3matp, ***p3matm, ***varhe;
5565: double **dnewm,**doldm;
5566: double *xp, *xm;
5567: double **gp, **gm;
5568: double ***gradg, ***trgradg;
5569: int theta;
5570:
5571: double eip, vip;
5572:
5573: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5574: xp=vector(1,npar);
5575: xm=vector(1,npar);
5576: dnewm=matrix(1,nlstate*nlstate,1,npar);
5577: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5578:
5579: pstamp(ficresstdeij);
5580: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5581: fprintf(ficresstdeij,"# Age");
5582: for(i=1; i<=nlstate;i++){
5583: for(j=1; j<=nlstate;j++)
5584: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5585: fprintf(ficresstdeij," e%1d. ",i);
5586: }
5587: fprintf(ficresstdeij,"\n");
5588:
5589: pstamp(ficrescveij);
5590: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5591: fprintf(ficrescveij,"# Age");
5592: for(i=1; i<=nlstate;i++)
5593: for(j=1; j<=nlstate;j++){
5594: cptj= (j-1)*nlstate+i;
5595: for(i2=1; i2<=nlstate;i2++)
5596: for(j2=1; j2<=nlstate;j2++){
5597: cptj2= (j2-1)*nlstate+i2;
5598: if(cptj2 <= cptj)
5599: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5600: }
5601: }
5602: fprintf(ficrescveij,"\n");
5603:
5604: if(estepm < stepm){
5605: printf ("Problem %d lower than %d\n",estepm, stepm);
5606: }
5607: else hstepm=estepm;
5608: /* We compute the life expectancy from trapezoids spaced every estepm months
5609: * This is mainly to measure the difference between two models: for example
5610: * if stepm=24 months pijx are given only every 2 years and by summing them
5611: * we are calculating an estimate of the Life Expectancy assuming a linear
5612: * progression in between and thus overestimating or underestimating according
5613: * to the curvature of the survival function. If, for the same date, we
5614: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5615: * to compare the new estimate of Life expectancy with the same linear
5616: * hypothesis. A more precise result, taking into account a more precise
5617: * curvature will be obtained if estepm is as small as stepm. */
5618:
5619: /* For example we decided to compute the life expectancy with the smallest unit */
5620: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5621: nhstepm is the number of hstepm from age to agelim
5622: nstepm is the number of stepm from age to agelin.
5623: Look at hpijx to understand the reason of that which relies in memory size
5624: and note for a fixed period like estepm months */
5625: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5626: survival function given by stepm (the optimization length). Unfortunately it
5627: means that if the survival funtion is printed only each two years of age and if
5628: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5629: results. So we changed our mind and took the option of the best precision.
5630: */
5631: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5632:
5633: /* If stepm=6 months */
5634: /* nhstepm age range expressed in number of stepm */
5635: agelim=AGESUP;
5636: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5637: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5638: /* if (stepm >= YEARM) hstepm=1;*/
5639: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5640:
5641: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5642: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5643: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5644: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5645: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5646: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5647:
5648: for (age=bage; age<=fage; age ++){
5649: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5650: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5651: /* if (stepm >= YEARM) hstepm=1;*/
5652: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5653:
1.126 brouard 5654: /* If stepm=6 months */
5655: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5656: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5657:
5658: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5659:
1.126 brouard 5660: /* Computing Variances of health expectancies */
5661: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5662: decrease memory allocation */
5663: for(theta=1; theta <=npar; theta++){
5664: for(i=1; i<=npar; i++){
1.222 brouard 5665: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5666: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5667: }
1.235 brouard 5668: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5669: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5670:
1.126 brouard 5671: for(j=1; j<= nlstate; j++){
1.222 brouard 5672: for(i=1; i<=nlstate; i++){
5673: for(h=0; h<=nhstepm-1; h++){
5674: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5675: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5676: }
5677: }
1.126 brouard 5678: }
1.218 brouard 5679:
1.126 brouard 5680: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5681: for(h=0; h<=nhstepm-1; h++){
5682: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5683: }
1.126 brouard 5684: }/* End theta */
5685:
5686:
5687: for(h=0; h<=nhstepm-1; h++)
5688: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5689: for(theta=1; theta <=npar; theta++)
5690: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5691:
1.218 brouard 5692:
1.222 brouard 5693: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5694: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5695: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5696:
1.222 brouard 5697: printf("%d|",(int)age);fflush(stdout);
5698: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5699: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5700: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5701: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5702: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5703: for(ij=1;ij<=nlstate*nlstate;ij++)
5704: for(ji=1;ji<=nlstate*nlstate;ji++)
5705: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5706: }
5707: }
1.218 brouard 5708:
1.126 brouard 5709: /* Computing expectancies */
1.235 brouard 5710: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5711: for(i=1; i<=nlstate;i++)
5712: for(j=1; j<=nlstate;j++)
1.222 brouard 5713: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5714: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5715:
1.222 brouard 5716: /* 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 5717:
1.222 brouard 5718: }
1.269 brouard 5719:
5720: /* Standard deviation of expectancies ij */
1.126 brouard 5721: fprintf(ficresstdeij,"%3.0f",age );
5722: for(i=1; i<=nlstate;i++){
5723: eip=0.;
5724: vip=0.;
5725: for(j=1; j<=nlstate;j++){
1.222 brouard 5726: eip += eij[i][j][(int)age];
5727: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5728: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5729: 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 5730: }
5731: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5732: }
5733: fprintf(ficresstdeij,"\n");
1.218 brouard 5734:
1.269 brouard 5735: /* Variance of expectancies ij */
1.126 brouard 5736: fprintf(ficrescveij,"%3.0f",age );
5737: for(i=1; i<=nlstate;i++)
5738: for(j=1; j<=nlstate;j++){
1.222 brouard 5739: cptj= (j-1)*nlstate+i;
5740: for(i2=1; i2<=nlstate;i2++)
5741: for(j2=1; j2<=nlstate;j2++){
5742: cptj2= (j2-1)*nlstate+i2;
5743: if(cptj2 <= cptj)
5744: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5745: }
1.126 brouard 5746: }
5747: fprintf(ficrescveij,"\n");
1.218 brouard 5748:
1.126 brouard 5749: }
5750: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5751: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5752: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5753: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5754: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5755: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5756: printf("\n");
5757: fprintf(ficlog,"\n");
1.218 brouard 5758:
1.126 brouard 5759: free_vector(xm,1,npar);
5760: free_vector(xp,1,npar);
5761: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5762: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5763: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5764: }
1.218 brouard 5765:
1.126 brouard 5766: /************ Variance ******************/
1.235 brouard 5767: 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 5768: {
5769: /* Variance of health expectancies */
5770: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5771: /* double **newm;*/
5772: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5773:
5774: /* int movingaverage(); */
5775: double **dnewm,**doldm;
5776: double **dnewmp,**doldmp;
5777: int i, j, nhstepm, hstepm, h, nstepm ;
5778: int k;
5779: double *xp;
5780: double **gp, **gm; /* for var eij */
5781: double ***gradg, ***trgradg; /*for var eij */
5782: double **gradgp, **trgradgp; /* for var p point j */
5783: double *gpp, *gmp; /* for var p point j */
5784: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5785: double ***p3mat;
5786: double age,agelim, hf;
5787: /* double ***mobaverage; */
5788: int theta;
5789: char digit[4];
5790: char digitp[25];
5791:
5792: char fileresprobmorprev[FILENAMELENGTH];
5793:
5794: if(popbased==1){
5795: if(mobilav!=0)
5796: strcpy(digitp,"-POPULBASED-MOBILAV_");
5797: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5798: }
5799: else
5800: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5801:
1.218 brouard 5802: /* if (mobilav!=0) { */
5803: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5804: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5805: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5806: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5807: /* } */
5808: /* } */
5809:
5810: strcpy(fileresprobmorprev,"PRMORPREV-");
5811: sprintf(digit,"%-d",ij);
5812: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5813: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5814: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5815: strcat(fileresprobmorprev,fileresu);
5816: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5817: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5818: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5819: }
5820: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5821: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5822: pstamp(ficresprobmorprev);
5823: 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 5824: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5825: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5826: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5827: }
5828: for(j=1;j<=cptcoveff;j++)
5829: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5830: fprintf(ficresprobmorprev,"\n");
5831:
1.218 brouard 5832: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5833: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5834: fprintf(ficresprobmorprev," p.%-d SE",j);
5835: for(i=1; i<=nlstate;i++)
5836: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5837: }
5838: fprintf(ficresprobmorprev,"\n");
5839:
5840: fprintf(ficgp,"\n# Routine varevsij");
5841: fprintf(ficgp,"\nunset title \n");
5842: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5843: 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");
5844: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5845: /* } */
5846: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5847: pstamp(ficresvij);
5848: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5849: if(popbased==1)
5850: 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);
5851: else
5852: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5853: fprintf(ficresvij,"# Age");
5854: for(i=1; i<=nlstate;i++)
5855: for(j=1; j<=nlstate;j++)
5856: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5857: fprintf(ficresvij,"\n");
5858:
5859: xp=vector(1,npar);
5860: dnewm=matrix(1,nlstate,1,npar);
5861: doldm=matrix(1,nlstate,1,nlstate);
5862: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5863: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5864:
5865: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5866: gpp=vector(nlstate+1,nlstate+ndeath);
5867: gmp=vector(nlstate+1,nlstate+ndeath);
5868: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5869:
1.218 brouard 5870: if(estepm < stepm){
5871: printf ("Problem %d lower than %d\n",estepm, stepm);
5872: }
5873: else hstepm=estepm;
5874: /* For example we decided to compute the life expectancy with the smallest unit */
5875: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5876: nhstepm is the number of hstepm from age to agelim
5877: nstepm is the number of stepm from age to agelim.
5878: Look at function hpijx to understand why because of memory size limitations,
5879: we decided (b) to get a life expectancy respecting the most precise curvature of the
5880: survival function given by stepm (the optimization length). Unfortunately it
5881: means that if the survival funtion is printed every two years of age and if
5882: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5883: results. So we changed our mind and took the option of the best precision.
5884: */
5885: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5886: agelim = AGESUP;
5887: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5888: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5889: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5890: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5891: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5892: gp=matrix(0,nhstepm,1,nlstate);
5893: gm=matrix(0,nhstepm,1,nlstate);
5894:
5895:
5896: for(theta=1; theta <=npar; theta++){
5897: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5898: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5899: }
5900:
1.242 brouard 5901: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5902:
5903: if (popbased==1) {
5904: if(mobilav ==0){
5905: for(i=1; i<=nlstate;i++)
5906: prlim[i][i]=probs[(int)age][i][ij];
5907: }else{ /* mobilav */
5908: for(i=1; i<=nlstate;i++)
5909: prlim[i][i]=mobaverage[(int)age][i][ij];
5910: }
5911: }
5912:
1.235 brouard 5913: 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 5914: for(j=1; j<= nlstate; j++){
5915: for(h=0; h<=nhstepm; h++){
5916: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5917: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5918: }
5919: }
5920: /* Next for computing probability of death (h=1 means
5921: computed over hstepm matrices product = hstepm*stepm months)
5922: as a weighted average of prlim.
5923: */
5924: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5925: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5926: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5927: }
5928: /* end probability of death */
5929:
5930: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5931: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5932:
1.242 brouard 5933: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5934:
5935: if (popbased==1) {
5936: if(mobilav ==0){
5937: for(i=1; i<=nlstate;i++)
5938: prlim[i][i]=probs[(int)age][i][ij];
5939: }else{ /* mobilav */
5940: for(i=1; i<=nlstate;i++)
5941: prlim[i][i]=mobaverage[(int)age][i][ij];
5942: }
5943: }
5944:
1.235 brouard 5945: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5946:
5947: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5948: for(h=0; h<=nhstepm; h++){
5949: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5950: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5951: }
5952: }
5953: /* This for computing probability of death (h=1 means
5954: computed over hstepm matrices product = hstepm*stepm months)
5955: as a weighted average of prlim.
5956: */
5957: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5958: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5959: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5960: }
5961: /* end probability of death */
5962:
5963: for(j=1; j<= nlstate; j++) /* vareij */
5964: for(h=0; h<=nhstepm; h++){
5965: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5966: }
5967:
5968: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5969: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5970: }
5971:
5972: } /* End theta */
5973:
5974: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5975:
5976: for(h=0; h<=nhstepm; h++) /* veij */
5977: for(j=1; j<=nlstate;j++)
5978: for(theta=1; theta <=npar; theta++)
5979: trgradg[h][j][theta]=gradg[h][theta][j];
5980:
5981: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5982: for(theta=1; theta <=npar; theta++)
5983: trgradgp[j][theta]=gradgp[theta][j];
5984:
5985:
5986: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5987: for(i=1;i<=nlstate;i++)
5988: for(j=1;j<=nlstate;j++)
5989: vareij[i][j][(int)age] =0.;
5990:
5991: for(h=0;h<=nhstepm;h++){
5992: for(k=0;k<=nhstepm;k++){
5993: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5994: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5995: for(i=1;i<=nlstate;i++)
5996: for(j=1;j<=nlstate;j++)
5997: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5998: }
5999: }
6000:
6001: /* pptj */
6002: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6003: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6004: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6005: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6006: varppt[j][i]=doldmp[j][i];
6007: /* end ppptj */
6008: /* x centered again */
6009:
1.242 brouard 6010: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6011:
6012: if (popbased==1) {
6013: if(mobilav ==0){
6014: for(i=1; i<=nlstate;i++)
6015: prlim[i][i]=probs[(int)age][i][ij];
6016: }else{ /* mobilav */
6017: for(i=1; i<=nlstate;i++)
6018: prlim[i][i]=mobaverage[(int)age][i][ij];
6019: }
6020: }
6021:
6022: /* This for computing probability of death (h=1 means
6023: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6024: as a weighted average of prlim.
6025: */
1.235 brouard 6026: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6027: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6028: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6029: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6030: }
6031: /* end probability of death */
6032:
6033: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6034: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6035: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6036: for(i=1; i<=nlstate;i++){
6037: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6038: }
6039: }
6040: fprintf(ficresprobmorprev,"\n");
6041:
6042: fprintf(ficresvij,"%.0f ",age );
6043: for(i=1; i<=nlstate;i++)
6044: for(j=1; j<=nlstate;j++){
6045: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6046: }
6047: fprintf(ficresvij,"\n");
6048: free_matrix(gp,0,nhstepm,1,nlstate);
6049: free_matrix(gm,0,nhstepm,1,nlstate);
6050: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6051: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6052: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6053: } /* End age */
6054: free_vector(gpp,nlstate+1,nlstate+ndeath);
6055: free_vector(gmp,nlstate+1,nlstate+ndeath);
6056: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6057: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6058: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6059: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6060: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6061: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6062: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6063: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6064: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6065: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6066: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6067: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6068: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6069: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6070: 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);
6071: /* 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 6072: */
1.218 brouard 6073: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6074: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6075:
1.218 brouard 6076: free_vector(xp,1,npar);
6077: free_matrix(doldm,1,nlstate,1,nlstate);
6078: free_matrix(dnewm,1,nlstate,1,npar);
6079: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6080: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6081: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6082: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6083: fclose(ficresprobmorprev);
6084: fflush(ficgp);
6085: fflush(fichtm);
6086: } /* end varevsij */
1.126 brouard 6087:
6088: /************ Variance of prevlim ******************/
1.269 brouard 6089: 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 6090: {
1.205 brouard 6091: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6092: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6093:
1.268 brouard 6094: double **dnewmpar,**doldm;
1.126 brouard 6095: int i, j, nhstepm, hstepm;
6096: double *xp;
6097: double *gp, *gm;
6098: double **gradg, **trgradg;
1.208 brouard 6099: double **mgm, **mgp;
1.126 brouard 6100: double age,agelim;
6101: int theta;
6102:
6103: pstamp(ficresvpl);
6104: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6105: fprintf(ficresvpl,"# Age ");
6106: if(nresult >=1)
6107: fprintf(ficresvpl," Result# ");
1.126 brouard 6108: for(i=1; i<=nlstate;i++)
6109: fprintf(ficresvpl," %1d-%1d",i,i);
6110: fprintf(ficresvpl,"\n");
6111:
6112: xp=vector(1,npar);
1.268 brouard 6113: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6114: doldm=matrix(1,nlstate,1,nlstate);
6115:
6116: hstepm=1*YEARM; /* Every year of age */
6117: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6118: agelim = AGESUP;
6119: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6120: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6121: if (stepm >= YEARM) hstepm=1;
6122: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6123: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6124: mgp=matrix(1,npar,1,nlstate);
6125: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6126: gp=vector(1,nlstate);
6127: gm=vector(1,nlstate);
6128:
6129: for(theta=1; theta <=npar; theta++){
6130: for(i=1; i<=npar; i++){ /* Computes gradient */
6131: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6132: }
1.209 brouard 6133: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6134: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6135: else
1.235 brouard 6136: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6137: for(i=1;i<=nlstate;i++){
1.126 brouard 6138: gp[i] = prlim[i][i];
1.208 brouard 6139: mgp[theta][i] = prlim[i][i];
6140: }
1.126 brouard 6141: for(i=1; i<=npar; i++) /* Computes gradient */
6142: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6143: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6144: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6145: else
1.235 brouard 6146: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6147: for(i=1;i<=nlstate;i++){
1.126 brouard 6148: gm[i] = prlim[i][i];
1.208 brouard 6149: mgm[theta][i] = prlim[i][i];
6150: }
1.126 brouard 6151: for(i=1;i<=nlstate;i++)
6152: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6153: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6154: } /* End theta */
6155:
6156: trgradg =matrix(1,nlstate,1,npar);
6157:
6158: for(j=1; j<=nlstate;j++)
6159: for(theta=1; theta <=npar; theta++)
6160: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6161: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6162: /* printf("\nmgm mgp %d ",(int)age); */
6163: /* for(j=1; j<=nlstate;j++){ */
6164: /* printf(" %d ",j); */
6165: /* for(theta=1; theta <=npar; theta++) */
6166: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6167: /* printf("\n "); */
6168: /* } */
6169: /* } */
6170: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6171: /* printf("\n gradg %d ",(int)age); */
6172: /* for(j=1; j<=nlstate;j++){ */
6173: /* printf("%d ",j); */
6174: /* for(theta=1; theta <=npar; theta++) */
6175: /* printf("%d %lf ",theta,gradg[theta][j]); */
6176: /* printf("\n "); */
6177: /* } */
6178: /* } */
1.126 brouard 6179:
6180: for(i=1;i<=nlstate;i++)
6181: varpl[i][(int)age] =0.;
1.209 brouard 6182: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
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: }else{
1.268 brouard 6186: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6187: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6188: }
1.126 brouard 6189: for(i=1;i<=nlstate;i++)
6190: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6191:
6192: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6193: if(nresult >=1)
6194: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6195: for(i=1; i<=nlstate;i++)
6196: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6197: fprintf(ficresvpl,"\n");
6198: free_vector(gp,1,nlstate);
6199: free_vector(gm,1,nlstate);
1.208 brouard 6200: free_matrix(mgm,1,npar,1,nlstate);
6201: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6202: free_matrix(gradg,1,npar,1,nlstate);
6203: free_matrix(trgradg,1,nlstate,1,npar);
6204: } /* End age */
6205:
6206: free_vector(xp,1,npar);
6207: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6208: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6209:
6210: }
6211:
6212:
6213: /************ Variance of backprevalence limit ******************/
1.269 brouard 6214: 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 6215: {
6216: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6217: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6218:
6219: double **dnewmpar,**doldm;
6220: int i, j, nhstepm, hstepm;
6221: double *xp;
6222: double *gp, *gm;
6223: double **gradg, **trgradg;
6224: double **mgm, **mgp;
6225: double age,agelim;
6226: int theta;
6227:
6228: pstamp(ficresvbl);
6229: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6230: fprintf(ficresvbl,"# Age ");
6231: if(nresult >=1)
6232: fprintf(ficresvbl," Result# ");
6233: for(i=1; i<=nlstate;i++)
6234: fprintf(ficresvbl," %1d-%1d",i,i);
6235: fprintf(ficresvbl,"\n");
6236:
6237: xp=vector(1,npar);
6238: dnewmpar=matrix(1,nlstate,1,npar);
6239: doldm=matrix(1,nlstate,1,nlstate);
6240:
6241: hstepm=1*YEARM; /* Every year of age */
6242: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6243: agelim = AGEINF;
6244: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6245: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6246: if (stepm >= YEARM) hstepm=1;
6247: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6248: gradg=matrix(1,npar,1,nlstate);
6249: mgp=matrix(1,npar,1,nlstate);
6250: mgm=matrix(1,npar,1,nlstate);
6251: gp=vector(1,nlstate);
6252: gm=vector(1,nlstate);
6253:
6254: for(theta=1; theta <=npar; theta++){
6255: for(i=1; i<=npar; i++){ /* Computes gradient */
6256: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6257: }
6258: if(mobilavproj > 0 )
6259: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6260: else
6261: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6262: for(i=1;i<=nlstate;i++){
6263: gp[i] = bprlim[i][i];
6264: mgp[theta][i] = bprlim[i][i];
6265: }
6266: for(i=1; i<=npar; i++) /* Computes gradient */
6267: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6268: if(mobilavproj > 0 )
6269: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6270: else
6271: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6272: for(i=1;i<=nlstate;i++){
6273: gm[i] = bprlim[i][i];
6274: mgm[theta][i] = bprlim[i][i];
6275: }
6276: for(i=1;i<=nlstate;i++)
6277: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6278: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6279: } /* End theta */
6280:
6281: trgradg =matrix(1,nlstate,1,npar);
6282:
6283: for(j=1; j<=nlstate;j++)
6284: for(theta=1; theta <=npar; theta++)
6285: trgradg[j][theta]=gradg[theta][j];
6286: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6287: /* printf("\nmgm mgp %d ",(int)age); */
6288: /* for(j=1; j<=nlstate;j++){ */
6289: /* printf(" %d ",j); */
6290: /* for(theta=1; theta <=npar; theta++) */
6291: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6292: /* printf("\n "); */
6293: /* } */
6294: /* } */
6295: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6296: /* printf("\n gradg %d ",(int)age); */
6297: /* for(j=1; j<=nlstate;j++){ */
6298: /* printf("%d ",j); */
6299: /* for(theta=1; theta <=npar; theta++) */
6300: /* printf("%d %lf ",theta,gradg[theta][j]); */
6301: /* printf("\n "); */
6302: /* } */
6303: /* } */
6304:
6305: for(i=1;i<=nlstate;i++)
6306: varbpl[i][(int)age] =0.;
6307: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6308: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6309: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6310: }else{
6311: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6312: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6313: }
6314: for(i=1;i<=nlstate;i++)
6315: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6316:
6317: fprintf(ficresvbl,"%.0f ",age );
6318: if(nresult >=1)
6319: fprintf(ficresvbl,"%d ",nres );
6320: for(i=1; i<=nlstate;i++)
6321: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6322: fprintf(ficresvbl,"\n");
6323: free_vector(gp,1,nlstate);
6324: free_vector(gm,1,nlstate);
6325: free_matrix(mgm,1,npar,1,nlstate);
6326: free_matrix(mgp,1,npar,1,nlstate);
6327: free_matrix(gradg,1,npar,1,nlstate);
6328: free_matrix(trgradg,1,nlstate,1,npar);
6329: } /* End age */
6330:
6331: free_vector(xp,1,npar);
6332: free_matrix(doldm,1,nlstate,1,npar);
6333: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6334:
6335: }
6336:
6337: /************ Variance of one-step probabilities ******************/
6338: 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 6339: {
6340: int i, j=0, k1, l1, tj;
6341: int k2, l2, j1, z1;
6342: int k=0, l;
6343: int first=1, first1, first2;
6344: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6345: double **dnewm,**doldm;
6346: double *xp;
6347: double *gp, *gm;
6348: double **gradg, **trgradg;
6349: double **mu;
6350: double age, cov[NCOVMAX+1];
6351: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6352: int theta;
6353: char fileresprob[FILENAMELENGTH];
6354: char fileresprobcov[FILENAMELENGTH];
6355: char fileresprobcor[FILENAMELENGTH];
6356: double ***varpij;
6357:
6358: strcpy(fileresprob,"PROB_");
6359: strcat(fileresprob,fileres);
6360: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6361: printf("Problem with resultfile: %s\n", fileresprob);
6362: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6363: }
6364: strcpy(fileresprobcov,"PROBCOV_");
6365: strcat(fileresprobcov,fileresu);
6366: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6367: printf("Problem with resultfile: %s\n", fileresprobcov);
6368: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6369: }
6370: strcpy(fileresprobcor,"PROBCOR_");
6371: strcat(fileresprobcor,fileresu);
6372: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6373: printf("Problem with resultfile: %s\n", fileresprobcor);
6374: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6375: }
6376: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6377: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6378: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6379: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6380: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6381: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6382: pstamp(ficresprob);
6383: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6384: fprintf(ficresprob,"# Age");
6385: pstamp(ficresprobcov);
6386: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6387: fprintf(ficresprobcov,"# Age");
6388: pstamp(ficresprobcor);
6389: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6390: fprintf(ficresprobcor,"# Age");
1.126 brouard 6391:
6392:
1.222 brouard 6393: for(i=1; i<=nlstate;i++)
6394: for(j=1; j<=(nlstate+ndeath);j++){
6395: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6396: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6397: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6398: }
6399: /* fprintf(ficresprob,"\n");
6400: fprintf(ficresprobcov,"\n");
6401: fprintf(ficresprobcor,"\n");
6402: */
6403: xp=vector(1,npar);
6404: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6405: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6406: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6407: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6408: first=1;
6409: fprintf(ficgp,"\n# Routine varprob");
6410: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6411: fprintf(fichtm,"\n");
6412:
1.266 brouard 6413: 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 6414: 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);
6415: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6416: and drawn. It helps understanding how is the covariance between two incidences.\
6417: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6418: 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 6419: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6420: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6421: standard deviations wide on each axis. <br>\
6422: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6423: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6424: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6425:
1.222 brouard 6426: cov[1]=1;
6427: /* tj=cptcoveff; */
1.225 brouard 6428: tj = (int) pow(2,cptcoveff);
1.222 brouard 6429: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6430: j1=0;
1.224 brouard 6431: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6432: if (cptcovn>0) {
6433: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6434: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6435: fprintf(ficresprob, "**********\n#\n");
6436: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6437: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6438: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6439:
1.222 brouard 6440: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6441: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6442: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6443:
6444:
1.222 brouard 6445: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6446: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6447: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6448:
1.222 brouard 6449: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6450: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6451: fprintf(ficresprobcor, "**********\n#");
6452: if(invalidvarcomb[j1]){
6453: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6454: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6455: continue;
6456: }
6457: }
6458: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6459: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6460: gp=vector(1,(nlstate)*(nlstate+ndeath));
6461: gm=vector(1,(nlstate)*(nlstate+ndeath));
6462: for (age=bage; age<=fage; age ++){
6463: cov[2]=age;
6464: if(nagesqr==1)
6465: cov[3]= age*age;
6466: for (k=1; k<=cptcovn;k++) {
6467: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6468: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6469: * 1 1 1 1 1
6470: * 2 2 1 1 1
6471: * 3 1 2 1 1
6472: */
6473: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6474: }
6475: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6476: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6477: for (k=1; k<=cptcovprod;k++)
6478: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6479:
6480:
1.222 brouard 6481: for(theta=1; theta <=npar; theta++){
6482: for(i=1; i<=npar; i++)
6483: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6484:
1.222 brouard 6485: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6486:
1.222 brouard 6487: k=0;
6488: for(i=1; i<= (nlstate); i++){
6489: for(j=1; j<=(nlstate+ndeath);j++){
6490: k=k+1;
6491: gp[k]=pmmij[i][j];
6492: }
6493: }
1.220 brouard 6494:
1.222 brouard 6495: for(i=1; i<=npar; i++)
6496: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6497:
1.222 brouard 6498: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6499: k=0;
6500: for(i=1; i<=(nlstate); i++){
6501: for(j=1; j<=(nlstate+ndeath);j++){
6502: k=k+1;
6503: gm[k]=pmmij[i][j];
6504: }
6505: }
1.220 brouard 6506:
1.222 brouard 6507: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6508: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6509: }
1.126 brouard 6510:
1.222 brouard 6511: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6512: for(theta=1; theta <=npar; theta++)
6513: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6514:
1.222 brouard 6515: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6516: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6517:
1.222 brouard 6518: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6519:
1.222 brouard 6520: k=0;
6521: for(i=1; i<=(nlstate); i++){
6522: for(j=1; j<=(nlstate+ndeath);j++){
6523: k=k+1;
6524: mu[k][(int) age]=pmmij[i][j];
6525: }
6526: }
6527: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6528: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6529: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6530:
1.222 brouard 6531: /*printf("\n%d ",(int)age);
6532: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6533: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6534: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6535: }*/
1.220 brouard 6536:
1.222 brouard 6537: fprintf(ficresprob,"\n%d ",(int)age);
6538: fprintf(ficresprobcov,"\n%d ",(int)age);
6539: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6540:
1.222 brouard 6541: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6542: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6543: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6544: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6545: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6546: }
6547: i=0;
6548: for (k=1; k<=(nlstate);k++){
6549: for (l=1; l<=(nlstate+ndeath);l++){
6550: i++;
6551: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6552: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6553: for (j=1; j<=i;j++){
6554: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6555: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6556: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6557: }
6558: }
6559: }/* end of loop for state */
6560: } /* end of loop for age */
6561: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6562: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6563: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6564: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6565:
6566: /* Confidence intervalle of pij */
6567: /*
6568: fprintf(ficgp,"\nunset parametric;unset label");
6569: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6570: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6571: 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);
6572: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6573: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6574: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6575: */
6576:
6577: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6578: first1=1;first2=2;
6579: for (k2=1; k2<=(nlstate);k2++){
6580: for (l2=1; l2<=(nlstate+ndeath);l2++){
6581: if(l2==k2) continue;
6582: j=(k2-1)*(nlstate+ndeath)+l2;
6583: for (k1=1; k1<=(nlstate);k1++){
6584: for (l1=1; l1<=(nlstate+ndeath);l1++){
6585: if(l1==k1) continue;
6586: i=(k1-1)*(nlstate+ndeath)+l1;
6587: if(i<=j) continue;
6588: for (age=bage; age<=fage; age ++){
6589: if ((int)age %5==0){
6590: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6591: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6592: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6593: mu1=mu[i][(int) age]/stepm*YEARM ;
6594: mu2=mu[j][(int) age]/stepm*YEARM;
6595: c12=cv12/sqrt(v1*v2);
6596: /* Computing eigen value of matrix of covariance */
6597: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6598: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6599: if ((lc2 <0) || (lc1 <0) ){
6600: if(first2==1){
6601: first1=0;
6602: 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);
6603: }
6604: 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);
6605: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6606: /* lc2=fabs(lc2); */
6607: }
1.220 brouard 6608:
1.222 brouard 6609: /* Eigen vectors */
6610: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6611: /*v21=sqrt(1.-v11*v11); *//* error */
6612: v21=(lc1-v1)/cv12*v11;
6613: v12=-v21;
6614: v22=v11;
6615: tnalp=v21/v11;
6616: if(first1==1){
6617: first1=0;
6618: 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);
6619: }
6620: 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);
6621: /*printf(fignu*/
6622: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6623: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6624: if(first==1){
6625: first=0;
6626: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6627: fprintf(ficgp,"\nset parametric;unset label");
6628: 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);
6629: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6630: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6631: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6632: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6633: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6634: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6635: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6636: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6637: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6638: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6639: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6640: 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 6641: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6642: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6643: }else{
6644: first=0;
6645: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6646: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6647: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6648: 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 6649: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6650: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6651: }/* if first */
6652: } /* age mod 5 */
6653: } /* end loop age */
6654: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6655: first=1;
6656: } /*l12 */
6657: } /* k12 */
6658: } /*l1 */
6659: }/* k1 */
6660: } /* loop on combination of covariates j1 */
6661: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6662: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6663: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6664: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6665: free_vector(xp,1,npar);
6666: fclose(ficresprob);
6667: fclose(ficresprobcov);
6668: fclose(ficresprobcor);
6669: fflush(ficgp);
6670: fflush(fichtmcov);
6671: }
1.126 brouard 6672:
6673:
6674: /******************* Printing html file ***********/
1.201 brouard 6675: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6676: int lastpass, int stepm, int weightopt, char model[],\
6677: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6678: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6679: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6680: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6681: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6682:
6683: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6684: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6685: </ul>");
1.237 brouard 6686: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6687: </ul>", model);
1.214 brouard 6688: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6689: 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",
6690: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6691: 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 6692: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6693: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6694: fprintf(fichtm,"\
6695: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6696: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6697: fprintf(fichtm,"\
1.217 brouard 6698: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6699: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6700: fprintf(fichtm,"\
1.126 brouard 6701: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6702: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6703: fprintf(fichtm,"\
1.217 brouard 6704: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6705: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6706: fprintf(fichtm,"\
1.211 brouard 6707: - (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 6708: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6709: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6710: if(prevfcast==1){
6711: fprintf(fichtm,"\
6712: - Prevalence projections by age and states: \
1.201 brouard 6713: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6714: }
1.126 brouard 6715:
6716:
1.225 brouard 6717: m=pow(2,cptcoveff);
1.222 brouard 6718: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6719:
1.264 brouard 6720: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6721:
6722: jj1=0;
6723:
6724: fprintf(fichtm," \n<ul>");
6725: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6726: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6727: if(m != 1 && TKresult[nres]!= k1)
6728: continue;
6729: jj1++;
6730: if (cptcovn > 0) {
6731: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6732: for (cpt=1; cpt<=cptcoveff;cpt++){
6733: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6734: }
6735: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6736: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6737: }
6738: fprintf(fichtm,"\">");
6739:
6740: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6741: fprintf(fichtm,"************ Results for covariates");
6742: for (cpt=1; cpt<=cptcoveff;cpt++){
6743: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6744: }
6745: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6746: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6747: }
6748: if(invalidvarcomb[k1]){
6749: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6750: continue;
6751: }
6752: fprintf(fichtm,"</a></li>");
6753: } /* cptcovn >0 */
6754: }
6755: fprintf(fichtm," \n</ul>");
6756:
1.222 brouard 6757: jj1=0;
1.237 brouard 6758:
6759: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6760: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6761: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6762: continue;
1.220 brouard 6763:
1.222 brouard 6764: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6765: jj1++;
6766: if (cptcovn > 0) {
1.264 brouard 6767: fprintf(fichtm,"\n<p><a name=\"rescov");
6768: for (cpt=1; cpt<=cptcoveff;cpt++){
6769: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6770: }
6771: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6772: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6773: }
6774: fprintf(fichtm,"\"</a>");
6775:
1.222 brouard 6776: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6777: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6778: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6779: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6780: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6781: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6782: }
1.237 brouard 6783: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6784: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6785: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6786: }
6787:
1.230 brouard 6788: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6789: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6790: if(invalidvarcomb[k1]){
6791: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6792: printf("\nCombination (%d) ignored because no cases \n",k1);
6793: continue;
6794: }
6795: }
6796: /* aij, bij */
1.259 brouard 6797: 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 6798: <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 6799: /* Pij */
1.241 brouard 6800: 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> \
6801: <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 6802: /* Quasi-incidences */
6803: 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 6804: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6805: 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 6806: 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> \
6807: <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 6808: /* Survival functions (period) in state j */
6809: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6810: 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> \
6811: <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 6812: }
6813: /* State specific survival functions (period) */
6814: for(cpt=1; cpt<=nlstate;cpt++){
6815: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6816: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6817: <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 6818: }
6819: /* Period (stable) prevalence in each health state */
6820: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6821: 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> \
6822: <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 6823: }
6824: if(backcast==1){
6825: /* Period (stable) back prevalence in each health state */
6826: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6827: 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 6828: <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 6829: }
1.217 brouard 6830: }
1.222 brouard 6831: if(prevfcast==1){
6832: /* Projection of prevalence up to period (stable) prevalence in each health state */
6833: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6834: 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> \
6835: <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 6836: }
6837: }
1.268 brouard 6838: if(backcast==1){
6839: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6840: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6841: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6842: 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 \
6843: 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) \
6844: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6845: <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 6846: }
6847: }
1.220 brouard 6848:
1.222 brouard 6849: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6850: 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> \
6851: <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 6852: }
6853: /* } /\* end i1 *\/ */
6854: }/* End k1 */
6855: fprintf(fichtm,"</ul>");
1.126 brouard 6856:
1.222 brouard 6857: fprintf(fichtm,"\
1.126 brouard 6858: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6859: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6860: - 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 6861: But because parameters are usually highly correlated (a higher incidence of disability \
6862: and a higher incidence of recovery can give very close observed transition) it might \
6863: be very useful to look not only at linear confidence intervals estimated from the \
6864: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6865: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6866: covariance matrix of the one-step probabilities. \
6867: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6868:
1.222 brouard 6869: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6870: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6871: fprintf(fichtm,"\
1.126 brouard 6872: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6873: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6874:
1.222 brouard 6875: fprintf(fichtm,"\
1.126 brouard 6876: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6877: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6878: fprintf(fichtm,"\
1.126 brouard 6879: - 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): \
6880: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6881: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6882: fprintf(fichtm,"\
1.126 brouard 6883: - (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): \
6884: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6885: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6886: fprintf(fichtm,"\
1.128 brouard 6887: - 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 6888: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6889: fprintf(fichtm,"\
1.128 brouard 6890: - 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 6891: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6892: fprintf(fichtm,"\
1.126 brouard 6893: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6894: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6895:
6896: /* if(popforecast==1) fprintf(fichtm,"\n */
6897: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6898: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6899: /* <br>",fileres,fileres,fileres,fileres); */
6900: /* else */
6901: /* 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 6902: fflush(fichtm);
6903: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6904:
1.225 brouard 6905: m=pow(2,cptcoveff);
1.222 brouard 6906: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6907:
1.222 brouard 6908: jj1=0;
1.237 brouard 6909:
1.241 brouard 6910: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6911: for(k1=1; k1<=m;k1++){
1.253 brouard 6912: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6913: continue;
1.222 brouard 6914: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6915: jj1++;
1.126 brouard 6916: if (cptcovn > 0) {
6917: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6918: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6919: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6920: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6921: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6922: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6923: }
6924:
1.126 brouard 6925: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6926:
1.222 brouard 6927: if(invalidvarcomb[k1]){
6928: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6929: continue;
6930: }
1.126 brouard 6931: }
6932: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6933: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6934: 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 6935: <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 6936: }
6937: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6938: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6939: true period expectancies (those weighted with period prevalences are also\
6940: drawn in addition to the population based expectancies computed using\
1.241 brouard 6941: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6942: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6943: /* } /\* end i1 *\/ */
6944: }/* End k1 */
1.241 brouard 6945: }/* End nres */
1.222 brouard 6946: fprintf(fichtm,"</ul>");
6947: fflush(fichtm);
1.126 brouard 6948: }
6949:
6950: /******************* Gnuplot file **************/
1.270 brouard 6951: 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 6952:
6953: char dirfileres[132],optfileres[132];
1.264 brouard 6954: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6955: 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 6956: int lv=0, vlv=0, kl=0;
1.130 brouard 6957: int ng=0;
1.201 brouard 6958: int vpopbased;
1.223 brouard 6959: int ioffset; /* variable offset for columns */
1.270 brouard 6960: int iyearc=1; /* variable column for year of projection */
6961: int iagec=1; /* variable column for age of projection */
1.235 brouard 6962: int nres=0; /* Index of resultline */
1.266 brouard 6963: int istart=1; /* For starting graphs in projections */
1.219 brouard 6964:
1.126 brouard 6965: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6966: /* printf("Problem with file %s",optionfilegnuplot); */
6967: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6968: /* } */
6969:
6970: /*#ifdef windows */
6971: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6972: /*#endif */
1.225 brouard 6973: m=pow(2,cptcoveff);
1.126 brouard 6974:
1.274 brouard 6975: /* diagram of the model */
6976: fprintf(ficgp,"\n#Diagram of the model \n");
6977: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
6978: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
6979: 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);
6980:
6981: 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);
6982: fprintf(ficgp,"\n#show arrow\nunset label\n");
6983: 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);
6984: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
6985: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
6986: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
6987: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
6988:
1.202 brouard 6989: /* Contribution to likelihood */
6990: /* Plot the probability implied in the likelihood */
1.223 brouard 6991: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6992: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6993: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6994: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6995: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6996: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6997: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6998: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6999: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7000: 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));
7001: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7002: 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));
7003: for (i=1; i<= nlstate ; i ++) {
7004: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7005: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7006: 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);
7007: for (j=2; j<= nlstate+ndeath ; j ++) {
7008: 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);
7009: }
7010: fprintf(ficgp,";\nset out; unset ylabel;\n");
7011: }
7012: /* 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 */
7013: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7014: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7015: fprintf(ficgp,"\nset out;unset log\n");
7016: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7017:
1.126 brouard 7018: strcpy(dirfileres,optionfilefiname);
7019: strcpy(optfileres,"vpl");
1.223 brouard 7020: /* 1eme*/
1.238 brouard 7021: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7022: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7023: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7024: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7025: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7026: continue;
7027: /* We are interested in selected combination by the resultline */
1.246 brouard 7028: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7029: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7030: strcpy(gplotlabel,"(");
1.238 brouard 7031: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7032: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7033: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7034: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7035: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7036: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7037: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7038: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7039: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7040: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7041: }
7042: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7043: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7044: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7045: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7046: }
7047: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7048: /* printf("\n#\n"); */
1.238 brouard 7049: fprintf(ficgp,"\n#\n");
7050: if(invalidvarcomb[k1]){
1.260 brouard 7051: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7052: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7053: continue;
7054: }
1.235 brouard 7055:
1.241 brouard 7056: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7057: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7058: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7059: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7060: 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);
7061: /* 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); */
7062: /* k1-1 error should be nres-1*/
1.238 brouard 7063: for (i=1; i<= nlstate ; i ++) {
7064: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7065: else fprintf(ficgp," %%*lf (%%*lf)");
7066: }
1.260 brouard 7067: 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 7068: for (i=1; i<= nlstate ; i ++) {
7069: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7070: else fprintf(ficgp," %%*lf (%%*lf)");
7071: }
1.260 brouard 7072: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7073: for (i=1; i<= nlstate ; i ++) {
7074: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7075: else fprintf(ficgp," %%*lf (%%*lf)");
7076: }
1.265 brouard 7077: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7078:
7079: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7080: if(cptcoveff ==0){
1.271 brouard 7081: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7082: }else{
7083: kl=0;
7084: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7085: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7086: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7087: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7088: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7089: vlv= nbcode[Tvaraff[k]][lv];
7090: kl++;
7091: /* 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 *\/ */
7092: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7093: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7094: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7095: if(k==cptcoveff){
7096: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7097: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7098: }else{
7099: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7100: kl++;
7101: }
7102: } /* end covariate */
7103: } /* end if no covariate */
7104:
1.238 brouard 7105: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7106: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7107: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7108: if(cptcoveff ==0){
1.245 brouard 7109: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7110: }else{
7111: kl=0;
7112: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7113: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7114: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7115: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7116: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7117: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7118: kl++;
1.238 brouard 7119: /* 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 *\/ */
7120: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7121: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7122: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7123: if(k==cptcoveff){
1.245 brouard 7124: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7125: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7126: }else{
7127: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7128: kl++;
7129: }
7130: } /* end covariate */
7131: } /* end if no covariate */
1.268 brouard 7132: if(backcast == 1){
7133: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7134: /* k1-1 error should be nres-1*/
7135: for (i=1; i<= nlstate ; i ++) {
7136: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7137: else fprintf(ficgp," %%*lf (%%*lf)");
7138: }
1.271 brouard 7139: 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 7140: for (i=1; i<= nlstate ; i ++) {
7141: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7142: else fprintf(ficgp," %%*lf (%%*lf)");
7143: }
1.276 brouard 7144: 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 7145: for (i=1; i<= nlstate ; i ++) {
7146: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7147: else fprintf(ficgp," %%*lf (%%*lf)");
7148: }
1.274 brouard 7149: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7150: } /* end if backprojcast */
1.238 brouard 7151: } /* end if backcast */
1.276 brouard 7152: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7153: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7154: } /* nres */
1.201 brouard 7155: } /* k1 */
7156: } /* cpt */
1.235 brouard 7157:
7158:
1.126 brouard 7159: /*2 eme*/
1.238 brouard 7160: for (k1=1; k1<= m ; k1 ++){
7161: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7162: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7163: continue;
7164: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7165: strcpy(gplotlabel,"(");
1.238 brouard 7166: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7167: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7168: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7169: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7170: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7171: vlv= nbcode[Tvaraff[k]][lv];
7172: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7173: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7174: }
1.237 brouard 7175: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7176: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7177: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7178: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7179: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7180: }
1.264 brouard 7181: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7182: fprintf(ficgp,"\n#\n");
1.223 brouard 7183: if(invalidvarcomb[k1]){
7184: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7185: continue;
7186: }
1.219 brouard 7187:
1.241 brouard 7188: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7189: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7190: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7191: if(vpopbased==0){
1.238 brouard 7192: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7193: }else
1.238 brouard 7194: fprintf(ficgp,"\nreplot ");
7195: for (i=1; i<= nlstate+1 ; i ++) {
7196: k=2*i;
1.261 brouard 7197: 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 7198: for (j=1; j<= nlstate+1 ; j ++) {
7199: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7200: else fprintf(ficgp," %%*lf (%%*lf)");
7201: }
7202: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7203: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7204: 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 7205: for (j=1; j<= nlstate+1 ; j ++) {
7206: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7207: else fprintf(ficgp," %%*lf (%%*lf)");
7208: }
7209: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7210: 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 7211: for (j=1; j<= nlstate+1 ; j ++) {
7212: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7213: else fprintf(ficgp," %%*lf (%%*lf)");
7214: }
7215: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7216: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7217: } /* state */
7218: } /* vpopbased */
1.264 brouard 7219: 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 7220: } /* end nres */
7221: } /* k1 end 2 eme*/
7222:
7223:
7224: /*3eme*/
7225: for (k1=1; k1<= m ; k1 ++){
7226: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7227: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7228: continue;
7229:
7230: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7231: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7232: strcpy(gplotlabel,"(");
1.238 brouard 7233: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7234: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7235: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7236: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7237: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7238: vlv= nbcode[Tvaraff[k]][lv];
7239: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7240: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7241: }
7242: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7243: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7244: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7245: }
1.264 brouard 7246: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7247: fprintf(ficgp,"\n#\n");
7248: if(invalidvarcomb[k1]){
7249: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7250: continue;
7251: }
7252:
7253: /* k=2+nlstate*(2*cpt-2); */
7254: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7255: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7256: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7257: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7258: 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 7259: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7260: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7261: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7262: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7263: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7264: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7265:
1.238 brouard 7266: */
7267: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7268: 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 7269: /* 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 7270:
1.238 brouard 7271: }
1.261 brouard 7272: 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 7273: }
1.264 brouard 7274: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7275: } /* end nres */
7276: } /* end kl 3eme */
1.126 brouard 7277:
1.223 brouard 7278: /* 4eme */
1.201 brouard 7279: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7280: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7281: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7282: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7283: continue;
1.238 brouard 7284: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7285: strcpy(gplotlabel,"(");
1.238 brouard 7286: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7287: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7288: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7289: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7290: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7291: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7292: vlv= nbcode[Tvaraff[k]][lv];
7293: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7294: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7295: }
7296: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7297: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7298: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7299: }
1.264 brouard 7300: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7301: fprintf(ficgp,"\n#\n");
7302: if(invalidvarcomb[k1]){
7303: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7304: continue;
1.223 brouard 7305: }
1.238 brouard 7306:
1.241 brouard 7307: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7308: 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 7309: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7310: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7311: k=3;
7312: for (i=1; i<= nlstate ; i ++){
7313: if(i==1){
7314: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7315: }else{
7316: fprintf(ficgp,", '' ");
7317: }
7318: l=(nlstate+ndeath)*(i-1)+1;
7319: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7320: for (j=2; j<= nlstate+ndeath ; j ++)
7321: fprintf(ficgp,"+$%d",k+l+j-1);
7322: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7323: } /* nlstate */
1.264 brouard 7324: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7325: } /* end cpt state*/
7326: } /* end nres */
7327: } /* end covariate k1 */
7328:
1.220 brouard 7329: /* 5eme */
1.201 brouard 7330: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7331: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7332: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7333: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7334: continue;
1.238 brouard 7335: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7336: strcpy(gplotlabel,"(");
1.238 brouard 7337: 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);
7338: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7339: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7340: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7341: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7342: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7343: vlv= nbcode[Tvaraff[k]][lv];
7344: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7345: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7346: }
7347: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7348: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7349: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7350: }
1.264 brouard 7351: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7352: fprintf(ficgp,"\n#\n");
7353: if(invalidvarcomb[k1]){
7354: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7355: continue;
7356: }
1.227 brouard 7357:
1.241 brouard 7358: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7359: 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 7360: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7361: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7362: k=3;
7363: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7364: if(j==1)
7365: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7366: else
7367: fprintf(ficgp,", '' ");
7368: l=(nlstate+ndeath)*(cpt-1) +j;
7369: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7370: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7371: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7372: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7373: } /* nlstate */
7374: fprintf(ficgp,", '' ");
7375: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7376: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7377: l=(nlstate+ndeath)*(cpt-1) +j;
7378: if(j < nlstate)
7379: fprintf(ficgp,"$%d +",k+l);
7380: else
7381: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7382: }
1.264 brouard 7383: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7384: } /* end cpt state*/
7385: } /* end covariate */
7386: } /* end nres */
1.227 brouard 7387:
1.220 brouard 7388: /* 6eme */
1.202 brouard 7389: /* CV preval stable (period) for each covariate */
1.237 brouard 7390: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7391: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7392: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7393: continue;
1.255 brouard 7394: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7395: strcpy(gplotlabel,"(");
1.211 brouard 7396: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7397: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7398: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7399: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7400: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7401: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7402: vlv= nbcode[Tvaraff[k]][lv];
7403: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7404: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7405: }
1.237 brouard 7406: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7407: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7408: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7409: }
1.264 brouard 7410: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7411: fprintf(ficgp,"\n#\n");
1.223 brouard 7412: if(invalidvarcomb[k1]){
1.227 brouard 7413: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7414: continue;
1.223 brouard 7415: }
1.227 brouard 7416:
1.241 brouard 7417: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7418: 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 7419: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7420: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7421: k=3; /* Offset */
1.255 brouard 7422: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7423: if(i==1)
7424: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7425: else
7426: fprintf(ficgp,", '' ");
1.255 brouard 7427: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7428: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7429: for (j=2; j<= nlstate ; j ++)
7430: fprintf(ficgp,"+$%d",k+l+j-1);
7431: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7432: } /* nlstate */
1.264 brouard 7433: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7434: } /* end cpt state*/
7435: } /* end covariate */
1.227 brouard 7436:
7437:
1.220 brouard 7438: /* 7eme */
1.218 brouard 7439: if(backcast == 1){
1.217 brouard 7440: /* CV back preval stable (period) for each covariate */
1.237 brouard 7441: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7442: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7443: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7444: continue;
1.268 brouard 7445: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7446: strcpy(gplotlabel,"(");
7447: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7448: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7449: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7450: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7451: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7452: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7453: vlv= nbcode[Tvaraff[k]][lv];
7454: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7455: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7456: }
1.237 brouard 7457: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7458: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7459: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7460: }
1.264 brouard 7461: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7462: fprintf(ficgp,"\n#\n");
7463: if(invalidvarcomb[k1]){
7464: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7465: continue;
7466: }
7467:
1.241 brouard 7468: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7469: 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 7470: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7471: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7472: k=3; /* Offset */
1.268 brouard 7473: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7474: if(i==1)
7475: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7476: else
7477: fprintf(ficgp,", '' ");
7478: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7479: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7480: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7481: /* 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 7482: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7483: /* for (j=2; j<= nlstate ; j ++) */
7484: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7485: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7486: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7487: } /* nlstate */
1.264 brouard 7488: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7489: } /* end cpt state*/
7490: } /* end covariate */
7491: } /* End if backcast */
7492:
1.223 brouard 7493: /* 8eme */
1.218 brouard 7494: if(prevfcast==1){
7495: /* Projection from cross-sectional to stable (period) for each covariate */
7496:
1.237 brouard 7497: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7498: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7499: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7500: continue;
1.211 brouard 7501: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7502: strcpy(gplotlabel,"(");
1.227 brouard 7503: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7504: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7505: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7506: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7507: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7508: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7509: vlv= nbcode[Tvaraff[k]][lv];
7510: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7511: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7512: }
1.237 brouard 7513: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7514: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7515: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7516: }
1.264 brouard 7517: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7518: fprintf(ficgp,"\n#\n");
7519: if(invalidvarcomb[k1]){
7520: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7521: continue;
7522: }
7523:
7524: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7525: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7526: 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 7527: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7528: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7529:
7530: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7531: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7532: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7533: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7534: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7535: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7536: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7537: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7538: if(i==istart){
1.227 brouard 7539: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7540: }else{
7541: fprintf(ficgp,",\\\n '' ");
7542: }
7543: if(cptcoveff ==0){ /* No covariate */
7544: ioffset=2; /* Age is in 2 */
7545: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7546: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7547: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7548: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7549: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7550: if(i==nlstate+1){
1.270 brouard 7551: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7552: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7553: fprintf(ficgp,",\\\n '' ");
7554: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7555: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7556: offyear, \
1.268 brouard 7557: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7558: }else
1.227 brouard 7559: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7560: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7561: }else{ /* more than 2 covariates */
1.270 brouard 7562: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7563: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7564: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7565: iyearc=ioffset-1;
7566: iagec=ioffset;
1.227 brouard 7567: fprintf(ficgp," u %d:(",ioffset);
7568: kl=0;
7569: strcpy(gplotcondition,"(");
7570: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7571: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7572: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7573: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7574: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7575: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7576: kl++;
7577: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7578: kl++;
7579: if(k <cptcoveff && cptcoveff>1)
7580: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7581: }
7582: strcpy(gplotcondition+strlen(gplotcondition),")");
7583: /* 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 *\/ */
7584: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7585: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7586: /* '' 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*/
7587: if(i==nlstate+1){
1.270 brouard 7588: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7589: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7590: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7591: fprintf(ficgp," u %d:(",iagec);
7592: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7593: iyearc, iagec, offyear, \
7594: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7595: /* '' 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 7596: }else{
7597: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7598: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7599: }
7600: } /* end if covariate */
7601: } /* nlstate */
1.264 brouard 7602: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7603: } /* end cpt state*/
7604: } /* end covariate */
7605: } /* End if prevfcast */
1.227 brouard 7606:
1.268 brouard 7607: if(backcast==1){
7608: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7609:
7610: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7611: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7612: if(m != 1 && TKresult[nres]!= k1)
7613: continue;
7614: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7615: strcpy(gplotlabel,"(");
7616: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7617: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7618: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7619: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7620: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7621: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7622: vlv= nbcode[Tvaraff[k]][lv];
7623: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7624: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7625: }
7626: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7627: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7628: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7629: }
7630: strcpy(gplotlabel+strlen(gplotlabel),")");
7631: fprintf(ficgp,"\n#\n");
7632: if(invalidvarcomb[k1]){
7633: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7634: continue;
7635: }
7636:
7637: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7638: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7639: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7640: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7641: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7642:
7643: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7644: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7645: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7646: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7647: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7648: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7649: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7650: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7651: if(i==istart){
7652: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7653: }else{
7654: fprintf(ficgp,",\\\n '' ");
7655: }
7656: if(cptcoveff ==0){ /* No covariate */
7657: ioffset=2; /* Age is in 2 */
7658: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7659: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7660: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7661: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7662: fprintf(ficgp," u %d:(", ioffset);
7663: if(i==nlstate+1){
1.270 brouard 7664: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7665: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7666: fprintf(ficgp,",\\\n '' ");
7667: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7668: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7669: offbyear, \
7670: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7671: }else
7672: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7673: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7674: }else{ /* more than 2 covariates */
1.270 brouard 7675: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7676: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7677: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7678: iyearc=ioffset-1;
7679: iagec=ioffset;
1.268 brouard 7680: fprintf(ficgp," u %d:(",ioffset);
7681: kl=0;
7682: strcpy(gplotcondition,"(");
7683: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7684: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7685: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7686: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7687: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7688: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7689: kl++;
7690: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7691: kl++;
7692: if(k <cptcoveff && cptcoveff>1)
7693: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7694: }
7695: strcpy(gplotcondition+strlen(gplotcondition),")");
7696: /* 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 *\/ */
7697: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7698: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7699: /* '' 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*/
7700: if(i==nlstate+1){
1.270 brouard 7701: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7702: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7703: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7704: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7705: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7706: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7707: iyearc,iagec,offbyear, \
7708: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7709: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7710: }else{
7711: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7712: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7713: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7714: }
7715: } /* end if covariate */
7716: } /* nlstate */
7717: fprintf(ficgp,"\nset out; unset label;\n");
7718: } /* end cpt state*/
7719: } /* end covariate */
7720: } /* End if backcast */
7721:
1.227 brouard 7722:
1.238 brouard 7723: /* 9eme writing MLE parameters */
7724: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7725: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7726: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7727: for(k=1; k <=(nlstate+ndeath); k++){
7728: if (k != i) {
1.227 brouard 7729: fprintf(ficgp,"# current state %d\n",k);
7730: for(j=1; j <=ncovmodel; j++){
7731: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7732: jk++;
7733: }
7734: fprintf(ficgp,"\n");
1.126 brouard 7735: }
7736: }
1.223 brouard 7737: }
1.187 brouard 7738: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7739:
1.145 brouard 7740: /*goto avoid;*/
1.238 brouard 7741: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7742: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7743: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7744: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7745: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7746: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7747: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7748: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7749: fprintf(ficgp,"# p11=1/(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,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7752: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7753: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7754: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7755: fprintf(ficgp,"#\n");
1.223 brouard 7756: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7757: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7758: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7759: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7760: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7761: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7762: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7763: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7764: continue;
1.264 brouard 7765: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7766: strcpy(gplotlabel,"(");
1.276 brouard 7767: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7768: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7769: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7770: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7771: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7772: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7773: vlv= nbcode[Tvaraff[k]][lv];
7774: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7775: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7776: }
1.237 brouard 7777: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7778: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7779: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7780: }
1.264 brouard 7781: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7782: fprintf(ficgp,"\n#\n");
1.264 brouard 7783: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7784: fprintf(ficgp,"\nset key outside ");
7785: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7786: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7787: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7788: if (ng==1){
7789: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7790: fprintf(ficgp,"\nunset log y");
7791: }else if (ng==2){
7792: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7793: fprintf(ficgp,"\nset log y");
7794: }else if (ng==3){
7795: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7796: fprintf(ficgp,"\nset log y");
7797: }else
7798: fprintf(ficgp,"\nunset title ");
7799: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7800: i=1;
7801: for(k2=1; k2<=nlstate; k2++) {
7802: k3=i;
7803: for(k=1; k<=(nlstate+ndeath); k++) {
7804: if (k != k2){
7805: switch( ng) {
7806: case 1:
7807: if(nagesqr==0)
7808: fprintf(ficgp," p%d+p%d*x",i,i+1);
7809: else /* nagesqr =1 */
7810: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7811: break;
7812: case 2: /* ng=2 */
7813: if(nagesqr==0)
7814: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7815: else /* nagesqr =1 */
7816: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7817: break;
7818: case 3:
7819: if(nagesqr==0)
7820: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7821: else /* nagesqr =1 */
7822: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7823: break;
7824: }
7825: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7826: ijp=1; /* product no age */
7827: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7828: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7829: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7830: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7831: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7832: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7833: if(DummyV[j]==0){
7834: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7835: }else{ /* quantitative */
7836: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7837: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7838: }
7839: ij++;
1.237 brouard 7840: }
1.268 brouard 7841: }
7842: }else if(cptcovprod >0){
7843: if(j==Tprod[ijp]) { /* */
7844: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7845: if(ijp <=cptcovprod) { /* Product */
7846: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7847: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7848: /* 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)]); */
7849: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7850: }else{ /* Vn is dummy and Vm is quanti */
7851: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7852: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7853: }
7854: }else{ /* Vn*Vm Vn is quanti */
7855: if(DummyV[Tvard[ijp][2]]==0){
7856: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7857: }else{ /* Both quanti */
7858: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7859: }
1.237 brouard 7860: }
1.268 brouard 7861: ijp++;
1.237 brouard 7862: }
1.268 brouard 7863: } /* end Tprod */
1.237 brouard 7864: } else{ /* simple covariate */
1.264 brouard 7865: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7866: if(Dummy[j]==0){
7867: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7868: }else{ /* quantitative */
7869: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7870: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7871: }
1.237 brouard 7872: } /* end simple */
7873: } /* end j */
1.223 brouard 7874: }else{
7875: i=i-ncovmodel;
7876: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7877: fprintf(ficgp," (1.");
7878: }
1.227 brouard 7879:
1.223 brouard 7880: if(ng != 1){
7881: fprintf(ficgp,")/(1");
1.227 brouard 7882:
1.264 brouard 7883: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7884: if(nagesqr==0)
1.264 brouard 7885: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7886: else /* nagesqr =1 */
1.264 brouard 7887: 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 7888:
1.223 brouard 7889: ij=1;
7890: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7891: if(cptcovage >0){
7892: if((j-2)==Tage[ij]) { /* Bug valgrind */
7893: if(ij <=cptcovage) { /* Bug valgrind */
7894: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7895: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7896: ij++;
7897: }
7898: }
7899: }else
7900: 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 7901: }
7902: fprintf(ficgp,")");
7903: }
7904: fprintf(ficgp,")");
7905: if(ng ==2)
1.276 brouard 7906: 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 7907: else /* ng= 3 */
1.276 brouard 7908: 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 7909: }else{ /* end ng <> 1 */
7910: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 7911: 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 7912: }
7913: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7914: fprintf(ficgp,",");
7915: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7916: fprintf(ficgp,",");
7917: i=i+ncovmodel;
7918: } /* end k */
7919: } /* end k2 */
1.276 brouard 7920: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
7921: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 7922: } /* end k1 */
1.223 brouard 7923: } /* end ng */
7924: /* avoid: */
7925: fflush(ficgp);
1.126 brouard 7926: } /* end gnuplot */
7927:
7928:
7929: /*************** Moving average **************/
1.219 brouard 7930: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7931: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7932:
1.222 brouard 7933: int i, cpt, cptcod;
7934: int modcovmax =1;
7935: int mobilavrange, mob;
7936: int iage=0;
7937:
1.266 brouard 7938: double sum=0., sumr=0.;
1.222 brouard 7939: double age;
1.266 brouard 7940: double *sumnewp, *sumnewm, *sumnewmr;
7941: double *agemingood, *agemaxgood;
7942: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7943:
7944:
1.225 brouard 7945: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7946: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7947:
7948: sumnewp = vector(1,ncovcombmax);
7949: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7950: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7951: agemingood = vector(1,ncovcombmax);
1.266 brouard 7952: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7953: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7954: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7955:
7956: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7957: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7958: sumnewp[cptcod]=0.;
1.266 brouard 7959: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
7960: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7961: }
7962: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7963:
1.266 brouard 7964: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7965: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7966: else mobilavrange=mobilav;
7967: for (age=bage; age<=fage; age++)
7968: for (i=1; i<=nlstate;i++)
7969: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7970: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7971: /* We keep the original values on the extreme ages bage, fage and for
7972: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7973: we use a 5 terms etc. until the borders are no more concerned.
7974: */
7975: for (mob=3;mob <=mobilavrange;mob=mob+2){
7976: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 7977: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7978: sumnewm[cptcod]=0.;
7979: for (i=1; i<=nlstate;i++){
1.222 brouard 7980: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7981: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7982: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7983: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7984: }
7985: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 7986: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7987: } /* end i */
7988: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
7989: } /* end cptcod */
1.222 brouard 7990: }/* end age */
7991: }/* end mob */
1.266 brouard 7992: }else{
7993: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 7994: return -1;
1.266 brouard 7995: }
7996:
7997: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 7998: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7999: if(invalidvarcomb[cptcod]){
8000: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8001: continue;
8002: }
1.219 brouard 8003:
1.266 brouard 8004: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8005: sumnewm[cptcod]=0.;
8006: sumnewmr[cptcod]=0.;
8007: for (i=1; i<=nlstate;i++){
8008: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8009: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8010: }
8011: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8012: agemingoodr[cptcod]=age;
8013: }
8014: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8015: agemingood[cptcod]=age;
8016: }
8017: } /* age */
8018: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8019: sumnewm[cptcod]=0.;
1.266 brouard 8020: sumnewmr[cptcod]=0.;
1.222 brouard 8021: for (i=1; i<=nlstate;i++){
8022: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8023: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8024: }
8025: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8026: agemaxgoodr[cptcod]=age;
1.222 brouard 8027: }
8028: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8029: agemaxgood[cptcod]=age;
8030: }
8031: } /* age */
8032: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8033: /* but they will change */
8034: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8035: sumnewm[cptcod]=0.;
8036: sumnewmr[cptcod]=0.;
8037: for (i=1; i<=nlstate;i++){
8038: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8039: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8040: }
8041: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8042: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8043: agemaxgoodr[cptcod]=age; /* age min */
8044: for (i=1; i<=nlstate;i++)
8045: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8046: }else{ /* bad we change the value with the values of good ages */
8047: for (i=1; i<=nlstate;i++){
8048: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8049: } /* i */
8050: } /* end bad */
8051: }else{
8052: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8053: agemaxgood[cptcod]=age;
8054: }else{ /* bad we change the value with the values of good ages */
8055: for (i=1; i<=nlstate;i++){
8056: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8057: } /* i */
8058: } /* end bad */
8059: }/* end else */
8060: sum=0.;sumr=0.;
8061: for (i=1; i<=nlstate;i++){
8062: sum+=mobaverage[(int)age][i][cptcod];
8063: sumr+=probs[(int)age][i][cptcod];
8064: }
8065: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8066: 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 8067: } /* end bad */
8068: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8069: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8070: 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 8071: } /* end bad */
8072: }/* age */
1.266 brouard 8073:
8074: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8075: sumnewm[cptcod]=0.;
1.266 brouard 8076: sumnewmr[cptcod]=0.;
1.222 brouard 8077: for (i=1; i<=nlstate;i++){
8078: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8079: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8080: }
8081: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8082: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8083: agemingoodr[cptcod]=age;
8084: for (i=1; i<=nlstate;i++)
8085: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8086: }else{ /* bad we change the value with the values of good ages */
8087: for (i=1; i<=nlstate;i++){
8088: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8089: } /* i */
8090: } /* end bad */
8091: }else{
8092: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8093: agemingood[cptcod]=age;
8094: }else{ /* bad */
8095: for (i=1; i<=nlstate;i++){
8096: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8097: } /* i */
8098: } /* end bad */
8099: }/* end else */
8100: sum=0.;sumr=0.;
8101: for (i=1; i<=nlstate;i++){
8102: sum+=mobaverage[(int)age][i][cptcod];
8103: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8104: }
1.266 brouard 8105: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8106: 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 8107: } /* end bad */
8108: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8109: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8110: printf("Moving average 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 8111: } /* end bad */
8112: }/* age */
1.266 brouard 8113:
1.222 brouard 8114:
8115: for (age=bage; age<=fage; age++){
1.235 brouard 8116: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8117: sumnewp[cptcod]=0.;
8118: sumnewm[cptcod]=0.;
8119: for (i=1; i<=nlstate;i++){
8120: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8121: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8122: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8123: }
8124: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8125: }
8126: /* printf("\n"); */
8127: /* } */
1.266 brouard 8128:
1.222 brouard 8129: /* brutal averaging */
1.266 brouard 8130: /* for (i=1; i<=nlstate;i++){ */
8131: /* for (age=1; age<=bage; age++){ */
8132: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8133: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8134: /* } */
8135: /* for (age=fage; age<=AGESUP; age++){ */
8136: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8137: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8138: /* } */
8139: /* } /\* end i status *\/ */
8140: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8141: /* for (age=1; age<=AGESUP; age++){ */
8142: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8143: /* mobaverage[(int)age][i][cptcod]=0.; */
8144: /* } */
8145: /* } */
1.222 brouard 8146: }/* end cptcod */
1.266 brouard 8147: free_vector(agemaxgoodr,1, ncovcombmax);
8148: free_vector(agemaxgood,1, ncovcombmax);
8149: free_vector(agemingood,1, ncovcombmax);
8150: free_vector(agemingoodr,1, ncovcombmax);
8151: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8152: free_vector(sumnewm,1, ncovcombmax);
8153: free_vector(sumnewp,1, ncovcombmax);
8154: return 0;
8155: }/* End movingaverage */
1.218 brouard 8156:
1.126 brouard 8157:
8158: /************** Forecasting ******************/
1.269 brouard 8159: 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 8160: /* proj1, year, month, day of starting projection
8161: agemin, agemax range of age
8162: dateprev1 dateprev2 range of dates during which prevalence is computed
8163: anproj2 year of en of projection (same day and month as proj1).
8164: */
1.267 brouard 8165: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8166: double agec; /* generic age */
8167: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8168: double *popeffectif,*popcount;
8169: double ***p3mat;
1.218 brouard 8170: /* double ***mobaverage; */
1.126 brouard 8171: char fileresf[FILENAMELENGTH];
8172:
8173: agelim=AGESUP;
1.211 brouard 8174: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8175: in each health status at the date of interview (if between dateprev1 and dateprev2).
8176: We still use firstpass and lastpass as another selection.
8177: */
1.214 brouard 8178: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8179: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8180:
1.201 brouard 8181: strcpy(fileresf,"F_");
8182: strcat(fileresf,fileresu);
1.126 brouard 8183: if((ficresf=fopen(fileresf,"w"))==NULL) {
8184: printf("Problem with forecast resultfile: %s\n", fileresf);
8185: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8186: }
1.235 brouard 8187: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8188: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8189:
1.225 brouard 8190: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8191:
8192:
8193: stepsize=(int) (stepm+YEARM-1)/YEARM;
8194: if (stepm<=12) stepsize=1;
8195: if(estepm < stepm){
8196: printf ("Problem %d lower than %d\n",estepm, stepm);
8197: }
1.270 brouard 8198: else{
8199: hstepm=estepm;
8200: }
8201: if(estepm > stepm){ /* Yes every two year */
8202: stepsize=2;
8203: }
1.126 brouard 8204:
8205: hstepm=hstepm/stepm;
8206: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8207: fractional in yp1 */
8208: anprojmean=yp;
8209: yp2=modf((yp1*12),&yp);
8210: mprojmean=yp;
8211: yp1=modf((yp2*30.5),&yp);
8212: jprojmean=yp;
8213: if(jprojmean==0) jprojmean=1;
8214: if(mprojmean==0) jprojmean=1;
8215:
1.227 brouard 8216: i1=pow(2,cptcoveff);
1.126 brouard 8217: if (cptcovn < 1){i1=1;}
8218:
8219: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8220:
8221: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8222:
1.126 brouard 8223: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8224: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8225: for(k=1; k<=i1;k++){
1.253 brouard 8226: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8227: continue;
1.227 brouard 8228: if(invalidvarcomb[k]){
8229: printf("\nCombination (%d) projection ignored because no cases \n",k);
8230: continue;
8231: }
8232: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8233: for(j=1;j<=cptcoveff;j++) {
8234: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8235: }
1.235 brouard 8236: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8237: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8238: }
1.227 brouard 8239: fprintf(ficresf," yearproj age");
8240: for(j=1; j<=nlstate+ndeath;j++){
8241: for(i=1; i<=nlstate;i++)
8242: fprintf(ficresf," p%d%d",i,j);
8243: fprintf(ficresf," wp.%d",j);
8244: }
8245: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8246: fprintf(ficresf,"\n");
8247: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8248: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8249: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8250: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8251: nhstepm = nhstepm/hstepm;
8252: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8253: oldm=oldms;savm=savms;
1.268 brouard 8254: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8255: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8256: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8257: for (h=0; h<=nhstepm; h++){
8258: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8259: break;
8260: }
8261: }
8262: fprintf(ficresf,"\n");
8263: for(j=1;j<=cptcoveff;j++)
8264: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8265: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8266:
8267: for(j=1; j<=nlstate+ndeath;j++) {
8268: ppij=0.;
8269: for(i=1; i<=nlstate;i++) {
8270: /* if (mobilav>=1) */
1.269 brouard 8271: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
1.268 brouard 8272: /* else { */ /* even if mobilav==-1 we use mobaverage */
8273: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8274: /* } */
8275: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8276: } /* end i */
8277: fprintf(ficresf," %.3f", ppij);
8278: }/* end j */
1.227 brouard 8279: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8280: } /* end agec */
1.266 brouard 8281: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8282: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8283: } /* end yearp */
8284: } /* end k */
1.219 brouard 8285:
1.126 brouard 8286: fclose(ficresf);
1.215 brouard 8287: printf("End of Computing forecasting \n");
8288: fprintf(ficlog,"End of Computing forecasting\n");
8289:
1.126 brouard 8290: }
8291:
1.269 brouard 8292: /************** Back Forecasting ******************/
8293: 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 8294: /* back1, year, month, day of starting backection
8295: agemin, agemax range of age
8296: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8297: anback2 year of end of backprojection (same day and month as back1).
8298: prevacurrent and prev are prevalences.
1.267 brouard 8299: */
8300: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8301: double agec; /* generic age */
1.268 brouard 8302: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8303: double *popeffectif,*popcount;
8304: double ***p3mat;
8305: /* double ***mobaverage; */
8306: char fileresfb[FILENAMELENGTH];
8307:
1.268 brouard 8308: agelim=AGEINF;
1.267 brouard 8309: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8310: in each health status at the date of interview (if between dateprev1 and dateprev2).
8311: We still use firstpass and lastpass as another selection.
8312: */
8313: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8314: /* firstpass, lastpass, stepm, weightopt, model); */
8315:
8316: /*Do we need to compute prevalence again?*/
8317:
8318: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8319:
8320: strcpy(fileresfb,"FB_");
8321: strcat(fileresfb,fileresu);
8322: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8323: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8324: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8325: }
8326: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8327: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8328:
8329: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8330:
8331:
8332: stepsize=(int) (stepm+YEARM-1)/YEARM;
8333: if (stepm<=12) stepsize=1;
8334: if(estepm < stepm){
8335: printf ("Problem %d lower than %d\n",estepm, stepm);
8336: }
1.270 brouard 8337: else{
8338: hstepm=estepm;
8339: }
8340: if(estepm >= stepm){ /* Yes every two year */
8341: stepsize=2;
8342: }
1.267 brouard 8343:
8344: hstepm=hstepm/stepm;
8345: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8346: fractional in yp1 */
8347: anprojmean=yp;
8348: yp2=modf((yp1*12),&yp);
8349: mprojmean=yp;
8350: yp1=modf((yp2*30.5),&yp);
8351: jprojmean=yp;
8352: if(jprojmean==0) jprojmean=1;
8353: if(mprojmean==0) jprojmean=1;
8354:
8355: i1=pow(2,cptcoveff);
8356: if (cptcovn < 1){i1=1;}
8357:
8358: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8359: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8360:
8361: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8362:
8363: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8364: for(k=1; k<=i1;k++){
8365: if(i1 != 1 && TKresult[nres]!= k)
8366: continue;
8367: if(invalidvarcomb[k]){
8368: printf("\nCombination (%d) projection ignored because no cases \n",k);
8369: continue;
8370: }
1.268 brouard 8371: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8372: for(j=1;j<=cptcoveff;j++) {
8373: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8374: }
8375: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8376: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8377: }
8378: fprintf(ficresfb," yearbproj age");
8379: for(j=1; j<=nlstate+ndeath;j++){
8380: for(i=1; i<=nlstate;i++)
1.268 brouard 8381: fprintf(ficresfb," b%d%d",i,j);
8382: fprintf(ficresfb," b.%d",j);
1.267 brouard 8383: }
8384: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8385: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8386: fprintf(ficresfb,"\n");
8387: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8388: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8389: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8390: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8391: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8392: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8393: nhstepm = nhstepm/hstepm;
8394: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8395: oldm=oldms;savm=savms;
1.268 brouard 8396: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8397: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8398: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8399: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8400: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8401: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8402: for (h=0; h<=nhstepm; h++){
1.268 brouard 8403: if (h*hstepm/YEARM*stepm ==-yearp) {
8404: break;
8405: }
8406: }
8407: fprintf(ficresfb,"\n");
8408: for(j=1;j<=cptcoveff;j++)
8409: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8410: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8411: for(i=1; i<=nlstate+ndeath;i++) {
8412: ppij=0.;ppi=0.;
8413: for(j=1; j<=nlstate;j++) {
8414: /* if (mobilav==1) */
1.269 brouard 8415: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8416: ppi=ppi+prevacurrent[(int)agec][j][k];
8417: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8418: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8419: /* else { */
8420: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8421: /* } */
1.268 brouard 8422: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8423: } /* end j */
8424: if(ppi <0.99){
8425: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8426: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8427: }
8428: fprintf(ficresfb," %.3f", ppij);
8429: }/* end j */
1.267 brouard 8430: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8431: } /* end agec */
8432: } /* end yearp */
8433: } /* end k */
1.217 brouard 8434:
1.267 brouard 8435: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8436:
1.267 brouard 8437: fclose(ficresfb);
8438: printf("End of Computing Back forecasting \n");
8439: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8440:
1.267 brouard 8441: }
1.217 brouard 8442:
1.269 brouard 8443: /* Variance of prevalence limit: varprlim */
8444: 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){
8445: /*------- Variance of period (stable) prevalence------*/
8446:
8447: char fileresvpl[FILENAMELENGTH];
8448: FILE *ficresvpl;
8449: double **oldm, **savm;
8450: double **varpl; /* Variances of prevalence limits by age */
8451: int i1, k, nres, j ;
8452:
8453: strcpy(fileresvpl,"VPL_");
8454: strcat(fileresvpl,fileresu);
8455: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8456: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8457: exit(0);
8458: }
8459: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8460: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8461:
8462: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8463: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8464:
8465: i1=pow(2,cptcoveff);
8466: if (cptcovn < 1){i1=1;}
8467:
8468: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8469: for(k=1; k<=i1;k++){
8470: if(i1 != 1 && TKresult[nres]!= k)
8471: continue;
8472: fprintf(ficresvpl,"\n#****** ");
8473: printf("\n#****** ");
8474: fprintf(ficlog,"\n#****** ");
8475: for(j=1;j<=cptcoveff;j++) {
8476: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8477: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8478: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8479: }
8480: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8481: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8482: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8483: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8484: }
8485: fprintf(ficresvpl,"******\n");
8486: printf("******\n");
8487: fprintf(ficlog,"******\n");
8488:
8489: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8490: oldm=oldms;savm=savms;
8491: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8492: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8493: /*}*/
8494: }
8495:
8496: fclose(ficresvpl);
8497: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8498: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8499:
8500: }
8501: /* Variance of back prevalence: varbprlim */
8502: 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){
8503: /*------- Variance of back (stable) prevalence------*/
8504:
8505: char fileresvbl[FILENAMELENGTH];
8506: FILE *ficresvbl;
8507:
8508: double **oldm, **savm;
8509: double **varbpl; /* Variances of back prevalence limits by age */
8510: int i1, k, nres, j ;
8511:
8512: strcpy(fileresvbl,"VBL_");
8513: strcat(fileresvbl,fileresu);
8514: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8515: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8516: exit(0);
8517: }
8518: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8519: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8520:
8521:
8522: i1=pow(2,cptcoveff);
8523: if (cptcovn < 1){i1=1;}
8524:
8525: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8526: for(k=1; k<=i1;k++){
8527: if(i1 != 1 && TKresult[nres]!= k)
8528: continue;
8529: fprintf(ficresvbl,"\n#****** ");
8530: printf("\n#****** ");
8531: fprintf(ficlog,"\n#****** ");
8532: for(j=1;j<=cptcoveff;j++) {
8533: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8534: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8535: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8536: }
8537: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8538: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8539: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8540: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8541: }
8542: fprintf(ficresvbl,"******\n");
8543: printf("******\n");
8544: fprintf(ficlog,"******\n");
8545:
8546: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8547: oldm=oldms;savm=savms;
8548:
8549: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8550: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8551: /*}*/
8552: }
8553:
8554: fclose(ficresvbl);
8555: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8556: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8557:
8558: } /* End of varbprlim */
8559:
1.126 brouard 8560: /************** Forecasting *****not tested NB*************/
1.227 brouard 8561: /* 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 8562:
1.227 brouard 8563: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8564: /* int *popage; */
8565: /* double calagedatem, agelim, kk1, kk2; */
8566: /* double *popeffectif,*popcount; */
8567: /* double ***p3mat,***tabpop,***tabpopprev; */
8568: /* /\* double ***mobaverage; *\/ */
8569: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8570:
1.227 brouard 8571: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8572: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8573: /* agelim=AGESUP; */
8574: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8575:
1.227 brouard 8576: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8577:
8578:
1.227 brouard 8579: /* strcpy(filerespop,"POP_"); */
8580: /* strcat(filerespop,fileresu); */
8581: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8582: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8583: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8584: /* } */
8585: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8586: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8587:
1.227 brouard 8588: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8589:
1.227 brouard 8590: /* /\* if (mobilav!=0) { *\/ */
8591: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8592: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8593: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8594: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8595: /* /\* } *\/ */
8596: /* /\* } *\/ */
1.126 brouard 8597:
1.227 brouard 8598: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8599: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8600:
1.227 brouard 8601: /* agelim=AGESUP; */
1.126 brouard 8602:
1.227 brouard 8603: /* hstepm=1; */
8604: /* hstepm=hstepm/stepm; */
1.218 brouard 8605:
1.227 brouard 8606: /* if (popforecast==1) { */
8607: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8608: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8609: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8610: /* } */
8611: /* popage=ivector(0,AGESUP); */
8612: /* popeffectif=vector(0,AGESUP); */
8613: /* popcount=vector(0,AGESUP); */
1.126 brouard 8614:
1.227 brouard 8615: /* i=1; */
8616: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8617:
1.227 brouard 8618: /* imx=i; */
8619: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8620: /* } */
1.218 brouard 8621:
1.227 brouard 8622: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8623: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8624: /* k=k+1; */
8625: /* fprintf(ficrespop,"\n#******"); */
8626: /* for(j=1;j<=cptcoveff;j++) { */
8627: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8628: /* } */
8629: /* fprintf(ficrespop,"******\n"); */
8630: /* fprintf(ficrespop,"# Age"); */
8631: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8632: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8633:
1.227 brouard 8634: /* for (cpt=0; cpt<=0;cpt++) { */
8635: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8636:
1.227 brouard 8637: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8638: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8639: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8640:
1.227 brouard 8641: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8642: /* oldm=oldms;savm=savms; */
8643: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8644:
1.227 brouard 8645: /* for (h=0; h<=nhstepm; h++){ */
8646: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8647: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8648: /* } */
8649: /* for(j=1; j<=nlstate+ndeath;j++) { */
8650: /* kk1=0.;kk2=0; */
8651: /* for(i=1; i<=nlstate;i++) { */
8652: /* if (mobilav==1) */
8653: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8654: /* else { */
8655: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8656: /* } */
8657: /* } */
8658: /* if (h==(int)(calagedatem+12*cpt)){ */
8659: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8660: /* /\*fprintf(ficrespop," %.3f", kk1); */
8661: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8662: /* } */
8663: /* } */
8664: /* for(i=1; i<=nlstate;i++){ */
8665: /* kk1=0.; */
8666: /* for(j=1; j<=nlstate;j++){ */
8667: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8668: /* } */
8669: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8670: /* } */
1.218 brouard 8671:
1.227 brouard 8672: /* if (h==(int)(calagedatem+12*cpt)) */
8673: /* for(j=1; j<=nlstate;j++) */
8674: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8675: /* } */
8676: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8677: /* } */
8678: /* } */
1.218 brouard 8679:
1.227 brouard 8680: /* /\******\/ */
1.218 brouard 8681:
1.227 brouard 8682: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8683: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8684: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8685: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8686: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8687:
1.227 brouard 8688: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8689: /* oldm=oldms;savm=savms; */
8690: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8691: /* for (h=0; h<=nhstepm; h++){ */
8692: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8693: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8694: /* } */
8695: /* for(j=1; j<=nlstate+ndeath;j++) { */
8696: /* kk1=0.;kk2=0; */
8697: /* for(i=1; i<=nlstate;i++) { */
8698: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8699: /* } */
8700: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8701: /* } */
8702: /* } */
8703: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8704: /* } */
8705: /* } */
8706: /* } */
8707: /* } */
1.218 brouard 8708:
1.227 brouard 8709: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8710:
1.227 brouard 8711: /* if (popforecast==1) { */
8712: /* free_ivector(popage,0,AGESUP); */
8713: /* free_vector(popeffectif,0,AGESUP); */
8714: /* free_vector(popcount,0,AGESUP); */
8715: /* } */
8716: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8717: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8718: /* fclose(ficrespop); */
8719: /* } /\* End of popforecast *\/ */
1.218 brouard 8720:
1.126 brouard 8721: int fileappend(FILE *fichier, char *optionfich)
8722: {
8723: if((fichier=fopen(optionfich,"a"))==NULL) {
8724: printf("Problem with file: %s\n", optionfich);
8725: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8726: return (0);
8727: }
8728: fflush(fichier);
8729: return (1);
8730: }
8731:
8732:
8733: /**************** function prwizard **********************/
8734: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8735: {
8736:
8737: /* Wizard to print covariance matrix template */
8738:
1.164 brouard 8739: char ca[32], cb[32];
8740: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8741: int numlinepar;
8742:
8743: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8744: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8745: for(i=1; i <=nlstate; i++){
8746: jj=0;
8747: for(j=1; j <=nlstate+ndeath; j++){
8748: if(j==i) continue;
8749: jj++;
8750: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8751: printf("%1d%1d",i,j);
8752: fprintf(ficparo,"%1d%1d",i,j);
8753: for(k=1; k<=ncovmodel;k++){
8754: /* printf(" %lf",param[i][j][k]); */
8755: /* fprintf(ficparo," %lf",param[i][j][k]); */
8756: printf(" 0.");
8757: fprintf(ficparo," 0.");
8758: }
8759: printf("\n");
8760: fprintf(ficparo,"\n");
8761: }
8762: }
8763: printf("# Scales (for hessian or gradient estimation)\n");
8764: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8765: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8766: for(i=1; i <=nlstate; i++){
8767: jj=0;
8768: for(j=1; j <=nlstate+ndeath; j++){
8769: if(j==i) continue;
8770: jj++;
8771: fprintf(ficparo,"%1d%1d",i,j);
8772: printf("%1d%1d",i,j);
8773: fflush(stdout);
8774: for(k=1; k<=ncovmodel;k++){
8775: /* printf(" %le",delti3[i][j][k]); */
8776: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8777: printf(" 0.");
8778: fprintf(ficparo," 0.");
8779: }
8780: numlinepar++;
8781: printf("\n");
8782: fprintf(ficparo,"\n");
8783: }
8784: }
8785: printf("# Covariance matrix\n");
8786: /* # 121 Var(a12)\n\ */
8787: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8788: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8789: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8790: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8791: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8792: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8793: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8794: fflush(stdout);
8795: fprintf(ficparo,"# Covariance matrix\n");
8796: /* # 121 Var(a12)\n\ */
8797: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8798: /* # ...\n\ */
8799: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8800:
8801: for(itimes=1;itimes<=2;itimes++){
8802: jj=0;
8803: for(i=1; i <=nlstate; i++){
8804: for(j=1; j <=nlstate+ndeath; j++){
8805: if(j==i) continue;
8806: for(k=1; k<=ncovmodel;k++){
8807: jj++;
8808: ca[0]= k+'a'-1;ca[1]='\0';
8809: if(itimes==1){
8810: printf("#%1d%1d%d",i,j,k);
8811: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8812: }else{
8813: printf("%1d%1d%d",i,j,k);
8814: fprintf(ficparo,"%1d%1d%d",i,j,k);
8815: /* printf(" %.5le",matcov[i][j]); */
8816: }
8817: ll=0;
8818: for(li=1;li <=nlstate; li++){
8819: for(lj=1;lj <=nlstate+ndeath; lj++){
8820: if(lj==li) continue;
8821: for(lk=1;lk<=ncovmodel;lk++){
8822: ll++;
8823: if(ll<=jj){
8824: cb[0]= lk +'a'-1;cb[1]='\0';
8825: if(ll<jj){
8826: if(itimes==1){
8827: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8828: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8829: }else{
8830: printf(" 0.");
8831: fprintf(ficparo," 0.");
8832: }
8833: }else{
8834: if(itimes==1){
8835: printf(" Var(%s%1d%1d)",ca,i,j);
8836: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8837: }else{
8838: printf(" 0.");
8839: fprintf(ficparo," 0.");
8840: }
8841: }
8842: }
8843: } /* end lk */
8844: } /* end lj */
8845: } /* end li */
8846: printf("\n");
8847: fprintf(ficparo,"\n");
8848: numlinepar++;
8849: } /* end k*/
8850: } /*end j */
8851: } /* end i */
8852: } /* end itimes */
8853:
8854: } /* end of prwizard */
8855: /******************* Gompertz Likelihood ******************************/
8856: double gompertz(double x[])
8857: {
8858: double A,B,L=0.0,sump=0.,num=0.;
8859: int i,n=0; /* n is the size of the sample */
8860:
1.220 brouard 8861: for (i=1;i<=imx ; i++) {
1.126 brouard 8862: sump=sump+weight[i];
8863: /* sump=sump+1;*/
8864: num=num+1;
8865: }
8866:
8867:
8868: /* for (i=0; i<=imx; i++)
8869: 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]);*/
8870:
8871: for (i=1;i<=imx ; i++)
8872: {
8873: if (cens[i] == 1 && wav[i]>1)
8874: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8875:
8876: if (cens[i] == 0 && wav[i]>1)
8877: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8878: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8879:
8880: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8881: if (wav[i] > 1 ) { /* ??? */
8882: L=L+A*weight[i];
8883: /* 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]);*/
8884: }
8885: }
8886:
8887: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8888:
8889: return -2*L*num/sump;
8890: }
8891:
1.136 brouard 8892: #ifdef GSL
8893: /******************* Gompertz_f Likelihood ******************************/
8894: double gompertz_f(const gsl_vector *v, void *params)
8895: {
8896: double A,B,LL=0.0,sump=0.,num=0.;
8897: double *x= (double *) v->data;
8898: int i,n=0; /* n is the size of the sample */
8899:
8900: for (i=0;i<=imx-1 ; i++) {
8901: sump=sump+weight[i];
8902: /* sump=sump+1;*/
8903: num=num+1;
8904: }
8905:
8906:
8907: /* for (i=0; i<=imx; i++)
8908: 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]);*/
8909: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8910: for (i=1;i<=imx ; i++)
8911: {
8912: if (cens[i] == 1 && wav[i]>1)
8913: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8914:
8915: if (cens[i] == 0 && wav[i]>1)
8916: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8917: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8918:
8919: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8920: if (wav[i] > 1 ) { /* ??? */
8921: LL=LL+A*weight[i];
8922: /* 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]);*/
8923: }
8924: }
8925:
8926: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8927: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8928:
8929: return -2*LL*num/sump;
8930: }
8931: #endif
8932:
1.126 brouard 8933: /******************* Printing html file ***********/
1.201 brouard 8934: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8935: int lastpass, int stepm, int weightopt, char model[],\
8936: int imx, double p[],double **matcov,double agemortsup){
8937: int i,k;
8938:
8939: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8940: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8941: for (i=1;i<=2;i++)
8942: 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 8943: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8944: fprintf(fichtm,"</ul>");
8945:
8946: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8947:
8948: 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>");
8949:
8950: for (k=agegomp;k<(agemortsup-2);k++)
8951: 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]);
8952:
8953:
8954: fflush(fichtm);
8955: }
8956:
8957: /******************* Gnuplot file **************/
1.201 brouard 8958: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8959:
8960: char dirfileres[132],optfileres[132];
1.164 brouard 8961:
1.126 brouard 8962: int ng;
8963:
8964:
8965: /*#ifdef windows */
8966: fprintf(ficgp,"cd \"%s\" \n",pathc);
8967: /*#endif */
8968:
8969:
8970: strcpy(dirfileres,optionfilefiname);
8971: strcpy(optfileres,"vpl");
1.199 brouard 8972: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8973: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8974: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8975: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8976: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8977:
8978: }
8979:
1.136 brouard 8980: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8981: {
1.126 brouard 8982:
1.136 brouard 8983: /*-------- data file ----------*/
8984: FILE *fic;
8985: char dummy[]=" ";
1.240 brouard 8986: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8987: int lstra;
1.136 brouard 8988: int linei, month, year,iout;
8989: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8990: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8991: char *stratrunc;
1.223 brouard 8992:
1.240 brouard 8993: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8994: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8995:
1.240 brouard 8996: for(v=1; v <=ncovcol;v++){
8997: DummyV[v]=0;
8998: FixedV[v]=0;
8999: }
9000: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9001: DummyV[v]=1;
9002: FixedV[v]=0;
9003: }
9004: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9005: DummyV[v]=0;
9006: FixedV[v]=1;
9007: }
9008: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9009: DummyV[v]=1;
9010: FixedV[v]=1;
9011: }
9012: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9013: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9014: 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]);
9015: }
1.126 brouard 9016:
1.136 brouard 9017: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9018: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9019: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9020: }
1.126 brouard 9021:
1.136 brouard 9022: i=1;
9023: linei=0;
9024: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9025: linei=linei+1;
9026: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9027: if(line[j] == '\t')
9028: line[j] = ' ';
9029: }
9030: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9031: ;
9032: };
9033: line[j+1]=0; /* Trims blanks at end of line */
9034: if(line[0]=='#'){
9035: fprintf(ficlog,"Comment line\n%s\n",line);
9036: printf("Comment line\n%s\n",line);
9037: continue;
9038: }
9039: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9040: strcpy(line, linetmp);
1.223 brouard 9041:
9042: /* Loops on waves */
9043: for (j=maxwav;j>=1;j--){
9044: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9045: cutv(stra, strb, line, ' ');
9046: if(strb[0]=='.') { /* Missing value */
9047: lval=-1;
9048: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9049: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9050: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9051: 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);
9052: 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);
9053: return 1;
9054: }
9055: }else{
9056: errno=0;
9057: /* what_kind_of_number(strb); */
9058: dval=strtod(strb,&endptr);
9059: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9060: /* if(strb != endptr && *endptr == '\0') */
9061: /* dval=dlval; */
9062: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9063: if( strb[0]=='\0' || (*endptr != '\0')){
9064: 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);
9065: 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);
9066: return 1;
9067: }
9068: cotqvar[j][iv][i]=dval;
9069: cotvar[j][ntv+iv][i]=dval;
9070: }
9071: strcpy(line,stra);
1.223 brouard 9072: }/* end loop ntqv */
1.225 brouard 9073:
1.223 brouard 9074: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9075: cutv(stra, strb, line, ' ');
9076: if(strb[0]=='.') { /* Missing value */
9077: lval=-1;
9078: }else{
9079: errno=0;
9080: lval=strtol(strb,&endptr,10);
9081: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9082: if( strb[0]=='\0' || (*endptr != '\0')){
9083: 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);
9084: 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);
9085: return 1;
9086: }
9087: }
9088: if(lval <-1 || lval >1){
9089: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9090: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9091: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9092: For example, for multinomial values like 1, 2 and 3,\n \
9093: build V1=0 V2=0 for the reference value (1),\n \
9094: V1=1 V2=0 for (2) \n \
1.223 brouard 9095: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9096: output of IMaCh is often meaningless.\n \
1.223 brouard 9097: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9098: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9099: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9100: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9101: For example, for multinomial values like 1, 2 and 3,\n \
9102: build V1=0 V2=0 for the reference value (1),\n \
9103: V1=1 V2=0 for (2) \n \
1.223 brouard 9104: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9105: output of IMaCh is often meaningless.\n \
1.223 brouard 9106: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9107: return 1;
9108: }
9109: cotvar[j][iv][i]=(double)(lval);
9110: strcpy(line,stra);
1.223 brouard 9111: }/* end loop ntv */
1.225 brouard 9112:
1.223 brouard 9113: /* Statuses at wave */
1.137 brouard 9114: cutv(stra, strb, line, ' ');
1.223 brouard 9115: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9116: lval=-1;
1.136 brouard 9117: }else{
1.238 brouard 9118: errno=0;
9119: lval=strtol(strb,&endptr,10);
9120: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9121: if( strb[0]=='\0' || (*endptr != '\0')){
9122: 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);
9123: 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);
9124: return 1;
9125: }
1.136 brouard 9126: }
1.225 brouard 9127:
1.136 brouard 9128: s[j][i]=lval;
1.225 brouard 9129:
1.223 brouard 9130: /* Date of Interview */
1.136 brouard 9131: strcpy(line,stra);
9132: cutv(stra, strb,line,' ');
1.169 brouard 9133: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9134: }
1.169 brouard 9135: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9136: month=99;
9137: year=9999;
1.136 brouard 9138: }else{
1.225 brouard 9139: 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);
9140: 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);
9141: return 1;
1.136 brouard 9142: }
9143: anint[j][i]= (double) year;
9144: mint[j][i]= (double)month;
9145: strcpy(line,stra);
1.223 brouard 9146: } /* End loop on waves */
1.225 brouard 9147:
1.223 brouard 9148: /* Date of death */
1.136 brouard 9149: cutv(stra, strb,line,' ');
1.169 brouard 9150: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9151: }
1.169 brouard 9152: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9153: month=99;
9154: year=9999;
9155: }else{
1.141 brouard 9156: 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 9157: 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);
9158: return 1;
1.136 brouard 9159: }
9160: andc[i]=(double) year;
9161: moisdc[i]=(double) month;
9162: strcpy(line,stra);
9163:
1.223 brouard 9164: /* Date of birth */
1.136 brouard 9165: cutv(stra, strb,line,' ');
1.169 brouard 9166: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9167: }
1.169 brouard 9168: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9169: month=99;
9170: year=9999;
9171: }else{
1.141 brouard 9172: 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);
9173: 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 9174: return 1;
1.136 brouard 9175: }
9176: if (year==9999) {
1.141 brouard 9177: 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);
9178: 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 9179: return 1;
9180:
1.136 brouard 9181: }
9182: annais[i]=(double)(year);
9183: moisnais[i]=(double)(month);
9184: strcpy(line,stra);
1.225 brouard 9185:
1.223 brouard 9186: /* Sample weight */
1.136 brouard 9187: cutv(stra, strb,line,' ');
9188: errno=0;
9189: dval=strtod(strb,&endptr);
9190: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9191: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9192: 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 9193: fflush(ficlog);
9194: return 1;
9195: }
9196: weight[i]=dval;
9197: strcpy(line,stra);
1.225 brouard 9198:
1.223 brouard 9199: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9200: cutv(stra, strb, line, ' ');
9201: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9202: lval=-1;
1.223 brouard 9203: }else{
1.225 brouard 9204: errno=0;
9205: /* what_kind_of_number(strb); */
9206: dval=strtod(strb,&endptr);
9207: /* if(strb != endptr && *endptr == '\0') */
9208: /* dval=dlval; */
9209: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9210: if( strb[0]=='\0' || (*endptr != '\0')){
9211: 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);
9212: 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);
9213: return 1;
9214: }
9215: coqvar[iv][i]=dval;
1.226 brouard 9216: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9217: }
9218: strcpy(line,stra);
9219: }/* end loop nqv */
1.136 brouard 9220:
1.223 brouard 9221: /* Covariate values */
1.136 brouard 9222: for (j=ncovcol;j>=1;j--){
9223: cutv(stra, strb,line,' ');
1.223 brouard 9224: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9225: lval=-1;
1.136 brouard 9226: }else{
1.225 brouard 9227: errno=0;
9228: lval=strtol(strb,&endptr,10);
9229: if( strb[0]=='\0' || (*endptr != '\0')){
9230: 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);
9231: 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);
9232: return 1;
9233: }
1.136 brouard 9234: }
9235: if(lval <-1 || lval >1){
1.225 brouard 9236: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9237: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9238: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9239: For example, for multinomial values like 1, 2 and 3,\n \
9240: build V1=0 V2=0 for the reference value (1),\n \
9241: V1=1 V2=0 for (2) \n \
1.136 brouard 9242: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9243: output of IMaCh is often meaningless.\n \
1.136 brouard 9244: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9245: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9246: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9247: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9248: For example, for multinomial values like 1, 2 and 3,\n \
9249: build V1=0 V2=0 for the reference value (1),\n \
9250: V1=1 V2=0 for (2) \n \
1.136 brouard 9251: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9252: output of IMaCh is often meaningless.\n \
1.136 brouard 9253: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9254: return 1;
1.136 brouard 9255: }
9256: covar[j][i]=(double)(lval);
9257: strcpy(line,stra);
9258: }
9259: lstra=strlen(stra);
1.225 brouard 9260:
1.136 brouard 9261: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9262: stratrunc = &(stra[lstra-9]);
9263: num[i]=atol(stratrunc);
9264: }
9265: else
9266: num[i]=atol(stra);
9267: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9268: 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;}*/
9269:
9270: i=i+1;
9271: } /* End loop reading data */
1.225 brouard 9272:
1.136 brouard 9273: *imax=i-1; /* Number of individuals */
9274: fclose(fic);
1.225 brouard 9275:
1.136 brouard 9276: return (0);
1.164 brouard 9277: /* endread: */
1.225 brouard 9278: printf("Exiting readdata: ");
9279: fclose(fic);
9280: return (1);
1.223 brouard 9281: }
1.126 brouard 9282:
1.234 brouard 9283: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9284: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9285: while (*p2 == ' ')
1.234 brouard 9286: p2++;
9287: /* while ((*p1++ = *p2++) !=0) */
9288: /* ; */
9289: /* do */
9290: /* while (*p2 == ' ') */
9291: /* p2++; */
9292: /* while (*p1++ == *p2++); */
9293: *stri=p2;
1.145 brouard 9294: }
9295:
1.235 brouard 9296: int decoderesult ( char resultline[], int nres)
1.230 brouard 9297: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9298: {
1.235 brouard 9299: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9300: char resultsav[MAXLINE];
1.234 brouard 9301: int resultmodel[MAXLINE];
9302: int modelresult[MAXLINE];
1.230 brouard 9303: char stra[80], strb[80], strc[80], strd[80],stre[80];
9304:
1.234 brouard 9305: removefirstspace(&resultline);
1.233 brouard 9306: printf("decoderesult:%s\n",resultline);
1.230 brouard 9307:
9308: if (strstr(resultline,"v") !=0){
9309: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9310: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9311: return 1;
9312: }
9313: trimbb(resultsav, resultline);
9314: if (strlen(resultsav) >1){
9315: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9316: }
1.253 brouard 9317: if(j == 0){ /* Resultline but no = */
9318: TKresult[nres]=0; /* Combination for the nresult and the model */
9319: return (0);
9320: }
9321:
1.234 brouard 9322: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9323: 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);
9324: 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);
9325: }
9326: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9327: if(nbocc(resultsav,'=') >1){
9328: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9329: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9330: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9331: }else
9332: cutl(strc,strd,resultsav,'=');
1.230 brouard 9333: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9334:
1.230 brouard 9335: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9336: Tvarsel[k]=atoi(strc);
9337: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9338: /* cptcovsel++; */
9339: if (nbocc(stra,'=') >0)
9340: strcpy(resultsav,stra); /* and analyzes it */
9341: }
1.235 brouard 9342: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9343: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9344: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9345: match=0;
1.236 brouard 9346: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9347: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9348: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9349: match=1;
9350: break;
9351: }
9352: }
9353: if(match == 0){
9354: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9355: }
9356: }
9357: }
1.235 brouard 9358: /* Checking for missing or useless values in comparison of current model needs */
9359: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9360: match=0;
1.235 brouard 9361: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9362: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9363: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9364: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9365: ++match;
9366: }
9367: }
9368: }
9369: if(match == 0){
9370: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9371: }else if(match > 1){
9372: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9373: }
9374: }
1.235 brouard 9375:
1.234 brouard 9376: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9377: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9378: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9379: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9380: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9381: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9382: /* 1 0 0 0 */
9383: /* 2 1 0 0 */
9384: /* 3 0 1 0 */
9385: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9386: /* 5 0 0 1 */
9387: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9388: /* 7 0 1 1 */
9389: /* 8 1 1 1 */
1.237 brouard 9390: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9391: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9392: /* V5*age V5 known which value for nres? */
9393: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9394: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9395: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9396: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9397: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9398: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9399: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9400: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9401: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9402: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9403: k4++;;
9404: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9405: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9406: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9407: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9408: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9409: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9410: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9411: k4q++;;
9412: }
9413: }
1.234 brouard 9414:
1.235 brouard 9415: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9416: return (0);
9417: }
1.235 brouard 9418:
1.230 brouard 9419: int decodemodel( char model[], int lastobs)
9420: /**< This routine decodes the model and returns:
1.224 brouard 9421: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9422: * - nagesqr = 1 if age*age in the model, otherwise 0.
9423: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9424: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9425: * - cptcovage number of covariates with age*products =2
9426: * - cptcovs number of simple covariates
9427: * - 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
9428: * which is a new column after the 9 (ncovcol) variables.
9429: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9430: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9431: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9432: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9433: */
1.136 brouard 9434: {
1.238 brouard 9435: int i, j, k, ks, v;
1.227 brouard 9436: int j1, k1, k2, k3, k4;
1.136 brouard 9437: char modelsav[80];
1.145 brouard 9438: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9439: char *strpt;
1.136 brouard 9440:
1.145 brouard 9441: /*removespace(model);*/
1.136 brouard 9442: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9443: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9444: if (strstr(model,"AGE") !=0){
1.192 brouard 9445: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9446: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9447: return 1;
9448: }
1.141 brouard 9449: if (strstr(model,"v") !=0){
9450: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9451: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9452: return 1;
9453: }
1.187 brouard 9454: strcpy(modelsav,model);
9455: if ((strpt=strstr(model,"age*age")) !=0){
9456: printf(" strpt=%s, model=%s\n",strpt, model);
9457: if(strpt != model){
1.234 brouard 9458: printf("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);
1.234 brouard 9461: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9462: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9463: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9464: return 1;
1.225 brouard 9465: }
1.187 brouard 9466: nagesqr=1;
9467: if (strstr(model,"+age*age") !=0)
1.234 brouard 9468: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9469: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9470: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9471: else
1.234 brouard 9472: substrchaine(modelsav, model, "age*age");
1.187 brouard 9473: }else
9474: nagesqr=0;
9475: if (strlen(modelsav) >1){
9476: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9477: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9478: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9479: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9480: * cst, age and age*age
9481: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9482: /* including age products which are counted in cptcovage.
9483: * but the covariates which are products must be treated
9484: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9485: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9486: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9487:
9488:
1.187 brouard 9489: /* Design
9490: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9491: * < ncovcol=8 >
9492: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9493: * k= 1 2 3 4 5 6 7 8
9494: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9495: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9496: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9497: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9498: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9499: * Tage[++cptcovage]=k
9500: * if products, new covar are created after ncovcol with k1
9501: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9502: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9503: * 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
9504: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9505: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9506: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9507: * < ncovcol=8 >
9508: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9509: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9510: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9511: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9512: * p Tprod[1]@2={ 6, 5}
9513: *p Tvard[1][1]@4= {7, 8, 5, 6}
9514: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9515: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9516: *How to reorganize?
9517: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9518: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9519: * {2, 1, 4, 8, 5, 6, 3, 7}
9520: * Struct []
9521: */
1.225 brouard 9522:
1.187 brouard 9523: /* This loop fills the array Tvar from the string 'model'.*/
9524: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9525: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9526: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9527: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9528: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9529: /* k=1 Tvar[1]=2 (from V2) */
9530: /* k=5 Tvar[5] */
9531: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9532: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9533: /* } */
1.198 brouard 9534: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9535: /*
9536: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9537: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9538: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9539: }
1.187 brouard 9540: cptcovage=0;
9541: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9542: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9543: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9544: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9545: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9546: /*scanf("%d",i);*/
9547: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9548: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9549: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9550: /* covar is not filled and then is empty */
9551: cptcovprod--;
9552: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9553: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9554: Typevar[k]=1; /* 1 for age product */
9555: cptcovage++; /* Sums the number of covariates which include age as a product */
9556: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9557: /*printf("stre=%s ", stre);*/
9558: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9559: cptcovprod--;
9560: cutl(stre,strb,strc,'V');
9561: Tvar[k]=atoi(stre);
9562: Typevar[k]=1; /* 1 for age product */
9563: cptcovage++;
9564: Tage[cptcovage]=k;
9565: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9566: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9567: cptcovn++;
9568: cptcovprodnoage++;k1++;
9569: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9570: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9571: because this model-covariate is a construction we invent a new column
9572: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9573: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9574: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9575: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9576: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9577: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9578: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9579: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9580: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9581: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9582: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9583: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9584: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9585: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9586: for (i=1; i<=lastobs;i++){
9587: /* Computes the new covariate which is a product of
9588: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9589: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9590: }
9591: } /* End age is not in the model */
9592: } /* End if model includes a product */
9593: else { /* no more sum */
9594: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9595: /* scanf("%d",i);*/
9596: cutl(strd,strc,strb,'V');
9597: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9598: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9599: Tvar[k]=atoi(strd);
9600: Typevar[k]=0; /* 0 for simple covariates */
9601: }
9602: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9603: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9604: scanf("%d",i);*/
1.187 brouard 9605: } /* end of loop + on total covariates */
9606: } /* end if strlen(modelsave == 0) age*age might exist */
9607: } /* end if strlen(model == 0) */
1.136 brouard 9608:
9609: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9610: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9611:
1.136 brouard 9612: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9613: printf("cptcovprod=%d ", cptcovprod);
9614: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9615: scanf("%d ",i);*/
9616:
9617:
1.230 brouard 9618: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9619: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9620: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9621: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9622: k = 1 2 3 4 5 6 7 8 9
9623: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9624: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9625: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9626: Dummy[k] 1 0 0 0 3 1 1 2 3
9627: Tmodelind[combination of covar]=k;
1.225 brouard 9628: */
9629: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9630: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9631: /* 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 9632: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9633: printf("Model=%s\n\
9634: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9635: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9636: 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);
9637: fprintf(ficlog,"Model=%s\n\
9638: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9639: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9640: 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 9641: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9642: 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 */
9643: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9644: Fixed[k]= 0;
9645: Dummy[k]= 0;
1.225 brouard 9646: ncoveff++;
1.232 brouard 9647: ncovf++;
1.234 brouard 9648: nsd++;
9649: modell[k].maintype= FTYPE;
9650: TvarsD[nsd]=Tvar[k];
9651: TvarsDind[nsd]=k;
9652: TvarF[ncovf]=Tvar[k];
9653: TvarFind[ncovf]=k;
9654: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9655: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9656: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9657: Fixed[k]= 0;
9658: Dummy[k]= 0;
9659: ncoveff++;
9660: ncovf++;
9661: modell[k].maintype= FTYPE;
9662: TvarF[ncovf]=Tvar[k];
9663: TvarFind[ncovf]=k;
1.230 brouard 9664: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9665: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9666: }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 9667: Fixed[k]= 0;
9668: Dummy[k]= 1;
1.230 brouard 9669: nqfveff++;
1.234 brouard 9670: modell[k].maintype= FTYPE;
9671: modell[k].subtype= FQ;
9672: nsq++;
9673: TvarsQ[nsq]=Tvar[k];
9674: TvarsQind[nsq]=k;
1.232 brouard 9675: ncovf++;
1.234 brouard 9676: TvarF[ncovf]=Tvar[k];
9677: TvarFind[ncovf]=k;
1.231 brouard 9678: 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 9679: 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 9680: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9681: Fixed[k]= 1;
9682: Dummy[k]= 0;
1.225 brouard 9683: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9684: modell[k].maintype= VTYPE;
9685: modell[k].subtype= VD;
9686: nsd++;
9687: TvarsD[nsd]=Tvar[k];
9688: TvarsDind[nsd]=k;
9689: ncovv++; /* Only simple time varying variables */
9690: TvarV[ncovv]=Tvar[k];
1.242 brouard 9691: 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 9692: 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 */
9693: 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 9694: 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);
9695: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9696: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9697: Fixed[k]= 1;
9698: Dummy[k]= 1;
9699: nqtveff++;
9700: modell[k].maintype= VTYPE;
9701: modell[k].subtype= VQ;
9702: ncovv++; /* Only simple time varying variables */
9703: nsq++;
9704: TvarsQ[nsq]=Tvar[k];
9705: TvarsQind[nsq]=k;
9706: TvarV[ncovv]=Tvar[k];
1.242 brouard 9707: 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 9708: 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 */
9709: 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 9710: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9711: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9712: 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 9713: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9714: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9715: ncova++;
9716: TvarA[ncova]=Tvar[k];
9717: TvarAind[ncova]=k;
1.231 brouard 9718: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9719: Fixed[k]= 2;
9720: Dummy[k]= 2;
9721: modell[k].maintype= ATYPE;
9722: modell[k].subtype= APFD;
9723: /* ncoveff++; */
1.227 brouard 9724: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9725: Fixed[k]= 2;
9726: Dummy[k]= 3;
9727: modell[k].maintype= ATYPE;
9728: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9729: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9730: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9731: Fixed[k]= 3;
9732: Dummy[k]= 2;
9733: modell[k].maintype= ATYPE;
9734: modell[k].subtype= APVD; /* Product age * varying dummy */
9735: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9736: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9737: Fixed[k]= 3;
9738: Dummy[k]= 3;
9739: modell[k].maintype= ATYPE;
9740: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9741: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9742: }
9743: }else if (Typevar[k] == 2) { /* product without age */
9744: k1=Tposprod[k];
9745: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9746: if(Tvard[k1][2] <=ncovcol){
9747: Fixed[k]= 1;
9748: Dummy[k]= 0;
9749: modell[k].maintype= FTYPE;
9750: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9751: ncovf++; /* Fixed variables without age */
9752: TvarF[ncovf]=Tvar[k];
9753: TvarFind[ncovf]=k;
9754: }else if(Tvard[k1][2] <=ncovcol+nqv){
9755: Fixed[k]= 0; /* or 2 ?*/
9756: Dummy[k]= 1;
9757: modell[k].maintype= FTYPE;
9758: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9759: ncovf++; /* Varying variables without age */
9760: TvarF[ncovf]=Tvar[k];
9761: TvarFind[ncovf]=k;
9762: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9763: Fixed[k]= 1;
9764: Dummy[k]= 0;
9765: modell[k].maintype= VTYPE;
9766: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9767: ncovv++; /* Varying variables without age */
9768: TvarV[ncovv]=Tvar[k];
9769: TvarVind[ncovv]=k;
9770: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9771: Fixed[k]= 1;
9772: Dummy[k]= 1;
9773: modell[k].maintype= VTYPE;
9774: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9775: ncovv++; /* Varying variables without age */
9776: TvarV[ncovv]=Tvar[k];
9777: TvarVind[ncovv]=k;
9778: }
1.227 brouard 9779: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9780: if(Tvard[k1][2] <=ncovcol){
9781: Fixed[k]= 0; /* or 2 ?*/
9782: Dummy[k]= 1;
9783: modell[k].maintype= FTYPE;
9784: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9785: ncovf++; /* Fixed variables without age */
9786: TvarF[ncovf]=Tvar[k];
9787: TvarFind[ncovf]=k;
9788: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9789: Fixed[k]= 1;
9790: Dummy[k]= 1;
9791: modell[k].maintype= VTYPE;
9792: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9793: ncovv++; /* Varying variables without age */
9794: TvarV[ncovv]=Tvar[k];
9795: TvarVind[ncovv]=k;
9796: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9797: Fixed[k]= 1;
9798: Dummy[k]= 1;
9799: modell[k].maintype= VTYPE;
9800: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9801: ncovv++; /* Varying variables without age */
9802: TvarV[ncovv]=Tvar[k];
9803: TvarVind[ncovv]=k;
9804: ncovv++; /* Varying variables without age */
9805: TvarV[ncovv]=Tvar[k];
9806: TvarVind[ncovv]=k;
9807: }
1.227 brouard 9808: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9809: if(Tvard[k1][2] <=ncovcol){
9810: Fixed[k]= 1;
9811: Dummy[k]= 1;
9812: modell[k].maintype= VTYPE;
9813: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9814: ncovv++; /* Varying variables without age */
9815: TvarV[ncovv]=Tvar[k];
9816: TvarVind[ncovv]=k;
9817: }else if(Tvard[k1][2] <=ncovcol+nqv){
9818: Fixed[k]= 1;
9819: Dummy[k]= 1;
9820: modell[k].maintype= VTYPE;
9821: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9822: ncovv++; /* Varying variables without age */
9823: TvarV[ncovv]=Tvar[k];
9824: TvarVind[ncovv]=k;
9825: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9826: Fixed[k]= 1;
9827: Dummy[k]= 0;
9828: modell[k].maintype= VTYPE;
9829: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9830: ncovv++; /* Varying variables without age */
9831: TvarV[ncovv]=Tvar[k];
9832: TvarVind[ncovv]=k;
9833: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9834: Fixed[k]= 1;
9835: Dummy[k]= 1;
9836: modell[k].maintype= VTYPE;
9837: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9838: ncovv++; /* Varying variables without age */
9839: TvarV[ncovv]=Tvar[k];
9840: TvarVind[ncovv]=k;
9841: }
1.227 brouard 9842: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9843: if(Tvard[k1][2] <=ncovcol){
9844: Fixed[k]= 1;
9845: Dummy[k]= 1;
9846: modell[k].maintype= VTYPE;
9847: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9848: ncovv++; /* Varying variables without age */
9849: TvarV[ncovv]=Tvar[k];
9850: TvarVind[ncovv]=k;
9851: }else if(Tvard[k1][2] <=ncovcol+nqv){
9852: Fixed[k]= 1;
9853: Dummy[k]= 1;
9854: modell[k].maintype= VTYPE;
9855: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9856: ncovv++; /* Varying variables without age */
9857: TvarV[ncovv]=Tvar[k];
9858: TvarVind[ncovv]=k;
9859: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9860: Fixed[k]= 1;
9861: Dummy[k]= 1;
9862: modell[k].maintype= VTYPE;
9863: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9864: ncovv++; /* Varying variables without age */
9865: TvarV[ncovv]=Tvar[k];
9866: TvarVind[ncovv]=k;
9867: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9868: Fixed[k]= 1;
9869: Dummy[k]= 1;
9870: modell[k].maintype= VTYPE;
9871: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9872: ncovv++; /* Varying variables without age */
9873: TvarV[ncovv]=Tvar[k];
9874: TvarVind[ncovv]=k;
9875: }
1.227 brouard 9876: }else{
1.240 brouard 9877: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9878: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9879: } /*end k1*/
1.225 brouard 9880: }else{
1.226 brouard 9881: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9882: 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 9883: }
1.227 brouard 9884: 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 9885: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9886: 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]);
9887: }
9888: /* Searching for doublons in the model */
9889: for(k1=1; k1<= cptcovt;k1++){
9890: for(k2=1; k2 <k1;k2++){
9891: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9892: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9893: if(Tvar[k1]==Tvar[k2]){
9894: 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]]);
9895: 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);
9896: return(1);
9897: }
9898: }else if (Typevar[k1] ==2){
9899: k3=Tposprod[k1];
9900: k4=Tposprod[k2];
9901: 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])) ){
9902: 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]]);
9903: 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);
9904: return(1);
9905: }
9906: }
1.227 brouard 9907: }
9908: }
1.225 brouard 9909: }
9910: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9911: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9912: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9913: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9914: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9915: /*endread:*/
1.225 brouard 9916: printf("Exiting decodemodel: ");
9917: return (1);
1.136 brouard 9918: }
9919:
1.169 brouard 9920: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9921: {/* Check ages at death */
1.136 brouard 9922: int i, m;
1.218 brouard 9923: int firstone=0;
9924:
1.136 brouard 9925: for (i=1; i<=imx; i++) {
9926: for(m=2; (m<= maxwav); m++) {
9927: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9928: anint[m][i]=9999;
1.216 brouard 9929: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9930: s[m][i]=-1;
1.136 brouard 9931: }
9932: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9933: *nberr = *nberr + 1;
1.218 brouard 9934: if(firstone == 0){
9935: firstone=1;
1.260 brouard 9936: 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 9937: }
1.262 brouard 9938: 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 9939: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9940: }
9941: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9942: (*nberr)++;
1.259 brouard 9943: 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 9944: 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 9945: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9946: }
9947: }
9948: }
9949:
9950: for (i=1; i<=imx; i++) {
9951: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9952: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9953: 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 9954: if (s[m][i] >= nlstate+1) {
1.169 brouard 9955: if(agedc[i]>0){
9956: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9957: agev[m][i]=agedc[i];
1.214 brouard 9958: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9959: }else {
1.136 brouard 9960: if ((int)andc[i]!=9999){
9961: nbwarn++;
9962: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9963: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9964: agev[m][i]=-1;
9965: }
9966: }
1.169 brouard 9967: } /* agedc > 0 */
1.214 brouard 9968: } /* end if */
1.136 brouard 9969: else if(s[m][i] !=9){ /* Standard case, age in fractional
9970: years but with the precision of a month */
9971: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9972: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9973: agev[m][i]=1;
9974: else if(agev[m][i] < *agemin){
9975: *agemin=agev[m][i];
9976: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9977: }
9978: else if(agev[m][i] >*agemax){
9979: *agemax=agev[m][i];
1.156 brouard 9980: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9981: }
9982: /*agev[m][i]=anint[m][i]-annais[i];*/
9983: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9984: } /* en if 9*/
1.136 brouard 9985: else { /* =9 */
1.214 brouard 9986: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9987: agev[m][i]=1;
9988: s[m][i]=-1;
9989: }
9990: }
1.214 brouard 9991: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9992: agev[m][i]=1;
1.214 brouard 9993: else{
9994: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9995: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9996: agev[m][i]=0;
9997: }
9998: } /* End for lastpass */
9999: }
1.136 brouard 10000:
10001: for (i=1; i<=imx; i++) {
10002: for(m=firstpass; (m<=lastpass); m++){
10003: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10004: (*nberr)++;
1.136 brouard 10005: 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);
10006: 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);
10007: return 1;
10008: }
10009: }
10010: }
10011:
10012: /*for (i=1; i<=imx; i++){
10013: for (m=firstpass; (m<lastpass); m++){
10014: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10015: }
10016:
10017: }*/
10018:
10019:
1.139 brouard 10020: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10021: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10022:
10023: return (0);
1.164 brouard 10024: /* endread:*/
1.136 brouard 10025: printf("Exiting calandcheckages: ");
10026: return (1);
10027: }
10028:
1.172 brouard 10029: #if defined(_MSC_VER)
10030: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10031: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10032: //#include "stdafx.h"
10033: //#include <stdio.h>
10034: //#include <tchar.h>
10035: //#include <windows.h>
10036: //#include <iostream>
10037: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10038:
10039: LPFN_ISWOW64PROCESS fnIsWow64Process;
10040:
10041: BOOL IsWow64()
10042: {
10043: BOOL bIsWow64 = FALSE;
10044:
10045: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10046: // (HANDLE, PBOOL);
10047:
10048: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10049:
10050: HMODULE module = GetModuleHandle(_T("kernel32"));
10051: const char funcName[] = "IsWow64Process";
10052: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10053: GetProcAddress(module, funcName);
10054:
10055: if (NULL != fnIsWow64Process)
10056: {
10057: if (!fnIsWow64Process(GetCurrentProcess(),
10058: &bIsWow64))
10059: //throw std::exception("Unknown error");
10060: printf("Unknown error\n");
10061: }
10062: return bIsWow64 != FALSE;
10063: }
10064: #endif
1.177 brouard 10065:
1.191 brouard 10066: void syscompilerinfo(int logged)
1.167 brouard 10067: {
10068: /* #include "syscompilerinfo.h"*/
1.185 brouard 10069: /* command line Intel compiler 32bit windows, XP compatible:*/
10070: /* /GS /W3 /Gy
10071: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10072: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10073: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10074: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10075: */
10076: /* 64 bits */
1.185 brouard 10077: /*
10078: /GS /W3 /Gy
10079: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10080: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10081: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10082: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10083: /* Optimization are useless and O3 is slower than O2 */
10084: /*
10085: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10086: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10087: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10088: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10089: */
1.186 brouard 10090: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10091: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10092: /PDB:"visual studio
10093: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10094: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10095: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10096: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10097: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10098: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10099: uiAccess='false'"
10100: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10101: /NOLOGO /TLBID:1
10102: */
1.177 brouard 10103: #if defined __INTEL_COMPILER
1.178 brouard 10104: #if defined(__GNUC__)
10105: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10106: #endif
1.177 brouard 10107: #elif defined(__GNUC__)
1.179 brouard 10108: #ifndef __APPLE__
1.174 brouard 10109: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10110: #endif
1.177 brouard 10111: struct utsname sysInfo;
1.178 brouard 10112: int cross = CROSS;
10113: if (cross){
10114: printf("Cross-");
1.191 brouard 10115: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10116: }
1.174 brouard 10117: #endif
10118:
1.171 brouard 10119: #include <stdint.h>
1.178 brouard 10120:
1.191 brouard 10121: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10122: #if defined(__clang__)
1.191 brouard 10123: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10124: #endif
10125: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10126: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10127: #endif
10128: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10129: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10130: #endif
10131: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10132: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10133: #endif
10134: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10135: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10136: #endif
10137: #if defined(_MSC_VER)
1.191 brouard 10138: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10139: #endif
10140: #if defined(__PGI)
1.191 brouard 10141: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10142: #endif
10143: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10144: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10145: #endif
1.191 brouard 10146: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10147:
1.167 brouard 10148: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10149: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10150: // Windows (x64 and x86)
1.191 brouard 10151: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10152: #elif __unix__ // all unices, not all compilers
10153: // Unix
1.191 brouard 10154: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10155: #elif __linux__
10156: // linux
1.191 brouard 10157: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10158: #elif __APPLE__
1.174 brouard 10159: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10160: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10161: #endif
10162:
10163: /* __MINGW32__ */
10164: /* __CYGWIN__ */
10165: /* __MINGW64__ */
10166: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10167: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10168: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10169: /* _WIN64 // Defined for applications for Win64. */
10170: /* _M_X64 // Defined for compilations that target x64 processors. */
10171: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10172:
1.167 brouard 10173: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10174: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10175: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10176: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10177: #else
1.191 brouard 10178: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10179: #endif
10180:
1.169 brouard 10181: #if defined(__GNUC__)
10182: # if defined(__GNUC_PATCHLEVEL__)
10183: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10184: + __GNUC_MINOR__ * 100 \
10185: + __GNUC_PATCHLEVEL__)
10186: # else
10187: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10188: + __GNUC_MINOR__ * 100)
10189: # endif
1.174 brouard 10190: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10191: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10192:
10193: if (uname(&sysInfo) != -1) {
10194: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10195: 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 10196: }
10197: else
10198: perror("uname() error");
1.179 brouard 10199: //#ifndef __INTEL_COMPILER
10200: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10201: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10202: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10203: #endif
1.169 brouard 10204: #endif
1.172 brouard 10205:
10206: // void main()
10207: // {
1.169 brouard 10208: #if defined(_MSC_VER)
1.174 brouard 10209: if (IsWow64()){
1.191 brouard 10210: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10211: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10212: }
10213: else{
1.191 brouard 10214: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10215: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10216: }
1.172 brouard 10217: // printf("\nPress Enter to continue...");
10218: // getchar();
10219: // }
10220:
1.169 brouard 10221: #endif
10222:
1.167 brouard 10223:
1.219 brouard 10224: }
1.136 brouard 10225:
1.219 brouard 10226: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10227: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10228: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10229: /* double ftolpl = 1.e-10; */
1.180 brouard 10230: double age, agebase, agelim;
1.203 brouard 10231: double tot;
1.180 brouard 10232:
1.202 brouard 10233: strcpy(filerespl,"PL_");
10234: strcat(filerespl,fileresu);
10235: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10236: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10237: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10238: }
1.227 brouard 10239: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10240: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10241: pstamp(ficrespl);
1.203 brouard 10242: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10243: fprintf(ficrespl,"#Age ");
10244: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10245: fprintf(ficrespl,"\n");
1.180 brouard 10246:
1.219 brouard 10247: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10248:
1.219 brouard 10249: agebase=ageminpar;
10250: agelim=agemaxpar;
1.180 brouard 10251:
1.227 brouard 10252: /* i1=pow(2,ncoveff); */
1.234 brouard 10253: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10254: if (cptcovn < 1){i1=1;}
1.180 brouard 10255:
1.238 brouard 10256: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10257: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10258: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10259: continue;
1.235 brouard 10260:
1.238 brouard 10261: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10262: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10263: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10264: /* k=k+1; */
10265: /* to clean */
10266: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10267: fprintf(ficrespl,"#******");
10268: printf("#******");
10269: fprintf(ficlog,"#******");
10270: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10271: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10272: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10273: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10274: }
10275: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10276: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10277: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10278: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10279: }
10280: fprintf(ficrespl,"******\n");
10281: printf("******\n");
10282: fprintf(ficlog,"******\n");
10283: if(invalidvarcomb[k]){
10284: printf("\nCombination (%d) ignored because no case \n",k);
10285: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10286: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10287: continue;
10288: }
1.219 brouard 10289:
1.238 brouard 10290: fprintf(ficrespl,"#Age ");
10291: for(j=1;j<=cptcoveff;j++) {
10292: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10293: }
10294: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10295: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10296:
1.238 brouard 10297: for (age=agebase; age<=agelim; age++){
10298: /* for (age=agebase; age<=agebase; age++){ */
10299: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10300: fprintf(ficrespl,"%.0f ",age );
10301: for(j=1;j<=cptcoveff;j++)
10302: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10303: tot=0.;
10304: for(i=1; i<=nlstate;i++){
10305: tot += prlim[i][i];
10306: fprintf(ficrespl," %.5f", prlim[i][i]);
10307: }
10308: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10309: } /* Age */
10310: /* was end of cptcod */
10311: } /* cptcov */
10312: } /* nres */
1.219 brouard 10313: return 0;
1.180 brouard 10314: }
10315:
1.218 brouard 10316: 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){
10317: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10318:
10319: /* Computes the back prevalence limit for any combination of covariate values
10320: * at any age between ageminpar and agemaxpar
10321: */
1.235 brouard 10322: int i, j, k, i1, nres=0 ;
1.217 brouard 10323: /* double ftolpl = 1.e-10; */
10324: double age, agebase, agelim;
10325: double tot;
1.218 brouard 10326: /* double ***mobaverage; */
10327: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10328:
10329: strcpy(fileresplb,"PLB_");
10330: strcat(fileresplb,fileresu);
10331: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10332: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10333: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10334: }
10335: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10336: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10337: pstamp(ficresplb);
10338: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10339: fprintf(ficresplb,"#Age ");
10340: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10341: fprintf(ficresplb,"\n");
10342:
1.218 brouard 10343:
10344: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10345:
10346: agebase=ageminpar;
10347: agelim=agemaxpar;
10348:
10349:
1.227 brouard 10350: i1=pow(2,cptcoveff);
1.218 brouard 10351: if (cptcovn < 1){i1=1;}
1.227 brouard 10352:
1.238 brouard 10353: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10354: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10355: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10356: continue;
10357: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10358: fprintf(ficresplb,"#******");
10359: printf("#******");
10360: fprintf(ficlog,"#******");
10361: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10362: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10363: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10364: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10365: }
10366: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10367: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10368: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10369: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10370: }
10371: fprintf(ficresplb,"******\n");
10372: printf("******\n");
10373: fprintf(ficlog,"******\n");
10374: if(invalidvarcomb[k]){
10375: printf("\nCombination (%d) ignored because no cases \n",k);
10376: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10377: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10378: continue;
10379: }
1.218 brouard 10380:
1.238 brouard 10381: fprintf(ficresplb,"#Age ");
10382: for(j=1;j<=cptcoveff;j++) {
10383: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10384: }
10385: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10386: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10387:
10388:
1.238 brouard 10389: for (age=agebase; age<=agelim; age++){
10390: /* for (age=agebase; age<=agebase; age++){ */
10391: if(mobilavproj > 0){
10392: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10393: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10394: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10395: }else if (mobilavproj == 0){
10396: 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);
10397: 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);
10398: exit(1);
10399: }else{
10400: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10401: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10402: /* printf("TOTOT\n"); */
10403: /* exit(1); */
1.238 brouard 10404: }
10405: fprintf(ficresplb,"%.0f ",age );
10406: for(j=1;j<=cptcoveff;j++)
10407: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10408: tot=0.;
10409: for(i=1; i<=nlstate;i++){
10410: tot += bprlim[i][i];
10411: fprintf(ficresplb," %.5f", bprlim[i][i]);
10412: }
10413: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10414: } /* Age */
10415: /* was end of cptcod */
1.255 brouard 10416: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10417: } /* end of any combination */
10418: } /* end of nres */
1.218 brouard 10419: /* hBijx(p, bage, fage); */
10420: /* fclose(ficrespijb); */
10421:
10422: return 0;
1.217 brouard 10423: }
1.218 brouard 10424:
1.180 brouard 10425: int hPijx(double *p, int bage, int fage){
10426: /*------------- h Pij x at various ages ------------*/
10427:
10428: int stepsize;
10429: int agelim;
10430: int hstepm;
10431: int nhstepm;
1.235 brouard 10432: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10433:
10434: double agedeb;
10435: double ***p3mat;
10436:
1.201 brouard 10437: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10438: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10439: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10440: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10441: }
10442: printf("Computing pij: result on file '%s' \n", filerespij);
10443: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10444:
10445: stepsize=(int) (stepm+YEARM-1)/YEARM;
10446: /*if (stepm<=24) stepsize=2;*/
10447:
10448: agelim=AGESUP;
10449: hstepm=stepsize*YEARM; /* Every year of age */
10450: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10451:
1.180 brouard 10452: /* hstepm=1; aff par mois*/
10453: pstamp(ficrespij);
10454: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10455: i1= pow(2,cptcoveff);
1.218 brouard 10456: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10457: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10458: /* k=k+1; */
1.235 brouard 10459: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10460: for(k=1; k<=i1;k++){
1.253 brouard 10461: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10462: continue;
1.183 brouard 10463: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10464: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10465: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10466: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10467: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10468: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10469: }
1.183 brouard 10470: fprintf(ficrespij,"******\n");
10471:
10472: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10473: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10474: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10475:
10476: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10477:
1.183 brouard 10478: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10479: oldm=oldms;savm=savms;
1.235 brouard 10480: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10481: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10482: for(i=1; i<=nlstate;i++)
10483: for(j=1; j<=nlstate+ndeath;j++)
10484: fprintf(ficrespij," %1d-%1d",i,j);
10485: fprintf(ficrespij,"\n");
10486: for (h=0; h<=nhstepm; h++){
10487: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10488: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10489: for(i=1; i<=nlstate;i++)
10490: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10491: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10492: fprintf(ficrespij,"\n");
10493: }
1.183 brouard 10494: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10495: fprintf(ficrespij,"\n");
10496: }
1.180 brouard 10497: /*}*/
10498: }
1.218 brouard 10499: return 0;
1.180 brouard 10500: }
1.218 brouard 10501:
10502: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10503: /*------------- h Bij x at various ages ------------*/
10504:
10505: int stepsize;
1.218 brouard 10506: /* int agelim; */
10507: int ageminl;
1.217 brouard 10508: int hstepm;
10509: int nhstepm;
1.238 brouard 10510: int h, i, i1, j, k, nres;
1.218 brouard 10511:
1.217 brouard 10512: double agedeb;
10513: double ***p3mat;
1.218 brouard 10514:
10515: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10516: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10517: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10518: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10519: }
10520: printf("Computing pij back: result on file '%s' \n", filerespijb);
10521: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10522:
10523: stepsize=(int) (stepm+YEARM-1)/YEARM;
10524: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10525:
1.218 brouard 10526: /* agelim=AGESUP; */
10527: ageminl=30;
10528: hstepm=stepsize*YEARM; /* Every year of age */
10529: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10530:
10531: /* hstepm=1; aff par mois*/
10532: pstamp(ficrespijb);
1.255 brouard 10533: 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 10534: i1= pow(2,cptcoveff);
1.218 brouard 10535: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10536: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10537: /* k=k+1; */
1.238 brouard 10538: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10539: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10540: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10541: continue;
10542: fprintf(ficrespijb,"\n#****** ");
10543: for(j=1;j<=cptcoveff;j++)
10544: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10545: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10546: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10547: }
10548: fprintf(ficrespijb,"******\n");
1.264 brouard 10549: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10550: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10551: continue;
10552: }
10553:
10554: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10555: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10556: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10557: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10558: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10559:
10560: /* nhstepm=nhstepm*YEARM; aff par mois*/
10561:
1.266 brouard 10562: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10563: /* and memory limitations if stepm is small */
10564:
1.238 brouard 10565: /* oldm=oldms;savm=savms; */
10566: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10567: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10568: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10569: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10570: for(i=1; i<=nlstate;i++)
10571: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10572: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10573: fprintf(ficrespijb,"\n");
1.238 brouard 10574: for (h=0; h<=nhstepm; h++){
10575: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10576: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10577: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10578: for(i=1; i<=nlstate;i++)
10579: for(j=1; j<=nlstate+ndeath;j++)
10580: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10581: fprintf(ficrespijb,"\n");
10582: }
10583: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10584: fprintf(ficrespijb,"\n");
10585: } /* end age deb */
10586: } /* end combination */
10587: } /* end nres */
1.218 brouard 10588: return 0;
10589: } /* hBijx */
1.217 brouard 10590:
1.180 brouard 10591:
1.136 brouard 10592: /***********************************************/
10593: /**************** Main Program *****************/
10594: /***********************************************/
10595:
10596: int main(int argc, char *argv[])
10597: {
10598: #ifdef GSL
10599: const gsl_multimin_fminimizer_type *T;
10600: size_t iteri = 0, it;
10601: int rval = GSL_CONTINUE;
10602: int status = GSL_SUCCESS;
10603: double ssval;
10604: #endif
10605: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10606: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10607: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10608: int jj, ll, li, lj, lk;
1.136 brouard 10609: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10610: int num_filled;
1.136 brouard 10611: int itimes;
10612: int NDIM=2;
10613: int vpopbased=0;
1.235 brouard 10614: int nres=0;
1.258 brouard 10615: int endishere=0;
1.277 ! brouard 10616: int noffset=0;
1.274 brouard 10617: int ncurrv=0; /* Temporary variable */
10618:
1.164 brouard 10619: char ca[32], cb[32];
1.136 brouard 10620: /* FILE *fichtm; *//* Html File */
10621: /* FILE *ficgp;*/ /*Gnuplot File */
10622: struct stat info;
1.191 brouard 10623: double agedeb=0.;
1.194 brouard 10624:
10625: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10626: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10627:
1.165 brouard 10628: double fret;
1.191 brouard 10629: double dum=0.; /* Dummy variable */
1.136 brouard 10630: double ***p3mat;
1.218 brouard 10631: /* double ***mobaverage; */
1.164 brouard 10632:
10633: char line[MAXLINE];
1.197 brouard 10634: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10635:
1.234 brouard 10636: char modeltemp[MAXLINE];
1.230 brouard 10637: char resultline[MAXLINE];
10638:
1.136 brouard 10639: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10640: char *tok, *val; /* pathtot */
1.136 brouard 10641: int firstobs=1, lastobs=10;
1.195 brouard 10642: int c, h , cpt, c2;
1.191 brouard 10643: int jl=0;
10644: int i1, j1, jk, stepsize=0;
1.194 brouard 10645: int count=0;
10646:
1.164 brouard 10647: int *tab;
1.136 brouard 10648: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10649: int backcast=0;
1.136 brouard 10650: int mobilav=0,popforecast=0;
1.191 brouard 10651: int hstepm=0, nhstepm=0;
1.136 brouard 10652: int agemortsup;
10653: float sumlpop=0.;
10654: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10655: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10656:
1.191 brouard 10657: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10658: double ftolpl=FTOL;
10659: double **prlim;
1.217 brouard 10660: double **bprlim;
1.136 brouard 10661: double ***param; /* Matrix of parameters */
1.251 brouard 10662: double ***paramstart; /* Matrix of starting parameter values */
10663: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10664: double **matcov; /* Matrix of covariance */
1.203 brouard 10665: double **hess; /* Hessian matrix */
1.136 brouard 10666: double ***delti3; /* Scale */
10667: double *delti; /* Scale */
10668: double ***eij, ***vareij;
10669: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10670:
1.136 brouard 10671: double *epj, vepp;
1.164 brouard 10672:
1.273 brouard 10673: double dateprev1, dateprev2;
10674: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10675: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10676:
1.136 brouard 10677: double **ximort;
1.145 brouard 10678: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10679: int *dcwave;
10680:
1.164 brouard 10681: char z[1]="c";
1.136 brouard 10682:
10683: /*char *strt;*/
10684: char strtend[80];
1.126 brouard 10685:
1.164 brouard 10686:
1.126 brouard 10687: /* setlocale (LC_ALL, ""); */
10688: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10689: /* textdomain (PACKAGE); */
10690: /* setlocale (LC_CTYPE, ""); */
10691: /* setlocale (LC_MESSAGES, ""); */
10692:
10693: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10694: rstart_time = time(NULL);
10695: /* (void) gettimeofday(&start_time,&tzp);*/
10696: start_time = *localtime(&rstart_time);
1.126 brouard 10697: curr_time=start_time;
1.157 brouard 10698: /*tml = *localtime(&start_time.tm_sec);*/
10699: /* strcpy(strstart,asctime(&tml)); */
10700: strcpy(strstart,asctime(&start_time));
1.126 brouard 10701:
10702: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10703: /* tp.tm_sec = tp.tm_sec +86400; */
10704: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10705: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10706: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10707: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10708: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10709: /* strt=asctime(&tmg); */
10710: /* printf("Time(after) =%s",strstart); */
10711: /* (void) time (&time_value);
10712: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10713: * tm = *localtime(&time_value);
10714: * strstart=asctime(&tm);
10715: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10716: */
10717:
10718: nberr=0; /* Number of errors and warnings */
10719: nbwarn=0;
1.184 brouard 10720: #ifdef WIN32
10721: _getcwd(pathcd, size);
10722: #else
1.126 brouard 10723: getcwd(pathcd, size);
1.184 brouard 10724: #endif
1.191 brouard 10725: syscompilerinfo(0);
1.196 brouard 10726: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10727: if(argc <=1){
10728: printf("\nEnter the parameter file name: ");
1.205 brouard 10729: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10730: printf("ERROR Empty parameter file name\n");
10731: goto end;
10732: }
1.126 brouard 10733: i=strlen(pathr);
10734: if(pathr[i-1]=='\n')
10735: pathr[i-1]='\0';
1.156 brouard 10736: i=strlen(pathr);
1.205 brouard 10737: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10738: pathr[i-1]='\0';
1.205 brouard 10739: }
10740: i=strlen(pathr);
10741: if( i==0 ){
10742: printf("ERROR Empty parameter file name\n");
10743: goto end;
10744: }
10745: for (tok = pathr; tok != NULL; ){
1.126 brouard 10746: printf("Pathr |%s|\n",pathr);
10747: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10748: printf("val= |%s| pathr=%s\n",val,pathr);
10749: strcpy (pathtot, val);
10750: if(pathr[0] == '\0') break; /* Dirty */
10751: }
10752: }
10753: else{
10754: strcpy(pathtot,argv[1]);
10755: }
10756: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10757: /*cygwin_split_path(pathtot,path,optionfile);
10758: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10759: /* cutv(path,optionfile,pathtot,'\\');*/
10760:
10761: /* Split argv[0], imach program to get pathimach */
10762: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10763: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10764: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10765: /* strcpy(pathimach,argv[0]); */
10766: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10767: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10768: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10769: #ifdef WIN32
10770: _chdir(path); /* Can be a relative path */
10771: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10772: #else
1.126 brouard 10773: chdir(path); /* Can be a relative path */
1.184 brouard 10774: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10775: #endif
10776: printf("Current directory %s!\n",pathcd);
1.126 brouard 10777: strcpy(command,"mkdir ");
10778: strcat(command,optionfilefiname);
10779: if((outcmd=system(command)) != 0){
1.169 brouard 10780: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10781: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10782: /* fclose(ficlog); */
10783: /* exit(1); */
10784: }
10785: /* if((imk=mkdir(optionfilefiname))<0){ */
10786: /* perror("mkdir"); */
10787: /* } */
10788:
10789: /*-------- arguments in the command line --------*/
10790:
1.186 brouard 10791: /* Main Log file */
1.126 brouard 10792: strcat(filelog, optionfilefiname);
10793: strcat(filelog,".log"); /* */
10794: if((ficlog=fopen(filelog,"w"))==NULL) {
10795: printf("Problem with logfile %s\n",filelog);
10796: goto end;
10797: }
10798: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10799: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10800: fprintf(ficlog,"\nEnter the parameter file name: \n");
10801: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10802: path=%s \n\
10803: optionfile=%s\n\
10804: optionfilext=%s\n\
1.156 brouard 10805: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10806:
1.197 brouard 10807: syscompilerinfo(1);
1.167 brouard 10808:
1.126 brouard 10809: printf("Local time (at start):%s",strstart);
10810: fprintf(ficlog,"Local time (at start): %s",strstart);
10811: fflush(ficlog);
10812: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10813: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10814:
10815: /* */
10816: strcpy(fileres,"r");
10817: strcat(fileres, optionfilefiname);
1.201 brouard 10818: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10819: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10820: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10821:
1.186 brouard 10822: /* Main ---------arguments file --------*/
1.126 brouard 10823:
10824: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10825: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10826: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10827: fflush(ficlog);
1.149 brouard 10828: /* goto end; */
10829: exit(70);
1.126 brouard 10830: }
10831:
10832:
10833:
10834: strcpy(filereso,"o");
1.201 brouard 10835: strcat(filereso,fileresu);
1.126 brouard 10836: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10837: printf("Problem with Output resultfile: %s\n", filereso);
10838: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10839: fflush(ficlog);
10840: goto end;
10841: }
10842:
10843: /* Reads comments: lines beginning with '#' */
10844: numlinepar=0;
1.277 ! brouard 10845: /* Is it a BOM UTF-8 Windows file? */
! 10846: /* First parameter line */
1.197 brouard 10847: while(fgets(line, MAXLINE, ficpar)) {
1.277 ! brouard 10848: noffset=0;
! 10849: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
! 10850: {
! 10851: noffset=noffset+3;
! 10852: printf("# File is an UTF8 Bom.\n"); // 0xBF
! 10853: }
! 10854: else if( line[0] == (char)0xFE && line[1] == (char)0xFF)
! 10855: {
! 10856: noffset=noffset+2;
! 10857: printf("# File is an UTF16BE BOM file\n");
! 10858: }
! 10859: else if( line[0] == 0 && line[1] == 0)
! 10860: {
! 10861: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
! 10862: noffset=noffset+4;
! 10863: printf("# File is an UTF16BE BOM file\n");
! 10864: }
! 10865: } else{
! 10866: ;/*printf(" Not a BOM file\n");*/
! 10867: }
! 10868:
1.197 brouard 10869: /* If line starts with a # it is a comment */
1.277 ! brouard 10870: if (line[noffset] == '#') {
1.197 brouard 10871: numlinepar++;
10872: fputs(line,stdout);
10873: fputs(line,ficparo);
10874: fputs(line,ficlog);
10875: continue;
10876: }else
10877: break;
10878: }
10879: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10880: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10881: if (num_filled != 5) {
10882: printf("Should be 5 parameters\n");
10883: }
1.126 brouard 10884: numlinepar++;
1.197 brouard 10885: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10886: }
10887: /* Second parameter line */
10888: while(fgets(line, MAXLINE, ficpar)) {
10889: /* If line starts with a # it is a comment */
10890: if (line[0] == '#') {
10891: numlinepar++;
10892: fputs(line,stdout);
10893: fputs(line,ficparo);
10894: fputs(line,ficlog);
10895: continue;
10896: }else
10897: break;
10898: }
1.223 brouard 10899: 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", \
10900: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10901: if (num_filled != 11) {
10902: 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 10903: printf("but line=%s\n",line);
1.197 brouard 10904: }
1.223 brouard 10905: 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 10906: }
1.203 brouard 10907: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10908: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10909: /* Third parameter line */
10910: while(fgets(line, MAXLINE, ficpar)) {
10911: /* If line starts with a # it is a comment */
10912: if (line[0] == '#') {
10913: numlinepar++;
10914: fputs(line,stdout);
10915: fputs(line,ficparo);
10916: fputs(line,ficlog);
10917: continue;
10918: }else
10919: break;
10920: }
1.201 brouard 10921: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 brouard 10922: if (num_filled == 0){
10923: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10924: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10925: model[0]='\0';
10926: goto end;
10927: } else if (num_filled != 1){
1.197 brouard 10928: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10929: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10930: model[0]='\0';
10931: goto end;
10932: }
10933: else{
10934: if (model[0]=='+'){
10935: for(i=1; i<=strlen(model);i++)
10936: modeltemp[i-1]=model[i];
1.201 brouard 10937: strcpy(model,modeltemp);
1.197 brouard 10938: }
10939: }
1.199 brouard 10940: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10941: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10942: }
10943: /* 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); */
10944: /* numlinepar=numlinepar+3; /\* In general *\/ */
10945: /* 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 10946: 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);
10947: 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 10948: fflush(ficlog);
1.190 brouard 10949: /* if(model[0]=='#'|| model[0]== '\0'){ */
10950: if(model[0]=='#'){
1.187 brouard 10951: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10952: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10953: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10954: if(mle != -1){
10955: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10956: exit(1);
10957: }
10958: }
1.126 brouard 10959: while((c=getc(ficpar))=='#' && c!= EOF){
10960: ungetc(c,ficpar);
10961: fgets(line, MAXLINE, ficpar);
10962: numlinepar++;
1.195 brouard 10963: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10964: z[0]=line[1];
10965: }
10966: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10967: fputs(line, stdout);
10968: //puts(line);
1.126 brouard 10969: fputs(line,ficparo);
10970: fputs(line,ficlog);
10971: }
10972: ungetc(c,ficpar);
10973:
10974:
1.145 brouard 10975: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 10976: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
10977: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
10978: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 10979: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10980: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10981: v1+v2*age+v2*v3 makes cptcovn = 3
10982: */
10983: if (strlen(model)>1)
1.187 brouard 10984: 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 10985: else
1.187 brouard 10986: ncovmodel=2; /* Constant and age */
1.133 brouard 10987: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10988: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10989: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10990: 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);
10991: 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);
10992: fflush(stdout);
10993: fclose (ficlog);
10994: goto end;
10995: }
1.126 brouard 10996: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10997: delti=delti3[1][1];
10998: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10999: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11000: /* We could also provide initial parameters values giving by simple logistic regression
11001: * only one way, that is without matrix product. We will have nlstate maximizations */
11002: /* for(i=1;i<nlstate;i++){ */
11003: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11004: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11005: /* } */
1.126 brouard 11006: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11007: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11008: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11009: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11010: fclose (ficparo);
11011: fclose (ficlog);
11012: goto end;
11013: exit(0);
1.220 brouard 11014: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11015: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11016: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11017: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11018: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11019: matcov=matrix(1,npar,1,npar);
1.203 brouard 11020: hess=matrix(1,npar,1,npar);
1.220 brouard 11021: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11022: /* Read guessed parameters */
1.126 brouard 11023: /* Reads comments: lines beginning with '#' */
11024: while((c=getc(ficpar))=='#' && c!= EOF){
11025: ungetc(c,ficpar);
11026: fgets(line, MAXLINE, ficpar);
11027: numlinepar++;
1.141 brouard 11028: fputs(line,stdout);
1.126 brouard 11029: fputs(line,ficparo);
11030: fputs(line,ficlog);
11031: }
11032: ungetc(c,ficpar);
11033:
11034: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11035: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11036: for(i=1; i <=nlstate; i++){
1.234 brouard 11037: j=0;
1.126 brouard 11038: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11039: if(jj==i) continue;
11040: j++;
11041: fscanf(ficpar,"%1d%1d",&i1,&j1);
11042: if ((i1 != i) || (j1 != jj)){
11043: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11044: It might be a problem of design; if ncovcol and the model are correct\n \
11045: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11046: exit(1);
11047: }
11048: fprintf(ficparo,"%1d%1d",i1,j1);
11049: if(mle==1)
11050: printf("%1d%1d",i,jj);
11051: fprintf(ficlog,"%1d%1d",i,jj);
11052: for(k=1; k<=ncovmodel;k++){
11053: fscanf(ficpar," %lf",¶m[i][j][k]);
11054: if(mle==1){
11055: printf(" %lf",param[i][j][k]);
11056: fprintf(ficlog," %lf",param[i][j][k]);
11057: }
11058: else
11059: fprintf(ficlog," %lf",param[i][j][k]);
11060: fprintf(ficparo," %lf",param[i][j][k]);
11061: }
11062: fscanf(ficpar,"\n");
11063: numlinepar++;
11064: if(mle==1)
11065: printf("\n");
11066: fprintf(ficlog,"\n");
11067: fprintf(ficparo,"\n");
1.126 brouard 11068: }
11069: }
11070: fflush(ficlog);
1.234 brouard 11071:
1.251 brouard 11072: /* Reads parameters values */
1.126 brouard 11073: p=param[1][1];
1.251 brouard 11074: pstart=paramstart[1][1];
1.126 brouard 11075:
11076: /* Reads comments: lines beginning with '#' */
11077: while((c=getc(ficpar))=='#' && c!= EOF){
11078: ungetc(c,ficpar);
11079: fgets(line, MAXLINE, ficpar);
11080: numlinepar++;
1.141 brouard 11081: fputs(line,stdout);
1.126 brouard 11082: fputs(line,ficparo);
11083: fputs(line,ficlog);
11084: }
11085: ungetc(c,ficpar);
11086:
11087: for(i=1; i <=nlstate; i++){
11088: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11089: fscanf(ficpar,"%1d%1d",&i1,&j1);
11090: if ( (i1-i) * (j1-j) != 0){
11091: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11092: exit(1);
11093: }
11094: printf("%1d%1d",i,j);
11095: fprintf(ficparo,"%1d%1d",i1,j1);
11096: fprintf(ficlog,"%1d%1d",i1,j1);
11097: for(k=1; k<=ncovmodel;k++){
11098: fscanf(ficpar,"%le",&delti3[i][j][k]);
11099: printf(" %le",delti3[i][j][k]);
11100: fprintf(ficparo," %le",delti3[i][j][k]);
11101: fprintf(ficlog," %le",delti3[i][j][k]);
11102: }
11103: fscanf(ficpar,"\n");
11104: numlinepar++;
11105: printf("\n");
11106: fprintf(ficparo,"\n");
11107: fprintf(ficlog,"\n");
1.126 brouard 11108: }
11109: }
11110: fflush(ficlog);
1.234 brouard 11111:
1.145 brouard 11112: /* Reads covariance matrix */
1.126 brouard 11113: delti=delti3[1][1];
1.220 brouard 11114:
11115:
1.126 brouard 11116: /* 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 11117:
1.126 brouard 11118: /* Reads comments: lines beginning with '#' */
11119: while((c=getc(ficpar))=='#' && c!= EOF){
11120: ungetc(c,ficpar);
11121: fgets(line, MAXLINE, ficpar);
11122: numlinepar++;
1.141 brouard 11123: fputs(line,stdout);
1.126 brouard 11124: fputs(line,ficparo);
11125: fputs(line,ficlog);
11126: }
11127: ungetc(c,ficpar);
1.220 brouard 11128:
1.126 brouard 11129: matcov=matrix(1,npar,1,npar);
1.203 brouard 11130: hess=matrix(1,npar,1,npar);
1.131 brouard 11131: for(i=1; i <=npar; i++)
11132: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11133:
1.194 brouard 11134: /* Scans npar lines */
1.126 brouard 11135: for(i=1; i <=npar; i++){
1.226 brouard 11136: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11137: if(count != 3){
1.226 brouard 11138: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11139: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11140: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11141: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11142: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11143: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11144: exit(1);
1.220 brouard 11145: }else{
1.226 brouard 11146: if(mle==1)
11147: printf("%1d%1d%d",i1,j1,jk);
11148: }
11149: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11150: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11151: for(j=1; j <=i; j++){
1.226 brouard 11152: fscanf(ficpar," %le",&matcov[i][j]);
11153: if(mle==1){
11154: printf(" %.5le",matcov[i][j]);
11155: }
11156: fprintf(ficlog," %.5le",matcov[i][j]);
11157: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11158: }
11159: fscanf(ficpar,"\n");
11160: numlinepar++;
11161: if(mle==1)
1.220 brouard 11162: printf("\n");
1.126 brouard 11163: fprintf(ficlog,"\n");
11164: fprintf(ficparo,"\n");
11165: }
1.194 brouard 11166: /* End of read covariance matrix npar lines */
1.126 brouard 11167: for(i=1; i <=npar; i++)
11168: for(j=i+1;j<=npar;j++)
1.226 brouard 11169: matcov[i][j]=matcov[j][i];
1.126 brouard 11170:
11171: if(mle==1)
11172: printf("\n");
11173: fprintf(ficlog,"\n");
11174:
11175: fflush(ficlog);
11176:
11177: /*-------- Rewriting parameter file ----------*/
11178: strcpy(rfileres,"r"); /* "Rparameterfile */
11179: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
11180: strcat(rfileres,"."); /* */
11181: strcat(rfileres,optionfilext); /* Other files have txt extension */
11182: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 11183: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11184: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 11185: }
11186: fprintf(ficres,"#%s\n",version);
11187: } /* End of mle != -3 */
1.218 brouard 11188:
1.186 brouard 11189: /* Main data
11190: */
1.126 brouard 11191: n= lastobs;
11192: num=lvector(1,n);
11193: moisnais=vector(1,n);
11194: annais=vector(1,n);
11195: moisdc=vector(1,n);
11196: andc=vector(1,n);
1.220 brouard 11197: weight=vector(1,n);
1.126 brouard 11198: agedc=vector(1,n);
11199: cod=ivector(1,n);
1.220 brouard 11200: for(i=1;i<=n;i++){
1.234 brouard 11201: num[i]=0;
11202: moisnais[i]=0;
11203: annais[i]=0;
11204: moisdc[i]=0;
11205: andc[i]=0;
11206: agedc[i]=0;
11207: cod[i]=0;
11208: weight[i]=1.0; /* Equal weights, 1 by default */
11209: }
1.126 brouard 11210: mint=matrix(1,maxwav,1,n);
11211: anint=matrix(1,maxwav,1,n);
1.131 brouard 11212: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11213: tab=ivector(1,NCOVMAX);
1.144 brouard 11214: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11215: 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 11216:
1.136 brouard 11217: /* Reads data from file datafile */
11218: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11219: goto end;
11220:
11221: /* Calculation of the number of parameters from char model */
1.234 brouard 11222: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11223: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11224: k=3 V4 Tvar[k=3]= 4 (from V4)
11225: k=2 V1 Tvar[k=2]= 1 (from V1)
11226: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11227: */
11228:
11229: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11230: TvarsDind=ivector(1,NCOVMAX); /* */
11231: TvarsD=ivector(1,NCOVMAX); /* */
11232: TvarsQind=ivector(1,NCOVMAX); /* */
11233: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11234: TvarF=ivector(1,NCOVMAX); /* */
11235: TvarFind=ivector(1,NCOVMAX); /* */
11236: TvarV=ivector(1,NCOVMAX); /* */
11237: TvarVind=ivector(1,NCOVMAX); /* */
11238: TvarA=ivector(1,NCOVMAX); /* */
11239: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11240: TvarFD=ivector(1,NCOVMAX); /* */
11241: TvarFDind=ivector(1,NCOVMAX); /* */
11242: TvarFQ=ivector(1,NCOVMAX); /* */
11243: TvarFQind=ivector(1,NCOVMAX); /* */
11244: TvarVD=ivector(1,NCOVMAX); /* */
11245: TvarVDind=ivector(1,NCOVMAX); /* */
11246: TvarVQ=ivector(1,NCOVMAX); /* */
11247: TvarVQind=ivector(1,NCOVMAX); /* */
11248:
1.230 brouard 11249: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11250: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11251: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11252: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11253: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11254: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11255: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11256: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11257: */
11258: /* For model-covariate k tells which data-covariate to use but
11259: because this model-covariate is a construction we invent a new column
11260: ncovcol + k1
11261: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11262: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11263: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11264: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11265: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11266: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11267: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11268: */
1.145 brouard 11269: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11270: 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 11271: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11272: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11273: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11274: 4 covariates (3 plus signs)
11275: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11276: */
1.230 brouard 11277: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11278: * individual dummy, fixed or varying:
11279: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11280: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11281: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11282: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11283: * Tmodelind[1]@9={9,0,3,2,}*/
11284: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11285: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11286: * individual quantitative, fixed or varying:
11287: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11288: * 3, 1, 0, 0, 0, 0, 0, 0},
11289: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11290: /* Main decodemodel */
11291:
1.187 brouard 11292:
1.223 brouard 11293: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11294: goto end;
11295:
1.137 brouard 11296: if((double)(lastobs-imx)/(double)imx > 1.10){
11297: nbwarn++;
11298: 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);
11299: 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);
11300: }
1.136 brouard 11301: /* if(mle==1){*/
1.137 brouard 11302: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11303: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11304: }
11305:
11306: /*-calculation of age at interview from date of interview and age at death -*/
11307: agev=matrix(1,maxwav,1,imx);
11308:
11309: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11310: goto end;
11311:
1.126 brouard 11312:
1.136 brouard 11313: agegomp=(int)agemin;
11314: free_vector(moisnais,1,n);
11315: free_vector(annais,1,n);
1.126 brouard 11316: /* free_matrix(mint,1,maxwav,1,n);
11317: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11318: /* free_vector(moisdc,1,n); */
11319: /* free_vector(andc,1,n); */
1.145 brouard 11320: /* */
11321:
1.126 brouard 11322: wav=ivector(1,imx);
1.214 brouard 11323: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11324: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11325: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11326: 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.*/
11327: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11328: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11329:
11330: /* Concatenates waves */
1.214 brouard 11331: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11332: Death is a valid wave (if date is known).
11333: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11334: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11335: and mw[mi+1][i]. dh depends on stepm.
11336: */
11337:
1.126 brouard 11338: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11339: /* Concatenates waves */
1.145 brouard 11340:
1.215 brouard 11341: free_vector(moisdc,1,n);
11342: free_vector(andc,1,n);
11343:
1.126 brouard 11344: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11345: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11346: ncodemax[1]=1;
1.145 brouard 11347: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11348: cptcoveff=0;
1.220 brouard 11349: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11350: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11351: }
11352:
11353: ncovcombmax=pow(2,cptcoveff);
11354: invalidvarcomb=ivector(1, ncovcombmax);
11355: for(i=1;i<ncovcombmax;i++)
11356: invalidvarcomb[i]=0;
11357:
1.211 brouard 11358: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11359: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11360: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11361:
1.200 brouard 11362: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11363: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11364: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11365: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11366: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11367: * (currently 0 or 1) in the data.
11368: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11369: * corresponding modality (h,j).
11370: */
11371:
1.145 brouard 11372: h=0;
11373: /*if (cptcovn > 0) */
1.126 brouard 11374: m=pow(2,cptcoveff);
11375:
1.144 brouard 11376: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11377: * For k=4 covariates, h goes from 1 to m=2**k
11378: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11379: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11380: * h\k 1 2 3 4
1.143 brouard 11381: *______________________________
11382: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11383: * 2 2 1 1 1
11384: * 3 i=2 1 2 1 1
11385: * 4 2 2 1 1
11386: * 5 i=3 1 i=2 1 2 1
11387: * 6 2 1 2 1
11388: * 7 i=4 1 2 2 1
11389: * 8 2 2 2 1
1.197 brouard 11390: * 9 i=5 1 i=3 1 i=2 1 2
11391: * 10 2 1 1 2
11392: * 11 i=6 1 2 1 2
11393: * 12 2 2 1 2
11394: * 13 i=7 1 i=4 1 2 2
11395: * 14 2 1 2 2
11396: * 15 i=8 1 2 2 2
11397: * 16 2 2 2 2
1.143 brouard 11398: */
1.212 brouard 11399: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11400: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11401: * and the value of each covariate?
11402: * V1=1, V2=1, V3=2, V4=1 ?
11403: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11404: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11405: * In order to get the real value in the data, we use nbcode
11406: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11407: * We are keeping this crazy system in order to be able (in the future?)
11408: * to have more than 2 values (0 or 1) for a covariate.
11409: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11410: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11411: * bbbbbbbb
11412: * 76543210
11413: * h-1 00000101 (6-1=5)
1.219 brouard 11414: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11415: * &
11416: * 1 00000001 (1)
1.219 brouard 11417: * 00000000 = 1 & ((h-1) >> (k-1))
11418: * +1= 00000001 =1
1.211 brouard 11419: *
11420: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11421: * h' 1101 =2^3+2^2+0x2^1+2^0
11422: * >>k' 11
11423: * & 00000001
11424: * = 00000001
11425: * +1 = 00000010=2 = codtabm(14,3)
11426: * Reverse h=6 and m=16?
11427: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11428: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11429: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11430: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11431: * V3=decodtabm(14,3,2**4)=2
11432: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11433: *(h-1) >> (j-1) 0011 =13 >> 2
11434: * &1 000000001
11435: * = 000000001
11436: * +1= 000000010 =2
11437: * 2211
11438: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11439: * V3=2
1.220 brouard 11440: * codtabm and decodtabm are identical
1.211 brouard 11441: */
11442:
1.145 brouard 11443:
11444: free_ivector(Ndum,-1,NCOVMAX);
11445:
11446:
1.126 brouard 11447:
1.186 brouard 11448: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11449: strcpy(optionfilegnuplot,optionfilefiname);
11450: if(mle==-3)
1.201 brouard 11451: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11452: strcat(optionfilegnuplot,".gp");
11453:
11454: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11455: printf("Problem with file %s",optionfilegnuplot);
11456: }
11457: else{
1.204 brouard 11458: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11459: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11460: //fprintf(ficgp,"set missing 'NaNq'\n");
11461: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11462: }
11463: /* fclose(ficgp);*/
1.186 brouard 11464:
11465:
11466: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11467:
11468: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11469: if(mle==-3)
1.201 brouard 11470: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11471: strcat(optionfilehtm,".htm");
11472: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11473: printf("Problem with %s \n",optionfilehtm);
11474: exit(0);
1.126 brouard 11475: }
11476:
11477: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11478: strcat(optionfilehtmcov,"-cov.htm");
11479: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11480: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11481: }
11482: else{
11483: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11484: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11485: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11486: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11487: }
11488:
1.213 brouard 11489: 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 11490: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11491: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11492: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11493: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11494: \n\
11495: <hr size=\"2\" color=\"#EC5E5E\">\
11496: <ul><li><h4>Parameter files</h4>\n\
11497: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11498: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11499: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11500: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11501: - Date and time at start: %s</ul>\n",\
11502: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11503: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11504: fileres,fileres,\
11505: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11506: fflush(fichtm);
11507:
11508: strcpy(pathr,path);
11509: strcat(pathr,optionfilefiname);
1.184 brouard 11510: #ifdef WIN32
11511: _chdir(optionfilefiname); /* Move to directory named optionfile */
11512: #else
1.126 brouard 11513: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11514: #endif
11515:
1.126 brouard 11516:
1.220 brouard 11517: /* Calculates basic frequencies. Computes observed prevalence at single age
11518: and for any valid combination of covariates
1.126 brouard 11519: and prints on file fileres'p'. */
1.251 brouard 11520: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11521: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11522:
11523: fprintf(fichtm,"\n");
1.274 brouard 11524: 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",\
11525: ftol, stepm);
11526: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11527: ncurrv=1;
11528: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11529: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11530: ncurrv=i;
11531: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
11532: fprintf(fichtm,"\n<li> Number of time varying (wave varying) covariates: ntv=%d ", ntv);
11533: ncurrv=i;
11534: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
11535: fprintf(fichtm,"\n<li>Number of quantitative time varying covariates: nqtv=%d ", nqtv);
11536: ncurrv=i;
11537: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11538: 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", \
11539: nlstate, ndeath, maxwav, mle, weightopt);
11540:
11541: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11542: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11543:
11544:
11545: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11546: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11547: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11548: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11549: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11550: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11551: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11552: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11553: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11554:
1.126 brouard 11555: /* For Powell, parameters are in a vector p[] starting at p[1]
11556: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11557: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11558:
11559: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11560: /* For mortality only */
1.126 brouard 11561: if (mle==-3){
1.136 brouard 11562: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11563: for(i=1;i<=NDIM;i++)
11564: for(j=1;j<=NDIM;j++)
11565: ximort[i][j]=0.;
1.186 brouard 11566: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11567: cens=ivector(1,n);
11568: ageexmed=vector(1,n);
11569: agecens=vector(1,n);
11570: dcwave=ivector(1,n);
1.223 brouard 11571:
1.126 brouard 11572: for (i=1; i<=imx; i++){
11573: dcwave[i]=-1;
11574: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11575: if (s[m][i]>nlstate) {
11576: dcwave[i]=m;
11577: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11578: break;
11579: }
1.126 brouard 11580: }
1.226 brouard 11581:
1.126 brouard 11582: for (i=1; i<=imx; i++) {
11583: if (wav[i]>0){
1.226 brouard 11584: ageexmed[i]=agev[mw[1][i]][i];
11585: j=wav[i];
11586: agecens[i]=1.;
11587:
11588: if (ageexmed[i]> 1 && wav[i] > 0){
11589: agecens[i]=agev[mw[j][i]][i];
11590: cens[i]= 1;
11591: }else if (ageexmed[i]< 1)
11592: cens[i]= -1;
11593: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11594: cens[i]=0 ;
1.126 brouard 11595: }
11596: else cens[i]=-1;
11597: }
11598:
11599: for (i=1;i<=NDIM;i++) {
11600: for (j=1;j<=NDIM;j++)
1.226 brouard 11601: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11602: }
11603:
1.145 brouard 11604: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11605: /*printf("%lf %lf", p[1], p[2]);*/
11606:
11607:
1.136 brouard 11608: #ifdef GSL
11609: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11610: #else
1.126 brouard 11611: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11612: #endif
1.201 brouard 11613: strcpy(filerespow,"POW-MORT_");
11614: strcat(filerespow,fileresu);
1.126 brouard 11615: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11616: printf("Problem with resultfile: %s\n", filerespow);
11617: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11618: }
1.136 brouard 11619: #ifdef GSL
11620: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11621: #else
1.126 brouard 11622: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11623: #endif
1.126 brouard 11624: /* for (i=1;i<=nlstate;i++)
11625: for(j=1;j<=nlstate+ndeath;j++)
11626: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11627: */
11628: fprintf(ficrespow,"\n");
1.136 brouard 11629: #ifdef GSL
11630: /* gsl starts here */
11631: T = gsl_multimin_fminimizer_nmsimplex;
11632: gsl_multimin_fminimizer *sfm = NULL;
11633: gsl_vector *ss, *x;
11634: gsl_multimin_function minex_func;
11635:
11636: /* Initial vertex size vector */
11637: ss = gsl_vector_alloc (NDIM);
11638:
11639: if (ss == NULL){
11640: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11641: }
11642: /* Set all step sizes to 1 */
11643: gsl_vector_set_all (ss, 0.001);
11644:
11645: /* Starting point */
1.126 brouard 11646:
1.136 brouard 11647: x = gsl_vector_alloc (NDIM);
11648:
11649: if (x == NULL){
11650: gsl_vector_free(ss);
11651: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11652: }
11653:
11654: /* Initialize method and iterate */
11655: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11656: /* gsl_vector_set(x, 0, 0.0268); */
11657: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11658: gsl_vector_set(x, 0, p[1]);
11659: gsl_vector_set(x, 1, p[2]);
11660:
11661: minex_func.f = &gompertz_f;
11662: minex_func.n = NDIM;
11663: minex_func.params = (void *)&p; /* ??? */
11664:
11665: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11666: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11667:
11668: printf("Iterations beginning .....\n\n");
11669: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11670:
11671: iteri=0;
11672: while (rval == GSL_CONTINUE){
11673: iteri++;
11674: status = gsl_multimin_fminimizer_iterate(sfm);
11675:
11676: if (status) printf("error: %s\n", gsl_strerror (status));
11677: fflush(0);
11678:
11679: if (status)
11680: break;
11681:
11682: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11683: ssval = gsl_multimin_fminimizer_size (sfm);
11684:
11685: if (rval == GSL_SUCCESS)
11686: printf ("converged to a local maximum at\n");
11687:
11688: printf("%5d ", iteri);
11689: for (it = 0; it < NDIM; it++){
11690: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11691: }
11692: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11693: }
11694:
11695: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11696:
11697: gsl_vector_free(x); /* initial values */
11698: gsl_vector_free(ss); /* inital step size */
11699: for (it=0; it<NDIM; it++){
11700: p[it+1]=gsl_vector_get(sfm->x,it);
11701: fprintf(ficrespow," %.12lf", p[it]);
11702: }
11703: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11704: #endif
11705: #ifdef POWELL
11706: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11707: #endif
1.126 brouard 11708: fclose(ficrespow);
11709:
1.203 brouard 11710: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11711:
11712: for(i=1; i <=NDIM; i++)
11713: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11714: matcov[i][j]=matcov[j][i];
1.126 brouard 11715:
11716: printf("\nCovariance matrix\n ");
1.203 brouard 11717: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11718: for(i=1; i <=NDIM; i++) {
11719: for(j=1;j<=NDIM;j++){
1.220 brouard 11720: printf("%f ",matcov[i][j]);
11721: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11722: }
1.203 brouard 11723: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11724: }
11725:
11726: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11727: for (i=1;i<=NDIM;i++) {
1.126 brouard 11728: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11729: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11730: }
1.126 brouard 11731: lsurv=vector(1,AGESUP);
11732: lpop=vector(1,AGESUP);
11733: tpop=vector(1,AGESUP);
11734: lsurv[agegomp]=100000;
11735:
11736: for (k=agegomp;k<=AGESUP;k++) {
11737: agemortsup=k;
11738: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11739: }
11740:
11741: for (k=agegomp;k<agemortsup;k++)
11742: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11743:
11744: for (k=agegomp;k<agemortsup;k++){
11745: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11746: sumlpop=sumlpop+lpop[k];
11747: }
11748:
11749: tpop[agegomp]=sumlpop;
11750: for (k=agegomp;k<(agemortsup-3);k++){
11751: /* tpop[k+1]=2;*/
11752: tpop[k+1]=tpop[k]-lpop[k];
11753: }
11754:
11755:
11756: printf("\nAge lx qx dx Lx Tx e(x)\n");
11757: for (k=agegomp;k<(agemortsup-2);k++)
11758: 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]);
11759:
11760:
11761: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11762: ageminpar=50;
11763: agemaxpar=100;
1.194 brouard 11764: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11765: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11766: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11767: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11768: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11769: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11770: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11771: }else{
11772: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11773: 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 11774: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11775: }
1.201 brouard 11776: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11777: stepm, weightopt,\
11778: model,imx,p,matcov,agemortsup);
11779:
11780: free_vector(lsurv,1,AGESUP);
11781: free_vector(lpop,1,AGESUP);
11782: free_vector(tpop,1,AGESUP);
1.220 brouard 11783: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11784: free_ivector(cens,1,n);
11785: free_vector(agecens,1,n);
11786: free_ivector(dcwave,1,n);
1.220 brouard 11787: #ifdef GSL
1.136 brouard 11788: #endif
1.186 brouard 11789: } /* Endof if mle==-3 mortality only */
1.205 brouard 11790: /* Standard */
11791: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11792: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11793: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11794: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11795: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11796: for (k=1; k<=npar;k++)
11797: printf(" %d %8.5f",k,p[k]);
11798: printf("\n");
1.205 brouard 11799: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11800: /* mlikeli uses func not funcone */
1.247 brouard 11801: /* for(i=1;i<nlstate;i++){ */
11802: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11803: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11804: /* } */
1.205 brouard 11805: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11806: }
11807: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11808: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11809: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11810: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11811: }
11812: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11813: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11814: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11815: for (k=1; k<=npar;k++)
11816: printf(" %d %8.5f",k,p[k]);
11817: printf("\n");
11818:
11819: /*--------- results files --------------*/
1.224 brouard 11820: 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 11821:
11822:
11823: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11824: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11825: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11826: for(i=1,jk=1; i <=nlstate; i++){
11827: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11828: if (k != i) {
11829: printf("%d%d ",i,k);
11830: fprintf(ficlog,"%d%d ",i,k);
11831: fprintf(ficres,"%1d%1d ",i,k);
11832: for(j=1; j <=ncovmodel; j++){
11833: printf("%12.7f ",p[jk]);
11834: fprintf(ficlog,"%12.7f ",p[jk]);
11835: fprintf(ficres,"%12.7f ",p[jk]);
11836: jk++;
11837: }
11838: printf("\n");
11839: fprintf(ficlog,"\n");
11840: fprintf(ficres,"\n");
11841: }
1.126 brouard 11842: }
11843: }
1.203 brouard 11844: if(mle != 0){
11845: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11846: ftolhess=ftol; /* Usually correct */
1.203 brouard 11847: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11848: 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");
11849: 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");
11850: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11851: for(k=1; k <=(nlstate+ndeath); k++){
11852: if (k != i) {
11853: printf("%d%d ",i,k);
11854: fprintf(ficlog,"%d%d ",i,k);
11855: for(j=1; j <=ncovmodel; j++){
11856: 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]));
11857: 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]));
11858: jk++;
11859: }
11860: printf("\n");
11861: fprintf(ficlog,"\n");
11862: }
11863: }
1.193 brouard 11864: }
1.203 brouard 11865: } /* end of hesscov and Wald tests */
1.225 brouard 11866:
1.203 brouard 11867: /* */
1.126 brouard 11868: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11869: printf("# Scales (for hessian or gradient estimation)\n");
11870: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11871: for(i=1,jk=1; i <=nlstate; i++){
11872: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11873: if (j!=i) {
11874: fprintf(ficres,"%1d%1d",i,j);
11875: printf("%1d%1d",i,j);
11876: fprintf(ficlog,"%1d%1d",i,j);
11877: for(k=1; k<=ncovmodel;k++){
11878: printf(" %.5e",delti[jk]);
11879: fprintf(ficlog," %.5e",delti[jk]);
11880: fprintf(ficres," %.5e",delti[jk]);
11881: jk++;
11882: }
11883: printf("\n");
11884: fprintf(ficlog,"\n");
11885: fprintf(ficres,"\n");
11886: }
1.126 brouard 11887: }
11888: }
11889:
11890: 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 11891: if(mle >= 1) /* To big for the screen */
1.126 brouard 11892: 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");
11893: 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");
11894: /* # 121 Var(a12)\n\ */
11895: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11896: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11897: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11898: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11899: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11900: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11901: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11902:
11903:
11904: /* Just to have a covariance matrix which will be more understandable
11905: even is we still don't want to manage dictionary of variables
11906: */
11907: for(itimes=1;itimes<=2;itimes++){
11908: jj=0;
11909: for(i=1; i <=nlstate; i++){
1.225 brouard 11910: for(j=1; j <=nlstate+ndeath; j++){
11911: if(j==i) continue;
11912: for(k=1; k<=ncovmodel;k++){
11913: jj++;
11914: ca[0]= k+'a'-1;ca[1]='\0';
11915: if(itimes==1){
11916: if(mle>=1)
11917: printf("#%1d%1d%d",i,j,k);
11918: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11919: fprintf(ficres,"#%1d%1d%d",i,j,k);
11920: }else{
11921: if(mle>=1)
11922: printf("%1d%1d%d",i,j,k);
11923: fprintf(ficlog,"%1d%1d%d",i,j,k);
11924: fprintf(ficres,"%1d%1d%d",i,j,k);
11925: }
11926: ll=0;
11927: for(li=1;li <=nlstate; li++){
11928: for(lj=1;lj <=nlstate+ndeath; lj++){
11929: if(lj==li) continue;
11930: for(lk=1;lk<=ncovmodel;lk++){
11931: ll++;
11932: if(ll<=jj){
11933: cb[0]= lk +'a'-1;cb[1]='\0';
11934: if(ll<jj){
11935: if(itimes==1){
11936: if(mle>=1)
11937: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11938: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11939: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11940: }else{
11941: if(mle>=1)
11942: printf(" %.5e",matcov[jj][ll]);
11943: fprintf(ficlog," %.5e",matcov[jj][ll]);
11944: fprintf(ficres," %.5e",matcov[jj][ll]);
11945: }
11946: }else{
11947: if(itimes==1){
11948: if(mle>=1)
11949: printf(" Var(%s%1d%1d)",ca,i,j);
11950: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11951: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11952: }else{
11953: if(mle>=1)
11954: printf(" %.7e",matcov[jj][ll]);
11955: fprintf(ficlog," %.7e",matcov[jj][ll]);
11956: fprintf(ficres," %.7e",matcov[jj][ll]);
11957: }
11958: }
11959: }
11960: } /* end lk */
11961: } /* end lj */
11962: } /* end li */
11963: if(mle>=1)
11964: printf("\n");
11965: fprintf(ficlog,"\n");
11966: fprintf(ficres,"\n");
11967: numlinepar++;
11968: } /* end k*/
11969: } /*end j */
1.126 brouard 11970: } /* end i */
11971: } /* end itimes */
11972:
11973: fflush(ficlog);
11974: fflush(ficres);
1.225 brouard 11975: while(fgets(line, MAXLINE, ficpar)) {
11976: /* If line starts with a # it is a comment */
11977: if (line[0] == '#') {
11978: numlinepar++;
11979: fputs(line,stdout);
11980: fputs(line,ficparo);
11981: fputs(line,ficlog);
11982: continue;
11983: }else
11984: break;
11985: }
11986:
1.209 brouard 11987: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11988: /* ungetc(c,ficpar); */
11989: /* fgets(line, MAXLINE, ficpar); */
11990: /* fputs(line,stdout); */
11991: /* fputs(line,ficparo); */
11992: /* } */
11993: /* ungetc(c,ficpar); */
1.126 brouard 11994:
11995: estepm=0;
1.209 brouard 11996: 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 11997:
11998: if (num_filled != 6) {
11999: 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);
12000: 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);
12001: goto end;
12002: }
12003: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12004: }
12005: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12006: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12007:
1.209 brouard 12008: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12009: if (estepm==0 || estepm < stepm) estepm=stepm;
12010: if (fage <= 2) {
12011: bage = ageminpar;
12012: fage = agemaxpar;
12013: }
12014:
12015: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12016: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12017: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12018:
1.186 brouard 12019: /* Other stuffs, more or less useful */
1.254 brouard 12020: while(fgets(line, MAXLINE, ficpar)) {
12021: /* If line starts with a # it is a comment */
12022: if (line[0] == '#') {
12023: numlinepar++;
12024: fputs(line,stdout);
12025: fputs(line,ficparo);
12026: fputs(line,ficlog);
12027: continue;
12028: }else
12029: break;
12030: }
12031:
12032: 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){
12033:
12034: if (num_filled != 7) {
12035: 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);
12036: 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);
12037: goto end;
12038: }
12039: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12040: 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);
12041: 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);
12042: 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 12043: }
1.254 brouard 12044:
12045: while(fgets(line, MAXLINE, ficpar)) {
12046: /* If line starts with a # it is a comment */
12047: if (line[0] == '#') {
12048: numlinepar++;
12049: fputs(line,stdout);
12050: fputs(line,ficparo);
12051: fputs(line,ficlog);
12052: continue;
12053: }else
12054: break;
1.126 brouard 12055: }
12056:
12057:
12058: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12059: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12060:
1.254 brouard 12061: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12062: if (num_filled != 1) {
12063: 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);
12064: 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);
12065: goto end;
12066: }
12067: printf("pop_based=%d\n",popbased);
12068: fprintf(ficlog,"pop_based=%d\n",popbased);
12069: fprintf(ficparo,"pop_based=%d\n",popbased);
12070: fprintf(ficres,"pop_based=%d\n",popbased);
12071: }
12072:
1.258 brouard 12073: /* Results */
12074: nresult=0;
12075: do{
12076: if(!fgets(line, MAXLINE, ficpar)){
12077: endishere=1;
12078: parameterline=14;
12079: }else if (line[0] == '#') {
12080: /* If line starts with a # it is a comment */
1.254 brouard 12081: numlinepar++;
12082: fputs(line,stdout);
12083: fputs(line,ficparo);
12084: fputs(line,ficlog);
12085: continue;
1.258 brouard 12086: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12087: parameterline=11;
12088: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12089: parameterline=12;
12090: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12091: parameterline=13;
12092: else{
12093: parameterline=14;
1.254 brouard 12094: }
1.258 brouard 12095: switch (parameterline){
12096: case 11:
12097: 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){
12098: if (num_filled != 8) {
12099: 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);
12100: 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);
12101: goto end;
12102: }
12103: 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);
12104: 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);
12105: 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);
12106: 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);
12107: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12108: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12109: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12110:
1.258 brouard 12111: }
1.254 brouard 12112: break;
1.258 brouard 12113: case 12:
12114: /*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);*/
12115: 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){
12116: if (num_filled != 8) {
1.262 brouard 12117: 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);
12118: 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 12119: goto end;
12120: }
12121: 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);
12122: 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);
12123: 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);
12124: 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);
12125: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12126: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12127: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12128: }
1.230 brouard 12129: break;
1.258 brouard 12130: case 13:
12131: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12132: if (num_filled == 0){
12133: resultline[0]='\0';
12134: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12135: 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);
12136: break;
12137: } else if (num_filled != 1){
12138: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12139: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12140: }
12141: nresult++; /* Sum of resultlines */
12142: printf("Result %d: result=%s\n",nresult, resultline);
12143: if(nresult > MAXRESULTLINES){
12144: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12145: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12146: goto end;
12147: }
12148: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12149: fprintf(ficparo,"result: %s\n",resultline);
12150: fprintf(ficres,"result: %s\n",resultline);
12151: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12152: break;
1.258 brouard 12153: case 14:
1.259 brouard 12154: if(ncovmodel >2 && nresult==0 ){
12155: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12156: goto end;
12157: }
1.259 brouard 12158: break;
1.258 brouard 12159: default:
12160: nresult=1;
12161: decoderesult(".",nresult ); /* No covariate */
12162: }
12163: } /* End switch parameterline */
12164: }while(endishere==0); /* End do */
1.126 brouard 12165:
1.230 brouard 12166: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12167: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12168:
12169: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12170: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12171: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12172: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12173: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12174: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12175: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12176: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12177: }else{
1.270 brouard 12178: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12179: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12180: }
12181: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12182: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12183: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12184:
1.225 brouard 12185: /*------------ free_vector -------------*/
12186: /* chdir(path); */
1.220 brouard 12187:
1.215 brouard 12188: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12189: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12190: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12191: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12192: free_lvector(num,1,n);
12193: free_vector(agedc,1,n);
12194: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12195: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12196: fclose(ficparo);
12197: fclose(ficres);
1.220 brouard 12198:
12199:
1.186 brouard 12200: /* Other results (useful)*/
1.220 brouard 12201:
12202:
1.126 brouard 12203: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12204: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12205: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12206: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12207: fclose(ficrespl);
12208:
12209: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12210: /*#include "hpijx.h"*/
12211: hPijx(p, bage, fage);
1.145 brouard 12212: fclose(ficrespij);
1.227 brouard 12213:
1.220 brouard 12214: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12215: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12216: k=1;
1.126 brouard 12217: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12218:
1.269 brouard 12219: /* Prevalence for each covariate combination in probs[age][status][cov] */
12220: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12221: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12222: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12223: for(k=1;k<=ncovcombmax;k++)
12224: probs[i][j][k]=0.;
1.269 brouard 12225: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12226: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12227: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12228: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12229: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12230: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12231: for(k=1;k<=ncovcombmax;k++)
12232: mobaverages[i][j][k]=0.;
1.219 brouard 12233: mobaverage=mobaverages;
12234: if (mobilav!=0) {
1.235 brouard 12235: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12236: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12237: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12238: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12239: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12240: }
1.269 brouard 12241: } else if (mobilavproj !=0) {
1.235 brouard 12242: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12243: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12244: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12245: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12246: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12247: }
1.269 brouard 12248: }else{
12249: printf("Internal error moving average\n");
12250: fflush(stdout);
12251: exit(1);
1.219 brouard 12252: }
12253: }/* end if moving average */
1.227 brouard 12254:
1.126 brouard 12255: /*---------- Forecasting ------------------*/
12256: if(prevfcast==1){
12257: /* if(stepm ==1){*/
1.269 brouard 12258: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12259: }
1.269 brouard 12260:
12261: /* Backcasting */
1.217 brouard 12262: if(backcast==1){
1.219 brouard 12263: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12264: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12265: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12266:
12267: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12268:
12269: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12270:
1.219 brouard 12271: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12272: fclose(ficresplb);
12273:
1.222 brouard 12274: hBijx(p, bage, fage, mobaverage);
12275: fclose(ficrespijb);
1.219 brouard 12276:
1.269 brouard 12277: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12278: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12279: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12280:
12281:
1.269 brouard 12282: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12283: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12284: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12285: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12286: } /* end Backcasting */
1.268 brouard 12287:
1.186 brouard 12288:
12289: /* ------ Other prevalence ratios------------ */
1.126 brouard 12290:
1.215 brouard 12291: free_ivector(wav,1,imx);
12292: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12293: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12294: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12295:
12296:
1.127 brouard 12297: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12298:
1.201 brouard 12299: strcpy(filerese,"E_");
12300: strcat(filerese,fileresu);
1.126 brouard 12301: if((ficreseij=fopen(filerese,"w"))==NULL) {
12302: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12303: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12304: }
1.208 brouard 12305: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12306: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12307:
12308: pstamp(ficreseij);
1.219 brouard 12309:
1.235 brouard 12310: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12311: if (cptcovn < 1){i1=1;}
12312:
12313: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12314: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12315: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12316: continue;
1.219 brouard 12317: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12318: printf("\n#****** ");
1.225 brouard 12319: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12320: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12321: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12322: }
12323: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12324: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12325: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12326: }
12327: fprintf(ficreseij,"******\n");
1.235 brouard 12328: printf("******\n");
1.219 brouard 12329:
12330: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12331: oldm=oldms;savm=savms;
1.235 brouard 12332: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12333:
1.219 brouard 12334: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12335: }
12336: fclose(ficreseij);
1.208 brouard 12337: printf("done evsij\n");fflush(stdout);
12338: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12339:
1.218 brouard 12340:
1.227 brouard 12341: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12342:
1.201 brouard 12343: strcpy(filerest,"T_");
12344: strcat(filerest,fileresu);
1.127 brouard 12345: if((ficrest=fopen(filerest,"w"))==NULL) {
12346: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12347: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12348: }
1.208 brouard 12349: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12350: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12351: strcpy(fileresstde,"STDE_");
12352: strcat(fileresstde,fileresu);
1.126 brouard 12353: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12354: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12355: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12356: }
1.227 brouard 12357: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12358: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12359:
1.201 brouard 12360: strcpy(filerescve,"CVE_");
12361: strcat(filerescve,fileresu);
1.126 brouard 12362: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12363: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12364: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12365: }
1.227 brouard 12366: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12367: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12368:
1.201 brouard 12369: strcpy(fileresv,"V_");
12370: strcat(fileresv,fileresu);
1.126 brouard 12371: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12372: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12373: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12374: }
1.227 brouard 12375: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12376: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12377:
1.235 brouard 12378: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12379: if (cptcovn < 1){i1=1;}
12380:
12381: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12382: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12383: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12384: continue;
1.242 brouard 12385: printf("\n#****** Result for:");
12386: fprintf(ficrest,"\n#****** Result for:");
12387: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12388: for(j=1;j<=cptcoveff;j++){
12389: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12390: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12391: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12392: }
1.235 brouard 12393: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12394: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12395: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12396: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12397: }
1.208 brouard 12398: fprintf(ficrest,"******\n");
1.227 brouard 12399: fprintf(ficlog,"******\n");
12400: printf("******\n");
1.208 brouard 12401:
12402: fprintf(ficresstdeij,"\n#****** ");
12403: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12404: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12405: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12406: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12407: }
1.235 brouard 12408: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12409: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12410: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12411: }
1.208 brouard 12412: fprintf(ficresstdeij,"******\n");
12413: fprintf(ficrescveij,"******\n");
12414:
12415: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12416: /* pstamp(ficresvij); */
1.225 brouard 12417: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12418: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12419: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12420: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12421: }
1.208 brouard 12422: fprintf(ficresvij,"******\n");
12423:
12424: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12425: oldm=oldms;savm=savms;
1.235 brouard 12426: printf(" cvevsij ");
12427: fprintf(ficlog, " cvevsij ");
12428: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12429: printf(" end cvevsij \n ");
12430: fprintf(ficlog, " end cvevsij \n ");
12431:
12432: /*
12433: */
12434: /* goto endfree; */
12435:
12436: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12437: pstamp(ficrest);
12438:
1.269 brouard 12439: epj=vector(1,nlstate+1);
1.208 brouard 12440: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12441: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12442: cptcod= 0; /* To be deleted */
12443: printf("varevsij vpopbased=%d \n",vpopbased);
12444: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12445: 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 12446: 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 ");
12447: if(vpopbased==1)
12448: 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);
12449: else
12450: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12451: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12452: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12453: fprintf(ficrest,"\n");
12454: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12455: printf("Computing age specific period (stable) prevalences in each health state \n");
12456: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12457: for(age=bage; age <=fage ;age++){
1.235 brouard 12458: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12459: if (vpopbased==1) {
12460: if(mobilav ==0){
12461: for(i=1; i<=nlstate;i++)
12462: prlim[i][i]=probs[(int)age][i][k];
12463: }else{ /* mobilav */
12464: for(i=1; i<=nlstate;i++)
12465: prlim[i][i]=mobaverage[(int)age][i][k];
12466: }
12467: }
1.219 brouard 12468:
1.227 brouard 12469: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12470: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12471: /* printf(" age %4.0f ",age); */
12472: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12473: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12474: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12475: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12476: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12477: }
12478: epj[nlstate+1] +=epj[j];
12479: }
12480: /* printf(" age %4.0f \n",age); */
1.219 brouard 12481:
1.227 brouard 12482: for(i=1, vepp=0.;i <=nlstate;i++)
12483: for(j=1;j <=nlstate;j++)
12484: vepp += vareij[i][j][(int)age];
12485: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12486: for(j=1;j <=nlstate;j++){
12487: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12488: }
12489: fprintf(ficrest,"\n");
12490: }
1.208 brouard 12491: } /* End vpopbased */
1.269 brouard 12492: free_vector(epj,1,nlstate+1);
1.208 brouard 12493: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12494: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12495: printf("done selection\n");fflush(stdout);
12496: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12497:
1.235 brouard 12498: } /* End k selection */
1.227 brouard 12499:
12500: printf("done State-specific expectancies\n");fflush(stdout);
12501: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12502:
1.269 brouard 12503: /* variance-covariance of period prevalence*/
12504: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12505:
1.227 brouard 12506:
12507: free_vector(weight,1,n);
12508: free_imatrix(Tvard,1,NCOVMAX,1,2);
12509: free_imatrix(s,1,maxwav+1,1,n);
12510: free_matrix(anint,1,maxwav,1,n);
12511: free_matrix(mint,1,maxwav,1,n);
12512: free_ivector(cod,1,n);
12513: free_ivector(tab,1,NCOVMAX);
12514: fclose(ficresstdeij);
12515: fclose(ficrescveij);
12516: fclose(ficresvij);
12517: fclose(ficrest);
12518: fclose(ficpar);
12519:
12520:
1.126 brouard 12521: /*---------- End : free ----------------*/
1.219 brouard 12522: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12523: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12524: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12525: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12526: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12527: } /* mle==-3 arrives here for freeing */
1.227 brouard 12528: /* endfree:*/
12529: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12530: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12531: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12532: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12533: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12534: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12535: free_matrix(covar,0,NCOVMAX,1,n);
12536: free_matrix(matcov,1,npar,1,npar);
12537: free_matrix(hess,1,npar,1,npar);
12538: /*free_vector(delti,1,npar);*/
12539: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12540: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12541: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12542: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12543:
12544: free_ivector(ncodemax,1,NCOVMAX);
12545: free_ivector(ncodemaxwundef,1,NCOVMAX);
12546: free_ivector(Dummy,-1,NCOVMAX);
12547: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12548: free_ivector(DummyV,1,NCOVMAX);
12549: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12550: free_ivector(Typevar,-1,NCOVMAX);
12551: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12552: free_ivector(TvarsQ,1,NCOVMAX);
12553: free_ivector(TvarsQind,1,NCOVMAX);
12554: free_ivector(TvarsD,1,NCOVMAX);
12555: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12556: free_ivector(TvarFD,1,NCOVMAX);
12557: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12558: free_ivector(TvarF,1,NCOVMAX);
12559: free_ivector(TvarFind,1,NCOVMAX);
12560: free_ivector(TvarV,1,NCOVMAX);
12561: free_ivector(TvarVind,1,NCOVMAX);
12562: free_ivector(TvarA,1,NCOVMAX);
12563: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12564: free_ivector(TvarFQ,1,NCOVMAX);
12565: free_ivector(TvarFQind,1,NCOVMAX);
12566: free_ivector(TvarVD,1,NCOVMAX);
12567: free_ivector(TvarVDind,1,NCOVMAX);
12568: free_ivector(TvarVQ,1,NCOVMAX);
12569: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12570: free_ivector(Tvarsel,1,NCOVMAX);
12571: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12572: free_ivector(Tposprod,1,NCOVMAX);
12573: free_ivector(Tprod,1,NCOVMAX);
12574: free_ivector(Tvaraff,1,NCOVMAX);
12575: free_ivector(invalidvarcomb,1,ncovcombmax);
12576: free_ivector(Tage,1,NCOVMAX);
12577: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12578: free_ivector(TmodelInvind,1,NCOVMAX);
12579: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12580:
12581: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12582: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12583: fflush(fichtm);
12584: fflush(ficgp);
12585:
1.227 brouard 12586:
1.126 brouard 12587: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12588: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12589: 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 12590: }else{
12591: printf("End of Imach\n");
12592: fprintf(ficlog,"End of Imach\n");
12593: }
12594: printf("See log file on %s\n",filelog);
12595: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12596: /*(void) gettimeofday(&end_time,&tzp);*/
12597: rend_time = time(NULL);
12598: end_time = *localtime(&rend_time);
12599: /* tml = *localtime(&end_time.tm_sec); */
12600: strcpy(strtend,asctime(&end_time));
1.126 brouard 12601: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12602: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12603: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12604:
1.157 brouard 12605: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12606: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12607: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12608: /* printf("Total time was %d uSec.\n", total_usecs);*/
12609: /* if(fileappend(fichtm,optionfilehtm)){ */
12610: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12611: fclose(fichtm);
12612: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12613: fclose(fichtmcov);
12614: fclose(ficgp);
12615: fclose(ficlog);
12616: /*------ End -----------*/
1.227 brouard 12617:
12618:
12619: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12620: #ifdef WIN32
1.227 brouard 12621: if (_chdir(pathcd) != 0)
12622: printf("Can't move to directory %s!\n",path);
12623: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12624: #else
1.227 brouard 12625: if(chdir(pathcd) != 0)
12626: printf("Can't move to directory %s!\n", path);
12627: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12628: #endif
1.126 brouard 12629: printf("Current directory %s!\n",pathcd);
12630: /*strcat(plotcmd,CHARSEPARATOR);*/
12631: sprintf(plotcmd,"gnuplot");
1.157 brouard 12632: #ifdef _WIN32
1.126 brouard 12633: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12634: #endif
12635: if(!stat(plotcmd,&info)){
1.158 brouard 12636: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12637: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12638: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12639: }else
12640: strcpy(pplotcmd,plotcmd);
1.157 brouard 12641: #ifdef __unix
1.126 brouard 12642: strcpy(plotcmd,GNUPLOTPROGRAM);
12643: if(!stat(plotcmd,&info)){
1.158 brouard 12644: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12645: }else
12646: strcpy(pplotcmd,plotcmd);
12647: #endif
12648: }else
12649: strcpy(pplotcmd,plotcmd);
12650:
12651: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12652: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12653:
1.126 brouard 12654: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12655: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12656: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12657: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12658: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12659: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12660: }
1.158 brouard 12661: printf(" Successful, please wait...");
1.126 brouard 12662: while (z[0] != 'q') {
12663: /* chdir(path); */
1.154 brouard 12664: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12665: scanf("%s",z);
12666: /* if (z[0] == 'c') system("./imach"); */
12667: if (z[0] == 'e') {
1.158 brouard 12668: #ifdef __APPLE__
1.152 brouard 12669: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12670: #elif __linux
12671: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12672: #else
1.152 brouard 12673: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12674: #endif
12675: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12676: system(pplotcmd);
1.126 brouard 12677: }
12678: else if (z[0] == 'g') system(plotcmd);
12679: else if (z[0] == 'q') exit(0);
12680: }
1.227 brouard 12681: end:
1.126 brouard 12682: while (z[0] != 'q') {
1.195 brouard 12683: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12684: scanf("%s",z);
12685: }
12686: }
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