Annotation of imach/src/imach.c, revision 1.274
1.274 ! brouard 1: /* $Id: imach.c,v 1.273 2017/06/27 11:06:02 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.274 ! brouard 4: Revision 1.273 2017/06/27 11:06:02 brouard
! 5: Summary: More documentation on projections
! 6:
1.273 brouard 7: Revision 1.272 2017/06/27 10:22:40 brouard
8: Summary: Color of backprojection changed from 6 to 5(yellow)
9:
1.272 brouard 10: Revision 1.271 2017/06/27 10:17:50 brouard
11: Summary: Some bug with rint
12:
1.271 brouard 13: Revision 1.270 2017/05/24 05:45:29 brouard
14: *** empty log message ***
15:
1.270 brouard 16: Revision 1.269 2017/05/23 08:39:25 brouard
17: Summary: Code into subroutine, cleanings
18:
1.269 brouard 19: Revision 1.268 2017/05/18 20:09:32 brouard
20: Summary: backprojection and confidence intervals of backprevalence
21:
1.268 brouard 22: Revision 1.267 2017/05/13 10:25:05 brouard
23: Summary: temporary save for backprojection
24:
1.267 brouard 25: Revision 1.266 2017/05/13 07:26:12 brouard
26: Summary: Version 0.99r13 (improvements and bugs fixed)
27:
1.266 brouard 28: Revision 1.265 2017/04/26 16:22:11 brouard
29: Summary: imach 0.99r13 Some bugs fixed
30:
1.265 brouard 31: Revision 1.264 2017/04/26 06:01:29 brouard
32: Summary: Labels in graphs
33:
1.264 brouard 34: Revision 1.263 2017/04/24 15:23:15 brouard
35: Summary: to save
36:
1.263 brouard 37: Revision 1.262 2017/04/18 16:48:12 brouard
38: *** empty log message ***
39:
1.262 brouard 40: Revision 1.261 2017/04/05 10:14:09 brouard
41: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
42:
1.261 brouard 43: Revision 1.260 2017/04/04 17:46:59 brouard
44: Summary: Gnuplot indexations fixed (humm)
45:
1.260 brouard 46: Revision 1.259 2017/04/04 13:01:16 brouard
47: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
48:
1.259 brouard 49: Revision 1.258 2017/04/03 10:17:47 brouard
50: Summary: Version 0.99r12
51:
52: Some cleanings, conformed with updated documentation.
53:
1.258 brouard 54: Revision 1.257 2017/03/29 16:53:30 brouard
55: Summary: Temp
56:
1.257 brouard 57: Revision 1.256 2017/03/27 05:50:23 brouard
58: Summary: Temporary
59:
1.256 brouard 60: Revision 1.255 2017/03/08 16:02:28 brouard
61: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
62:
1.255 brouard 63: Revision 1.254 2017/03/08 07:13:00 brouard
64: Summary: Fixing data parameter line
65:
1.254 brouard 66: Revision 1.253 2016/12/15 11:59:41 brouard
67: Summary: 0.99 in progress
68:
1.253 brouard 69: Revision 1.252 2016/09/15 21:15:37 brouard
70: *** empty log message ***
71:
1.252 brouard 72: Revision 1.251 2016/09/15 15:01:13 brouard
73: Summary: not working
74:
1.251 brouard 75: Revision 1.250 2016/09/08 16:07:27 brouard
76: Summary: continue
77:
1.250 brouard 78: Revision 1.249 2016/09/07 17:14:18 brouard
79: Summary: Starting values from frequencies
80:
1.249 brouard 81: Revision 1.248 2016/09/07 14:10:18 brouard
82: *** empty log message ***
83:
1.248 brouard 84: Revision 1.247 2016/09/02 11:11:21 brouard
85: *** empty log message ***
86:
1.247 brouard 87: Revision 1.246 2016/09/02 08:49:22 brouard
88: *** empty log message ***
89:
1.246 brouard 90: Revision 1.245 2016/09/02 07:25:01 brouard
91: *** empty log message ***
92:
1.245 brouard 93: Revision 1.244 2016/09/02 07:17:34 brouard
94: *** empty log message ***
95:
1.244 brouard 96: Revision 1.243 2016/09/02 06:45:35 brouard
97: *** empty log message ***
98:
1.243 brouard 99: Revision 1.242 2016/08/30 15:01:20 brouard
100: Summary: Fixing a lots
101:
1.242 brouard 102: Revision 1.241 2016/08/29 17:17:25 brouard
103: Summary: gnuplot problem in Back projection to fix
104:
1.241 brouard 105: Revision 1.240 2016/08/29 07:53:18 brouard
106: Summary: Better
107:
1.240 brouard 108: Revision 1.239 2016/08/26 15:51:03 brouard
109: Summary: Improvement in Powell output in order to copy and paste
110:
111: Author:
112:
1.239 brouard 113: Revision 1.238 2016/08/26 14:23:35 brouard
114: Summary: Starting tests of 0.99
115:
1.238 brouard 116: Revision 1.237 2016/08/26 09:20:19 brouard
117: Summary: to valgrind
118:
1.237 brouard 119: Revision 1.236 2016/08/25 10:50:18 brouard
120: *** empty log message ***
121:
1.236 brouard 122: Revision 1.235 2016/08/25 06:59:23 brouard
123: *** empty log message ***
124:
1.235 brouard 125: Revision 1.234 2016/08/23 16:51:20 brouard
126: *** empty log message ***
127:
1.234 brouard 128: Revision 1.233 2016/08/23 07:40:50 brouard
129: Summary: not working
130:
1.233 brouard 131: Revision 1.232 2016/08/22 14:20:21 brouard
132: Summary: not working
133:
1.232 brouard 134: Revision 1.231 2016/08/22 07:17:15 brouard
135: Summary: not working
136:
1.231 brouard 137: Revision 1.230 2016/08/22 06:55:53 brouard
138: Summary: Not working
139:
1.230 brouard 140: Revision 1.229 2016/07/23 09:45:53 brouard
141: Summary: Completing for func too
142:
1.229 brouard 143: Revision 1.228 2016/07/22 17:45:30 brouard
144: Summary: Fixing some arrays, still debugging
145:
1.227 brouard 146: Revision 1.226 2016/07/12 18:42:34 brouard
147: Summary: temp
148:
1.226 brouard 149: Revision 1.225 2016/07/12 08:40:03 brouard
150: Summary: saving but not running
151:
1.225 brouard 152: Revision 1.224 2016/07/01 13:16:01 brouard
153: Summary: Fixes
154:
1.224 brouard 155: Revision 1.223 2016/02/19 09:23:35 brouard
156: Summary: temporary
157:
1.223 brouard 158: Revision 1.222 2016/02/17 08:14:50 brouard
159: Summary: Probably last 0.98 stable version 0.98r6
160:
1.222 brouard 161: Revision 1.221 2016/02/15 23:35:36 brouard
162: Summary: minor bug
163:
1.220 brouard 164: Revision 1.219 2016/02/15 00:48:12 brouard
165: *** empty log message ***
166:
1.219 brouard 167: Revision 1.218 2016/02/12 11:29:23 brouard
168: Summary: 0.99 Back projections
169:
1.218 brouard 170: Revision 1.217 2015/12/23 17:18:31 brouard
171: Summary: Experimental backcast
172:
1.217 brouard 173: Revision 1.216 2015/12/18 17:32:11 brouard
174: Summary: 0.98r4 Warning and status=-2
175:
176: Version 0.98r4 is now:
177: - displaying an error when status is -1, date of interview unknown and date of death known;
178: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
179: Older changes concerning s=-2, dating from 2005 have been supersed.
180:
1.216 brouard 181: Revision 1.215 2015/12/16 08:52:24 brouard
182: Summary: 0.98r4 working
183:
1.215 brouard 184: Revision 1.214 2015/12/16 06:57:54 brouard
185: Summary: temporary not working
186:
1.214 brouard 187: Revision 1.213 2015/12/11 18:22:17 brouard
188: Summary: 0.98r4
189:
1.213 brouard 190: Revision 1.212 2015/11/21 12:47:24 brouard
191: Summary: minor typo
192:
1.212 brouard 193: Revision 1.211 2015/11/21 12:41:11 brouard
194: Summary: 0.98r3 with some graph of projected cross-sectional
195:
196: Author: Nicolas Brouard
197:
1.211 brouard 198: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 199: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 200: Summary: Adding ftolpl parameter
201: Author: N Brouard
202:
203: We had difficulties to get smoothed confidence intervals. It was due
204: to the period prevalence which wasn't computed accurately. The inner
205: parameter ftolpl is now an outer parameter of the .imach parameter
206: file after estepm. If ftolpl is small 1.e-4 and estepm too,
207: computation are long.
208:
1.209 brouard 209: Revision 1.208 2015/11/17 14:31:57 brouard
210: Summary: temporary
211:
1.208 brouard 212: Revision 1.207 2015/10/27 17:36:57 brouard
213: *** empty log message ***
214:
1.207 brouard 215: Revision 1.206 2015/10/24 07:14:11 brouard
216: *** empty log message ***
217:
1.206 brouard 218: Revision 1.205 2015/10/23 15:50:53 brouard
219: Summary: 0.98r3 some clarification for graphs on likelihood contributions
220:
1.205 brouard 221: Revision 1.204 2015/10/01 16:20:26 brouard
222: Summary: Some new graphs of contribution to likelihood
223:
1.204 brouard 224: Revision 1.203 2015/09/30 17:45:14 brouard
225: Summary: looking at better estimation of the hessian
226:
227: Also a better criteria for convergence to the period prevalence And
228: therefore adding the number of years needed to converge. (The
229: prevalence in any alive state shold sum to one
230:
1.203 brouard 231: Revision 1.202 2015/09/22 19:45:16 brouard
232: Summary: Adding some overall graph on contribution to likelihood. Might change
233:
1.202 brouard 234: Revision 1.201 2015/09/15 17:34:58 brouard
235: Summary: 0.98r0
236:
237: - Some new graphs like suvival functions
238: - Some bugs fixed like model=1+age+V2.
239:
1.201 brouard 240: Revision 1.200 2015/09/09 16:53:55 brouard
241: Summary: Big bug thanks to Flavia
242:
243: Even model=1+age+V2. did not work anymore
244:
1.200 brouard 245: Revision 1.199 2015/09/07 14:09:23 brouard
246: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
247:
1.199 brouard 248: Revision 1.198 2015/09/03 07:14:39 brouard
249: Summary: 0.98q5 Flavia
250:
1.198 brouard 251: Revision 1.197 2015/09/01 18:24:39 brouard
252: *** empty log message ***
253:
1.197 brouard 254: Revision 1.196 2015/08/18 23:17:52 brouard
255: Summary: 0.98q5
256:
1.196 brouard 257: Revision 1.195 2015/08/18 16:28:39 brouard
258: Summary: Adding a hack for testing purpose
259:
260: After reading the title, ftol and model lines, if the comment line has
261: a q, starting with #q, the answer at the end of the run is quit. It
262: permits to run test files in batch with ctest. The former workaround was
263: $ echo q | imach foo.imach
264:
1.195 brouard 265: Revision 1.194 2015/08/18 13:32:00 brouard
266: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
267:
1.194 brouard 268: Revision 1.193 2015/08/04 07:17:42 brouard
269: Summary: 0.98q4
270:
1.193 brouard 271: Revision 1.192 2015/07/16 16:49:02 brouard
272: Summary: Fixing some outputs
273:
1.192 brouard 274: Revision 1.191 2015/07/14 10:00:33 brouard
275: Summary: Some fixes
276:
1.191 brouard 277: Revision 1.190 2015/05/05 08:51:13 brouard
278: Summary: Adding digits in output parameters (7 digits instead of 6)
279:
280: Fix 1+age+.
281:
1.190 brouard 282: Revision 1.189 2015/04/30 14:45:16 brouard
283: Summary: 0.98q2
284:
1.189 brouard 285: Revision 1.188 2015/04/30 08:27:53 brouard
286: *** empty log message ***
287:
1.188 brouard 288: Revision 1.187 2015/04/29 09:11:15 brouard
289: *** empty log message ***
290:
1.187 brouard 291: Revision 1.186 2015/04/23 12:01:52 brouard
292: Summary: V1*age is working now, version 0.98q1
293:
294: Some codes had been disabled in order to simplify and Vn*age was
295: working in the optimization phase, ie, giving correct MLE parameters,
296: but, as usual, outputs were not correct and program core dumped.
297:
1.186 brouard 298: Revision 1.185 2015/03/11 13:26:42 brouard
299: Summary: Inclusion of compile and links command line for Intel Compiler
300:
1.185 brouard 301: Revision 1.184 2015/03/11 11:52:39 brouard
302: Summary: Back from Windows 8. Intel Compiler
303:
1.184 brouard 304: Revision 1.183 2015/03/10 20:34:32 brouard
305: Summary: 0.98q0, trying with directest, mnbrak fixed
306:
307: We use directest instead of original Powell test; probably no
308: incidence on the results, but better justifications;
309: We fixed Numerical Recipes mnbrak routine which was wrong and gave
310: wrong results.
311:
1.183 brouard 312: Revision 1.182 2015/02/12 08:19:57 brouard
313: Summary: Trying to keep directest which seems simpler and more general
314: Author: Nicolas Brouard
315:
1.182 brouard 316: Revision 1.181 2015/02/11 23:22:24 brouard
317: Summary: Comments on Powell added
318:
319: Author:
320:
1.181 brouard 321: Revision 1.180 2015/02/11 17:33:45 brouard
322: Summary: Finishing move from main to function (hpijx and prevalence_limit)
323:
1.180 brouard 324: Revision 1.179 2015/01/04 09:57:06 brouard
325: Summary: back to OS/X
326:
1.179 brouard 327: Revision 1.178 2015/01/04 09:35:48 brouard
328: *** empty log message ***
329:
1.178 brouard 330: Revision 1.177 2015/01/03 18:40:56 brouard
331: Summary: Still testing ilc32 on OSX
332:
1.177 brouard 333: Revision 1.176 2015/01/03 16:45:04 brouard
334: *** empty log message ***
335:
1.176 brouard 336: Revision 1.175 2015/01/03 16:33:42 brouard
337: *** empty log message ***
338:
1.175 brouard 339: Revision 1.174 2015/01/03 16:15:49 brouard
340: Summary: Still in cross-compilation
341:
1.174 brouard 342: Revision 1.173 2015/01/03 12:06:26 brouard
343: Summary: trying to detect cross-compilation
344:
1.173 brouard 345: Revision 1.172 2014/12/27 12:07:47 brouard
346: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
347:
1.172 brouard 348: Revision 1.171 2014/12/23 13:26:59 brouard
349: Summary: Back from Visual C
350:
351: Still problem with utsname.h on Windows
352:
1.171 brouard 353: Revision 1.170 2014/12/23 11:17:12 brouard
354: Summary: Cleaning some \%% back to %%
355:
356: The escape was mandatory for a specific compiler (which one?), but too many warnings.
357:
1.170 brouard 358: Revision 1.169 2014/12/22 23:08:31 brouard
359: Summary: 0.98p
360:
361: Outputs some informations on compiler used, OS etc. Testing on different platforms.
362:
1.169 brouard 363: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 364: Summary: update
1.169 brouard 365:
1.168 brouard 366: Revision 1.167 2014/12/22 13:50:56 brouard
367: Summary: Testing uname and compiler version and if compiled 32 or 64
368:
369: Testing on Linux 64
370:
1.167 brouard 371: Revision 1.166 2014/12/22 11:40:47 brouard
372: *** empty log message ***
373:
1.166 brouard 374: Revision 1.165 2014/12/16 11:20:36 brouard
375: Summary: After compiling on Visual C
376:
377: * imach.c (Module): Merging 1.61 to 1.162
378:
1.165 brouard 379: Revision 1.164 2014/12/16 10:52:11 brouard
380: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
381:
382: * imach.c (Module): Merging 1.61 to 1.162
383:
1.164 brouard 384: Revision 1.163 2014/12/16 10:30:11 brouard
385: * imach.c (Module): Merging 1.61 to 1.162
386:
1.163 brouard 387: Revision 1.162 2014/09/25 11:43:39 brouard
388: Summary: temporary backup 0.99!
389:
1.162 brouard 390: Revision 1.1 2014/09/16 11:06:58 brouard
391: Summary: With some code (wrong) for nlopt
392:
393: Author:
394:
395: Revision 1.161 2014/09/15 20:41:41 brouard
396: Summary: Problem with macro SQR on Intel compiler
397:
1.161 brouard 398: Revision 1.160 2014/09/02 09:24:05 brouard
399: *** empty log message ***
400:
1.160 brouard 401: Revision 1.159 2014/09/01 10:34:10 brouard
402: Summary: WIN32
403: Author: Brouard
404:
1.159 brouard 405: Revision 1.158 2014/08/27 17:11:51 brouard
406: *** empty log message ***
407:
1.158 brouard 408: Revision 1.157 2014/08/27 16:26:55 brouard
409: Summary: Preparing windows Visual studio version
410: Author: Brouard
411:
412: In order to compile on Visual studio, time.h is now correct and time_t
413: and tm struct should be used. difftime should be used but sometimes I
414: just make the differences in raw time format (time(&now).
415: Trying to suppress #ifdef LINUX
416: Add xdg-open for __linux in order to open default browser.
417:
1.157 brouard 418: Revision 1.156 2014/08/25 20:10:10 brouard
419: *** empty log message ***
420:
1.156 brouard 421: Revision 1.155 2014/08/25 18:32:34 brouard
422: Summary: New compile, minor changes
423: Author: Brouard
424:
1.155 brouard 425: Revision 1.154 2014/06/20 17:32:08 brouard
426: Summary: Outputs now all graphs of convergence to period prevalence
427:
1.154 brouard 428: Revision 1.153 2014/06/20 16:45:46 brouard
429: Summary: If 3 live state, convergence to period prevalence on same graph
430: Author: Brouard
431:
1.153 brouard 432: Revision 1.152 2014/06/18 17:54:09 brouard
433: Summary: open browser, use gnuplot on same dir than imach if not found in the path
434:
1.152 brouard 435: Revision 1.151 2014/06/18 16:43:30 brouard
436: *** empty log message ***
437:
1.151 brouard 438: Revision 1.150 2014/06/18 16:42:35 brouard
439: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
440: Author: brouard
441:
1.150 brouard 442: Revision 1.149 2014/06/18 15:51:14 brouard
443: Summary: Some fixes in parameter files errors
444: Author: Nicolas Brouard
445:
1.149 brouard 446: Revision 1.148 2014/06/17 17:38:48 brouard
447: Summary: Nothing new
448: Author: Brouard
449:
450: Just a new packaging for OS/X version 0.98nS
451:
1.148 brouard 452: Revision 1.147 2014/06/16 10:33:11 brouard
453: *** empty log message ***
454:
1.147 brouard 455: Revision 1.146 2014/06/16 10:20:28 brouard
456: Summary: Merge
457: Author: Brouard
458:
459: Merge, before building revised version.
460:
1.146 brouard 461: Revision 1.145 2014/06/10 21:23:15 brouard
462: Summary: Debugging with valgrind
463: Author: Nicolas Brouard
464:
465: Lot of changes in order to output the results with some covariates
466: After the Edimburgh REVES conference 2014, it seems mandatory to
467: improve the code.
468: No more memory valgrind error but a lot has to be done in order to
469: continue the work of splitting the code into subroutines.
470: Also, decodemodel has been improved. Tricode is still not
471: optimal. nbcode should be improved. Documentation has been added in
472: the source code.
473:
1.144 brouard 474: Revision 1.143 2014/01/26 09:45:38 brouard
475: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
476:
477: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
478: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
479:
1.143 brouard 480: Revision 1.142 2014/01/26 03:57:36 brouard
481: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
482:
483: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
484:
1.142 brouard 485: Revision 1.141 2014/01/26 02:42:01 brouard
486: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
487:
1.141 brouard 488: Revision 1.140 2011/09/02 10:37:54 brouard
489: Summary: times.h is ok with mingw32 now.
490:
1.140 brouard 491: Revision 1.139 2010/06/14 07:50:17 brouard
492: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
493: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
494:
1.139 brouard 495: Revision 1.138 2010/04/30 18:19:40 brouard
496: *** empty log message ***
497:
1.138 brouard 498: Revision 1.137 2010/04/29 18:11:38 brouard
499: (Module): Checking covariates for more complex models
500: than V1+V2. A lot of change to be done. Unstable.
501:
1.137 brouard 502: Revision 1.136 2010/04/26 20:30:53 brouard
503: (Module): merging some libgsl code. Fixing computation
504: of likelione (using inter/intrapolation if mle = 0) in order to
505: get same likelihood as if mle=1.
506: Some cleaning of code and comments added.
507:
1.136 brouard 508: Revision 1.135 2009/10/29 15:33:14 brouard
509: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
510:
1.135 brouard 511: Revision 1.134 2009/10/29 13:18:53 brouard
512: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
513:
1.134 brouard 514: Revision 1.133 2009/07/06 10:21:25 brouard
515: just nforces
516:
1.133 brouard 517: Revision 1.132 2009/07/06 08:22:05 brouard
518: Many tings
519:
1.132 brouard 520: Revision 1.131 2009/06/20 16:22:47 brouard
521: Some dimensions resccaled
522:
1.131 brouard 523: Revision 1.130 2009/05/26 06:44:34 brouard
524: (Module): Max Covariate is now set to 20 instead of 8. A
525: lot of cleaning with variables initialized to 0. Trying to make
526: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
527:
1.130 brouard 528: Revision 1.129 2007/08/31 13:49:27 lievre
529: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
530:
1.129 lievre 531: Revision 1.128 2006/06/30 13:02:05 brouard
532: (Module): Clarifications on computing e.j
533:
1.128 brouard 534: Revision 1.127 2006/04/28 18:11:50 brouard
535: (Module): Yes the sum of survivors was wrong since
536: imach-114 because nhstepm was no more computed in the age
537: loop. Now we define nhstepma in the age loop.
538: (Module): In order to speed up (in case of numerous covariates) we
539: compute health expectancies (without variances) in a first step
540: and then all the health expectancies with variances or standard
541: deviation (needs data from the Hessian matrices) which slows the
542: computation.
543: In the future we should be able to stop the program is only health
544: expectancies and graph are needed without standard deviations.
545:
1.127 brouard 546: Revision 1.126 2006/04/28 17:23:28 brouard
547: (Module): Yes the sum of survivors was wrong since
548: imach-114 because nhstepm was no more computed in the age
549: loop. Now we define nhstepma in the age loop.
550: Version 0.98h
551:
1.126 brouard 552: Revision 1.125 2006/04/04 15:20:31 lievre
553: Errors in calculation of health expectancies. Age was not initialized.
554: Forecasting file added.
555:
556: Revision 1.124 2006/03/22 17:13:53 lievre
557: Parameters are printed with %lf instead of %f (more numbers after the comma).
558: The log-likelihood is printed in the log file
559:
560: Revision 1.123 2006/03/20 10:52:43 brouard
561: * imach.c (Module): <title> changed, corresponds to .htm file
562: name. <head> headers where missing.
563:
564: * imach.c (Module): Weights can have a decimal point as for
565: English (a comma might work with a correct LC_NUMERIC environment,
566: otherwise the weight is truncated).
567: Modification of warning when the covariates values are not 0 or
568: 1.
569: Version 0.98g
570:
571: Revision 1.122 2006/03/20 09:45:41 brouard
572: (Module): Weights can have a decimal point as for
573: English (a comma might work with a correct LC_NUMERIC environment,
574: otherwise the weight is truncated).
575: Modification of warning when the covariates values are not 0 or
576: 1.
577: Version 0.98g
578:
579: Revision 1.121 2006/03/16 17:45:01 lievre
580: * imach.c (Module): Comments concerning covariates added
581:
582: * imach.c (Module): refinements in the computation of lli if
583: status=-2 in order to have more reliable computation if stepm is
584: not 1 month. Version 0.98f
585:
586: Revision 1.120 2006/03/16 15:10:38 lievre
587: (Module): refinements in the computation of lli if
588: status=-2 in order to have more reliable computation if stepm is
589: not 1 month. Version 0.98f
590:
591: Revision 1.119 2006/03/15 17:42:26 brouard
592: (Module): Bug if status = -2, the loglikelihood was
593: computed as likelihood omitting the logarithm. Version O.98e
594:
595: Revision 1.118 2006/03/14 18:20:07 brouard
596: (Module): varevsij Comments added explaining the second
597: table of variances if popbased=1 .
598: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
599: (Module): Function pstamp added
600: (Module): Version 0.98d
601:
602: Revision 1.117 2006/03/14 17:16:22 brouard
603: (Module): varevsij Comments added explaining the second
604: table of variances if popbased=1 .
605: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
606: (Module): Function pstamp added
607: (Module): Version 0.98d
608:
609: Revision 1.116 2006/03/06 10:29:27 brouard
610: (Module): Variance-covariance wrong links and
611: varian-covariance of ej. is needed (Saito).
612:
613: Revision 1.115 2006/02/27 12:17:45 brouard
614: (Module): One freematrix added in mlikeli! 0.98c
615:
616: Revision 1.114 2006/02/26 12:57:58 brouard
617: (Module): Some improvements in processing parameter
618: filename with strsep.
619:
620: Revision 1.113 2006/02/24 14:20:24 brouard
621: (Module): Memory leaks checks with valgrind and:
622: datafile was not closed, some imatrix were not freed and on matrix
623: allocation too.
624:
625: Revision 1.112 2006/01/30 09:55:26 brouard
626: (Module): Back to gnuplot.exe instead of wgnuplot.exe
627:
628: Revision 1.111 2006/01/25 20:38:18 brouard
629: (Module): Lots of cleaning and bugs added (Gompertz)
630: (Module): Comments can be added in data file. Missing date values
631: can be a simple dot '.'.
632:
633: Revision 1.110 2006/01/25 00:51:50 brouard
634: (Module): Lots of cleaning and bugs added (Gompertz)
635:
636: Revision 1.109 2006/01/24 19:37:15 brouard
637: (Module): Comments (lines starting with a #) are allowed in data.
638:
639: Revision 1.108 2006/01/19 18:05:42 lievre
640: Gnuplot problem appeared...
641: To be fixed
642:
643: Revision 1.107 2006/01/19 16:20:37 brouard
644: Test existence of gnuplot in imach path
645:
646: Revision 1.106 2006/01/19 13:24:36 brouard
647: Some cleaning and links added in html output
648:
649: Revision 1.105 2006/01/05 20:23:19 lievre
650: *** empty log message ***
651:
652: Revision 1.104 2005/09/30 16:11:43 lievre
653: (Module): sump fixed, loop imx fixed, and simplifications.
654: (Module): If the status is missing at the last wave but we know
655: that the person is alive, then we can code his/her status as -2
656: (instead of missing=-1 in earlier versions) and his/her
657: contributions to the likelihood is 1 - Prob of dying from last
658: health status (= 1-p13= p11+p12 in the easiest case of somebody in
659: the healthy state at last known wave). Version is 0.98
660:
661: Revision 1.103 2005/09/30 15:54:49 lievre
662: (Module): sump fixed, loop imx fixed, and simplifications.
663:
664: Revision 1.102 2004/09/15 17:31:30 brouard
665: Add the possibility to read data file including tab characters.
666:
667: Revision 1.101 2004/09/15 10:38:38 brouard
668: Fix on curr_time
669:
670: Revision 1.100 2004/07/12 18:29:06 brouard
671: Add version for Mac OS X. Just define UNIX in Makefile
672:
673: Revision 1.99 2004/06/05 08:57:40 brouard
674: *** empty log message ***
675:
676: Revision 1.98 2004/05/16 15:05:56 brouard
677: New version 0.97 . First attempt to estimate force of mortality
678: directly from the data i.e. without the need of knowing the health
679: state at each age, but using a Gompertz model: log u =a + b*age .
680: This is the basic analysis of mortality and should be done before any
681: other analysis, in order to test if the mortality estimated from the
682: cross-longitudinal survey is different from the mortality estimated
683: from other sources like vital statistic data.
684:
685: The same imach parameter file can be used but the option for mle should be -3.
686:
1.133 brouard 687: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 688: former routines in order to include the new code within the former code.
689:
690: The output is very simple: only an estimate of the intercept and of
691: the slope with 95% confident intervals.
692:
693: Current limitations:
694: A) Even if you enter covariates, i.e. with the
695: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
696: B) There is no computation of Life Expectancy nor Life Table.
697:
698: Revision 1.97 2004/02/20 13:25:42 lievre
699: Version 0.96d. Population forecasting command line is (temporarily)
700: suppressed.
701:
702: Revision 1.96 2003/07/15 15:38:55 brouard
703: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
704: rewritten within the same printf. Workaround: many printfs.
705:
706: Revision 1.95 2003/07/08 07:54:34 brouard
707: * imach.c (Repository):
708: (Repository): Using imachwizard code to output a more meaningful covariance
709: matrix (cov(a12,c31) instead of numbers.
710:
711: Revision 1.94 2003/06/27 13:00:02 brouard
712: Just cleaning
713:
714: Revision 1.93 2003/06/25 16:33:55 brouard
715: (Module): On windows (cygwin) function asctime_r doesn't
716: exist so I changed back to asctime which exists.
717: (Module): Version 0.96b
718:
719: Revision 1.92 2003/06/25 16:30:45 brouard
720: (Module): On windows (cygwin) function asctime_r doesn't
721: exist so I changed back to asctime which exists.
722:
723: Revision 1.91 2003/06/25 15:30:29 brouard
724: * imach.c (Repository): Duplicated warning errors corrected.
725: (Repository): Elapsed time after each iteration is now output. It
726: helps to forecast when convergence will be reached. Elapsed time
727: is stamped in powell. We created a new html file for the graphs
728: concerning matrix of covariance. It has extension -cov.htm.
729:
730: Revision 1.90 2003/06/24 12:34:15 brouard
731: (Module): Some bugs corrected for windows. Also, when
732: mle=-1 a template is output in file "or"mypar.txt with the design
733: of the covariance matrix to be input.
734:
735: Revision 1.89 2003/06/24 12:30:52 brouard
736: (Module): Some bugs corrected for windows. Also, when
737: mle=-1 a template is output in file "or"mypar.txt with the design
738: of the covariance matrix to be input.
739:
740: Revision 1.88 2003/06/23 17:54:56 brouard
741: * 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.
742:
743: Revision 1.87 2003/06/18 12:26:01 brouard
744: Version 0.96
745:
746: Revision 1.86 2003/06/17 20:04:08 brouard
747: (Module): Change position of html and gnuplot routines and added
748: routine fileappend.
749:
750: Revision 1.85 2003/06/17 13:12:43 brouard
751: * imach.c (Repository): Check when date of death was earlier that
752: current date of interview. It may happen when the death was just
753: prior to the death. In this case, dh was negative and likelihood
754: was wrong (infinity). We still send an "Error" but patch by
755: assuming that the date of death was just one stepm after the
756: interview.
757: (Repository): Because some people have very long ID (first column)
758: we changed int to long in num[] and we added a new lvector for
759: memory allocation. But we also truncated to 8 characters (left
760: truncation)
761: (Repository): No more line truncation errors.
762:
763: Revision 1.84 2003/06/13 21:44:43 brouard
764: * imach.c (Repository): Replace "freqsummary" at a correct
765: place. It differs from routine "prevalence" which may be called
766: many times. Probs is memory consuming and must be used with
767: parcimony.
768: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
769:
770: Revision 1.83 2003/06/10 13:39:11 lievre
771: *** empty log message ***
772:
773: Revision 1.82 2003/06/05 15:57:20 brouard
774: Add log in imach.c and fullversion number is now printed.
775:
776: */
777: /*
778: Interpolated Markov Chain
779:
780: Short summary of the programme:
781:
1.227 brouard 782: This program computes Healthy Life Expectancies or State-specific
783: (if states aren't health statuses) Expectancies from
784: cross-longitudinal data. Cross-longitudinal data consist in:
785:
786: -1- a first survey ("cross") where individuals from different ages
787: are interviewed on their health status or degree of disability (in
788: the case of a health survey which is our main interest)
789:
790: -2- at least a second wave of interviews ("longitudinal") which
791: measure each change (if any) in individual health status. Health
792: expectancies are computed from the time spent in each health state
793: according to a model. More health states you consider, more time is
794: necessary to reach the Maximum Likelihood of the parameters involved
795: in the model. The simplest model is the multinomial logistic model
796: where pij is the probability to be observed in state j at the second
797: wave conditional to be observed in state i at the first
798: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
799: etc , where 'age' is age and 'sex' is a covariate. If you want to
800: have a more complex model than "constant and age", you should modify
801: the program where the markup *Covariates have to be included here
802: again* invites you to do it. More covariates you add, slower the
1.126 brouard 803: convergence.
804:
805: The advantage of this computer programme, compared to a simple
806: multinomial logistic model, is clear when the delay between waves is not
807: identical for each individual. Also, if a individual missed an
808: intermediate interview, the information is lost, but taken into
809: account using an interpolation or extrapolation.
810:
811: hPijx is the probability to be observed in state i at age x+h
812: conditional to the observed state i at age x. The delay 'h' can be
813: split into an exact number (nh*stepm) of unobserved intermediate
814: states. This elementary transition (by month, quarter,
815: semester or year) is modelled as a multinomial logistic. The hPx
816: matrix is simply the matrix product of nh*stepm elementary matrices
817: and the contribution of each individual to the likelihood is simply
818: hPijx.
819:
820: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 821: of the life expectancies. It also computes the period (stable) prevalence.
822:
823: Back prevalence and projections:
1.227 brouard 824:
825: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
826: double agemaxpar, double ftolpl, int *ncvyearp, double
827: dateprev1,double dateprev2, int firstpass, int lastpass, int
828: mobilavproj)
829:
830: Computes the back prevalence limit for any combination of
831: covariate values k at any age between ageminpar and agemaxpar and
832: returns it in **bprlim. In the loops,
833:
834: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
835: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
836:
837: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 838: Computes for any combination of covariates k and any age between bage and fage
839: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
840: oldm=oldms;savm=savms;
1.227 brouard 841:
1.267 brouard 842: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 843: Computes the transition matrix starting at age 'age' over
844: 'nhstepm*hstepm*stepm' months (i.e. until
845: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 846: nhstepm*hstepm matrices.
847:
848: Returns p3mat[i][j][h] after calling
849: p3mat[i][j][h]=matprod2(newm,
850: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
851: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
852: oldm);
1.226 brouard 853:
854: Important routines
855:
856: - func (or funcone), computes logit (pij) distinguishing
857: o fixed variables (single or product dummies or quantitative);
858: o varying variables by:
859: (1) wave (single, product dummies, quantitative),
860: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
861: % fixed dummy (treated) or quantitative (not done because time-consuming);
862: % varying dummy (not done) or quantitative (not done);
863: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
864: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
865: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
866: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
867: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 868:
1.226 brouard 869:
870:
1.133 brouard 871: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
872: Institut national d'études démographiques, Paris.
1.126 brouard 873: This software have been partly granted by Euro-REVES, a concerted action
874: from the European Union.
875: It is copyrighted identically to a GNU software product, ie programme and
876: software can be distributed freely for non commercial use. Latest version
877: can be accessed at http://euroreves.ined.fr/imach .
878:
879: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
880: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
881:
882: **********************************************************************/
883: /*
884: main
885: read parameterfile
886: read datafile
887: concatwav
888: freqsummary
889: if (mle >= 1)
890: mlikeli
891: print results files
892: if mle==1
893: computes hessian
894: read end of parameter file: agemin, agemax, bage, fage, estepm
895: begin-prev-date,...
896: open gnuplot file
897: open html file
1.145 brouard 898: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
899: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
900: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
901: freexexit2 possible for memory heap.
902:
903: h Pij x | pij_nom ficrestpij
904: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
905: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
906: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
907:
908: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
909: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
910: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
911: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
912: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
913:
1.126 brouard 914: forecasting if prevfcast==1 prevforecast call prevalence()
915: health expectancies
916: Variance-covariance of DFLE
917: prevalence()
918: movingaverage()
919: varevsij()
920: if popbased==1 varevsij(,popbased)
921: total life expectancies
922: Variance of period (stable) prevalence
923: end
924: */
925:
1.187 brouard 926: /* #define DEBUG */
927: /* #define DEBUGBRENT */
1.203 brouard 928: /* #define DEBUGLINMIN */
929: /* #define DEBUGHESS */
930: #define DEBUGHESSIJ
1.224 brouard 931: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 932: #define POWELL /* Instead of NLOPT */
1.224 brouard 933: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 934: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
935: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 936:
937: #include <math.h>
938: #include <stdio.h>
939: #include <stdlib.h>
940: #include <string.h>
1.226 brouard 941: #include <ctype.h>
1.159 brouard 942:
943: #ifdef _WIN32
944: #include <io.h>
1.172 brouard 945: #include <windows.h>
946: #include <tchar.h>
1.159 brouard 947: #else
1.126 brouard 948: #include <unistd.h>
1.159 brouard 949: #endif
1.126 brouard 950:
951: #include <limits.h>
952: #include <sys/types.h>
1.171 brouard 953:
954: #if defined(__GNUC__)
955: #include <sys/utsname.h> /* Doesn't work on Windows */
956: #endif
957:
1.126 brouard 958: #include <sys/stat.h>
959: #include <errno.h>
1.159 brouard 960: /* extern int errno; */
1.126 brouard 961:
1.157 brouard 962: /* #ifdef LINUX */
963: /* #include <time.h> */
964: /* #include "timeval.h" */
965: /* #else */
966: /* #include <sys/time.h> */
967: /* #endif */
968:
1.126 brouard 969: #include <time.h>
970:
1.136 brouard 971: #ifdef GSL
972: #include <gsl/gsl_errno.h>
973: #include <gsl/gsl_multimin.h>
974: #endif
975:
1.167 brouard 976:
1.162 brouard 977: #ifdef NLOPT
978: #include <nlopt.h>
979: typedef struct {
980: double (* function)(double [] );
981: } myfunc_data ;
982: #endif
983:
1.126 brouard 984: /* #include <libintl.h> */
985: /* #define _(String) gettext (String) */
986:
1.251 brouard 987: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 988:
989: #define GNUPLOTPROGRAM "gnuplot"
990: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
991: #define FILENAMELENGTH 132
992:
993: #define GLOCK_ERROR_NOPATH -1 /* empty path */
994: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
995:
1.144 brouard 996: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
997: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 998:
999: #define NINTERVMAX 8
1.144 brouard 1000: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1001: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1002: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1003: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1004: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1005: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 1006: #define MAXN 20000
1.144 brouard 1007: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1008: /* #define AGESUP 130 */
1009: #define AGESUP 150
1.268 brouard 1010: #define AGEINF 0
1.218 brouard 1011: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1012: #define AGEBASE 40
1.194 brouard 1013: #define AGEOVERFLOW 1.e20
1.164 brouard 1014: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1015: #ifdef _WIN32
1016: #define DIRSEPARATOR '\\'
1017: #define CHARSEPARATOR "\\"
1018: #define ODIRSEPARATOR '/'
1019: #else
1.126 brouard 1020: #define DIRSEPARATOR '/'
1021: #define CHARSEPARATOR "/"
1022: #define ODIRSEPARATOR '\\'
1023: #endif
1024:
1.274 ! brouard 1025: /* $Id: imach.c,v 1.273 2017/06/27 11:06:02 brouard Exp $ */
1.126 brouard 1026: /* $State: Exp $ */
1.196 brouard 1027: #include "version.h"
1028: char version[]=__IMACH_VERSION__;
1.224 brouard 1029: 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";
1.274 ! brouard 1030: char fullversion[]="$Revision: 1.273 $ $Date: 2017/06/27 11:06:02 $";
1.126 brouard 1031: char strstart[80];
1032: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1033: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1034: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1035: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1036: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1037: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1038: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1039: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1040: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1041: int cptcovprodnoage=0; /**< Number of covariate products without age */
1042: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1043: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1044: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1045: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1046: int nsd=0; /**< Total number of single dummy variables (output) */
1047: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1048: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1049: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1050: int ntveff=0; /**< ntveff number of effective time varying variables */
1051: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1052: int cptcov=0; /* Working variable */
1.218 brouard 1053: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1054: int npar=NPARMAX;
1055: int nlstate=2; /* Number of live states */
1056: int ndeath=1; /* Number of dead states */
1.130 brouard 1057: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1058: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1059: int popbased=0;
1060:
1061: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1062: int maxwav=0; /* Maxim number of waves */
1063: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1064: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1065: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1066: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1067: int mle=1, weightopt=0;
1.126 brouard 1068: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1069: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1070: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1071: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1072: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1073: int selected(int kvar); /* Is covariate kvar selected for printing results */
1074:
1.130 brouard 1075: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1076: double **matprod2(); /* test */
1.126 brouard 1077: double **oldm, **newm, **savm; /* Working pointers to matrices */
1078: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1079: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1080:
1.136 brouard 1081: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1082: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1083: FILE *ficlog, *ficrespow;
1.130 brouard 1084: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1085: double fretone; /* Only one call to likelihood */
1.130 brouard 1086: long ipmx=0; /* Number of contributions */
1.126 brouard 1087: double sw; /* Sum of weights */
1088: char filerespow[FILENAMELENGTH];
1089: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1090: FILE *ficresilk;
1091: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1092: FILE *ficresprobmorprev;
1093: FILE *fichtm, *fichtmcov; /* Html File */
1094: FILE *ficreseij;
1095: char filerese[FILENAMELENGTH];
1096: FILE *ficresstdeij;
1097: char fileresstde[FILENAMELENGTH];
1098: FILE *ficrescveij;
1099: char filerescve[FILENAMELENGTH];
1100: FILE *ficresvij;
1101: char fileresv[FILENAMELENGTH];
1.269 brouard 1102:
1.126 brouard 1103: char title[MAXLINE];
1.234 brouard 1104: char model[MAXLINE]; /**< The model line */
1.217 brouard 1105: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1106: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1107: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1108: char command[FILENAMELENGTH];
1109: int outcmd=0;
1110:
1.217 brouard 1111: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1112: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1113: char filelog[FILENAMELENGTH]; /* Log file */
1114: char filerest[FILENAMELENGTH];
1115: char fileregp[FILENAMELENGTH];
1116: char popfile[FILENAMELENGTH];
1117:
1118: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1119:
1.157 brouard 1120: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1121: /* struct timezone tzp; */
1122: /* extern int gettimeofday(); */
1123: struct tm tml, *gmtime(), *localtime();
1124:
1125: extern time_t time();
1126:
1127: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1128: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1129: struct tm tm;
1130:
1.126 brouard 1131: char strcurr[80], strfor[80];
1132:
1133: char *endptr;
1134: long lval;
1135: double dval;
1136:
1137: #define NR_END 1
1138: #define FREE_ARG char*
1139: #define FTOL 1.0e-10
1140:
1141: #define NRANSI
1.240 brouard 1142: #define ITMAX 200
1143: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1144:
1145: #define TOL 2.0e-4
1146:
1147: #define CGOLD 0.3819660
1148: #define ZEPS 1.0e-10
1149: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1150:
1151: #define GOLD 1.618034
1152: #define GLIMIT 100.0
1153: #define TINY 1.0e-20
1154:
1155: static double maxarg1,maxarg2;
1156: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1157: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1158:
1159: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1160: #define rint(a) floor(a+0.5)
1.166 brouard 1161: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1162: #define mytinydouble 1.0e-16
1.166 brouard 1163: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1164: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1165: /* static double dsqrarg; */
1166: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1167: static double sqrarg;
1168: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1169: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1170: int agegomp= AGEGOMP;
1171:
1172: int imx;
1173: int stepm=1;
1174: /* Stepm, step in month: minimum step interpolation*/
1175:
1176: int estepm;
1177: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1178:
1179: int m,nb;
1180: long *num;
1.197 brouard 1181: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1182: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1183: covariate for which somebody answered excluding
1184: undefined. Usually 2: 0 and 1. */
1185: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1186: covariate for which somebody answered including
1187: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1188: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1189: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1190: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1191: double *ageexmed,*agecens;
1192: double dateintmean=0;
1193:
1194: double *weight;
1195: int **s; /* Status */
1.141 brouard 1196: double *agedc;
1.145 brouard 1197: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1198: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1199: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1200: double **coqvar; /* Fixed quantitative covariate nqv */
1201: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1202: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1203: double idx;
1204: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1205: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1206: /*k 1 2 3 4 5 6 7 8 9 */
1207: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1208: /* Tndvar[k] 1 2 3 4 5 */
1209: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1210: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1211: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1212: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1213: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1214: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1215: /* Tprod[i]=k 4 7 */
1216: /* Tage[i]=k 5 8 */
1217: /* */
1218: /* Type */
1219: /* V 1 2 3 4 5 */
1220: /* F F V V V */
1221: /* D Q D D Q */
1222: /* */
1223: int *TvarsD;
1224: int *TvarsDind;
1225: int *TvarsQ;
1226: int *TvarsQind;
1227:
1.235 brouard 1228: #define MAXRESULTLINES 10
1229: int nresult=0;
1.258 brouard 1230: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1231: int TKresult[MAXRESULTLINES];
1.237 brouard 1232: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1233: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1234: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1235: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1236: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1237: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1238:
1.234 brouard 1239: /* 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 1240: 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 */
1241: 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 */
1242: 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 */
1243: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1244: 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 */
1245: 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 1246: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1247: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1248: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1249: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1250: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1251: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1252: 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 */
1253: 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 */
1254:
1.230 brouard 1255: int *Tvarsel; /**< Selected covariates for output */
1256: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1257: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1258: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1259: 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 1260: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1261: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1262: int *Tage;
1.227 brouard 1263: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1264: 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 1265: 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*/
1266: 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 1267: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1268: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1269: int **Tvard;
1270: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1271: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1272: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1273: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1274: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1275: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1276: double *lsurv, *lpop, *tpop;
1277:
1.231 brouard 1278: #define FD 1; /* Fixed dummy covariate */
1279: #define FQ 2; /* Fixed quantitative covariate */
1280: #define FP 3; /* Fixed product covariate */
1281: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1282: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1283: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1284: #define VD 10; /* Varying dummy covariate */
1285: #define VQ 11; /* Varying quantitative covariate */
1286: #define VP 12; /* Varying product covariate */
1287: #define VPDD 13; /* Varying product dummy*dummy covariate */
1288: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1289: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1290: #define APFD 16; /* Age product * fixed dummy covariate */
1291: #define APFQ 17; /* Age product * fixed quantitative covariate */
1292: #define APVD 18; /* Age product * varying dummy covariate */
1293: #define APVQ 19; /* Age product * varying quantitative covariate */
1294:
1295: #define FTYPE 1; /* Fixed covariate */
1296: #define VTYPE 2; /* Varying covariate (loop in wave) */
1297: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1298:
1299: struct kmodel{
1300: int maintype; /* main type */
1301: int subtype; /* subtype */
1302: };
1303: struct kmodel modell[NCOVMAX];
1304:
1.143 brouard 1305: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1306: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1307:
1308: /**************** split *************************/
1309: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1310: {
1311: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1312: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1313: */
1314: char *ss; /* pointer */
1.186 brouard 1315: int l1=0, l2=0; /* length counters */
1.126 brouard 1316:
1317: l1 = strlen(path ); /* length of path */
1318: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1319: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1320: if ( ss == NULL ) { /* no directory, so determine current directory */
1321: strcpy( name, path ); /* we got the fullname name because no directory */
1322: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1323: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1324: /* get current working directory */
1325: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1326: #ifdef WIN32
1327: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1328: #else
1329: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1330: #endif
1.126 brouard 1331: return( GLOCK_ERROR_GETCWD );
1332: }
1333: /* got dirc from getcwd*/
1334: printf(" DIRC = %s \n",dirc);
1.205 brouard 1335: } else { /* strip directory from path */
1.126 brouard 1336: ss++; /* after this, the filename */
1337: l2 = strlen( ss ); /* length of filename */
1338: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1339: strcpy( name, ss ); /* save file name */
1340: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1341: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1342: printf(" DIRC2 = %s \n",dirc);
1343: }
1344: /* We add a separator at the end of dirc if not exists */
1345: l1 = strlen( dirc ); /* length of directory */
1346: if( dirc[l1-1] != DIRSEPARATOR ){
1347: dirc[l1] = DIRSEPARATOR;
1348: dirc[l1+1] = 0;
1349: printf(" DIRC3 = %s \n",dirc);
1350: }
1351: ss = strrchr( name, '.' ); /* find last / */
1352: if (ss >0){
1353: ss++;
1354: strcpy(ext,ss); /* save extension */
1355: l1= strlen( name);
1356: l2= strlen(ss)+1;
1357: strncpy( finame, name, l1-l2);
1358: finame[l1-l2]= 0;
1359: }
1360:
1361: return( 0 ); /* we're done */
1362: }
1363:
1364:
1365: /******************************************/
1366:
1367: void replace_back_to_slash(char *s, char*t)
1368: {
1369: int i;
1370: int lg=0;
1371: i=0;
1372: lg=strlen(t);
1373: for(i=0; i<= lg; i++) {
1374: (s[i] = t[i]);
1375: if (t[i]== '\\') s[i]='/';
1376: }
1377: }
1378:
1.132 brouard 1379: char *trimbb(char *out, char *in)
1.137 brouard 1380: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1381: char *s;
1382: s=out;
1383: while (*in != '\0'){
1.137 brouard 1384: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1385: in++;
1386: }
1387: *out++ = *in++;
1388: }
1389: *out='\0';
1390: return s;
1391: }
1392:
1.187 brouard 1393: /* char *substrchaine(char *out, char *in, char *chain) */
1394: /* { */
1395: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1396: /* char *s, *t; */
1397: /* t=in;s=out; */
1398: /* while ((*in != *chain) && (*in != '\0')){ */
1399: /* *out++ = *in++; */
1400: /* } */
1401:
1402: /* /\* *in matches *chain *\/ */
1403: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1404: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1405: /* } */
1406: /* in--; chain--; */
1407: /* while ( (*in != '\0')){ */
1408: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1409: /* *out++ = *in++; */
1410: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1411: /* } */
1412: /* *out='\0'; */
1413: /* out=s; */
1414: /* return out; */
1415: /* } */
1416: char *substrchaine(char *out, char *in, char *chain)
1417: {
1418: /* Substract chain 'chain' from 'in', return and output 'out' */
1419: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1420:
1421: char *strloc;
1422:
1423: strcpy (out, in);
1424: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1425: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1426: if(strloc != NULL){
1427: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1428: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1429: /* strcpy (strloc, strloc +strlen(chain));*/
1430: }
1431: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1432: return out;
1433: }
1434:
1435:
1.145 brouard 1436: char *cutl(char *blocc, char *alocc, char *in, char occ)
1437: {
1.187 brouard 1438: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1439: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1440: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1441: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1442: */
1.160 brouard 1443: char *s, *t;
1.145 brouard 1444: t=in;s=in;
1445: while ((*in != occ) && (*in != '\0')){
1446: *alocc++ = *in++;
1447: }
1448: if( *in == occ){
1449: *(alocc)='\0';
1450: s=++in;
1451: }
1452:
1453: if (s == t) {/* occ not found */
1454: *(alocc-(in-s))='\0';
1455: in=s;
1456: }
1457: while ( *in != '\0'){
1458: *blocc++ = *in++;
1459: }
1460:
1461: *blocc='\0';
1462: return t;
1463: }
1.137 brouard 1464: char *cutv(char *blocc, char *alocc, char *in, char occ)
1465: {
1.187 brouard 1466: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1467: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1468: gives blocc="abcdef2ghi" and alocc="j".
1469: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1470: */
1471: char *s, *t;
1472: t=in;s=in;
1473: while (*in != '\0'){
1474: while( *in == occ){
1475: *blocc++ = *in++;
1476: s=in;
1477: }
1478: *blocc++ = *in++;
1479: }
1480: if (s == t) /* occ not found */
1481: *(blocc-(in-s))='\0';
1482: else
1483: *(blocc-(in-s)-1)='\0';
1484: in=s;
1485: while ( *in != '\0'){
1486: *alocc++ = *in++;
1487: }
1488:
1489: *alocc='\0';
1490: return s;
1491: }
1492:
1.126 brouard 1493: int nbocc(char *s, char occ)
1494: {
1495: int i,j=0;
1496: int lg=20;
1497: i=0;
1498: lg=strlen(s);
1499: for(i=0; i<= lg; i++) {
1.234 brouard 1500: if (s[i] == occ ) j++;
1.126 brouard 1501: }
1502: return j;
1503: }
1504:
1.137 brouard 1505: /* void cutv(char *u,char *v, char*t, char occ) */
1506: /* { */
1507: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1508: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1509: /* gives u="abcdef2ghi" and v="j" *\/ */
1510: /* int i,lg,j,p=0; */
1511: /* i=0; */
1512: /* lg=strlen(t); */
1513: /* for(j=0; j<=lg-1; j++) { */
1514: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1515: /* } */
1.126 brouard 1516:
1.137 brouard 1517: /* for(j=0; j<p; j++) { */
1518: /* (u[j] = t[j]); */
1519: /* } */
1520: /* u[p]='\0'; */
1.126 brouard 1521:
1.137 brouard 1522: /* for(j=0; j<= lg; j++) { */
1523: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1524: /* } */
1525: /* } */
1.126 brouard 1526:
1.160 brouard 1527: #ifdef _WIN32
1528: char * strsep(char **pp, const char *delim)
1529: {
1530: char *p, *q;
1531:
1532: if ((p = *pp) == NULL)
1533: return 0;
1534: if ((q = strpbrk (p, delim)) != NULL)
1535: {
1536: *pp = q + 1;
1537: *q = '\0';
1538: }
1539: else
1540: *pp = 0;
1541: return p;
1542: }
1543: #endif
1544:
1.126 brouard 1545: /********************** nrerror ********************/
1546:
1547: void nrerror(char error_text[])
1548: {
1549: fprintf(stderr,"ERREUR ...\n");
1550: fprintf(stderr,"%s\n",error_text);
1551: exit(EXIT_FAILURE);
1552: }
1553: /*********************** vector *******************/
1554: double *vector(int nl, int nh)
1555: {
1556: double *v;
1557: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1558: if (!v) nrerror("allocation failure in vector");
1559: return v-nl+NR_END;
1560: }
1561:
1562: /************************ free vector ******************/
1563: void free_vector(double*v, int nl, int nh)
1564: {
1565: free((FREE_ARG)(v+nl-NR_END));
1566: }
1567:
1568: /************************ivector *******************************/
1569: int *ivector(long nl,long nh)
1570: {
1571: int *v;
1572: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1573: if (!v) nrerror("allocation failure in ivector");
1574: return v-nl+NR_END;
1575: }
1576:
1577: /******************free ivector **************************/
1578: void free_ivector(int *v, long nl, long nh)
1579: {
1580: free((FREE_ARG)(v+nl-NR_END));
1581: }
1582:
1583: /************************lvector *******************************/
1584: long *lvector(long nl,long nh)
1585: {
1586: long *v;
1587: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1588: if (!v) nrerror("allocation failure in ivector");
1589: return v-nl+NR_END;
1590: }
1591:
1592: /******************free lvector **************************/
1593: void free_lvector(long *v, long nl, long nh)
1594: {
1595: free((FREE_ARG)(v+nl-NR_END));
1596: }
1597:
1598: /******************* imatrix *******************************/
1599: int **imatrix(long nrl, long nrh, long ncl, long nch)
1600: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1601: {
1602: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1603: int **m;
1604:
1605: /* allocate pointers to rows */
1606: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1607: if (!m) nrerror("allocation failure 1 in matrix()");
1608: m += NR_END;
1609: m -= nrl;
1610:
1611:
1612: /* allocate rows and set pointers to them */
1613: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1614: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1615: m[nrl] += NR_END;
1616: m[nrl] -= ncl;
1617:
1618: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1619:
1620: /* return pointer to array of pointers to rows */
1621: return m;
1622: }
1623:
1624: /****************** free_imatrix *************************/
1625: void free_imatrix(m,nrl,nrh,ncl,nch)
1626: int **m;
1627: long nch,ncl,nrh,nrl;
1628: /* free an int matrix allocated by imatrix() */
1629: {
1630: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1631: free((FREE_ARG) (m+nrl-NR_END));
1632: }
1633:
1634: /******************* matrix *******************************/
1635: double **matrix(long nrl, long nrh, long ncl, long nch)
1636: {
1637: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1638: double **m;
1639:
1640: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1641: if (!m) nrerror("allocation failure 1 in matrix()");
1642: m += NR_END;
1643: m -= nrl;
1644:
1645: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1646: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1647: m[nrl] += NR_END;
1648: m[nrl] -= ncl;
1649:
1650: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1651: return m;
1.145 brouard 1652: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1653: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1654: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1655: */
1656: }
1657:
1658: /*************************free matrix ************************/
1659: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1660: {
1661: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1662: free((FREE_ARG)(m+nrl-NR_END));
1663: }
1664:
1665: /******************* ma3x *******************************/
1666: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1667: {
1668: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1669: double ***m;
1670:
1671: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1672: if (!m) nrerror("allocation failure 1 in matrix()");
1673: m += NR_END;
1674: m -= nrl;
1675:
1676: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1677: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1678: m[nrl] += NR_END;
1679: m[nrl] -= ncl;
1680:
1681: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1682:
1683: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1684: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1685: m[nrl][ncl] += NR_END;
1686: m[nrl][ncl] -= nll;
1687: for (j=ncl+1; j<=nch; j++)
1688: m[nrl][j]=m[nrl][j-1]+nlay;
1689:
1690: for (i=nrl+1; i<=nrh; i++) {
1691: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1692: for (j=ncl+1; j<=nch; j++)
1693: m[i][j]=m[i][j-1]+nlay;
1694: }
1695: return m;
1696: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1697: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1698: */
1699: }
1700:
1701: /*************************free ma3x ************************/
1702: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1703: {
1704: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1705: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1706: free((FREE_ARG)(m+nrl-NR_END));
1707: }
1708:
1709: /*************** function subdirf ***********/
1710: char *subdirf(char fileres[])
1711: {
1712: /* Caution optionfilefiname is hidden */
1713: strcpy(tmpout,optionfilefiname);
1714: strcat(tmpout,"/"); /* Add to the right */
1715: strcat(tmpout,fileres);
1716: return tmpout;
1717: }
1718:
1719: /*************** function subdirf2 ***********/
1720: char *subdirf2(char fileres[], char *preop)
1721: {
1722:
1723: /* Caution optionfilefiname is hidden */
1724: strcpy(tmpout,optionfilefiname);
1725: strcat(tmpout,"/");
1726: strcat(tmpout,preop);
1727: strcat(tmpout,fileres);
1728: return tmpout;
1729: }
1730:
1731: /*************** function subdirf3 ***********/
1732: char *subdirf3(char fileres[], char *preop, char *preop2)
1733: {
1734:
1735: /* Caution optionfilefiname is hidden */
1736: strcpy(tmpout,optionfilefiname);
1737: strcat(tmpout,"/");
1738: strcat(tmpout,preop);
1739: strcat(tmpout,preop2);
1740: strcat(tmpout,fileres);
1741: return tmpout;
1742: }
1.213 brouard 1743:
1744: /*************** function subdirfext ***********/
1745: char *subdirfext(char fileres[], char *preop, char *postop)
1746: {
1747:
1748: strcpy(tmpout,preop);
1749: strcat(tmpout,fileres);
1750: strcat(tmpout,postop);
1751: return tmpout;
1752: }
1.126 brouard 1753:
1.213 brouard 1754: /*************** function subdirfext3 ***********/
1755: char *subdirfext3(char fileres[], char *preop, char *postop)
1756: {
1757:
1758: /* Caution optionfilefiname is hidden */
1759: strcpy(tmpout,optionfilefiname);
1760: strcat(tmpout,"/");
1761: strcat(tmpout,preop);
1762: strcat(tmpout,fileres);
1763: strcat(tmpout,postop);
1764: return tmpout;
1765: }
1766:
1.162 brouard 1767: char *asc_diff_time(long time_sec, char ascdiff[])
1768: {
1769: long sec_left, days, hours, minutes;
1770: days = (time_sec) / (60*60*24);
1771: sec_left = (time_sec) % (60*60*24);
1772: hours = (sec_left) / (60*60) ;
1773: sec_left = (sec_left) %(60*60);
1774: minutes = (sec_left) /60;
1775: sec_left = (sec_left) % (60);
1776: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1777: return ascdiff;
1778: }
1779:
1.126 brouard 1780: /***************** f1dim *************************/
1781: extern int ncom;
1782: extern double *pcom,*xicom;
1783: extern double (*nrfunc)(double []);
1784:
1785: double f1dim(double x)
1786: {
1787: int j;
1788: double f;
1789: double *xt;
1790:
1791: xt=vector(1,ncom);
1792: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1793: f=(*nrfunc)(xt);
1794: free_vector(xt,1,ncom);
1795: return f;
1796: }
1797:
1798: /*****************brent *************************/
1799: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1800: {
1801: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1802: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1803: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1804: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1805: * returned function value.
1806: */
1.126 brouard 1807: int iter;
1808: double a,b,d,etemp;
1.159 brouard 1809: double fu=0,fv,fw,fx;
1.164 brouard 1810: double ftemp=0.;
1.126 brouard 1811: double p,q,r,tol1,tol2,u,v,w,x,xm;
1812: double e=0.0;
1813:
1814: a=(ax < cx ? ax : cx);
1815: b=(ax > cx ? ax : cx);
1816: x=w=v=bx;
1817: fw=fv=fx=(*f)(x);
1818: for (iter=1;iter<=ITMAX;iter++) {
1819: xm=0.5*(a+b);
1820: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1821: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1822: printf(".");fflush(stdout);
1823: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1824: #ifdef DEBUGBRENT
1.126 brouard 1825: 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);
1826: 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);
1827: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1828: #endif
1829: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1830: *xmin=x;
1831: return fx;
1832: }
1833: ftemp=fu;
1834: if (fabs(e) > tol1) {
1835: r=(x-w)*(fx-fv);
1836: q=(x-v)*(fx-fw);
1837: p=(x-v)*q-(x-w)*r;
1838: q=2.0*(q-r);
1839: if (q > 0.0) p = -p;
1840: q=fabs(q);
1841: etemp=e;
1842: e=d;
1843: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1844: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1845: else {
1.224 brouard 1846: d=p/q;
1847: u=x+d;
1848: if (u-a < tol2 || b-u < tol2)
1849: d=SIGN(tol1,xm-x);
1.126 brouard 1850: }
1851: } else {
1852: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1853: }
1854: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1855: fu=(*f)(u);
1856: if (fu <= fx) {
1857: if (u >= x) a=x; else b=x;
1858: SHFT(v,w,x,u)
1.183 brouard 1859: SHFT(fv,fw,fx,fu)
1860: } else {
1861: if (u < x) a=u; else b=u;
1862: if (fu <= fw || w == x) {
1.224 brouard 1863: v=w;
1864: w=u;
1865: fv=fw;
1866: fw=fu;
1.183 brouard 1867: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1868: v=u;
1869: fv=fu;
1.183 brouard 1870: }
1871: }
1.126 brouard 1872: }
1873: nrerror("Too many iterations in brent");
1874: *xmin=x;
1875: return fx;
1876: }
1877:
1878: /****************** mnbrak ***********************/
1879:
1880: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1881: double (*func)(double))
1.183 brouard 1882: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1883: the downhill direction (defined by the function as evaluated at the initial points) and returns
1884: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1885: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1886: */
1.126 brouard 1887: double ulim,u,r,q, dum;
1888: double fu;
1.187 brouard 1889:
1890: double scale=10.;
1891: int iterscale=0;
1892:
1893: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1894: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1895:
1896:
1897: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1898: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1899: /* *bx = *ax - (*ax - *bx)/scale; */
1900: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1901: /* } */
1902:
1.126 brouard 1903: if (*fb > *fa) {
1904: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1905: SHFT(dum,*fb,*fa,dum)
1906: }
1.126 brouard 1907: *cx=(*bx)+GOLD*(*bx-*ax);
1908: *fc=(*func)(*cx);
1.183 brouard 1909: #ifdef DEBUG
1.224 brouard 1910: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1911: 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 1912: #endif
1.224 brouard 1913: 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 1914: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1915: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1916: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1917: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1918: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1919: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1920: fu=(*func)(u);
1.163 brouard 1921: #ifdef DEBUG
1922: /* f(x)=A(x-u)**2+f(u) */
1923: double A, fparabu;
1924: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1925: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1926: 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);
1927: 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 1928: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1929: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1930: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1931: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1932: #endif
1.184 brouard 1933: #ifdef MNBRAKORIGINAL
1.183 brouard 1934: #else
1.191 brouard 1935: /* if (fu > *fc) { */
1936: /* #ifdef DEBUG */
1937: /* printf("mnbrak4 fu > fc \n"); */
1938: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1939: /* #endif */
1940: /* /\* 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 *\\/ *\/ */
1941: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1942: /* dum=u; /\* Shifting c and u *\/ */
1943: /* u = *cx; */
1944: /* *cx = dum; */
1945: /* dum = fu; */
1946: /* fu = *fc; */
1947: /* *fc =dum; */
1948: /* } else { /\* end *\/ */
1949: /* #ifdef DEBUG */
1950: /* printf("mnbrak3 fu < fc \n"); */
1951: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1952: /* #endif */
1953: /* dum=u; /\* Shifting c and u *\/ */
1954: /* u = *cx; */
1955: /* *cx = dum; */
1956: /* dum = fu; */
1957: /* fu = *fc; */
1958: /* *fc =dum; */
1959: /* } */
1.224 brouard 1960: #ifdef DEBUGMNBRAK
1961: double A, fparabu;
1962: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1963: fparabu= *fa - A*(*ax-u)*(*ax-u);
1964: 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);
1965: 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 1966: #endif
1.191 brouard 1967: dum=u; /* Shifting c and u */
1968: u = *cx;
1969: *cx = dum;
1970: dum = fu;
1971: fu = *fc;
1972: *fc =dum;
1.183 brouard 1973: #endif
1.162 brouard 1974: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1975: #ifdef DEBUG
1.224 brouard 1976: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1977: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1978: #endif
1.126 brouard 1979: fu=(*func)(u);
1980: if (fu < *fc) {
1.183 brouard 1981: #ifdef DEBUG
1.224 brouard 1982: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1983: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1984: #endif
1985: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1986: SHFT(*fb,*fc,fu,(*func)(u))
1987: #ifdef DEBUG
1988: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1989: #endif
1990: }
1.162 brouard 1991: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1992: #ifdef DEBUG
1.224 brouard 1993: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1994: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1995: #endif
1.126 brouard 1996: u=ulim;
1997: fu=(*func)(u);
1.183 brouard 1998: } else { /* u could be left to b (if r > q parabola has a maximum) */
1999: #ifdef DEBUG
1.224 brouard 2000: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2001: 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 2002: #endif
1.126 brouard 2003: u=(*cx)+GOLD*(*cx-*bx);
2004: fu=(*func)(u);
1.224 brouard 2005: #ifdef DEBUG
2006: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2007: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2008: #endif
1.183 brouard 2009: } /* end tests */
1.126 brouard 2010: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2011: SHFT(*fa,*fb,*fc,fu)
2012: #ifdef DEBUG
1.224 brouard 2013: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2014: 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 2015: #endif
2016: } /* 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 2017: }
2018:
2019: /*************** linmin ************************/
1.162 brouard 2020: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2021: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2022: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2023: the value of func at the returned location p . This is actually all accomplished by calling the
2024: routines mnbrak and brent .*/
1.126 brouard 2025: int ncom;
2026: double *pcom,*xicom;
2027: double (*nrfunc)(double []);
2028:
1.224 brouard 2029: #ifdef LINMINORIGINAL
1.126 brouard 2030: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2031: #else
2032: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2033: #endif
1.126 brouard 2034: {
2035: double brent(double ax, double bx, double cx,
2036: double (*f)(double), double tol, double *xmin);
2037: double f1dim(double x);
2038: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2039: double *fc, double (*func)(double));
2040: int j;
2041: double xx,xmin,bx,ax;
2042: double fx,fb,fa;
1.187 brouard 2043:
1.203 brouard 2044: #ifdef LINMINORIGINAL
2045: #else
2046: double scale=10., axs, xxs; /* Scale added for infinity */
2047: #endif
2048:
1.126 brouard 2049: ncom=n;
2050: pcom=vector(1,n);
2051: xicom=vector(1,n);
2052: nrfunc=func;
2053: for (j=1;j<=n;j++) {
2054: pcom[j]=p[j];
1.202 brouard 2055: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2056: }
1.187 brouard 2057:
1.203 brouard 2058: #ifdef LINMINORIGINAL
2059: xx=1.;
2060: #else
2061: axs=0.0;
2062: xxs=1.;
2063: do{
2064: xx= xxs;
2065: #endif
1.187 brouard 2066: ax=0.;
2067: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2068: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2069: /* 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)) */
2070: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2071: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2072: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2073: /* 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 2074: #ifdef LINMINORIGINAL
2075: #else
2076: if (fx != fx){
1.224 brouard 2077: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2078: printf("|");
2079: fprintf(ficlog,"|");
1.203 brouard 2080: #ifdef DEBUGLINMIN
1.224 brouard 2081: 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 2082: #endif
2083: }
1.224 brouard 2084: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2085: #endif
2086:
1.191 brouard 2087: #ifdef DEBUGLINMIN
2088: 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 2089: 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 2090: #endif
1.224 brouard 2091: #ifdef LINMINORIGINAL
2092: #else
2093: if(fb == fx){ /* Flat function in the direction */
2094: xmin=xx;
2095: *flat=1;
2096: }else{
2097: *flat=0;
2098: #endif
2099: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2100: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2101: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2102: /* fmin = f(p[j] + xmin * xi[j]) */
2103: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2104: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2105: #ifdef DEBUG
1.224 brouard 2106: 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);
2107: 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);
2108: #endif
2109: #ifdef LINMINORIGINAL
2110: #else
2111: }
1.126 brouard 2112: #endif
1.191 brouard 2113: #ifdef DEBUGLINMIN
2114: printf("linmin end ");
1.202 brouard 2115: fprintf(ficlog,"linmin end ");
1.191 brouard 2116: #endif
1.126 brouard 2117: for (j=1;j<=n;j++) {
1.203 brouard 2118: #ifdef LINMINORIGINAL
2119: xi[j] *= xmin;
2120: #else
2121: #ifdef DEBUGLINMIN
2122: if(xxs <1.0)
2123: printf(" before xi[%d]=%12.8f", j,xi[j]);
2124: #endif
2125: 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) */
2126: #ifdef DEBUGLINMIN
2127: if(xxs <1.0)
2128: 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 );
2129: #endif
2130: #endif
1.187 brouard 2131: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2132: }
1.191 brouard 2133: #ifdef DEBUGLINMIN
1.203 brouard 2134: printf("\n");
1.191 brouard 2135: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2136: 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 2137: for (j=1;j<=n;j++) {
1.202 brouard 2138: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2139: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2140: if(j % ncovmodel == 0){
1.191 brouard 2141: printf("\n");
1.202 brouard 2142: fprintf(ficlog,"\n");
2143: }
1.191 brouard 2144: }
1.203 brouard 2145: #else
1.191 brouard 2146: #endif
1.126 brouard 2147: free_vector(xicom,1,n);
2148: free_vector(pcom,1,n);
2149: }
2150:
2151:
2152: /*************** powell ************************/
1.162 brouard 2153: /*
2154: Minimization of a function func of n variables. Input consists of an initial starting point
2155: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2156: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2157: such that failure to decrease by more than this amount on one iteration signals doneness. On
2158: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2159: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2160: */
1.224 brouard 2161: #ifdef LINMINORIGINAL
2162: #else
2163: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2164: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2165: #endif
1.126 brouard 2166: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2167: double (*func)(double []))
2168: {
1.224 brouard 2169: #ifdef LINMINORIGINAL
2170: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2171: double (*func)(double []));
1.224 brouard 2172: #else
1.241 brouard 2173: void linmin(double p[], double xi[], int n, double *fret,
2174: double (*func)(double []),int *flat);
1.224 brouard 2175: #endif
1.239 brouard 2176: int i,ibig,j,jk,k;
1.126 brouard 2177: double del,t,*pt,*ptt,*xit;
1.181 brouard 2178: double directest;
1.126 brouard 2179: double fp,fptt;
2180: double *xits;
2181: int niterf, itmp;
1.224 brouard 2182: #ifdef LINMINORIGINAL
2183: #else
2184:
2185: flatdir=ivector(1,n);
2186: for (j=1;j<=n;j++) flatdir[j]=0;
2187: #endif
1.126 brouard 2188:
2189: pt=vector(1,n);
2190: ptt=vector(1,n);
2191: xit=vector(1,n);
2192: xits=vector(1,n);
2193: *fret=(*func)(p);
2194: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2195: rcurr_time = time(NULL);
1.126 brouard 2196: for (*iter=1;;++(*iter)) {
1.187 brouard 2197: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2198: ibig=0;
2199: del=0.0;
1.157 brouard 2200: rlast_time=rcurr_time;
2201: /* (void) gettimeofday(&curr_time,&tzp); */
2202: rcurr_time = time(NULL);
2203: curr_time = *localtime(&rcurr_time);
2204: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2205: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2206: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2207: for (i=1;i<=n;i++) {
1.126 brouard 2208: fprintf(ficrespow," %.12lf", p[i]);
2209: }
1.239 brouard 2210: fprintf(ficrespow,"\n");fflush(ficrespow);
2211: printf("\n#model= 1 + age ");
2212: fprintf(ficlog,"\n#model= 1 + age ");
2213: if(nagesqr==1){
1.241 brouard 2214: printf(" + age*age ");
2215: fprintf(ficlog," + age*age ");
1.239 brouard 2216: }
2217: for(j=1;j <=ncovmodel-2;j++){
2218: if(Typevar[j]==0) {
2219: printf(" + V%d ",Tvar[j]);
2220: fprintf(ficlog," + V%d ",Tvar[j]);
2221: }else if(Typevar[j]==1) {
2222: printf(" + V%d*age ",Tvar[j]);
2223: fprintf(ficlog," + V%d*age ",Tvar[j]);
2224: }else if(Typevar[j]==2) {
2225: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2226: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2227: }
2228: }
1.126 brouard 2229: printf("\n");
1.239 brouard 2230: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2231: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2232: fprintf(ficlog,"\n");
1.239 brouard 2233: for(i=1,jk=1; i <=nlstate; i++){
2234: for(k=1; k <=(nlstate+ndeath); k++){
2235: if (k != i) {
2236: printf("%d%d ",i,k);
2237: fprintf(ficlog,"%d%d ",i,k);
2238: for(j=1; j <=ncovmodel; j++){
2239: printf("%12.7f ",p[jk]);
2240: fprintf(ficlog,"%12.7f ",p[jk]);
2241: jk++;
2242: }
2243: printf("\n");
2244: fprintf(ficlog,"\n");
2245: }
2246: }
2247: }
1.241 brouard 2248: if(*iter <=3 && *iter >1){
1.157 brouard 2249: tml = *localtime(&rcurr_time);
2250: strcpy(strcurr,asctime(&tml));
2251: rforecast_time=rcurr_time;
1.126 brouard 2252: itmp = strlen(strcurr);
2253: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2254: strcurr[itmp-1]='\0';
1.162 brouard 2255: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2256: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2257: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2258: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2259: forecast_time = *localtime(&rforecast_time);
2260: strcpy(strfor,asctime(&forecast_time));
2261: itmp = strlen(strfor);
2262: if(strfor[itmp-1]=='\n')
2263: strfor[itmp-1]='\0';
2264: 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);
2265: 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 2266: }
2267: }
1.187 brouard 2268: for (i=1;i<=n;i++) { /* For each direction i */
2269: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2270: fptt=(*fret);
2271: #ifdef DEBUG
1.203 brouard 2272: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2273: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2274: #endif
1.203 brouard 2275: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2276: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2277: #ifdef LINMINORIGINAL
1.188 brouard 2278: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2279: #else
2280: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2281: flatdir[i]=flat; /* Function is vanishing in that direction i */
2282: #endif
2283: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2284: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2285: /* because that direction will be replaced unless the gain del is small */
2286: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2287: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2288: /* with the new direction. */
2289: del=fabs(fptt-(*fret));
2290: ibig=i;
1.126 brouard 2291: }
2292: #ifdef DEBUG
2293: printf("%d %.12e",i,(*fret));
2294: fprintf(ficlog,"%d %.12e",i,(*fret));
2295: for (j=1;j<=n;j++) {
1.224 brouard 2296: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2297: printf(" x(%d)=%.12e",j,xit[j]);
2298: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2299: }
2300: for(j=1;j<=n;j++) {
1.225 brouard 2301: printf(" p(%d)=%.12e",j,p[j]);
2302: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2303: }
2304: printf("\n");
2305: fprintf(ficlog,"\n");
2306: #endif
1.187 brouard 2307: } /* end loop on each direction i */
2308: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2309: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2310: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2311: for(j=1;j<=n;j++) {
1.225 brouard 2312: if(flatdir[j] >0){
2313: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2314: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2315: }
2316: /* printf("\n"); */
2317: /* fprintf(ficlog,"\n"); */
2318: }
1.243 brouard 2319: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2320: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2321: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2322: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2323: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2324: /* decreased of more than 3.84 */
2325: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2326: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2327: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2328:
1.188 brouard 2329: /* Starting the program with initial values given by a former maximization will simply change */
2330: /* the scales of the directions and the directions, because the are reset to canonical directions */
2331: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2332: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2333: #ifdef DEBUG
2334: int k[2],l;
2335: k[0]=1;
2336: k[1]=-1;
2337: printf("Max: %.12e",(*func)(p));
2338: fprintf(ficlog,"Max: %.12e",(*func)(p));
2339: for (j=1;j<=n;j++) {
2340: printf(" %.12e",p[j]);
2341: fprintf(ficlog," %.12e",p[j]);
2342: }
2343: printf("\n");
2344: fprintf(ficlog,"\n");
2345: for(l=0;l<=1;l++) {
2346: for (j=1;j<=n;j++) {
2347: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2348: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2349: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2350: }
2351: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2352: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2353: }
2354: #endif
2355:
1.224 brouard 2356: #ifdef LINMINORIGINAL
2357: #else
2358: free_ivector(flatdir,1,n);
2359: #endif
1.126 brouard 2360: free_vector(xit,1,n);
2361: free_vector(xits,1,n);
2362: free_vector(ptt,1,n);
2363: free_vector(pt,1,n);
2364: return;
1.192 brouard 2365: } /* enough precision */
1.240 brouard 2366: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2367: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2368: ptt[j]=2.0*p[j]-pt[j];
2369: xit[j]=p[j]-pt[j];
2370: pt[j]=p[j];
2371: }
1.181 brouard 2372: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2373: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2374: if (*iter <=4) {
1.225 brouard 2375: #else
2376: #endif
1.224 brouard 2377: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2378: #else
1.161 brouard 2379: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2380: #endif
1.162 brouard 2381: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2382: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2383: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2384: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2385: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2386: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2387: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2388: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2389: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2390: /* Even if f3 <f1, directest can be negative and t >0 */
2391: /* mu² and del² are equal when f3=f1 */
2392: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2393: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2394: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2395: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2396: #ifdef NRCORIGINAL
2397: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2398: #else
2399: 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 2400: t= t- del*SQR(fp-fptt);
1.183 brouard 2401: #endif
1.202 brouard 2402: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2403: #ifdef DEBUG
1.181 brouard 2404: 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);
2405: 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 2406: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2407: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2408: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2409: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2410: 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);
2411: 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);
2412: #endif
1.183 brouard 2413: #ifdef POWELLORIGINAL
2414: if (t < 0.0) { /* Then we use it for new direction */
2415: #else
1.182 brouard 2416: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2417: 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 2418: 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 2419: 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 2420: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2421: }
1.181 brouard 2422: if (directest < 0.0) { /* Then we use it for new direction */
2423: #endif
1.191 brouard 2424: #ifdef DEBUGLINMIN
1.234 brouard 2425: printf("Before linmin in direction P%d-P0\n",n);
2426: for (j=1;j<=n;j++) {
2427: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2428: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2429: if(j % ncovmodel == 0){
2430: printf("\n");
2431: fprintf(ficlog,"\n");
2432: }
2433: }
1.224 brouard 2434: #endif
2435: #ifdef LINMINORIGINAL
1.234 brouard 2436: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2437: #else
1.234 brouard 2438: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2439: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2440: #endif
1.234 brouard 2441:
1.191 brouard 2442: #ifdef DEBUGLINMIN
1.234 brouard 2443: for (j=1;j<=n;j++) {
2444: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2445: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2446: if(j % ncovmodel == 0){
2447: printf("\n");
2448: fprintf(ficlog,"\n");
2449: }
2450: }
1.224 brouard 2451: #endif
1.234 brouard 2452: for (j=1;j<=n;j++) {
2453: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2454: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2455: }
1.224 brouard 2456: #ifdef LINMINORIGINAL
2457: #else
1.234 brouard 2458: for (j=1, flatd=0;j<=n;j++) {
2459: if(flatdir[j]>0)
2460: flatd++;
2461: }
2462: if(flatd >0){
1.255 brouard 2463: printf("%d flat directions: ",flatd);
2464: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2465: for (j=1;j<=n;j++) {
2466: if(flatdir[j]>0){
2467: printf("%d ",j);
2468: fprintf(ficlog,"%d ",j);
2469: }
2470: }
2471: printf("\n");
2472: fprintf(ficlog,"\n");
2473: }
1.191 brouard 2474: #endif
1.234 brouard 2475: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2476: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2477:
1.126 brouard 2478: #ifdef DEBUG
1.234 brouard 2479: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2480: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2481: for(j=1;j<=n;j++){
2482: printf(" %lf",xit[j]);
2483: fprintf(ficlog," %lf",xit[j]);
2484: }
2485: printf("\n");
2486: fprintf(ficlog,"\n");
1.126 brouard 2487: #endif
1.192 brouard 2488: } /* end of t or directest negative */
1.224 brouard 2489: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2490: #else
1.234 brouard 2491: } /* end if (fptt < fp) */
1.192 brouard 2492: #endif
1.225 brouard 2493: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2494: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2495: #else
1.224 brouard 2496: #endif
1.234 brouard 2497: } /* loop iteration */
1.126 brouard 2498: }
1.234 brouard 2499:
1.126 brouard 2500: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2501:
1.235 brouard 2502: 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 2503: {
1.235 brouard 2504: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2505: (and selected quantitative values in nres)
2506: by left multiplying the unit
1.234 brouard 2507: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2508: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2509: /* Wx is row vector: population in state 1, population in state 2, population dead */
2510: /* or prevalence in state 1, prevalence in state 2, 0 */
2511: /* newm is the matrix after multiplications, its rows are identical at a factor */
2512: /* Initial matrix pimij */
2513: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2514: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2515: /* 0, 0 , 1} */
2516: /*
2517: * and after some iteration: */
2518: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2519: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2520: /* 0, 0 , 1} */
2521: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2522: /* {0.51571254859325999, 0.4842874514067399, */
2523: /* 0.51326036147820708, 0.48673963852179264} */
2524: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2525:
1.126 brouard 2526: int i, ii,j,k;
1.209 brouard 2527: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2528: /* double **matprod2(); */ /* test */
1.218 brouard 2529: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2530: double **newm;
1.209 brouard 2531: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2532: int ncvloop=0;
1.169 brouard 2533:
1.209 brouard 2534: min=vector(1,nlstate);
2535: max=vector(1,nlstate);
2536: meandiff=vector(1,nlstate);
2537:
1.218 brouard 2538: /* Starting with matrix unity */
1.126 brouard 2539: for (ii=1;ii<=nlstate+ndeath;ii++)
2540: for (j=1;j<=nlstate+ndeath;j++){
2541: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2542: }
1.169 brouard 2543:
2544: cov[1]=1.;
2545:
2546: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2547: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2548: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2549: ncvloop++;
1.126 brouard 2550: newm=savm;
2551: /* Covariates have to be included here again */
1.138 brouard 2552: cov[2]=agefin;
1.187 brouard 2553: if(nagesqr==1)
2554: cov[3]= agefin*agefin;;
1.234 brouard 2555: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2556: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2557: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2558: /* 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 2559: }
2560: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2561: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2562: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2563: /* 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 2564: }
1.237 brouard 2565: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2566: if(Dummy[Tvar[Tage[k]]]){
2567: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2568: } else{
1.235 brouard 2569: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2570: }
1.235 brouard 2571: /* 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 2572: }
1.237 brouard 2573: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2574: /* 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 2575: if(Dummy[Tvard[k][1]==0]){
2576: if(Dummy[Tvard[k][2]==0]){
2577: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2578: }else{
2579: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2580: }
2581: }else{
2582: if(Dummy[Tvard[k][2]==0]){
2583: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2584: }else{
2585: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2586: }
2587: }
1.234 brouard 2588: }
1.138 brouard 2589: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2590: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2591: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2592: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2593: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2594: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2595: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2596:
1.126 brouard 2597: savm=oldm;
2598: oldm=newm;
1.209 brouard 2599:
2600: for(j=1; j<=nlstate; j++){
2601: max[j]=0.;
2602: min[j]=1.;
2603: }
2604: for(i=1;i<=nlstate;i++){
2605: sumnew=0;
2606: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2607: for(j=1; j<=nlstate; j++){
2608: prlim[i][j]= newm[i][j]/(1-sumnew);
2609: max[j]=FMAX(max[j],prlim[i][j]);
2610: min[j]=FMIN(min[j],prlim[i][j]);
2611: }
2612: }
2613:
1.126 brouard 2614: maxmax=0.;
1.209 brouard 2615: for(j=1; j<=nlstate; j++){
2616: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2617: maxmax=FMAX(maxmax,meandiff[j]);
2618: /* 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 2619: } /* j loop */
1.203 brouard 2620: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2621: /* 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 2622: if(maxmax < ftolpl){
1.209 brouard 2623: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2624: free_vector(min,1,nlstate);
2625: free_vector(max,1,nlstate);
2626: free_vector(meandiff,1,nlstate);
1.126 brouard 2627: return prlim;
2628: }
1.169 brouard 2629: } /* age loop */
1.208 brouard 2630: /* After some age loop it doesn't converge */
1.209 brouard 2631: 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 2632: 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 2633: /* 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); */
2634: free_vector(min,1,nlstate);
2635: free_vector(max,1,nlstate);
2636: free_vector(meandiff,1,nlstate);
1.208 brouard 2637:
1.169 brouard 2638: return prlim; /* should not reach here */
1.126 brouard 2639: }
2640:
1.217 brouard 2641:
2642: /**** Back Prevalence limit (stable or period prevalence) ****************/
2643:
1.218 brouard 2644: /* 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) */
2645: /* 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 2646: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2647: {
1.264 brouard 2648: /* 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 2649: matrix by transitions matrix until convergence is reached with precision ftolpl */
2650: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2651: /* Wx is row vector: population in state 1, population in state 2, population dead */
2652: /* or prevalence in state 1, prevalence in state 2, 0 */
2653: /* newm is the matrix after multiplications, its rows are identical at a factor */
2654: /* Initial matrix pimij */
2655: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2656: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2657: /* 0, 0 , 1} */
2658: /*
2659: * and after some iteration: */
2660: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2661: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2662: /* 0, 0 , 1} */
2663: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2664: /* {0.51571254859325999, 0.4842874514067399, */
2665: /* 0.51326036147820708, 0.48673963852179264} */
2666: /* If we start from prlim again, prlim tends to a constant matrix */
2667:
2668: int i, ii,j,k;
1.247 brouard 2669: int first=0;
1.217 brouard 2670: double *min, *max, *meandiff, maxmax,sumnew=0.;
2671: /* double **matprod2(); */ /* test */
2672: double **out, cov[NCOVMAX+1], **bmij();
2673: double **newm;
1.218 brouard 2674: double **dnewm, **doldm, **dsavm; /* for use */
2675: double **oldm, **savm; /* for use */
2676:
1.217 brouard 2677: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2678: int ncvloop=0;
2679:
2680: min=vector(1,nlstate);
2681: max=vector(1,nlstate);
2682: meandiff=vector(1,nlstate);
2683:
1.266 brouard 2684: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2685: oldm=oldms; savm=savms;
2686:
2687: /* Starting with matrix unity */
2688: for (ii=1;ii<=nlstate+ndeath;ii++)
2689: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2690: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2691: }
2692:
2693: cov[1]=1.;
2694:
2695: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2696: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2697: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2698: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2699: ncvloop++;
1.218 brouard 2700: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2701: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2702: /* Covariates have to be included here again */
2703: cov[2]=agefin;
2704: if(nagesqr==1)
2705: cov[3]= agefin*agefin;;
1.242 brouard 2706: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2707: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2708: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2709: /* 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 2710: }
2711: /* for (k=1; k<=cptcovn;k++) { */
2712: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2713: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2714: /* /\* 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])]); *\/ */
2715: /* } */
2716: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2717: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2718: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2719: /* 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]); */
2720: }
2721: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2722: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2723: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2724: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2725: for (k=1; k<=cptcovage;k++){ /* For product with age */
2726: if(Dummy[Tvar[Tage[k]]]){
2727: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2728: } else{
2729: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2730: }
2731: /* 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]); */
2732: }
2733: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2734: /* 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]); */
2735: if(Dummy[Tvard[k][1]==0]){
2736: if(Dummy[Tvard[k][2]==0]){
2737: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2738: }else{
2739: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2740: }
2741: }else{
2742: if(Dummy[Tvard[k][2]==0]){
2743: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2744: }else{
2745: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2746: }
2747: }
1.217 brouard 2748: }
2749:
2750: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2751: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2752: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2753: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2754: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2755: /* ij should be linked to the correct index of cov */
2756: /* age and covariate values ij are in 'cov', but we need to pass
2757: * ij for the observed prevalence at age and status and covariate
2758: * number: prevacurrent[(int)agefin][ii][ij]
2759: */
2760: /* 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 *\/ */
2761: /* 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 *\/ */
2762: 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 2763: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2764: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2765: /* for(i=1; i<=nlstate+ndeath; i++) { */
2766: /* printf("%d newm= ",i); */
2767: /* for(j=1;j<=nlstate+ndeath;j++) { */
2768: /* printf("%f ",newm[i][j]); */
2769: /* } */
2770: /* printf("oldm * "); */
2771: /* for(j=1;j<=nlstate+ndeath;j++) { */
2772: /* printf("%f ",oldm[i][j]); */
2773: /* } */
1.268 brouard 2774: /* printf(" bmmij "); */
1.266 brouard 2775: /* for(j=1;j<=nlstate+ndeath;j++) { */
2776: /* printf("%f ",pmmij[i][j]); */
2777: /* } */
2778: /* printf("\n"); */
2779: /* } */
2780: /* } */
1.217 brouard 2781: savm=oldm;
2782: oldm=newm;
1.266 brouard 2783:
1.217 brouard 2784: for(j=1; j<=nlstate; j++){
2785: max[j]=0.;
2786: min[j]=1.;
2787: }
2788: for(j=1; j<=nlstate; j++){
2789: for(i=1;i<=nlstate;i++){
1.234 brouard 2790: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2791: bprlim[i][j]= newm[i][j];
2792: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2793: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2794: }
2795: }
1.218 brouard 2796:
1.217 brouard 2797: maxmax=0.;
2798: for(i=1; i<=nlstate; i++){
2799: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2800: maxmax=FMAX(maxmax,meandiff[i]);
2801: /* 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 2802: } /* i loop */
1.217 brouard 2803: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2804: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2805: if(maxmax < ftolpl){
1.220 brouard 2806: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2807: free_vector(min,1,nlstate);
2808: free_vector(max,1,nlstate);
2809: free_vector(meandiff,1,nlstate);
2810: return bprlim;
2811: }
2812: } /* age loop */
2813: /* After some age loop it doesn't converge */
1.247 brouard 2814: if(first){
2815: first=1;
2816: 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\
2817: 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);
2818: }
2819: 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 2820: Oldest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
2821: /* 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); */
2822: free_vector(min,1,nlstate);
2823: free_vector(max,1,nlstate);
2824: free_vector(meandiff,1,nlstate);
2825:
2826: return bprlim; /* should not reach here */
2827: }
2828:
1.126 brouard 2829: /*************** transition probabilities ***************/
2830:
2831: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2832: {
1.138 brouard 2833: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2834: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2835: model to the ncovmodel covariates (including constant and age).
2836: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2837: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2838: ncth covariate in the global vector x is given by the formula:
2839: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2840: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2841: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2842: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2843: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2844: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2845: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2846: */
2847: double s1, lnpijopii;
1.126 brouard 2848: /*double t34;*/
1.164 brouard 2849: int i,j, nc, ii, jj;
1.126 brouard 2850:
1.223 brouard 2851: for(i=1; i<= nlstate; i++){
2852: for(j=1; j<i;j++){
2853: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2854: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2855: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2856: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2857: }
2858: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2859: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2860: }
2861: for(j=i+1; j<=nlstate+ndeath;j++){
2862: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2863: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2864: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2865: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2866: }
2867: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2868: }
2869: }
1.218 brouard 2870:
1.223 brouard 2871: for(i=1; i<= nlstate; i++){
2872: s1=0;
2873: for(j=1; j<i; j++){
2874: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2875: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2876: }
2877: for(j=i+1; j<=nlstate+ndeath; j++){
2878: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2879: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2880: }
2881: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2882: ps[i][i]=1./(s1+1.);
2883: /* Computing other pijs */
2884: for(j=1; j<i; j++)
2885: ps[i][j]= exp(ps[i][j])*ps[i][i];
2886: for(j=i+1; j<=nlstate+ndeath; j++)
2887: ps[i][j]= exp(ps[i][j])*ps[i][i];
2888: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2889: } /* end i */
1.218 brouard 2890:
1.223 brouard 2891: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2892: for(jj=1; jj<= nlstate+ndeath; jj++){
2893: ps[ii][jj]=0;
2894: ps[ii][ii]=1;
2895: }
2896: }
1.218 brouard 2897:
2898:
1.223 brouard 2899: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2900: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2901: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2902: /* } */
2903: /* printf("\n "); */
2904: /* } */
2905: /* printf("\n ");printf("%lf ",cov[2]);*/
2906: /*
2907: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2908: goto end;*/
1.266 brouard 2909: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2910: }
2911:
1.218 brouard 2912: /*************** backward transition probabilities ***************/
2913:
2914: /* 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 ) */
2915: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2916: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2917: {
1.266 brouard 2918: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2919: * 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 2920: */
1.218 brouard 2921: int i, ii, j,k;
1.222 brouard 2922:
2923: double **out, **pmij();
2924: double sumnew=0.;
1.218 brouard 2925: double agefin;
1.268 brouard 2926: 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 2927: double **dnewm, **dsavm, **doldm;
2928: double **bbmij;
2929:
1.218 brouard 2930: doldm=ddoldms; /* global pointers */
1.222 brouard 2931: dnewm=ddnewms;
2932: dsavm=ddsavms;
2933:
2934: agefin=cov[2];
1.268 brouard 2935: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 2936: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2937: the observed prevalence (with this covariate ij) at beginning of transition */
2938: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 2939:
2940: /* P_x */
1.266 brouard 2941: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 2942: /* outputs pmmij which is a stochastic matrix in row */
2943:
2944: /* Diag(w_x) */
2945: /* Problem with prevacurrent which can be zero */
2946: sumnew=0.;
1.269 brouard 2947: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 2948: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.269 brouard 2949: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 2950: sumnew+=prevacurrent[(int)agefin][ii][ij];
2951: }
2952: if(sumnew >0.01){ /* At least some value in the prevalence */
2953: for (ii=1;ii<=nlstate+ndeath;ii++){
2954: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 2955: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 2956: }
2957: }else{
2958: for (ii=1;ii<=nlstate+ndeath;ii++){
2959: for (j=1;j<=nlstate+ndeath;j++)
2960: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
2961: }
2962: /* if(sumnew <0.9){ */
2963: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2964: /* } */
2965: }
2966: k3=0.0; /* We put the last diagonal to 0 */
2967: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
2968: doldm[ii][ii]= k3;
2969: }
2970: /* End doldm, At the end doldm is diag[(w_i)] */
2971:
2972: /* left Product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm) */
2973: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* Bug Valgrind */
2974:
2975: /* Diag(Sum_i w^i_x p^ij_x */
2976: /* 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 2977: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 2978: sumnew=0.;
1.222 brouard 2979: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 2980: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 2981: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 2982: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 2983: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 2984: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 2985: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2986: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 2987: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 2988: /* }else */
1.268 brouard 2989: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2990: } /*End ii */
2991: } /* 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 */
2992:
2993: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* Bug Valgrind */
2994: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 2995: /* end bmij */
1.266 brouard 2996: return ps; /*pointer is unchanged */
1.218 brouard 2997: }
1.217 brouard 2998: /*************** transition probabilities ***************/
2999:
1.218 brouard 3000: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3001: {
3002: /* According to parameters values stored in x and the covariate's values stored in cov,
3003: computes the probability to be observed in state j being in state i by appying the
3004: model to the ncovmodel covariates (including constant and age).
3005: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3006: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3007: ncth covariate in the global vector x is given by the formula:
3008: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3009: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3010: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3011: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3012: Outputs ps[i][j] the probability to be observed in j being in j according to
3013: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3014: */
3015: double s1, lnpijopii;
3016: /*double t34;*/
3017: int i,j, nc, ii, jj;
3018:
1.234 brouard 3019: for(i=1; i<= nlstate; i++){
3020: for(j=1; j<i;j++){
3021: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3022: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3023: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3024: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3025: }
3026: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3027: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3028: }
3029: for(j=i+1; j<=nlstate+ndeath;j++){
3030: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3031: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3032: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3033: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3034: }
3035: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3036: }
3037: }
3038:
3039: for(i=1; i<= nlstate; i++){
3040: s1=0;
3041: for(j=1; j<i; j++){
3042: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3043: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3044: }
3045: for(j=i+1; j<=nlstate+ndeath; j++){
3046: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3047: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3048: }
3049: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3050: ps[i][i]=1./(s1+1.);
3051: /* Computing other pijs */
3052: for(j=1; j<i; j++)
3053: ps[i][j]= exp(ps[i][j])*ps[i][i];
3054: for(j=i+1; j<=nlstate+ndeath; j++)
3055: ps[i][j]= exp(ps[i][j])*ps[i][i];
3056: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3057: } /* end i */
3058:
3059: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3060: for(jj=1; jj<= nlstate+ndeath; jj++){
3061: ps[ii][jj]=0;
3062: ps[ii][ii]=1;
3063: }
3064: }
3065: /* Added for backcast */ /* Transposed matrix too */
3066: for(jj=1; jj<= nlstate+ndeath; jj++){
3067: s1=0.;
3068: for(ii=1; ii<= nlstate+ndeath; ii++){
3069: s1+=ps[ii][jj];
3070: }
3071: for(ii=1; ii<= nlstate; ii++){
3072: ps[ii][jj]=ps[ii][jj]/s1;
3073: }
3074: }
3075: /* Transposition */
3076: for(jj=1; jj<= nlstate+ndeath; jj++){
3077: for(ii=jj; ii<= nlstate+ndeath; ii++){
3078: s1=ps[ii][jj];
3079: ps[ii][jj]=ps[jj][ii];
3080: ps[jj][ii]=s1;
3081: }
3082: }
3083: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3084: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3085: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3086: /* } */
3087: /* printf("\n "); */
3088: /* } */
3089: /* printf("\n ");printf("%lf ",cov[2]);*/
3090: /*
3091: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3092: goto end;*/
3093: return ps;
1.217 brouard 3094: }
3095:
3096:
1.126 brouard 3097: /**************** Product of 2 matrices ******************/
3098:
1.145 brouard 3099: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3100: {
3101: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3102: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3103: /* in, b, out are matrice of pointers which should have been initialized
3104: before: only the contents of out is modified. The function returns
3105: a pointer to pointers identical to out */
1.145 brouard 3106: int i, j, k;
1.126 brouard 3107: for(i=nrl; i<= nrh; i++)
1.145 brouard 3108: for(k=ncolol; k<=ncoloh; k++){
3109: out[i][k]=0.;
3110: for(j=ncl; j<=nch; j++)
3111: out[i][k] +=in[i][j]*b[j][k];
3112: }
1.126 brouard 3113: return out;
3114: }
3115:
3116:
3117: /************* Higher Matrix Product ***************/
3118:
1.235 brouard 3119: 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 3120: {
1.218 brouard 3121: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3122: 'nhstepm*hstepm*stepm' months (i.e. until
3123: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3124: nhstepm*hstepm matrices.
3125: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3126: (typically every 2 years instead of every month which is too big
3127: for the memory).
3128: Model is determined by parameters x and covariates have to be
3129: included manually here.
3130:
3131: */
3132:
3133: int i, j, d, h, k;
1.131 brouard 3134: double **out, cov[NCOVMAX+1];
1.126 brouard 3135: double **newm;
1.187 brouard 3136: double agexact;
1.214 brouard 3137: double agebegin, ageend;
1.126 brouard 3138:
3139: /* Hstepm could be zero and should return the unit matrix */
3140: for (i=1;i<=nlstate+ndeath;i++)
3141: for (j=1;j<=nlstate+ndeath;j++){
3142: oldm[i][j]=(i==j ? 1.0 : 0.0);
3143: po[i][j][0]=(i==j ? 1.0 : 0.0);
3144: }
3145: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3146: for(h=1; h <=nhstepm; h++){
3147: for(d=1; d <=hstepm; d++){
3148: newm=savm;
3149: /* Covariates have to be included here again */
3150: cov[1]=1.;
1.214 brouard 3151: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3152: cov[2]=agexact;
3153: if(nagesqr==1)
1.227 brouard 3154: cov[3]= agexact*agexact;
1.235 brouard 3155: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3156: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3157: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3158: /* 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)); */
3159: }
3160: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3161: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3162: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3163: /* 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]); */
3164: }
3165: for (k=1; k<=cptcovage;k++){
3166: if(Dummy[Tvar[Tage[k]]]){
3167: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3168: } else{
3169: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3170: }
3171: /* 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]); */
3172: }
3173: for (k=1; k<=cptcovprod;k++){ /* */
3174: /* 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]); */
3175: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3176: }
3177: /* for (k=1; k<=cptcovn;k++) */
3178: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3179: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3180: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3181: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3182: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3183:
3184:
1.126 brouard 3185: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3186: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3187: /* right multiplication of oldm by the current matrix */
1.126 brouard 3188: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3189: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3190: /* if((int)age == 70){ */
3191: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3192: /* for(i=1; i<=nlstate+ndeath; i++) { */
3193: /* printf("%d pmmij ",i); */
3194: /* for(j=1;j<=nlstate+ndeath;j++) { */
3195: /* printf("%f ",pmmij[i][j]); */
3196: /* } */
3197: /* printf(" oldm "); */
3198: /* for(j=1;j<=nlstate+ndeath;j++) { */
3199: /* printf("%f ",oldm[i][j]); */
3200: /* } */
3201: /* printf("\n"); */
3202: /* } */
3203: /* } */
1.126 brouard 3204: savm=oldm;
3205: oldm=newm;
3206: }
3207: for(i=1; i<=nlstate+ndeath; i++)
3208: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3209: po[i][j][h]=newm[i][j];
3210: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3211: }
1.128 brouard 3212: /*printf("h=%d ",h);*/
1.126 brouard 3213: } /* end h */
1.267 brouard 3214: /* printf("\n H=%d \n",h); */
1.126 brouard 3215: return po;
3216: }
3217:
1.217 brouard 3218: /************* Higher Back Matrix Product ***************/
1.218 brouard 3219: /* 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 3220: 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 3221: {
1.266 brouard 3222: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3223: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3224: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3225: nhstepm*hstepm matrices.
3226: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3227: (typically every 2 years instead of every month which is too big
1.217 brouard 3228: for the memory).
1.218 brouard 3229: Model is determined by parameters x and covariates have to be
1.266 brouard 3230: included manually here. Then we use a call to bmij(x and cov)
3231: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3232: */
1.217 brouard 3233:
3234: int i, j, d, h, k;
1.266 brouard 3235: double **out, cov[NCOVMAX+1], **bmij();
3236: double **newm, ***newmm;
1.217 brouard 3237: double agexact;
3238: double agebegin, ageend;
1.222 brouard 3239: double **oldm, **savm;
1.217 brouard 3240:
1.266 brouard 3241: newmm=po; /* To be saved */
3242: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3243: /* Hstepm could be zero and should return the unit matrix */
3244: for (i=1;i<=nlstate+ndeath;i++)
3245: for (j=1;j<=nlstate+ndeath;j++){
3246: oldm[i][j]=(i==j ? 1.0 : 0.0);
3247: po[i][j][0]=(i==j ? 1.0 : 0.0);
3248: }
3249: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3250: for(h=1; h <=nhstepm; h++){
3251: for(d=1; d <=hstepm; d++){
3252: newm=savm;
3253: /* Covariates have to be included here again */
3254: cov[1]=1.;
1.271 brouard 3255: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3256: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3257: cov[2]=agexact;
3258: if(nagesqr==1)
1.222 brouard 3259: cov[3]= agexact*agexact;
1.266 brouard 3260: for (k=1; k<=cptcovn;k++){
3261: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3262: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3263: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3264: /* 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)); */
3265: }
1.267 brouard 3266: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3267: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3268: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3269: /* 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]); */
3270: }
3271: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3272: if(Dummy[Tvar[Tage[k]]]){
3273: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3274: } else{
3275: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3276: }
3277: /* 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]); */
3278: }
3279: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3280: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3281: }
1.217 brouard 3282: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3283: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3284:
1.218 brouard 3285: /* Careful transposed matrix */
1.266 brouard 3286: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3287: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3288: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3289: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3290: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3291: /* if((int)age == 70){ */
3292: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3293: /* for(i=1; i<=nlstate+ndeath; i++) { */
3294: /* printf("%d pmmij ",i); */
3295: /* for(j=1;j<=nlstate+ndeath;j++) { */
3296: /* printf("%f ",pmmij[i][j]); */
3297: /* } */
3298: /* printf(" oldm "); */
3299: /* for(j=1;j<=nlstate+ndeath;j++) { */
3300: /* printf("%f ",oldm[i][j]); */
3301: /* } */
3302: /* printf("\n"); */
3303: /* } */
3304: /* } */
3305: savm=oldm;
3306: oldm=newm;
3307: }
3308: for(i=1; i<=nlstate+ndeath; i++)
3309: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3310: po[i][j][h]=newm[i][j];
1.268 brouard 3311: /* if(h==nhstepm) */
3312: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3313: }
1.268 brouard 3314: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3315: } /* end h */
1.268 brouard 3316: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3317: return po;
3318: }
3319:
3320:
1.162 brouard 3321: #ifdef NLOPT
3322: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3323: double fret;
3324: double *xt;
3325: int j;
3326: myfunc_data *d2 = (myfunc_data *) pd;
3327: /* xt = (p1-1); */
3328: xt=vector(1,n);
3329: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3330:
3331: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3332: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3333: printf("Function = %.12lf ",fret);
3334: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3335: printf("\n");
3336: free_vector(xt,1,n);
3337: return fret;
3338: }
3339: #endif
1.126 brouard 3340:
3341: /*************** log-likelihood *************/
3342: double func( double *x)
3343: {
1.226 brouard 3344: int i, ii, j, k, mi, d, kk;
3345: int ioffset=0;
3346: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3347: double **out;
3348: double lli; /* Individual log likelihood */
3349: int s1, s2;
1.228 brouard 3350: 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 3351: double bbh, survp;
3352: long ipmx;
3353: double agexact;
3354: /*extern weight */
3355: /* We are differentiating ll according to initial status */
3356: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3357: /*for(i=1;i<imx;i++)
3358: printf(" %d\n",s[4][i]);
3359: */
1.162 brouard 3360:
1.226 brouard 3361: ++countcallfunc;
1.162 brouard 3362:
1.226 brouard 3363: cov[1]=1.;
1.126 brouard 3364:
1.226 brouard 3365: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3366: ioffset=0;
1.226 brouard 3367: if(mle==1){
3368: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3369: /* Computes the values of the ncovmodel covariates of the model
3370: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3371: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3372: to be observed in j being in i according to the model.
3373: */
1.243 brouard 3374: ioffset=2+nagesqr ;
1.233 brouard 3375: /* Fixed */
1.234 brouard 3376: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3377: 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)*/
3378: }
1.226 brouard 3379: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3380: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3381: has been calculated etc */
3382: /* For an individual i, wav[i] gives the number of effective waves */
3383: /* We compute the contribution to Likelihood of each effective transition
3384: mw[mi][i] is real wave of the mi th effectve wave */
3385: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3386: s2=s[mw[mi+1][i]][i];
3387: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3388: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3389: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3390: */
3391: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3392: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3393: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3394: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3395: }
3396: for (ii=1;ii<=nlstate+ndeath;ii++)
3397: for (j=1;j<=nlstate+ndeath;j++){
3398: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3399: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3400: }
3401: for(d=0; d<dh[mi][i]; d++){
3402: newm=savm;
3403: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3404: cov[2]=agexact;
3405: if(nagesqr==1)
3406: cov[3]= agexact*agexact; /* Should be changed here */
3407: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3408: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3409: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3410: else
3411: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3412: }
3413: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3414: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3415: savm=oldm;
3416: oldm=newm;
3417: } /* end mult */
3418:
3419: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3420: /* But now since version 0.9 we anticipate for bias at large stepm.
3421: * If stepm is larger than one month (smallest stepm) and if the exact delay
3422: * (in months) between two waves is not a multiple of stepm, we rounded to
3423: * the nearest (and in case of equal distance, to the lowest) interval but now
3424: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3425: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3426: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3427: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3428: * -stepm/2 to stepm/2 .
3429: * For stepm=1 the results are the same as for previous versions of Imach.
3430: * For stepm > 1 the results are less biased than in previous versions.
3431: */
1.234 brouard 3432: s1=s[mw[mi][i]][i];
3433: s2=s[mw[mi+1][i]][i];
3434: bbh=(double)bh[mi][i]/(double)stepm;
3435: /* bias bh is positive if real duration
3436: * is higher than the multiple of stepm and negative otherwise.
3437: */
3438: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3439: if( s2 > nlstate){
3440: /* i.e. if s2 is a death state and if the date of death is known
3441: then the contribution to the likelihood is the probability to
3442: die between last step unit time and current step unit time,
3443: which is also equal to probability to die before dh
3444: minus probability to die before dh-stepm .
3445: In version up to 0.92 likelihood was computed
3446: as if date of death was unknown. Death was treated as any other
3447: health state: the date of the interview describes the actual state
3448: and not the date of a change in health state. The former idea was
3449: to consider that at each interview the state was recorded
3450: (healthy, disable or death) and IMaCh was corrected; but when we
3451: introduced the exact date of death then we should have modified
3452: the contribution of an exact death to the likelihood. This new
3453: contribution is smaller and very dependent of the step unit
3454: stepm. It is no more the probability to die between last interview
3455: and month of death but the probability to survive from last
3456: interview up to one month before death multiplied by the
3457: probability to die within a month. Thanks to Chris
3458: Jackson for correcting this bug. Former versions increased
3459: mortality artificially. The bad side is that we add another loop
3460: which slows down the processing. The difference can be up to 10%
3461: lower mortality.
3462: */
3463: /* If, at the beginning of the maximization mostly, the
3464: cumulative probability or probability to be dead is
3465: constant (ie = 1) over time d, the difference is equal to
3466: 0. out[s1][3] = savm[s1][3]: probability, being at state
3467: s1 at precedent wave, to be dead a month before current
3468: wave is equal to probability, being at state s1 at
3469: precedent wave, to be dead at mont of the current
3470: wave. Then the observed probability (that this person died)
3471: is null according to current estimated parameter. In fact,
3472: it should be very low but not zero otherwise the log go to
3473: infinity.
3474: */
1.183 brouard 3475: /* #ifdef INFINITYORIGINAL */
3476: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3477: /* #else */
3478: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3479: /* lli=log(mytinydouble); */
3480: /* else */
3481: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3482: /* #endif */
1.226 brouard 3483: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3484:
1.226 brouard 3485: } else if ( s2==-1 ) { /* alive */
3486: for (j=1,survp=0. ; j<=nlstate; j++)
3487: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3488: /*survp += out[s1][j]; */
3489: lli= log(survp);
3490: }
3491: else if (s2==-4) {
3492: for (j=3,survp=0. ; j<=nlstate; j++)
3493: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3494: lli= log(survp);
3495: }
3496: else if (s2==-5) {
3497: for (j=1,survp=0. ; j<=2; j++)
3498: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3499: lli= log(survp);
3500: }
3501: else{
3502: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3503: /* 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 */
3504: }
3505: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3506: /*if(lli ==000.0)*/
3507: /*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); */
3508: ipmx +=1;
3509: sw += weight[i];
3510: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3511: /* if (lli < log(mytinydouble)){ */
3512: /* 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); */
3513: /* 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]); */
3514: /* } */
3515: } /* end of wave */
3516: } /* end of individual */
3517: } else if(mle==2){
3518: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3519: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3520: for(mi=1; mi<= wav[i]-1; mi++){
3521: for (ii=1;ii<=nlstate+ndeath;ii++)
3522: for (j=1;j<=nlstate+ndeath;j++){
3523: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3524: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3525: }
3526: for(d=0; d<=dh[mi][i]; d++){
3527: newm=savm;
3528: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3529: cov[2]=agexact;
3530: if(nagesqr==1)
3531: cov[3]= agexact*agexact;
3532: for (kk=1; kk<=cptcovage;kk++) {
3533: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3534: }
3535: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3536: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3537: savm=oldm;
3538: oldm=newm;
3539: } /* end mult */
3540:
3541: s1=s[mw[mi][i]][i];
3542: s2=s[mw[mi+1][i]][i];
3543: bbh=(double)bh[mi][i]/(double)stepm;
3544: 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 */
3545: ipmx +=1;
3546: sw += weight[i];
3547: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3548: } /* end of wave */
3549: } /* end of individual */
3550: } else if(mle==3){ /* exponential inter-extrapolation */
3551: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3552: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3553: for(mi=1; mi<= wav[i]-1; mi++){
3554: for (ii=1;ii<=nlstate+ndeath;ii++)
3555: for (j=1;j<=nlstate+ndeath;j++){
3556: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3557: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3558: }
3559: for(d=0; d<dh[mi][i]; d++){
3560: newm=savm;
3561: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3562: cov[2]=agexact;
3563: if(nagesqr==1)
3564: cov[3]= agexact*agexact;
3565: for (kk=1; kk<=cptcovage;kk++) {
3566: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3567: }
3568: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3569: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3570: savm=oldm;
3571: oldm=newm;
3572: } /* end mult */
3573:
3574: s1=s[mw[mi][i]][i];
3575: s2=s[mw[mi+1][i]][i];
3576: bbh=(double)bh[mi][i]/(double)stepm;
3577: 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 */
3578: ipmx +=1;
3579: sw += weight[i];
3580: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3581: } /* end of wave */
3582: } /* end of individual */
3583: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3584: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3585: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3586: for(mi=1; mi<= wav[i]-1; mi++){
3587: for (ii=1;ii<=nlstate+ndeath;ii++)
3588: for (j=1;j<=nlstate+ndeath;j++){
3589: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3590: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3591: }
3592: for(d=0; d<dh[mi][i]; d++){
3593: newm=savm;
3594: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3595: cov[2]=agexact;
3596: if(nagesqr==1)
3597: cov[3]= agexact*agexact;
3598: for (kk=1; kk<=cptcovage;kk++) {
3599: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3600: }
1.126 brouard 3601:
1.226 brouard 3602: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3603: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3604: savm=oldm;
3605: oldm=newm;
3606: } /* end mult */
3607:
3608: s1=s[mw[mi][i]][i];
3609: s2=s[mw[mi+1][i]][i];
3610: if( s2 > nlstate){
3611: lli=log(out[s1][s2] - savm[s1][s2]);
3612: } else if ( s2==-1 ) { /* alive */
3613: for (j=1,survp=0. ; j<=nlstate; j++)
3614: survp += out[s1][j];
3615: lli= log(survp);
3616: }else{
3617: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3618: }
3619: ipmx +=1;
3620: sw += weight[i];
3621: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3622: /* 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 3623: } /* end of wave */
3624: } /* end of individual */
3625: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3626: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3627: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3628: for(mi=1; mi<= wav[i]-1; mi++){
3629: for (ii=1;ii<=nlstate+ndeath;ii++)
3630: for (j=1;j<=nlstate+ndeath;j++){
3631: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3632: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3633: }
3634: for(d=0; d<dh[mi][i]; d++){
3635: newm=savm;
3636: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3637: cov[2]=agexact;
3638: if(nagesqr==1)
3639: cov[3]= agexact*agexact;
3640: for (kk=1; kk<=cptcovage;kk++) {
3641: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3642: }
1.126 brouard 3643:
1.226 brouard 3644: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3645: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3646: savm=oldm;
3647: oldm=newm;
3648: } /* end mult */
3649:
3650: s1=s[mw[mi][i]][i];
3651: s2=s[mw[mi+1][i]][i];
3652: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3653: ipmx +=1;
3654: sw += weight[i];
3655: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3656: /*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]);*/
3657: } /* end of wave */
3658: } /* end of individual */
3659: } /* End of if */
3660: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3661: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3662: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3663: return -l;
1.126 brouard 3664: }
3665:
3666: /*************** log-likelihood *************/
3667: double funcone( double *x)
3668: {
1.228 brouard 3669: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3670: int i, ii, j, k, mi, d, kk;
1.228 brouard 3671: int ioffset=0;
1.131 brouard 3672: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3673: double **out;
3674: double lli; /* Individual log likelihood */
3675: double llt;
3676: int s1, s2;
1.228 brouard 3677: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3678:
1.126 brouard 3679: double bbh, survp;
1.187 brouard 3680: double agexact;
1.214 brouard 3681: double agebegin, ageend;
1.126 brouard 3682: /*extern weight */
3683: /* We are differentiating ll according to initial status */
3684: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3685: /*for(i=1;i<imx;i++)
3686: printf(" %d\n",s[4][i]);
3687: */
3688: cov[1]=1.;
3689:
3690: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3691: ioffset=0;
3692: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3693: /* ioffset=2+nagesqr+cptcovage; */
3694: ioffset=2+nagesqr;
1.232 brouard 3695: /* Fixed */
1.224 brouard 3696: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3697: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3698: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3699: 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)*/
3700: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3701: /* cov[2+6]=covar[Tvar[6]][i]; */
3702: /* cov[2+6]=covar[2][i]; V2 */
3703: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3704: /* cov[2+7]=covar[Tvar[7]][i]; */
3705: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3706: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3707: /* cov[2+9]=covar[Tvar[9]][i]; */
3708: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3709: }
1.232 brouard 3710: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3711: /* 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?)*\/ */
3712: /* } */
1.231 brouard 3713: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3714: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3715: /* } */
1.225 brouard 3716:
1.233 brouard 3717:
3718: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3719: /* Wave varying (but not age varying) */
3720: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3721: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3722: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3723: }
1.232 brouard 3724: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3725: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3726: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3727: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3728: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3729: /* 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 3730: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3731: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3732: /* /\* 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]); *\/ */
3733: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3734: /* } */
1.126 brouard 3735: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3736: for (j=1;j<=nlstate+ndeath;j++){
3737: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3738: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3739: }
1.214 brouard 3740:
3741: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3742: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3743: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3744: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3745: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3746: and mw[mi+1][i]. dh depends on stepm.*/
3747: newm=savm;
1.247 brouard 3748: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3749: cov[2]=agexact;
3750: if(nagesqr==1)
3751: cov[3]= agexact*agexact;
3752: for (kk=1; kk<=cptcovage;kk++) {
3753: if(!FixedV[Tvar[Tage[kk]]])
3754: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3755: else
3756: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3757: }
3758: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3759: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3760: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3761: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3762: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3763: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3764: savm=oldm;
3765: oldm=newm;
1.126 brouard 3766: } /* end mult */
3767:
3768: s1=s[mw[mi][i]][i];
3769: s2=s[mw[mi+1][i]][i];
1.217 brouard 3770: /* if(s2==-1){ */
1.268 brouard 3771: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3772: /* /\* exit(1); *\/ */
3773: /* } */
1.126 brouard 3774: bbh=(double)bh[mi][i]/(double)stepm;
3775: /* bias is positive if real duration
3776: * is higher than the multiple of stepm and negative otherwise.
3777: */
3778: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3779: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3780: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3781: for (j=1,survp=0. ; j<=nlstate; j++)
3782: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3783: lli= log(survp);
1.126 brouard 3784: }else if (mle==1){
1.242 brouard 3785: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3786: } else if(mle==2){
1.242 brouard 3787: 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 3788: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3789: 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 3790: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3791: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3792: } else{ /* mle=0 back to 1 */
1.242 brouard 3793: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3794: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3795: } /* End of if */
3796: ipmx +=1;
3797: sw += weight[i];
3798: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3799: /*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 3800: if(globpr){
1.246 brouard 3801: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3802: %11.6f %11.6f %11.6f ", \
1.242 brouard 3803: 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 3804: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3805: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3806: llt +=ll[k]*gipmx/gsw;
3807: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3808: }
3809: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3810: }
1.232 brouard 3811: } /* end of wave */
3812: } /* end of individual */
3813: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3814: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3815: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3816: if(globpr==0){ /* First time we count the contributions and weights */
3817: gipmx=ipmx;
3818: gsw=sw;
3819: }
3820: return -l;
1.126 brouard 3821: }
3822:
3823:
3824: /*************** function likelione ***********/
3825: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3826: {
3827: /* This routine should help understanding what is done with
3828: the selection of individuals/waves and
3829: to check the exact contribution to the likelihood.
3830: Plotting could be done.
3831: */
3832: int k;
3833:
3834: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3835: strcpy(fileresilk,"ILK_");
1.202 brouard 3836: strcat(fileresilk,fileresu);
1.126 brouard 3837: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3838: printf("Problem with resultfile: %s\n", fileresilk);
3839: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3840: }
1.214 brouard 3841: 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");
3842: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3843: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3844: for(k=1; k<=nlstate; k++)
3845: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3846: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3847: }
3848:
3849: *fretone=(*funcone)(p);
3850: if(*globpri !=0){
3851: fclose(ficresilk);
1.205 brouard 3852: if (mle ==0)
3853: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3854: else if(mle >=1)
3855: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3856: 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 3857: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3858:
3859: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3860: 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 3861: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3862: }
1.207 brouard 3863: 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 3864: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3865: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3866: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3867: fflush(fichtm);
1.205 brouard 3868: }
1.126 brouard 3869: return;
3870: }
3871:
3872:
3873: /*********** Maximum Likelihood Estimation ***************/
3874:
3875: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3876: {
1.165 brouard 3877: int i,j, iter=0;
1.126 brouard 3878: double **xi;
3879: double fret;
3880: double fretone; /* Only one call to likelihood */
3881: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3882:
3883: #ifdef NLOPT
3884: int creturn;
3885: nlopt_opt opt;
3886: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3887: double *lb;
3888: double minf; /* the minimum objective value, upon return */
3889: double * p1; /* Shifted parameters from 0 instead of 1 */
3890: myfunc_data dinst, *d = &dinst;
3891: #endif
3892:
3893:
1.126 brouard 3894: xi=matrix(1,npar,1,npar);
3895: for (i=1;i<=npar;i++)
3896: for (j=1;j<=npar;j++)
3897: xi[i][j]=(i==j ? 1.0 : 0.0);
3898: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3899: strcpy(filerespow,"POW_");
1.126 brouard 3900: strcat(filerespow,fileres);
3901: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3902: printf("Problem with resultfile: %s\n", filerespow);
3903: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3904: }
3905: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3906: for (i=1;i<=nlstate;i++)
3907: for(j=1;j<=nlstate+ndeath;j++)
3908: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3909: fprintf(ficrespow,"\n");
1.162 brouard 3910: #ifdef POWELL
1.126 brouard 3911: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3912: #endif
1.126 brouard 3913:
1.162 brouard 3914: #ifdef NLOPT
3915: #ifdef NEWUOA
3916: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3917: #else
3918: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3919: #endif
3920: lb=vector(0,npar-1);
3921: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3922: nlopt_set_lower_bounds(opt, lb);
3923: nlopt_set_initial_step1(opt, 0.1);
3924:
3925: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3926: d->function = func;
3927: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3928: nlopt_set_min_objective(opt, myfunc, d);
3929: nlopt_set_xtol_rel(opt, ftol);
3930: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3931: printf("nlopt failed! %d\n",creturn);
3932: }
3933: else {
3934: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3935: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3936: iter=1; /* not equal */
3937: }
3938: nlopt_destroy(opt);
3939: #endif
1.126 brouard 3940: free_matrix(xi,1,npar,1,npar);
3941: fclose(ficrespow);
1.203 brouard 3942: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3943: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3944: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3945:
3946: }
3947:
3948: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3949: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3950: {
3951: double **a,**y,*x,pd;
1.203 brouard 3952: /* double **hess; */
1.164 brouard 3953: int i, j;
1.126 brouard 3954: int *indx;
3955:
3956: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3957: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3958: void lubksb(double **a, int npar, int *indx, double b[]) ;
3959: void ludcmp(double **a, int npar, int *indx, double *d) ;
3960: double gompertz(double p[]);
1.203 brouard 3961: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3962:
3963: printf("\nCalculation of the hessian matrix. Wait...\n");
3964: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3965: for (i=1;i<=npar;i++){
1.203 brouard 3966: printf("%d-",i);fflush(stdout);
3967: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3968:
3969: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3970:
3971: /* printf(" %f ",p[i]);
3972: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3973: }
3974:
3975: for (i=1;i<=npar;i++) {
3976: for (j=1;j<=npar;j++) {
3977: if (j>i) {
1.203 brouard 3978: printf(".%d-%d",i,j);fflush(stdout);
3979: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3980: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3981:
3982: hess[j][i]=hess[i][j];
3983: /*printf(" %lf ",hess[i][j]);*/
3984: }
3985: }
3986: }
3987: printf("\n");
3988: fprintf(ficlog,"\n");
3989:
3990: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3991: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3992:
3993: a=matrix(1,npar,1,npar);
3994: y=matrix(1,npar,1,npar);
3995: x=vector(1,npar);
3996: indx=ivector(1,npar);
3997: for (i=1;i<=npar;i++)
3998: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3999: ludcmp(a,npar,indx,&pd);
4000:
4001: for (j=1;j<=npar;j++) {
4002: for (i=1;i<=npar;i++) x[i]=0;
4003: x[j]=1;
4004: lubksb(a,npar,indx,x);
4005: for (i=1;i<=npar;i++){
4006: matcov[i][j]=x[i];
4007: }
4008: }
4009:
4010: printf("\n#Hessian matrix#\n");
4011: fprintf(ficlog,"\n#Hessian matrix#\n");
4012: for (i=1;i<=npar;i++) {
4013: for (j=1;j<=npar;j++) {
1.203 brouard 4014: printf("%.6e ",hess[i][j]);
4015: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4016: }
4017: printf("\n");
4018: fprintf(ficlog,"\n");
4019: }
4020:
1.203 brouard 4021: /* printf("\n#Covariance matrix#\n"); */
4022: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4023: /* for (i=1;i<=npar;i++) { */
4024: /* for (j=1;j<=npar;j++) { */
4025: /* printf("%.6e ",matcov[i][j]); */
4026: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4027: /* } */
4028: /* printf("\n"); */
4029: /* fprintf(ficlog,"\n"); */
4030: /* } */
4031:
1.126 brouard 4032: /* Recompute Inverse */
1.203 brouard 4033: /* for (i=1;i<=npar;i++) */
4034: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4035: /* ludcmp(a,npar,indx,&pd); */
4036:
4037: /* printf("\n#Hessian matrix recomputed#\n"); */
4038:
4039: /* for (j=1;j<=npar;j++) { */
4040: /* for (i=1;i<=npar;i++) x[i]=0; */
4041: /* x[j]=1; */
4042: /* lubksb(a,npar,indx,x); */
4043: /* for (i=1;i<=npar;i++){ */
4044: /* y[i][j]=x[i]; */
4045: /* printf("%.3e ",y[i][j]); */
4046: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4047: /* } */
4048: /* printf("\n"); */
4049: /* fprintf(ficlog,"\n"); */
4050: /* } */
4051:
4052: /* Verifying the inverse matrix */
4053: #ifdef DEBUGHESS
4054: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4055:
1.203 brouard 4056: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4057: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4058:
4059: for (j=1;j<=npar;j++) {
4060: for (i=1;i<=npar;i++){
1.203 brouard 4061: printf("%.2f ",y[i][j]);
4062: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4063: }
4064: printf("\n");
4065: fprintf(ficlog,"\n");
4066: }
1.203 brouard 4067: #endif
1.126 brouard 4068:
4069: free_matrix(a,1,npar,1,npar);
4070: free_matrix(y,1,npar,1,npar);
4071: free_vector(x,1,npar);
4072: free_ivector(indx,1,npar);
1.203 brouard 4073: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4074:
4075:
4076: }
4077:
4078: /*************** hessian matrix ****************/
4079: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4080: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4081: int i;
4082: int l=1, lmax=20;
1.203 brouard 4083: double k1,k2, res, fx;
1.132 brouard 4084: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4085: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4086: int k=0,kmax=10;
4087: double l1;
4088:
4089: fx=func(x);
4090: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4091: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4092: l1=pow(10,l);
4093: delts=delt;
4094: for(k=1 ; k <kmax; k=k+1){
4095: delt = delta*(l1*k);
4096: p2[theta]=x[theta] +delt;
1.145 brouard 4097: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4098: p2[theta]=x[theta]-delt;
4099: k2=func(p2)-fx;
4100: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4101: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4102:
1.203 brouard 4103: #ifdef DEBUGHESSII
1.126 brouard 4104: 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);
4105: 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);
4106: #endif
4107: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4108: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4109: k=kmax;
4110: }
4111: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4112: k=kmax; l=lmax*10;
1.126 brouard 4113: }
4114: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4115: delts=delt;
4116: }
1.203 brouard 4117: } /* End loop k */
1.126 brouard 4118: }
4119: delti[theta]=delts;
4120: return res;
4121:
4122: }
4123:
1.203 brouard 4124: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4125: {
4126: int i;
1.164 brouard 4127: int l=1, lmax=20;
1.126 brouard 4128: double k1,k2,k3,k4,res,fx;
1.132 brouard 4129: double p2[MAXPARM+1];
1.203 brouard 4130: int k, kmax=1;
4131: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4132:
4133: int firstime=0;
1.203 brouard 4134:
1.126 brouard 4135: fx=func(x);
1.203 brouard 4136: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4137: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4138: p2[thetai]=x[thetai]+delti[thetai]*k;
4139: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4140: k1=func(p2)-fx;
4141:
1.203 brouard 4142: p2[thetai]=x[thetai]+delti[thetai]*k;
4143: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4144: k2=func(p2)-fx;
4145:
1.203 brouard 4146: p2[thetai]=x[thetai]-delti[thetai]*k;
4147: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4148: k3=func(p2)-fx;
4149:
1.203 brouard 4150: p2[thetai]=x[thetai]-delti[thetai]*k;
4151: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4152: k4=func(p2)-fx;
1.203 brouard 4153: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4154: if(k1*k2*k3*k4 <0.){
1.208 brouard 4155: firstime=1;
1.203 brouard 4156: kmax=kmax+10;
1.208 brouard 4157: }
4158: if(kmax >=10 || firstime ==1){
1.246 brouard 4159: 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);
4160: 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 4161: 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);
4162: 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);
4163: }
4164: #ifdef DEBUGHESSIJ
4165: v1=hess[thetai][thetai];
4166: v2=hess[thetaj][thetaj];
4167: cv12=res;
4168: /* Computing eigen value of Hessian matrix */
4169: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4170: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4171: if ((lc2 <0) || (lc1 <0) ){
4172: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4173: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4174: 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);
4175: 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);
4176: }
1.126 brouard 4177: #endif
4178: }
4179: return res;
4180: }
4181:
1.203 brouard 4182: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4183: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4184: /* { */
4185: /* int i; */
4186: /* int l=1, lmax=20; */
4187: /* double k1,k2,k3,k4,res,fx; */
4188: /* double p2[MAXPARM+1]; */
4189: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4190: /* int k=0,kmax=10; */
4191: /* double l1; */
4192:
4193: /* fx=func(x); */
4194: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4195: /* l1=pow(10,l); */
4196: /* delts=delt; */
4197: /* for(k=1 ; k <kmax; k=k+1){ */
4198: /* delt = delti*(l1*k); */
4199: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4200: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4201: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4202: /* k1=func(p2)-fx; */
4203:
4204: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4205: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4206: /* k2=func(p2)-fx; */
4207:
4208: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4209: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4210: /* k3=func(p2)-fx; */
4211:
4212: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4213: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4214: /* k4=func(p2)-fx; */
4215: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4216: /* #ifdef DEBUGHESSIJ */
4217: /* 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); */
4218: /* 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); */
4219: /* #endif */
4220: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4221: /* k=kmax; */
4222: /* } */
4223: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4224: /* k=kmax; l=lmax*10; */
4225: /* } */
4226: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4227: /* delts=delt; */
4228: /* } */
4229: /* } /\* End loop k *\/ */
4230: /* } */
4231: /* delti[theta]=delts; */
4232: /* return res; */
4233: /* } */
4234:
4235:
1.126 brouard 4236: /************** Inverse of matrix **************/
4237: void ludcmp(double **a, int n, int *indx, double *d)
4238: {
4239: int i,imax,j,k;
4240: double big,dum,sum,temp;
4241: double *vv;
4242:
4243: vv=vector(1,n);
4244: *d=1.0;
4245: for (i=1;i<=n;i++) {
4246: big=0.0;
4247: for (j=1;j<=n;j++)
4248: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4249: if (big == 0.0){
4250: printf(" Singular Hessian matrix at row %d:\n",i);
4251: for (j=1;j<=n;j++) {
4252: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4253: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4254: }
4255: fflush(ficlog);
4256: fclose(ficlog);
4257: nrerror("Singular matrix in routine ludcmp");
4258: }
1.126 brouard 4259: vv[i]=1.0/big;
4260: }
4261: for (j=1;j<=n;j++) {
4262: for (i=1;i<j;i++) {
4263: sum=a[i][j];
4264: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4265: a[i][j]=sum;
4266: }
4267: big=0.0;
4268: for (i=j;i<=n;i++) {
4269: sum=a[i][j];
4270: for (k=1;k<j;k++)
4271: sum -= a[i][k]*a[k][j];
4272: a[i][j]=sum;
4273: if ( (dum=vv[i]*fabs(sum)) >= big) {
4274: big=dum;
4275: imax=i;
4276: }
4277: }
4278: if (j != imax) {
4279: for (k=1;k<=n;k++) {
4280: dum=a[imax][k];
4281: a[imax][k]=a[j][k];
4282: a[j][k]=dum;
4283: }
4284: *d = -(*d);
4285: vv[imax]=vv[j];
4286: }
4287: indx[j]=imax;
4288: if (a[j][j] == 0.0) a[j][j]=TINY;
4289: if (j != n) {
4290: dum=1.0/(a[j][j]);
4291: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4292: }
4293: }
4294: free_vector(vv,1,n); /* Doesn't work */
4295: ;
4296: }
4297:
4298: void lubksb(double **a, int n, int *indx, double b[])
4299: {
4300: int i,ii=0,ip,j;
4301: double sum;
4302:
4303: for (i=1;i<=n;i++) {
4304: ip=indx[i];
4305: sum=b[ip];
4306: b[ip]=b[i];
4307: if (ii)
4308: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4309: else if (sum) ii=i;
4310: b[i]=sum;
4311: }
4312: for (i=n;i>=1;i--) {
4313: sum=b[i];
4314: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4315: b[i]=sum/a[i][i];
4316: }
4317: }
4318:
4319: void pstamp(FILE *fichier)
4320: {
1.196 brouard 4321: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4322: }
4323:
1.253 brouard 4324:
4325:
1.126 brouard 4326: /************ Frequencies ********************/
1.251 brouard 4327: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4328: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4329: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4330: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4331:
1.265 brouard 4332: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4333: int iind=0, iage=0;
4334: int mi; /* Effective wave */
4335: int first;
4336: double ***freq; /* Frequencies */
1.268 brouard 4337: 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 */
4338: 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 4339: double *meanq;
4340: double **meanqt;
4341: double *pp, **prop, *posprop, *pospropt;
4342: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4343: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4344: double agebegin, ageend;
4345:
4346: pp=vector(1,nlstate);
1.251 brouard 4347: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4348: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4349: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4350: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4351: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4352: meanqt=matrix(1,lastpass,1,nqtveff);
4353: strcpy(fileresp,"P_");
4354: strcat(fileresp,fileresu);
4355: /*strcat(fileresphtm,fileresu);*/
4356: if((ficresp=fopen(fileresp,"w"))==NULL) {
4357: printf("Problem with prevalence resultfile: %s\n", fileresp);
4358: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4359: exit(0);
4360: }
1.240 brouard 4361:
1.226 brouard 4362: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4363: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4364: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4365: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4366: fflush(ficlog);
4367: exit(70);
4368: }
4369: else{
4370: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4371: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4372: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4373: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4374: }
1.237 brouard 4375: 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 4376:
1.226 brouard 4377: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4378: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4379: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4380: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4381: fflush(ficlog);
4382: exit(70);
1.240 brouard 4383: } else{
1.226 brouard 4384: 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 4385: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4386: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4387: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4388: }
1.240 brouard 4389: 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);
4390:
1.253 brouard 4391: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4392: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4393: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4394: j1=0;
1.126 brouard 4395:
1.227 brouard 4396: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4397: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4398: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4399:
4400:
1.226 brouard 4401: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4402: reference=low_education V1=0,V2=0
4403: med_educ V1=1 V2=0,
4404: high_educ V1=0 V2=1
4405: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4406: */
1.249 brouard 4407: dateintsum=0;
4408: k2cpt=0;
4409:
1.253 brouard 4410: if(cptcoveff == 0 )
1.265 brouard 4411: nl=1; /* Constant and age model only */
1.253 brouard 4412: else
4413: nl=2;
1.265 brouard 4414:
4415: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4416: /* Loop on nj=1 or 2 if dummy covariates j!=0
4417: * Loop on j1(1 to 2**cptcoveff) covariate combination
4418: * freq[s1][s2][iage] =0.
4419: * Loop on iind
4420: * ++freq[s1][s2][iage] weighted
4421: * end iind
4422: * if covariate and j!0
4423: * headers Variable on one line
4424: * endif cov j!=0
4425: * header of frequency table by age
4426: * Loop on age
4427: * pp[s1]+=freq[s1][s2][iage] weighted
4428: * pos+=freq[s1][s2][iage] weighted
4429: * Loop on s1 initial state
4430: * fprintf(ficresp
4431: * end s1
4432: * end age
4433: * if j!=0 computes starting values
4434: * end compute starting values
4435: * end j1
4436: * end nl
4437: */
1.253 brouard 4438: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4439: if(nj==1)
4440: j=0; /* First pass for the constant */
1.265 brouard 4441: else{
1.253 brouard 4442: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4443: }
1.251 brouard 4444: first=1;
1.265 brouard 4445: 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 4446: posproptt=0.;
4447: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4448: scanf("%d", i);*/
4449: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4450: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4451: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4452: freq[i][s2][m]=0;
1.251 brouard 4453:
4454: for (i=1; i<=nlstate; i++) {
1.240 brouard 4455: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4456: prop[i][m]=0;
4457: posprop[i]=0;
4458: pospropt[i]=0;
4459: }
4460: /* for (z1=1; z1<= nqfveff; z1++) { */
4461: /* meanq[z1]+=0.; */
4462: /* for(m=1;m<=lastpass;m++){ */
4463: /* meanqt[m][z1]=0.; */
4464: /* } */
4465: /* } */
4466:
4467: /* dateintsum=0; */
4468: /* k2cpt=0; */
4469:
1.265 brouard 4470: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4471: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4472: bool=1;
4473: if(j !=0){
4474: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4475: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4476: /* for (z1=1; z1<= nqfveff; z1++) { */
4477: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4478: /* } */
4479: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4480: /* if(Tvaraff[z1] ==-20){ */
4481: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4482: /* }else if(Tvaraff[z1] ==-10){ */
4483: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4484: /* }else */
4485: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4486: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4487: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4488: /* 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",
4489: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4490: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4491: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4492: } /* Onlyf fixed */
4493: } /* end z1 */
4494: } /* cptcovn > 0 */
4495: } /* end any */
4496: }/* end j==0 */
1.265 brouard 4497: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4498: /* for(m=firstpass; m<=lastpass; m++){ */
4499: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4500: m=mw[mi][iind];
4501: if(j!=0){
4502: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4503: for (z1=1; z1<=cptcoveff; z1++) {
4504: if( Fixed[Tmodelind[z1]]==1){
4505: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4506: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4507: value is -1, we don't select. It differs from the
4508: constant and age model which counts them. */
4509: bool=0; /* not selected */
4510: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4511: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4512: bool=0;
4513: }
4514: }
4515: }
4516: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4517: } /* end j==0 */
4518: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4519: if(bool==1){
4520: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4521: and mw[mi+1][iind]. dh depends on stepm. */
4522: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4523: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4524: if(m >=firstpass && m <=lastpass){
4525: k2=anint[m][iind]+(mint[m][iind]/12.);
4526: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4527: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4528: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4529: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4530: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4531: if (m<lastpass) {
4532: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4533: /* 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]); */
4534: if(s[m][iind]==-1)
4535: 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.));
4536: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4537: /* if((int)agev[m][iind] == 55) */
4538: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4539: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4540: 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 4541: }
1.251 brouard 4542: } /* end if between passes */
4543: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4544: dateintsum=dateintsum+k2; /* on all covariates ?*/
4545: k2cpt++;
4546: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4547: }
1.251 brouard 4548: }else{
4549: bool=1;
4550: }/* end bool 2 */
4551: } /* end m */
4552: } /* end bool */
4553: } /* end iind = 1 to imx */
4554: /* prop[s][age] is feeded for any initial and valid live state as well as
4555: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4556:
4557:
4558: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4559: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4560: pstamp(ficresp);
1.251 brouard 4561: if (cptcoveff>0 && j!=0){
1.265 brouard 4562: pstamp(ficresp);
1.251 brouard 4563: printf( "\n#********** Variable ");
4564: fprintf(ficresp, "\n#********** Variable ");
4565: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4566: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4567: fprintf(ficlog, "\n#********** Variable ");
4568: for (z1=1; z1<=cptcoveff; z1++){
4569: if(!FixedV[Tvaraff[z1]]){
4570: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4571: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4572: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4573: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4574: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4575: }else{
1.251 brouard 4576: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4577: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4578: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4579: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4580: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4581: }
4582: }
4583: printf( "**********\n#");
4584: fprintf(ficresp, "**********\n#");
4585: fprintf(ficresphtm, "**********</h3>\n");
4586: fprintf(ficresphtmfr, "**********</h3>\n");
4587: fprintf(ficlog, "**********\n");
4588: }
4589: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4590: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4591: fprintf(ficresp, " Age");
4592: 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 4593: for(i=1; i<=nlstate;i++) {
1.265 brouard 4594: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4595: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4596: }
1.265 brouard 4597: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4598: fprintf(ficresphtm, "\n");
4599:
4600: /* Header of frequency table by age */
4601: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4602: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4603: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4604: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4605: if(s2!=0 && m!=0)
4606: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4607: }
1.226 brouard 4608: }
1.251 brouard 4609: fprintf(ficresphtmfr, "\n");
4610:
4611: /* For each age */
4612: for(iage=iagemin; iage <= iagemax+3; iage++){
4613: fprintf(ficresphtm,"<tr>");
4614: if(iage==iagemax+1){
4615: fprintf(ficlog,"1");
4616: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4617: }else if(iage==iagemax+2){
4618: fprintf(ficlog,"0");
4619: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4620: }else if(iage==iagemax+3){
4621: fprintf(ficlog,"Total");
4622: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4623: }else{
1.240 brouard 4624: if(first==1){
1.251 brouard 4625: first=0;
4626: printf("See log file for details...\n");
4627: }
4628: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4629: fprintf(ficlog,"Age %d", iage);
4630: }
1.265 brouard 4631: for(s1=1; s1 <=nlstate ; s1++){
4632: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4633: pp[s1] += freq[s1][m][iage];
1.251 brouard 4634: }
1.265 brouard 4635: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4636: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4637: pos += freq[s1][m][iage];
4638: if(pp[s1]>=1.e-10){
1.251 brouard 4639: if(first==1){
1.265 brouard 4640: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4641: }
1.265 brouard 4642: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4643: }else{
4644: if(first==1)
1.265 brouard 4645: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4646: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4647: }
4648: }
4649:
1.265 brouard 4650: for(s1=1; s1 <=nlstate ; s1++){
4651: /* posprop[s1]=0; */
4652: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4653: pp[s1] += freq[s1][m][iage];
4654: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4655:
4656: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4657: pos += pp[s1]; /* pos is the total number of transitions until this age */
4658: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4659: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4660: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4661: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4662: }
4663:
4664: /* Writing ficresp */
4665: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4666: if( iage <= iagemax){
4667: fprintf(ficresp," %d",iage);
4668: }
4669: }else if( nj==2){
4670: if( iage <= iagemax){
4671: fprintf(ficresp," %d",iage);
4672: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4673: }
1.240 brouard 4674: }
1.265 brouard 4675: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4676: if(pos>=1.e-5){
1.251 brouard 4677: if(first==1)
1.265 brouard 4678: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4679: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4680: }else{
4681: if(first==1)
1.265 brouard 4682: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4683: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4684: }
4685: if( iage <= iagemax){
4686: if(pos>=1.e-5){
1.265 brouard 4687: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4688: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4689: }else if( nj==2){
4690: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4691: }
4692: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4693: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4694: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4695: } else{
4696: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4697: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4698: }
1.240 brouard 4699: }
1.265 brouard 4700: pospropt[s1] +=posprop[s1];
4701: } /* end loop s1 */
1.251 brouard 4702: /* pospropt=0.; */
1.265 brouard 4703: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4704: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4705: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4706: if(first==1){
1.265 brouard 4707: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4708: }
1.265 brouard 4709: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4710: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4711: }
1.265 brouard 4712: if(s1!=0 && m!=0)
4713: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4714: }
1.265 brouard 4715: } /* end loop s1 */
1.251 brouard 4716: posproptt=0.;
1.265 brouard 4717: for(s1=1; s1 <=nlstate; s1++){
4718: posproptt += pospropt[s1];
1.251 brouard 4719: }
4720: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4721: fprintf(ficresphtm,"</tr>\n");
4722: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4723: if(iage <= iagemax)
4724: fprintf(ficresp,"\n");
1.240 brouard 4725: }
1.251 brouard 4726: if(first==1)
4727: printf("Others in log...\n");
4728: fprintf(ficlog,"\n");
4729: } /* end loop age iage */
1.265 brouard 4730:
1.251 brouard 4731: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4732: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4733: if(posproptt < 1.e-5){
1.265 brouard 4734: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4735: }else{
1.265 brouard 4736: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4737: }
1.226 brouard 4738: }
1.251 brouard 4739: fprintf(ficresphtm,"</tr>\n");
4740: fprintf(ficresphtm,"</table>\n");
4741: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4742: if(posproptt < 1.e-5){
1.251 brouard 4743: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4744: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4745: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4746: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4747: invalidvarcomb[j1]=1;
1.226 brouard 4748: }else{
1.251 brouard 4749: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4750: invalidvarcomb[j1]=0;
1.226 brouard 4751: }
1.251 brouard 4752: fprintf(ficresphtmfr,"</table>\n");
4753: fprintf(ficlog,"\n");
4754: if(j!=0){
4755: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4756: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4757: for(k=1; k <=(nlstate+ndeath); k++){
4758: if (k != i) {
1.265 brouard 4759: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4760: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4761: if(j1==1){ /* All dummy covariates to zero */
4762: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4763: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4764: printf("%d%d ",i,k);
4765: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4766: 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]));
4767: 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]));
4768: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4769: }
1.253 brouard 4770: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4771: for(iage=iagemin; iage <= iagemax+3; iage++){
4772: x[iage]= (double)iage;
4773: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4774: /* 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 4775: }
1.268 brouard 4776: /* Some are not finite, but linreg will ignore these ages */
4777: no=0;
1.253 brouard 4778: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4779: pstart[s1]=b;
4780: pstart[s1-1]=a;
1.252 brouard 4781: }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 */
4782: 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]);
4783: 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 4784: 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 4785: printf("%d%d ",i,k);
4786: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4787: 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 4788: }else{ /* Other cases, like quantitative fixed or varying covariates */
4789: ;
4790: }
4791: /* printf("%12.7f )", param[i][jj][k]); */
4792: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4793: s1++;
1.251 brouard 4794: } /* end jj */
4795: } /* end k!= i */
4796: } /* end k */
1.265 brouard 4797: } /* end i, s1 */
1.251 brouard 4798: } /* end j !=0 */
4799: } /* end selected combination of covariate j1 */
4800: if(j==0){ /* We can estimate starting values from the occurences in each case */
4801: printf("#Freqsummary: Starting values for the constants:\n");
4802: fprintf(ficlog,"\n");
1.265 brouard 4803: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4804: for(k=1; k <=(nlstate+ndeath); k++){
4805: if (k != i) {
4806: printf("%d%d ",i,k);
4807: fprintf(ficlog,"%d%d ",i,k);
4808: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4809: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4810: if(jj==1){ /* Age has to be done */
1.265 brouard 4811: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4812: 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]));
4813: 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 4814: }
4815: /* printf("%12.7f )", param[i][jj][k]); */
4816: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4817: s1++;
1.250 brouard 4818: }
1.251 brouard 4819: printf("\n");
4820: fprintf(ficlog,"\n");
1.250 brouard 4821: }
4822: }
4823: }
1.251 brouard 4824: printf("#Freqsummary\n");
4825: fprintf(ficlog,"\n");
1.265 brouard 4826: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4827: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4828: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4829: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4830: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4831: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4832: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4833: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4834: /* } */
4835: }
1.265 brouard 4836: } /* end loop s1 */
1.251 brouard 4837:
4838: printf("\n");
4839: fprintf(ficlog,"\n");
4840: } /* end j=0 */
1.249 brouard 4841: } /* end j */
1.252 brouard 4842:
1.253 brouard 4843: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4844: for(i=1, jk=1; i <=nlstate; i++){
4845: for(j=1; j <=nlstate+ndeath; j++){
4846: if(j!=i){
4847: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4848: printf("%1d%1d",i,j);
4849: fprintf(ficparo,"%1d%1d",i,j);
4850: for(k=1; k<=ncovmodel;k++){
4851: /* printf(" %lf",param[i][j][k]); */
4852: /* fprintf(ficparo," %lf",param[i][j][k]); */
4853: p[jk]=pstart[jk];
4854: printf(" %f ",pstart[jk]);
4855: fprintf(ficparo," %f ",pstart[jk]);
4856: jk++;
4857: }
4858: printf("\n");
4859: fprintf(ficparo,"\n");
4860: }
4861: }
4862: }
4863: } /* end mle=-2 */
1.226 brouard 4864: dateintmean=dateintsum/k2cpt;
1.240 brouard 4865:
1.226 brouard 4866: fclose(ficresp);
4867: fclose(ficresphtm);
4868: fclose(ficresphtmfr);
4869: free_vector(meanq,1,nqfveff);
4870: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4871: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4872: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4873: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4874: free_vector(pospropt,1,nlstate);
4875: free_vector(posprop,1,nlstate);
1.251 brouard 4876: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4877: free_vector(pp,1,nlstate);
4878: /* End of freqsummary */
4879: }
1.126 brouard 4880:
1.268 brouard 4881: /* Simple linear regression */
4882: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4883:
4884: /* y=a+bx regression */
4885: double sumx = 0.0; /* sum of x */
4886: double sumx2 = 0.0; /* sum of x**2 */
4887: double sumxy = 0.0; /* sum of x * y */
4888: double sumy = 0.0; /* sum of y */
4889: double sumy2 = 0.0; /* sum of y**2 */
4890: double sume2 = 0.0; /* sum of square or residuals */
4891: double yhat;
4892:
4893: double denom=0;
4894: int i;
4895: int ne=*no;
4896:
4897: for ( i=ifi, ne=0;i<=ila;i++) {
4898: if(!isfinite(x[i]) || !isfinite(y[i])){
4899: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4900: continue;
4901: }
4902: ne=ne+1;
4903: sumx += x[i];
4904: sumx2 += x[i]*x[i];
4905: sumxy += x[i] * y[i];
4906: sumy += y[i];
4907: sumy2 += y[i]*y[i];
4908: denom = (ne * sumx2 - sumx*sumx);
4909: /* 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); */
4910: }
4911:
4912: denom = (ne * sumx2 - sumx*sumx);
4913: if (denom == 0) {
4914: // vertical, slope m is infinity
4915: *b = INFINITY;
4916: *a = 0;
4917: if (r) *r = 0;
4918: return 1;
4919: }
4920:
4921: *b = (ne * sumxy - sumx * sumy) / denom;
4922: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4923: if (r!=NULL) {
4924: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4925: sqrt((sumx2 - sumx*sumx/ne) *
4926: (sumy2 - sumy*sumy/ne));
4927: }
4928: *no=ne;
4929: for ( i=ifi, ne=0;i<=ila;i++) {
4930: if(!isfinite(x[i]) || !isfinite(y[i])){
4931: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4932: continue;
4933: }
4934: ne=ne+1;
4935: yhat = y[i] - *a -*b* x[i];
4936: sume2 += yhat * yhat ;
4937:
4938: denom = (ne * sumx2 - sumx*sumx);
4939: /* 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); */
4940: }
4941: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
4942: *sa= *sb * sqrt(sumx2/ne);
4943:
4944: return 0;
4945: }
4946:
1.126 brouard 4947: /************ Prevalence ********************/
1.227 brouard 4948: 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)
4949: {
4950: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4951: in each health status at the date of interview (if between dateprev1 and dateprev2).
4952: We still use firstpass and lastpass as another selection.
4953: */
1.126 brouard 4954:
1.227 brouard 4955: int i, m, jk, j1, bool, z1,j, iv;
4956: int mi; /* Effective wave */
4957: int iage;
4958: double agebegin, ageend;
4959:
4960: double **prop;
4961: double posprop;
4962: double y2; /* in fractional years */
4963: int iagemin, iagemax;
4964: int first; /** to stop verbosity which is redirected to log file */
4965:
4966: iagemin= (int) agemin;
4967: iagemax= (int) agemax;
4968: /*pp=vector(1,nlstate);*/
1.251 brouard 4969: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4970: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4971: j1=0;
1.222 brouard 4972:
1.227 brouard 4973: /*j=cptcoveff;*/
4974: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4975:
1.227 brouard 4976: first=1;
4977: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4978: for (i=1; i<=nlstate; i++)
1.251 brouard 4979: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4980: prop[i][iage]=0.0;
4981: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4982: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4983: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4984:
4985: for (i=1; i<=imx; i++) { /* Each individual */
4986: bool=1;
4987: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4988: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4989: m=mw[mi][i];
4990: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4991: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4992: for (z1=1; z1<=cptcoveff; z1++){
4993: if( Fixed[Tmodelind[z1]]==1){
4994: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4995: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4996: bool=0;
4997: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4998: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4999: bool=0;
5000: }
5001: }
5002: if(bool==1){ /* Otherwise we skip that wave/person */
5003: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5004: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5005: if(m >=firstpass && m <=lastpass){
5006: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5007: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5008: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5009: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5010: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5011: 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);
5012: exit(1);
5013: }
5014: if (s[m][i]>0 && s[m][i]<=nlstate) {
5015: /*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]]);*/
5016: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5017: prop[s[m][i]][iagemax+3] += weight[i];
5018: } /* end valid statuses */
5019: } /* end selection of dates */
5020: } /* end selection of waves */
5021: } /* end bool */
5022: } /* end wave */
5023: } /* end individual */
5024: for(i=iagemin; i <= iagemax+3; i++){
5025: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5026: posprop += prop[jk][i];
5027: }
5028:
5029: for(jk=1; jk <=nlstate ; jk++){
5030: if( i <= iagemax){
5031: if(posprop>=1.e-5){
5032: probs[i][jk][j1]= prop[jk][i]/posprop;
5033: } else{
5034: if(first==1){
5035: first=0;
1.266 brouard 5036: 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]);
5037: 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]);
5038: }else{
5039: 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 5040: }
5041: }
5042: }
5043: }/* end jk */
5044: }/* end i */
1.222 brouard 5045: /*} *//* end i1 */
1.227 brouard 5046: } /* end j1 */
1.222 brouard 5047:
1.227 brouard 5048: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5049: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5050: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5051: } /* End of prevalence */
1.126 brouard 5052:
5053: /************* Waves Concatenation ***************/
5054:
5055: 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)
5056: {
5057: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5058: Death is a valid wave (if date is known).
5059: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5060: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5061: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5062: */
1.126 brouard 5063:
1.224 brouard 5064: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5065: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5066: double sum=0., jmean=0.;*/
1.224 brouard 5067: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5068: int j, k=0,jk, ju, jl;
5069: double sum=0.;
5070: first=0;
1.214 brouard 5071: firstwo=0;
1.217 brouard 5072: firsthree=0;
1.218 brouard 5073: firstfour=0;
1.164 brouard 5074: jmin=100000;
1.126 brouard 5075: jmax=-1;
5076: jmean=0.;
1.224 brouard 5077:
5078: /* Treating live states */
1.214 brouard 5079: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5080: mi=0; /* First valid wave */
1.227 brouard 5081: mli=0; /* Last valid wave */
1.126 brouard 5082: m=firstpass;
1.214 brouard 5083: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5084: 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 */
5085: mli=m-1;/* mw[++mi][i]=m-1; */
5086: }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 */
5087: mw[++mi][i]=m;
5088: mli=m;
1.224 brouard 5089: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5090: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5091: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5092: }
1.227 brouard 5093: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5094: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5095: break;
1.224 brouard 5096: #else
1.227 brouard 5097: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5098: if(firsthree == 0){
1.262 brouard 5099: 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 5100: firsthree=1;
5101: }
1.262 brouard 5102: 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 5103: mw[++mi][i]=m;
5104: mli=m;
5105: }
5106: if(s[m][i]==-2){ /* Vital status is really unknown */
5107: nbwarn++;
5108: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5109: 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);
5110: 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);
5111: }
5112: break;
5113: }
5114: break;
1.224 brouard 5115: #endif
1.227 brouard 5116: }/* End m >= lastpass */
1.126 brouard 5117: }/* end while */
1.224 brouard 5118:
1.227 brouard 5119: /* 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 5120: /* After last pass */
1.224 brouard 5121: /* Treating death states */
1.214 brouard 5122: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5123: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5124: /* } */
1.126 brouard 5125: mi++; /* Death is another wave */
5126: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5127: /* Only death is a correct wave */
1.126 brouard 5128: mw[mi][i]=m;
1.257 brouard 5129: } /* else not in a death state */
1.224 brouard 5130: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5131: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5132: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5133: 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 */
5134: nbwarn++;
5135: if(firstfiv==0){
5136: 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 );
5137: firstfiv=1;
5138: }else{
5139: 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 );
5140: }
5141: }else{ /* Death occured afer last wave potential bias */
5142: nberr++;
5143: if(firstwo==0){
1.257 brouard 5144: 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 5145: firstwo=1;
5146: }
1.257 brouard 5147: 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 5148: }
1.257 brouard 5149: }else{ /* if date of interview is unknown */
1.227 brouard 5150: /* death is known but not confirmed by death status at any wave */
5151: if(firstfour==0){
5152: 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 );
5153: firstfour=1;
5154: }
5155: 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 5156: }
1.224 brouard 5157: } /* end if date of death is known */
5158: #endif
5159: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5160: /* wav[i]=mw[mi][i]; */
1.126 brouard 5161: if(mi==0){
5162: nbwarn++;
5163: if(first==0){
1.227 brouard 5164: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5165: first=1;
1.126 brouard 5166: }
5167: if(first==1){
1.227 brouard 5168: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5169: }
5170: } /* end mi==0 */
5171: } /* End individuals */
1.214 brouard 5172: /* wav and mw are no more changed */
1.223 brouard 5173:
1.214 brouard 5174:
1.126 brouard 5175: for(i=1; i<=imx; i++){
5176: for(mi=1; mi<wav[i];mi++){
5177: if (stepm <=0)
1.227 brouard 5178: dh[mi][i]=1;
1.126 brouard 5179: else{
1.260 brouard 5180: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5181: if (agedc[i] < 2*AGESUP) {
5182: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5183: if(j==0) j=1; /* Survives at least one month after exam */
5184: else if(j<0){
5185: nberr++;
5186: 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]);
5187: j=1; /* Temporary Dangerous patch */
5188: 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);
5189: 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]);
5190: 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);
5191: }
5192: k=k+1;
5193: if (j >= jmax){
5194: jmax=j;
5195: ijmax=i;
5196: }
5197: if (j <= jmin){
5198: jmin=j;
5199: ijmin=i;
5200: }
5201: sum=sum+j;
5202: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5203: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5204: }
5205: }
5206: else{
5207: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5208: /* 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 5209:
1.227 brouard 5210: k=k+1;
5211: if (j >= jmax) {
5212: jmax=j;
5213: ijmax=i;
5214: }
5215: else if (j <= jmin){
5216: jmin=j;
5217: ijmin=i;
5218: }
5219: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5220: /*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]);*/
5221: if(j<0){
5222: nberr++;
5223: 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]);
5224: 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]);
5225: }
5226: sum=sum+j;
5227: }
5228: jk= j/stepm;
5229: jl= j -jk*stepm;
5230: ju= j -(jk+1)*stepm;
5231: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5232: if(jl==0){
5233: dh[mi][i]=jk;
5234: bh[mi][i]=0;
5235: }else{ /* We want a negative bias in order to only have interpolation ie
5236: * to avoid the price of an extra matrix product in likelihood */
5237: dh[mi][i]=jk+1;
5238: bh[mi][i]=ju;
5239: }
5240: }else{
5241: if(jl <= -ju){
5242: dh[mi][i]=jk;
5243: bh[mi][i]=jl; /* bias is positive if real duration
5244: * is higher than the multiple of stepm and negative otherwise.
5245: */
5246: }
5247: else{
5248: dh[mi][i]=jk+1;
5249: bh[mi][i]=ju;
5250: }
5251: if(dh[mi][i]==0){
5252: dh[mi][i]=1; /* At least one step */
5253: bh[mi][i]=ju; /* At least one step */
5254: /* 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);*/
5255: }
5256: } /* end if mle */
1.126 brouard 5257: }
5258: } /* end wave */
5259: }
5260: jmean=sum/k;
5261: 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 5262: 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 5263: }
1.126 brouard 5264:
5265: /*********** Tricode ****************************/
1.220 brouard 5266: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5267: {
5268: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5269: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5270: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5271: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5272: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5273: */
1.130 brouard 5274:
1.242 brouard 5275: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5276: int modmaxcovj=0; /* Modality max of covariates j */
5277: int cptcode=0; /* Modality max of covariates j */
5278: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5279:
5280:
1.242 brouard 5281: /* cptcoveff=0; */
5282: /* *cptcov=0; */
1.126 brouard 5283:
1.242 brouard 5284: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5285:
1.242 brouard 5286: /* Loop on covariates without age and products and no quantitative variable */
5287: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5288: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5289: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5290: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5291: switch(Fixed[k]) {
5292: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5293: 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*/
5294: ij=(int)(covar[Tvar[k]][i]);
5295: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5296: * If product of Vn*Vm, still boolean *:
5297: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5298: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5299: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5300: modality of the nth covariate of individual i. */
5301: if (ij > modmaxcovj)
5302: modmaxcovj=ij;
5303: else if (ij < modmincovj)
5304: modmincovj=ij;
5305: if ((ij < -1) && (ij > NCOVMAX)){
5306: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5307: exit(1);
5308: }else
5309: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5310: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5311: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5312: /* getting the maximum value of the modality of the covariate
5313: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5314: female ies 1, then modmaxcovj=1.
5315: */
5316: } /* end for loop on individuals i */
5317: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5318: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5319: cptcode=modmaxcovj;
5320: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5321: /*for (i=0; i<=cptcode; i++) {*/
5322: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5323: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5324: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5325: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5326: if( j != -1){
5327: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5328: covariate for which somebody answered excluding
5329: undefined. Usually 2: 0 and 1. */
5330: }
5331: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5332: covariate for which somebody answered including
5333: undefined. Usually 3: -1, 0 and 1. */
5334: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5335: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5336: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5337:
1.242 brouard 5338: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5339: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5340: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5341: /* modmincovj=3; modmaxcovj = 7; */
5342: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5343: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5344: /* defining two dummy variables: variables V1_1 and V1_2.*/
5345: /* nbcode[Tvar[j]][ij]=k; */
5346: /* nbcode[Tvar[j]][1]=0; */
5347: /* nbcode[Tvar[j]][2]=1; */
5348: /* nbcode[Tvar[j]][3]=2; */
5349: /* To be continued (not working yet). */
5350: ij=0; /* ij is similar to i but can jump over null modalities */
5351: 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*/
5352: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5353: break;
5354: }
5355: ij++;
5356: 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*/
5357: cptcode = ij; /* New max modality for covar j */
5358: } /* end of loop on modality i=-1 to 1 or more */
5359: break;
5360: case 1: /* Testing on varying covariate, could be simple and
5361: * should look at waves or product of fixed *
5362: * varying. No time to test -1, assuming 0 and 1 only */
5363: ij=0;
5364: for(i=0; i<=1;i++){
5365: nbcode[Tvar[k]][++ij]=i;
5366: }
5367: break;
5368: default:
5369: break;
5370: } /* end switch */
5371: } /* end dummy test */
5372:
5373: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5374: /* /\*recode from 0 *\/ */
5375: /* k is a modality. If we have model=V1+V1*sex */
5376: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5377: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5378: /* } */
5379: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5380: /* if (ij > ncodemax[j]) { */
5381: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5382: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5383: /* break; */
5384: /* } */
5385: /* } /\* end of loop on modality k *\/ */
5386: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5387:
5388: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5389: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5390: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5391: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5392: 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 */
5393: 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 */
5394: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5395: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5396:
5397: ij=0;
5398: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5399: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5400: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5401: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5402: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5403: /* If product not in single variable we don't print results */
5404: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5405: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5406: 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*/
5407: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5408: 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 */
5409: if(Fixed[k]!=0)
5410: anyvaryingduminmodel=1;
5411: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5412: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5413: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5414: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5415: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5416: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5417: }
5418: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5419: /* ij--; */
5420: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5421: *cptcov=ij; /*Number of total real effective covariates: effective
5422: * because they can be excluded from the model and real
5423: * if in the model but excluded because missing values, but how to get k from ij?*/
5424: for(j=ij+1; j<= cptcovt; j++){
5425: Tvaraff[j]=0;
5426: Tmodelind[j]=0;
5427: }
5428: for(j=ntveff+1; j<= cptcovt; j++){
5429: TmodelInvind[j]=0;
5430: }
5431: /* To be sorted */
5432: ;
5433: }
1.126 brouard 5434:
1.145 brouard 5435:
1.126 brouard 5436: /*********** Health Expectancies ****************/
5437:
1.235 brouard 5438: 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 5439:
5440: {
5441: /* Health expectancies, no variances */
1.164 brouard 5442: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5443: int nhstepma, nstepma; /* Decreasing with age */
5444: double age, agelim, hf;
5445: double ***p3mat;
5446: double eip;
5447:
1.238 brouard 5448: /* pstamp(ficreseij); */
1.126 brouard 5449: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5450: fprintf(ficreseij,"# Age");
5451: for(i=1; i<=nlstate;i++){
5452: for(j=1; j<=nlstate;j++){
5453: fprintf(ficreseij," e%1d%1d ",i,j);
5454: }
5455: fprintf(ficreseij," e%1d. ",i);
5456: }
5457: fprintf(ficreseij,"\n");
5458:
5459:
5460: if(estepm < stepm){
5461: printf ("Problem %d lower than %d\n",estepm, stepm);
5462: }
5463: else hstepm=estepm;
5464: /* We compute the life expectancy from trapezoids spaced every estepm months
5465: * This is mainly to measure the difference between two models: for example
5466: * if stepm=24 months pijx are given only every 2 years and by summing them
5467: * we are calculating an estimate of the Life Expectancy assuming a linear
5468: * progression in between and thus overestimating or underestimating according
5469: * to the curvature of the survival function. If, for the same date, we
5470: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5471: * to compare the new estimate of Life expectancy with the same linear
5472: * hypothesis. A more precise result, taking into account a more precise
5473: * curvature will be obtained if estepm is as small as stepm. */
5474:
5475: /* For example we decided to compute the life expectancy with the smallest unit */
5476: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5477: nhstepm is the number of hstepm from age to agelim
5478: nstepm is the number of stepm from age to agelin.
1.270 brouard 5479: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5480: and note for a fixed period like estepm months */
5481: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5482: survival function given by stepm (the optimization length). Unfortunately it
5483: means that if the survival funtion is printed only each two years of age and if
5484: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5485: results. So we changed our mind and took the option of the best precision.
5486: */
5487: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5488:
5489: agelim=AGESUP;
5490: /* If stepm=6 months */
5491: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5492: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5493:
5494: /* nhstepm age range expressed in number of stepm */
5495: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5496: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5497: /* if (stepm >= YEARM) hstepm=1;*/
5498: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5499: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5500:
5501: for (age=bage; age<=fage; age ++){
5502: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5503: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5504: /* if (stepm >= YEARM) hstepm=1;*/
5505: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5506:
5507: /* If stepm=6 months */
5508: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5509: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5510:
1.235 brouard 5511: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5512:
5513: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5514:
5515: printf("%d|",(int)age);fflush(stdout);
5516: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5517:
5518: /* Computing expectancies */
5519: for(i=1; i<=nlstate;i++)
5520: for(j=1; j<=nlstate;j++)
5521: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5522: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5523:
5524: /* 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]);*/
5525:
5526: }
5527:
5528: fprintf(ficreseij,"%3.0f",age );
5529: for(i=1; i<=nlstate;i++){
5530: eip=0;
5531: for(j=1; j<=nlstate;j++){
5532: eip +=eij[i][j][(int)age];
5533: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5534: }
5535: fprintf(ficreseij,"%9.4f", eip );
5536: }
5537: fprintf(ficreseij,"\n");
5538:
5539: }
5540: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5541: printf("\n");
5542: fprintf(ficlog,"\n");
5543:
5544: }
5545:
1.235 brouard 5546: 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 5547:
5548: {
5549: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5550: to initial status i, ei. .
1.126 brouard 5551: */
5552: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5553: int nhstepma, nstepma; /* Decreasing with age */
5554: double age, agelim, hf;
5555: double ***p3matp, ***p3matm, ***varhe;
5556: double **dnewm,**doldm;
5557: double *xp, *xm;
5558: double **gp, **gm;
5559: double ***gradg, ***trgradg;
5560: int theta;
5561:
5562: double eip, vip;
5563:
5564: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5565: xp=vector(1,npar);
5566: xm=vector(1,npar);
5567: dnewm=matrix(1,nlstate*nlstate,1,npar);
5568: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5569:
5570: pstamp(ficresstdeij);
5571: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5572: fprintf(ficresstdeij,"# Age");
5573: for(i=1; i<=nlstate;i++){
5574: for(j=1; j<=nlstate;j++)
5575: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5576: fprintf(ficresstdeij," e%1d. ",i);
5577: }
5578: fprintf(ficresstdeij,"\n");
5579:
5580: pstamp(ficrescveij);
5581: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5582: fprintf(ficrescveij,"# Age");
5583: for(i=1; i<=nlstate;i++)
5584: for(j=1; j<=nlstate;j++){
5585: cptj= (j-1)*nlstate+i;
5586: for(i2=1; i2<=nlstate;i2++)
5587: for(j2=1; j2<=nlstate;j2++){
5588: cptj2= (j2-1)*nlstate+i2;
5589: if(cptj2 <= cptj)
5590: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5591: }
5592: }
5593: fprintf(ficrescveij,"\n");
5594:
5595: if(estepm < stepm){
5596: printf ("Problem %d lower than %d\n",estepm, stepm);
5597: }
5598: else hstepm=estepm;
5599: /* We compute the life expectancy from trapezoids spaced every estepm months
5600: * This is mainly to measure the difference between two models: for example
5601: * if stepm=24 months pijx are given only every 2 years and by summing them
5602: * we are calculating an estimate of the Life Expectancy assuming a linear
5603: * progression in between and thus overestimating or underestimating according
5604: * to the curvature of the survival function. If, for the same date, we
5605: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5606: * to compare the new estimate of Life expectancy with the same linear
5607: * hypothesis. A more precise result, taking into account a more precise
5608: * curvature will be obtained if estepm is as small as stepm. */
5609:
5610: /* For example we decided to compute the life expectancy with the smallest unit */
5611: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5612: nhstepm is the number of hstepm from age to agelim
5613: nstepm is the number of stepm from age to agelin.
5614: Look at hpijx to understand the reason of that which relies in memory size
5615: and note for a fixed period like estepm months */
5616: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5617: survival function given by stepm (the optimization length). Unfortunately it
5618: means that if the survival funtion is printed only each two years of age and if
5619: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5620: results. So we changed our mind and took the option of the best precision.
5621: */
5622: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5623:
5624: /* If stepm=6 months */
5625: /* nhstepm age range expressed in number of stepm */
5626: agelim=AGESUP;
5627: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5628: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5629: /* if (stepm >= YEARM) hstepm=1;*/
5630: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5631:
5632: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5633: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5634: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5635: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5636: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5637: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5638:
5639: for (age=bage; age<=fage; age ++){
5640: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5641: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5642: /* if (stepm >= YEARM) hstepm=1;*/
5643: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5644:
1.126 brouard 5645: /* If stepm=6 months */
5646: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5647: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5648:
5649: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5650:
1.126 brouard 5651: /* Computing Variances of health expectancies */
5652: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5653: decrease memory allocation */
5654: for(theta=1; theta <=npar; theta++){
5655: for(i=1; i<=npar; i++){
1.222 brouard 5656: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5657: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5658: }
1.235 brouard 5659: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5660: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5661:
1.126 brouard 5662: for(j=1; j<= nlstate; j++){
1.222 brouard 5663: for(i=1; i<=nlstate; i++){
5664: for(h=0; h<=nhstepm-1; h++){
5665: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5666: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5667: }
5668: }
1.126 brouard 5669: }
1.218 brouard 5670:
1.126 brouard 5671: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5672: for(h=0; h<=nhstepm-1; h++){
5673: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5674: }
1.126 brouard 5675: }/* End theta */
5676:
5677:
5678: for(h=0; h<=nhstepm-1; h++)
5679: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5680: for(theta=1; theta <=npar; theta++)
5681: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5682:
1.218 brouard 5683:
1.222 brouard 5684: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5685: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5686: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5687:
1.222 brouard 5688: printf("%d|",(int)age);fflush(stdout);
5689: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5690: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5691: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5692: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5693: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5694: for(ij=1;ij<=nlstate*nlstate;ij++)
5695: for(ji=1;ji<=nlstate*nlstate;ji++)
5696: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5697: }
5698: }
1.218 brouard 5699:
1.126 brouard 5700: /* Computing expectancies */
1.235 brouard 5701: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5702: for(i=1; i<=nlstate;i++)
5703: for(j=1; j<=nlstate;j++)
1.222 brouard 5704: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5705: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5706:
1.222 brouard 5707: /* 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 5708:
1.222 brouard 5709: }
1.269 brouard 5710:
5711: /* Standard deviation of expectancies ij */
1.126 brouard 5712: fprintf(ficresstdeij,"%3.0f",age );
5713: for(i=1; i<=nlstate;i++){
5714: eip=0.;
5715: vip=0.;
5716: for(j=1; j<=nlstate;j++){
1.222 brouard 5717: eip += eij[i][j][(int)age];
5718: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5719: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5720: 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 5721: }
5722: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5723: }
5724: fprintf(ficresstdeij,"\n");
1.218 brouard 5725:
1.269 brouard 5726: /* Variance of expectancies ij */
1.126 brouard 5727: fprintf(ficrescveij,"%3.0f",age );
5728: for(i=1; i<=nlstate;i++)
5729: for(j=1; j<=nlstate;j++){
1.222 brouard 5730: cptj= (j-1)*nlstate+i;
5731: for(i2=1; i2<=nlstate;i2++)
5732: for(j2=1; j2<=nlstate;j2++){
5733: cptj2= (j2-1)*nlstate+i2;
5734: if(cptj2 <= cptj)
5735: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5736: }
1.126 brouard 5737: }
5738: fprintf(ficrescveij,"\n");
1.218 brouard 5739:
1.126 brouard 5740: }
5741: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5742: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5743: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5744: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5745: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5746: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5747: printf("\n");
5748: fprintf(ficlog,"\n");
1.218 brouard 5749:
1.126 brouard 5750: free_vector(xm,1,npar);
5751: free_vector(xp,1,npar);
5752: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5753: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5754: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5755: }
1.218 brouard 5756:
1.126 brouard 5757: /************ Variance ******************/
1.235 brouard 5758: 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 5759: {
5760: /* Variance of health expectancies */
5761: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5762: /* double **newm;*/
5763: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5764:
5765: /* int movingaverage(); */
5766: double **dnewm,**doldm;
5767: double **dnewmp,**doldmp;
5768: int i, j, nhstepm, hstepm, h, nstepm ;
5769: int k;
5770: double *xp;
5771: double **gp, **gm; /* for var eij */
5772: double ***gradg, ***trgradg; /*for var eij */
5773: double **gradgp, **trgradgp; /* for var p point j */
5774: double *gpp, *gmp; /* for var p point j */
5775: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5776: double ***p3mat;
5777: double age,agelim, hf;
5778: /* double ***mobaverage; */
5779: int theta;
5780: char digit[4];
5781: char digitp[25];
5782:
5783: char fileresprobmorprev[FILENAMELENGTH];
5784:
5785: if(popbased==1){
5786: if(mobilav!=0)
5787: strcpy(digitp,"-POPULBASED-MOBILAV_");
5788: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5789: }
5790: else
5791: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5792:
1.218 brouard 5793: /* if (mobilav!=0) { */
5794: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5795: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5796: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5797: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5798: /* } */
5799: /* } */
5800:
5801: strcpy(fileresprobmorprev,"PRMORPREV-");
5802: sprintf(digit,"%-d",ij);
5803: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5804: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5805: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5806: strcat(fileresprobmorprev,fileresu);
5807: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5808: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5809: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5810: }
5811: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5812: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5813: pstamp(ficresprobmorprev);
5814: 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 5815: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5816: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5817: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5818: }
5819: for(j=1;j<=cptcoveff;j++)
5820: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5821: fprintf(ficresprobmorprev,"\n");
5822:
1.218 brouard 5823: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5824: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5825: fprintf(ficresprobmorprev," p.%-d SE",j);
5826: for(i=1; i<=nlstate;i++)
5827: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5828: }
5829: fprintf(ficresprobmorprev,"\n");
5830:
5831: fprintf(ficgp,"\n# Routine varevsij");
5832: fprintf(ficgp,"\nunset title \n");
5833: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5834: 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");
5835: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5836: /* } */
5837: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5838: pstamp(ficresvij);
5839: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5840: if(popbased==1)
5841: 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);
5842: else
5843: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5844: fprintf(ficresvij,"# Age");
5845: for(i=1; i<=nlstate;i++)
5846: for(j=1; j<=nlstate;j++)
5847: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5848: fprintf(ficresvij,"\n");
5849:
5850: xp=vector(1,npar);
5851: dnewm=matrix(1,nlstate,1,npar);
5852: doldm=matrix(1,nlstate,1,nlstate);
5853: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5854: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5855:
5856: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5857: gpp=vector(nlstate+1,nlstate+ndeath);
5858: gmp=vector(nlstate+1,nlstate+ndeath);
5859: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5860:
1.218 brouard 5861: if(estepm < stepm){
5862: printf ("Problem %d lower than %d\n",estepm, stepm);
5863: }
5864: else hstepm=estepm;
5865: /* For example we decided to compute the life expectancy with the smallest unit */
5866: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5867: nhstepm is the number of hstepm from age to agelim
5868: nstepm is the number of stepm from age to agelim.
5869: Look at function hpijx to understand why because of memory size limitations,
5870: we decided (b) to get a life expectancy respecting the most precise curvature of the
5871: survival function given by stepm (the optimization length). Unfortunately it
5872: means that if the survival funtion is printed every two years of age and if
5873: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5874: results. So we changed our mind and took the option of the best precision.
5875: */
5876: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5877: agelim = AGESUP;
5878: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5879: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5880: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5881: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5882: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5883: gp=matrix(0,nhstepm,1,nlstate);
5884: gm=matrix(0,nhstepm,1,nlstate);
5885:
5886:
5887: for(theta=1; theta <=npar; theta++){
5888: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5889: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5890: }
5891:
1.242 brouard 5892: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5893:
5894: if (popbased==1) {
5895: if(mobilav ==0){
5896: for(i=1; i<=nlstate;i++)
5897: prlim[i][i]=probs[(int)age][i][ij];
5898: }else{ /* mobilav */
5899: for(i=1; i<=nlstate;i++)
5900: prlim[i][i]=mobaverage[(int)age][i][ij];
5901: }
5902: }
5903:
1.235 brouard 5904: 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 5905: for(j=1; j<= nlstate; j++){
5906: for(h=0; h<=nhstepm; h++){
5907: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5908: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5909: }
5910: }
5911: /* Next for computing probability of death (h=1 means
5912: computed over hstepm matrices product = hstepm*stepm months)
5913: as a weighted average of prlim.
5914: */
5915: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5916: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5917: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5918: }
5919: /* end probability of death */
5920:
5921: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5922: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5923:
1.242 brouard 5924: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5925:
5926: if (popbased==1) {
5927: if(mobilav ==0){
5928: for(i=1; i<=nlstate;i++)
5929: prlim[i][i]=probs[(int)age][i][ij];
5930: }else{ /* mobilav */
5931: for(i=1; i<=nlstate;i++)
5932: prlim[i][i]=mobaverage[(int)age][i][ij];
5933: }
5934: }
5935:
1.235 brouard 5936: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5937:
5938: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5939: for(h=0; h<=nhstepm; h++){
5940: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5941: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5942: }
5943: }
5944: /* This for computing probability of death (h=1 means
5945: computed over hstepm matrices product = hstepm*stepm months)
5946: as a weighted average of prlim.
5947: */
5948: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5949: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5950: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5951: }
5952: /* end probability of death */
5953:
5954: for(j=1; j<= nlstate; j++) /* vareij */
5955: for(h=0; h<=nhstepm; h++){
5956: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5957: }
5958:
5959: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5960: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5961: }
5962:
5963: } /* End theta */
5964:
5965: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5966:
5967: for(h=0; h<=nhstepm; h++) /* veij */
5968: for(j=1; j<=nlstate;j++)
5969: for(theta=1; theta <=npar; theta++)
5970: trgradg[h][j][theta]=gradg[h][theta][j];
5971:
5972: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5973: for(theta=1; theta <=npar; theta++)
5974: trgradgp[j][theta]=gradgp[theta][j];
5975:
5976:
5977: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5978: for(i=1;i<=nlstate;i++)
5979: for(j=1;j<=nlstate;j++)
5980: vareij[i][j][(int)age] =0.;
5981:
5982: for(h=0;h<=nhstepm;h++){
5983: for(k=0;k<=nhstepm;k++){
5984: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5985: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5986: for(i=1;i<=nlstate;i++)
5987: for(j=1;j<=nlstate;j++)
5988: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5989: }
5990: }
5991:
5992: /* pptj */
5993: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5994: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5995: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5996: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5997: varppt[j][i]=doldmp[j][i];
5998: /* end ppptj */
5999: /* x centered again */
6000:
1.242 brouard 6001: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6002:
6003: if (popbased==1) {
6004: if(mobilav ==0){
6005: for(i=1; i<=nlstate;i++)
6006: prlim[i][i]=probs[(int)age][i][ij];
6007: }else{ /* mobilav */
6008: for(i=1; i<=nlstate;i++)
6009: prlim[i][i]=mobaverage[(int)age][i][ij];
6010: }
6011: }
6012:
6013: /* This for computing probability of death (h=1 means
6014: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6015: as a weighted average of prlim.
6016: */
1.235 brouard 6017: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6018: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6019: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6020: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6021: }
6022: /* end probability of death */
6023:
6024: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6025: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6026: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6027: for(i=1; i<=nlstate;i++){
6028: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6029: }
6030: }
6031: fprintf(ficresprobmorprev,"\n");
6032:
6033: fprintf(ficresvij,"%.0f ",age );
6034: for(i=1; i<=nlstate;i++)
6035: for(j=1; j<=nlstate;j++){
6036: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6037: }
6038: fprintf(ficresvij,"\n");
6039: free_matrix(gp,0,nhstepm,1,nlstate);
6040: free_matrix(gm,0,nhstepm,1,nlstate);
6041: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6042: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6043: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6044: } /* End age */
6045: free_vector(gpp,nlstate+1,nlstate+ndeath);
6046: free_vector(gmp,nlstate+1,nlstate+ndeath);
6047: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6048: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6049: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6050: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6051: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6052: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6053: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6054: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6055: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6056: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6057: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6058: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6059: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6060: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6061: 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);
6062: /* 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 6063: */
1.218 brouard 6064: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6065: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6066:
1.218 brouard 6067: free_vector(xp,1,npar);
6068: free_matrix(doldm,1,nlstate,1,nlstate);
6069: free_matrix(dnewm,1,nlstate,1,npar);
6070: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6071: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6072: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6073: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6074: fclose(ficresprobmorprev);
6075: fflush(ficgp);
6076: fflush(fichtm);
6077: } /* end varevsij */
1.126 brouard 6078:
6079: /************ Variance of prevlim ******************/
1.269 brouard 6080: 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 6081: {
1.205 brouard 6082: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6083: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6084:
1.268 brouard 6085: double **dnewmpar,**doldm;
1.126 brouard 6086: int i, j, nhstepm, hstepm;
6087: double *xp;
6088: double *gp, *gm;
6089: double **gradg, **trgradg;
1.208 brouard 6090: double **mgm, **mgp;
1.126 brouard 6091: double age,agelim;
6092: int theta;
6093:
6094: pstamp(ficresvpl);
6095: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6096: fprintf(ficresvpl,"# Age ");
6097: if(nresult >=1)
6098: fprintf(ficresvpl," Result# ");
1.126 brouard 6099: for(i=1; i<=nlstate;i++)
6100: fprintf(ficresvpl," %1d-%1d",i,i);
6101: fprintf(ficresvpl,"\n");
6102:
6103: xp=vector(1,npar);
1.268 brouard 6104: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6105: doldm=matrix(1,nlstate,1,nlstate);
6106:
6107: hstepm=1*YEARM; /* Every year of age */
6108: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6109: agelim = AGESUP;
6110: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6111: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6112: if (stepm >= YEARM) hstepm=1;
6113: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6114: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6115: mgp=matrix(1,npar,1,nlstate);
6116: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6117: gp=vector(1,nlstate);
6118: gm=vector(1,nlstate);
6119:
6120: for(theta=1; theta <=npar; theta++){
6121: for(i=1; i<=npar; i++){ /* Computes gradient */
6122: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6123: }
1.209 brouard 6124: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6125: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6126: else
1.235 brouard 6127: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6128: for(i=1;i<=nlstate;i++){
1.126 brouard 6129: gp[i] = prlim[i][i];
1.208 brouard 6130: mgp[theta][i] = prlim[i][i];
6131: }
1.126 brouard 6132: for(i=1; i<=npar; i++) /* Computes gradient */
6133: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6134: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6135: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6136: else
1.235 brouard 6137: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6138: for(i=1;i<=nlstate;i++){
1.126 brouard 6139: gm[i] = prlim[i][i];
1.208 brouard 6140: mgm[theta][i] = prlim[i][i];
6141: }
1.126 brouard 6142: for(i=1;i<=nlstate;i++)
6143: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6144: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6145: } /* End theta */
6146:
6147: trgradg =matrix(1,nlstate,1,npar);
6148:
6149: for(j=1; j<=nlstate;j++)
6150: for(theta=1; theta <=npar; theta++)
6151: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6152: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6153: /* printf("\nmgm mgp %d ",(int)age); */
6154: /* for(j=1; j<=nlstate;j++){ */
6155: /* printf(" %d ",j); */
6156: /* for(theta=1; theta <=npar; theta++) */
6157: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6158: /* printf("\n "); */
6159: /* } */
6160: /* } */
6161: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6162: /* printf("\n gradg %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 ",theta,gradg[theta][j]); */
6167: /* printf("\n "); */
6168: /* } */
6169: /* } */
1.126 brouard 6170:
6171: for(i=1;i<=nlstate;i++)
6172: varpl[i][(int)age] =0.;
1.209 brouard 6173: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6174: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6175: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6176: }else{
1.268 brouard 6177: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6178: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6179: }
1.126 brouard 6180: for(i=1;i<=nlstate;i++)
6181: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6182:
6183: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6184: if(nresult >=1)
6185: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6186: for(i=1; i<=nlstate;i++)
6187: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6188: fprintf(ficresvpl,"\n");
6189: free_vector(gp,1,nlstate);
6190: free_vector(gm,1,nlstate);
1.208 brouard 6191: free_matrix(mgm,1,npar,1,nlstate);
6192: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6193: free_matrix(gradg,1,npar,1,nlstate);
6194: free_matrix(trgradg,1,nlstate,1,npar);
6195: } /* End age */
6196:
6197: free_vector(xp,1,npar);
6198: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6199: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6200:
6201: }
6202:
6203:
6204: /************ Variance of backprevalence limit ******************/
1.269 brouard 6205: 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 6206: {
6207: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6208: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6209:
6210: double **dnewmpar,**doldm;
6211: int i, j, nhstepm, hstepm;
6212: double *xp;
6213: double *gp, *gm;
6214: double **gradg, **trgradg;
6215: double **mgm, **mgp;
6216: double age,agelim;
6217: int theta;
6218:
6219: pstamp(ficresvbl);
6220: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6221: fprintf(ficresvbl,"# Age ");
6222: if(nresult >=1)
6223: fprintf(ficresvbl," Result# ");
6224: for(i=1; i<=nlstate;i++)
6225: fprintf(ficresvbl," %1d-%1d",i,i);
6226: fprintf(ficresvbl,"\n");
6227:
6228: xp=vector(1,npar);
6229: dnewmpar=matrix(1,nlstate,1,npar);
6230: doldm=matrix(1,nlstate,1,nlstate);
6231:
6232: hstepm=1*YEARM; /* Every year of age */
6233: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6234: agelim = AGEINF;
6235: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6236: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6237: if (stepm >= YEARM) hstepm=1;
6238: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6239: gradg=matrix(1,npar,1,nlstate);
6240: mgp=matrix(1,npar,1,nlstate);
6241: mgm=matrix(1,npar,1,nlstate);
6242: gp=vector(1,nlstate);
6243: gm=vector(1,nlstate);
6244:
6245: for(theta=1; theta <=npar; theta++){
6246: for(i=1; i<=npar; i++){ /* Computes gradient */
6247: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6248: }
6249: if(mobilavproj > 0 )
6250: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6251: else
6252: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6253: for(i=1;i<=nlstate;i++){
6254: gp[i] = bprlim[i][i];
6255: mgp[theta][i] = bprlim[i][i];
6256: }
6257: for(i=1; i<=npar; i++) /* Computes gradient */
6258: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6259: if(mobilavproj > 0 )
6260: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6261: else
6262: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6263: for(i=1;i<=nlstate;i++){
6264: gm[i] = bprlim[i][i];
6265: mgm[theta][i] = bprlim[i][i];
6266: }
6267: for(i=1;i<=nlstate;i++)
6268: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6269: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6270: } /* End theta */
6271:
6272: trgradg =matrix(1,nlstate,1,npar);
6273:
6274: for(j=1; j<=nlstate;j++)
6275: for(theta=1; theta <=npar; theta++)
6276: trgradg[j][theta]=gradg[theta][j];
6277: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6278: /* printf("\nmgm mgp %d ",(int)age); */
6279: /* for(j=1; j<=nlstate;j++){ */
6280: /* printf(" %d ",j); */
6281: /* for(theta=1; theta <=npar; theta++) */
6282: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6283: /* printf("\n "); */
6284: /* } */
6285: /* } */
6286: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6287: /* printf("\n gradg %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 ",theta,gradg[theta][j]); */
6292: /* printf("\n "); */
6293: /* } */
6294: /* } */
6295:
6296: for(i=1;i<=nlstate;i++)
6297: varbpl[i][(int)age] =0.;
6298: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6299: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6300: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6301: }else{
6302: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6303: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6304: }
6305: for(i=1;i<=nlstate;i++)
6306: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6307:
6308: fprintf(ficresvbl,"%.0f ",age );
6309: if(nresult >=1)
6310: fprintf(ficresvbl,"%d ",nres );
6311: for(i=1; i<=nlstate;i++)
6312: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6313: fprintf(ficresvbl,"\n");
6314: free_vector(gp,1,nlstate);
6315: free_vector(gm,1,nlstate);
6316: free_matrix(mgm,1,npar,1,nlstate);
6317: free_matrix(mgp,1,npar,1,nlstate);
6318: free_matrix(gradg,1,npar,1,nlstate);
6319: free_matrix(trgradg,1,nlstate,1,npar);
6320: } /* End age */
6321:
6322: free_vector(xp,1,npar);
6323: free_matrix(doldm,1,nlstate,1,npar);
6324: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6325:
6326: }
6327:
6328: /************ Variance of one-step probabilities ******************/
6329: 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 6330: {
6331: int i, j=0, k1, l1, tj;
6332: int k2, l2, j1, z1;
6333: int k=0, l;
6334: int first=1, first1, first2;
6335: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6336: double **dnewm,**doldm;
6337: double *xp;
6338: double *gp, *gm;
6339: double **gradg, **trgradg;
6340: double **mu;
6341: double age, cov[NCOVMAX+1];
6342: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6343: int theta;
6344: char fileresprob[FILENAMELENGTH];
6345: char fileresprobcov[FILENAMELENGTH];
6346: char fileresprobcor[FILENAMELENGTH];
6347: double ***varpij;
6348:
6349: strcpy(fileresprob,"PROB_");
6350: strcat(fileresprob,fileres);
6351: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6352: printf("Problem with resultfile: %s\n", fileresprob);
6353: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6354: }
6355: strcpy(fileresprobcov,"PROBCOV_");
6356: strcat(fileresprobcov,fileresu);
6357: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6358: printf("Problem with resultfile: %s\n", fileresprobcov);
6359: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6360: }
6361: strcpy(fileresprobcor,"PROBCOR_");
6362: strcat(fileresprobcor,fileresu);
6363: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6364: printf("Problem with resultfile: %s\n", fileresprobcor);
6365: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6366: }
6367: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6368: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6369: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6370: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6371: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6372: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6373: pstamp(ficresprob);
6374: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6375: fprintf(ficresprob,"# Age");
6376: pstamp(ficresprobcov);
6377: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6378: fprintf(ficresprobcov,"# Age");
6379: pstamp(ficresprobcor);
6380: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6381: fprintf(ficresprobcor,"# Age");
1.126 brouard 6382:
6383:
1.222 brouard 6384: for(i=1; i<=nlstate;i++)
6385: for(j=1; j<=(nlstate+ndeath);j++){
6386: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6387: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6388: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6389: }
6390: /* fprintf(ficresprob,"\n");
6391: fprintf(ficresprobcov,"\n");
6392: fprintf(ficresprobcor,"\n");
6393: */
6394: xp=vector(1,npar);
6395: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6396: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6397: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6398: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6399: first=1;
6400: fprintf(ficgp,"\n# Routine varprob");
6401: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6402: fprintf(fichtm,"\n");
6403:
1.266 brouard 6404: 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 6405: 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);
6406: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6407: and drawn. It helps understanding how is the covariance between two incidences.\
6408: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6409: 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 6410: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6411: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6412: standard deviations wide on each axis. <br>\
6413: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6414: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6415: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6416:
1.222 brouard 6417: cov[1]=1;
6418: /* tj=cptcoveff; */
1.225 brouard 6419: tj = (int) pow(2,cptcoveff);
1.222 brouard 6420: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6421: j1=0;
1.224 brouard 6422: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6423: if (cptcovn>0) {
6424: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6425: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6426: fprintf(ficresprob, "**********\n#\n");
6427: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6428: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6429: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6430:
1.222 brouard 6431: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6432: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6433: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6434:
6435:
1.222 brouard 6436: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6437: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6438: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6439:
1.222 brouard 6440: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6441: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6442: fprintf(ficresprobcor, "**********\n#");
6443: if(invalidvarcomb[j1]){
6444: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6445: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6446: continue;
6447: }
6448: }
6449: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6450: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6451: gp=vector(1,(nlstate)*(nlstate+ndeath));
6452: gm=vector(1,(nlstate)*(nlstate+ndeath));
6453: for (age=bage; age<=fage; age ++){
6454: cov[2]=age;
6455: if(nagesqr==1)
6456: cov[3]= age*age;
6457: for (k=1; k<=cptcovn;k++) {
6458: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6459: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6460: * 1 1 1 1 1
6461: * 2 2 1 1 1
6462: * 3 1 2 1 1
6463: */
6464: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6465: }
6466: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6467: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6468: for (k=1; k<=cptcovprod;k++)
6469: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6470:
6471:
1.222 brouard 6472: for(theta=1; theta <=npar; theta++){
6473: for(i=1; i<=npar; i++)
6474: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6475:
1.222 brouard 6476: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6477:
1.222 brouard 6478: k=0;
6479: for(i=1; i<= (nlstate); i++){
6480: for(j=1; j<=(nlstate+ndeath);j++){
6481: k=k+1;
6482: gp[k]=pmmij[i][j];
6483: }
6484: }
1.220 brouard 6485:
1.222 brouard 6486: for(i=1; i<=npar; i++)
6487: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6488:
1.222 brouard 6489: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6490: k=0;
6491: for(i=1; i<=(nlstate); i++){
6492: for(j=1; j<=(nlstate+ndeath);j++){
6493: k=k+1;
6494: gm[k]=pmmij[i][j];
6495: }
6496: }
1.220 brouard 6497:
1.222 brouard 6498: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6499: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6500: }
1.126 brouard 6501:
1.222 brouard 6502: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6503: for(theta=1; theta <=npar; theta++)
6504: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6505:
1.222 brouard 6506: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6507: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6508:
1.222 brouard 6509: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6510:
1.222 brouard 6511: k=0;
6512: for(i=1; i<=(nlstate); i++){
6513: for(j=1; j<=(nlstate+ndeath);j++){
6514: k=k+1;
6515: mu[k][(int) age]=pmmij[i][j];
6516: }
6517: }
6518: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6519: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6520: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6521:
1.222 brouard 6522: /*printf("\n%d ",(int)age);
6523: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6524: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6525: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6526: }*/
1.220 brouard 6527:
1.222 brouard 6528: fprintf(ficresprob,"\n%d ",(int)age);
6529: fprintf(ficresprobcov,"\n%d ",(int)age);
6530: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6531:
1.222 brouard 6532: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6533: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6534: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6535: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6536: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6537: }
6538: i=0;
6539: for (k=1; k<=(nlstate);k++){
6540: for (l=1; l<=(nlstate+ndeath);l++){
6541: i++;
6542: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6543: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6544: for (j=1; j<=i;j++){
6545: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6546: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6547: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6548: }
6549: }
6550: }/* end of loop for state */
6551: } /* end of loop for age */
6552: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6553: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6554: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6555: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6556:
6557: /* Confidence intervalle of pij */
6558: /*
6559: fprintf(ficgp,"\nunset parametric;unset label");
6560: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6561: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6562: 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);
6563: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6564: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6565: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6566: */
6567:
6568: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6569: first1=1;first2=2;
6570: for (k2=1; k2<=(nlstate);k2++){
6571: for (l2=1; l2<=(nlstate+ndeath);l2++){
6572: if(l2==k2) continue;
6573: j=(k2-1)*(nlstate+ndeath)+l2;
6574: for (k1=1; k1<=(nlstate);k1++){
6575: for (l1=1; l1<=(nlstate+ndeath);l1++){
6576: if(l1==k1) continue;
6577: i=(k1-1)*(nlstate+ndeath)+l1;
6578: if(i<=j) continue;
6579: for (age=bage; age<=fage; age ++){
6580: if ((int)age %5==0){
6581: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6582: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6583: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6584: mu1=mu[i][(int) age]/stepm*YEARM ;
6585: mu2=mu[j][(int) age]/stepm*YEARM;
6586: c12=cv12/sqrt(v1*v2);
6587: /* Computing eigen value of matrix of covariance */
6588: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6589: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6590: if ((lc2 <0) || (lc1 <0) ){
6591: if(first2==1){
6592: first1=0;
6593: 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);
6594: }
6595: 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);
6596: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6597: /* lc2=fabs(lc2); */
6598: }
1.220 brouard 6599:
1.222 brouard 6600: /* Eigen vectors */
6601: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6602: /*v21=sqrt(1.-v11*v11); *//* error */
6603: v21=(lc1-v1)/cv12*v11;
6604: v12=-v21;
6605: v22=v11;
6606: tnalp=v21/v11;
6607: if(first1==1){
6608: first1=0;
6609: 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);
6610: }
6611: 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);
6612: /*printf(fignu*/
6613: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6614: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6615: if(first==1){
6616: first=0;
6617: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6618: fprintf(ficgp,"\nset parametric;unset label");
6619: 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);
6620: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6621: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6622: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6623: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6624: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6625: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6626: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6627: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6628: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6629: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6630: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6631: 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 6632: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6633: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6634: }else{
6635: first=0;
6636: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6637: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6638: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6639: 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 6640: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6641: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6642: }/* if first */
6643: } /* age mod 5 */
6644: } /* end loop age */
6645: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6646: first=1;
6647: } /*l12 */
6648: } /* k12 */
6649: } /*l1 */
6650: }/* k1 */
6651: } /* loop on combination of covariates j1 */
6652: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6653: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6654: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6655: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6656: free_vector(xp,1,npar);
6657: fclose(ficresprob);
6658: fclose(ficresprobcov);
6659: fclose(ficresprobcor);
6660: fflush(ficgp);
6661: fflush(fichtmcov);
6662: }
1.126 brouard 6663:
6664:
6665: /******************* Printing html file ***********/
1.201 brouard 6666: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6667: int lastpass, int stepm, int weightopt, char model[],\
6668: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6669: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.273 brouard 6670: double jprev1, double mprev1,double anprev1, double dateprev1, double dateproj1, double dateback1, \
6671: double jprev2, double mprev2,double anprev2, double dateprev2, double dateproj2, double dateback2){
1.237 brouard 6672: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6673:
6674: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6675: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6676: </ul>");
1.237 brouard 6677: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6678: </ul>", model);
1.214 brouard 6679: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6680: 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",
6681: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6682: 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 6683: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6684: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6685: fprintf(fichtm,"\
6686: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6687: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6688: fprintf(fichtm,"\
1.217 brouard 6689: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6690: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6691: fprintf(fichtm,"\
1.126 brouard 6692: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6693: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6694: fprintf(fichtm,"\
1.217 brouard 6695: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6696: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6697: fprintf(fichtm,"\
1.211 brouard 6698: - (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 6699: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6700: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6701: if(prevfcast==1){
6702: fprintf(fichtm,"\
6703: - Prevalence projections by age and states: \
1.201 brouard 6704: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6705: }
1.126 brouard 6706:
6707:
1.225 brouard 6708: m=pow(2,cptcoveff);
1.222 brouard 6709: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6710:
1.264 brouard 6711: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6712:
6713: jj1=0;
6714:
6715: fprintf(fichtm," \n<ul>");
6716: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6717: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6718: if(m != 1 && TKresult[nres]!= k1)
6719: continue;
6720: jj1++;
6721: if (cptcovn > 0) {
6722: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6723: for (cpt=1; cpt<=cptcoveff;cpt++){
6724: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6725: }
6726: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6727: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6728: }
6729: fprintf(fichtm,"\">");
6730:
6731: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6732: fprintf(fichtm,"************ Results for covariates");
6733: for (cpt=1; cpt<=cptcoveff;cpt++){
6734: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6735: }
6736: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6737: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6738: }
6739: if(invalidvarcomb[k1]){
6740: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6741: continue;
6742: }
6743: fprintf(fichtm,"</a></li>");
6744: } /* cptcovn >0 */
6745: }
6746: fprintf(fichtm," \n</ul>");
6747:
1.222 brouard 6748: jj1=0;
1.237 brouard 6749:
6750: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6751: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6752: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6753: continue;
1.220 brouard 6754:
1.222 brouard 6755: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6756: jj1++;
6757: if (cptcovn > 0) {
1.264 brouard 6758: fprintf(fichtm,"\n<p><a name=\"rescov");
6759: for (cpt=1; cpt<=cptcoveff;cpt++){
6760: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6761: }
6762: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6763: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6764: }
6765: fprintf(fichtm,"\"</a>");
6766:
1.222 brouard 6767: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6768: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6769: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6770: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6771: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6772: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6773: }
1.237 brouard 6774: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6775: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6776: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6777: }
6778:
1.230 brouard 6779: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6780: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6781: if(invalidvarcomb[k1]){
6782: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6783: printf("\nCombination (%d) ignored because no cases \n",k1);
6784: continue;
6785: }
6786: }
6787: /* aij, bij */
1.259 brouard 6788: 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 6789: <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 6790: /* Pij */
1.241 brouard 6791: 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> \
6792: <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 6793: /* Quasi-incidences */
6794: 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 6795: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6796: 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 6797: 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> \
6798: <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 6799: /* Survival functions (period) in state j */
6800: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6801: 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> \
6802: <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 6803: }
6804: /* State specific survival functions (period) */
6805: for(cpt=1; cpt<=nlstate;cpt++){
6806: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6807: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6808: <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 6809: }
6810: /* Period (stable) prevalence in each health state */
6811: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6812: 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> \
6813: <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 6814: }
6815: if(backcast==1){
6816: /* Period (stable) back prevalence in each health state */
6817: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6818: 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 6819: <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 6820: }
1.217 brouard 6821: }
1.222 brouard 6822: if(prevfcast==1){
6823: /* Projection of prevalence up to period (stable) prevalence in each health state */
6824: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6825: 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> \
6826: <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 6827: }
6828: }
1.268 brouard 6829: if(backcast==1){
6830: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
6831: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 6832: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
6833: 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 \
6834: 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) \
6835: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6836: <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 6837: }
6838: }
1.220 brouard 6839:
1.222 brouard 6840: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6841: 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> \
6842: <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 6843: }
6844: /* } /\* end i1 *\/ */
6845: }/* End k1 */
6846: fprintf(fichtm,"</ul>");
1.126 brouard 6847:
1.222 brouard 6848: fprintf(fichtm,"\
1.126 brouard 6849: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6850: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6851: - 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 6852: But because parameters are usually highly correlated (a higher incidence of disability \
6853: and a higher incidence of recovery can give very close observed transition) it might \
6854: be very useful to look not only at linear confidence intervals estimated from the \
6855: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6856: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6857: covariance matrix of the one-step probabilities. \
6858: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6859:
1.222 brouard 6860: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6861: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6862: fprintf(fichtm,"\
1.126 brouard 6863: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6864: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6865:
1.222 brouard 6866: fprintf(fichtm,"\
1.126 brouard 6867: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6868: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6869: fprintf(fichtm,"\
1.126 brouard 6870: - 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): \
6871: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6872: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6873: fprintf(fichtm,"\
1.126 brouard 6874: - (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): \
6875: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6876: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6877: fprintf(fichtm,"\
1.128 brouard 6878: - 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 6879: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6880: fprintf(fichtm,"\
1.128 brouard 6881: - 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 6882: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6883: fprintf(fichtm,"\
1.126 brouard 6884: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6885: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6886:
6887: /* if(popforecast==1) fprintf(fichtm,"\n */
6888: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6889: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6890: /* <br>",fileres,fileres,fileres,fileres); */
6891: /* else */
6892: /* 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 6893: fflush(fichtm);
6894: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6895:
1.225 brouard 6896: m=pow(2,cptcoveff);
1.222 brouard 6897: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6898:
1.222 brouard 6899: jj1=0;
1.237 brouard 6900:
1.241 brouard 6901: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6902: for(k1=1; k1<=m;k1++){
1.253 brouard 6903: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6904: continue;
1.222 brouard 6905: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6906: jj1++;
1.126 brouard 6907: if (cptcovn > 0) {
6908: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6909: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6910: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6911: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6912: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6913: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6914: }
6915:
1.126 brouard 6916: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6917:
1.222 brouard 6918: if(invalidvarcomb[k1]){
6919: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6920: continue;
6921: }
1.126 brouard 6922: }
6923: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6924: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6925: 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 6926: <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 6927: }
6928: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6929: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6930: true period expectancies (those weighted with period prevalences are also\
6931: drawn in addition to the population based expectancies computed using\
1.241 brouard 6932: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6933: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6934: /* } /\* end i1 *\/ */
6935: }/* End k1 */
1.241 brouard 6936: }/* End nres */
1.222 brouard 6937: fprintf(fichtm,"</ul>");
6938: fflush(fichtm);
1.126 brouard 6939: }
6940:
6941: /******************* Gnuplot file **************/
1.270 brouard 6942: 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 6943:
6944: char dirfileres[132],optfileres[132];
1.264 brouard 6945: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6946: 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 6947: int lv=0, vlv=0, kl=0;
1.130 brouard 6948: int ng=0;
1.201 brouard 6949: int vpopbased;
1.223 brouard 6950: int ioffset; /* variable offset for columns */
1.270 brouard 6951: int iyearc=1; /* variable column for year of projection */
6952: int iagec=1; /* variable column for age of projection */
1.235 brouard 6953: int nres=0; /* Index of resultline */
1.266 brouard 6954: int istart=1; /* For starting graphs in projections */
1.219 brouard 6955:
1.126 brouard 6956: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6957: /* printf("Problem with file %s",optionfilegnuplot); */
6958: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6959: /* } */
6960:
6961: /*#ifdef windows */
6962: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6963: /*#endif */
1.225 brouard 6964: m=pow(2,cptcoveff);
1.126 brouard 6965:
1.274 ! brouard 6966: /* diagram of the model */
! 6967: fprintf(ficgp,"\n#Diagram of the model \n");
! 6968: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
! 6969: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
! 6970: 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);
! 6971:
! 6972: 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);
! 6973: fprintf(ficgp,"\n#show arrow\nunset label\n");
! 6974: 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);
! 6975: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
! 6976: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
! 6977: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
! 6978: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
! 6979:
1.202 brouard 6980: /* Contribution to likelihood */
6981: /* Plot the probability implied in the likelihood */
1.223 brouard 6982: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6983: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6984: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6985: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6986: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6987: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6988: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6989: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6990: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6991: 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));
6992: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6993: 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));
6994: for (i=1; i<= nlstate ; i ++) {
6995: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6996: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6997: 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);
6998: for (j=2; j<= nlstate+ndeath ; j ++) {
6999: 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);
7000: }
7001: fprintf(ficgp,";\nset out; unset ylabel;\n");
7002: }
7003: /* 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 */
7004: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7005: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7006: fprintf(ficgp,"\nset out;unset log\n");
7007: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7008:
1.126 brouard 7009: strcpy(dirfileres,optionfilefiname);
7010: strcpy(optfileres,"vpl");
1.223 brouard 7011: /* 1eme*/
1.238 brouard 7012: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7013: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7014: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7015: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7016: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7017: continue;
7018: /* We are interested in selected combination by the resultline */
1.246 brouard 7019: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 7020: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7021: strcpy(gplotlabel,"(");
1.238 brouard 7022: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7023: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7024: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7025: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7026: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7027: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7028: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7029: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7030: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7031: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7032: }
7033: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7034: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7035: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7036: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7037: }
7038: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7039: /* printf("\n#\n"); */
1.238 brouard 7040: fprintf(ficgp,"\n#\n");
7041: if(invalidvarcomb[k1]){
1.260 brouard 7042: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7043: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7044: continue;
7045: }
1.235 brouard 7046:
1.241 brouard 7047: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7048: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.264 brouard 7049: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7050: 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);
7051: /* 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); */
7052: /* k1-1 error should be nres-1*/
1.238 brouard 7053: for (i=1; i<= nlstate ; i ++) {
7054: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7055: else fprintf(ficgp," %%*lf (%%*lf)");
7056: }
1.260 brouard 7057: 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 7058: for (i=1; i<= nlstate ; i ++) {
7059: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7060: else fprintf(ficgp," %%*lf (%%*lf)");
7061: }
1.260 brouard 7062: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7063: for (i=1; i<= nlstate ; i ++) {
7064: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7065: else fprintf(ficgp," %%*lf (%%*lf)");
7066: }
1.265 brouard 7067: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7068:
7069: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7070: if(cptcoveff ==0){
1.271 brouard 7071: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7072: }else{
7073: kl=0;
7074: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7075: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7076: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7077: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7078: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7079: vlv= nbcode[Tvaraff[k]][lv];
7080: kl++;
7081: /* 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 *\/ */
7082: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7083: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7084: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7085: if(k==cptcoveff){
7086: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7087: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7088: }else{
7089: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7090: kl++;
7091: }
7092: } /* end covariate */
7093: } /* end if no covariate */
7094:
1.238 brouard 7095: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
7096: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7097: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7098: if(cptcoveff ==0){
1.245 brouard 7099: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7100: }else{
7101: kl=0;
7102: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7103: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7104: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7105: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7106: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7107: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7108: kl++;
1.238 brouard 7109: /* 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 *\/ */
7110: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7111: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7112: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7113: if(k==cptcoveff){
1.245 brouard 7114: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7115: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7116: }else{
7117: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7118: kl++;
7119: }
7120: } /* end covariate */
7121: } /* end if no covariate */
1.268 brouard 7122: if(backcast == 1){
7123: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7124: /* k1-1 error should be nres-1*/
7125: for (i=1; i<= nlstate ; i ++) {
7126: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7127: else fprintf(ficgp," %%*lf (%%*lf)");
7128: }
1.271 brouard 7129: 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 7130: for (i=1; i<= nlstate ; i ++) {
7131: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7132: else fprintf(ficgp," %%*lf (%%*lf)");
7133: }
1.272 brouard 7134: fprintf(ficgp,"\" t\"95%% CI\" w l lt 5,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 7135: for (i=1; i<= nlstate ; i ++) {
7136: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7137: else fprintf(ficgp," %%*lf (%%*lf)");
7138: }
1.274 ! brouard 7139: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7140: } /* end if backprojcast */
1.238 brouard 7141: } /* end if backcast */
1.264 brouard 7142: fprintf(ficgp,"\nset out ;unset label;\n");
1.238 brouard 7143: } /* nres */
1.201 brouard 7144: } /* k1 */
7145: } /* cpt */
1.235 brouard 7146:
7147:
1.126 brouard 7148: /*2 eme*/
1.238 brouard 7149: for (k1=1; k1<= m ; k1 ++){
7150: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7151: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7152: continue;
7153: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7154: strcpy(gplotlabel,"(");
1.238 brouard 7155: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7156: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7157: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7158: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7159: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7160: vlv= nbcode[Tvaraff[k]][lv];
7161: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7162: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7163: }
1.237 brouard 7164: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7165: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7166: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7167: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7168: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7169: }
1.264 brouard 7170: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7171: fprintf(ficgp,"\n#\n");
1.223 brouard 7172: if(invalidvarcomb[k1]){
7173: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7174: continue;
7175: }
1.219 brouard 7176:
1.241 brouard 7177: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7178: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7179: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7180: if(vpopbased==0){
1.238 brouard 7181: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7182: }else
1.238 brouard 7183: fprintf(ficgp,"\nreplot ");
7184: for (i=1; i<= nlstate+1 ; i ++) {
7185: k=2*i;
1.261 brouard 7186: 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 7187: for (j=1; j<= nlstate+1 ; j ++) {
7188: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7189: else fprintf(ficgp," %%*lf (%%*lf)");
7190: }
7191: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7192: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7193: 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 7194: for (j=1; j<= nlstate+1 ; j ++) {
7195: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7196: else fprintf(ficgp," %%*lf (%%*lf)");
7197: }
7198: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7199: 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 7200: for (j=1; j<= nlstate+1 ; j ++) {
7201: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7202: else fprintf(ficgp," %%*lf (%%*lf)");
7203: }
7204: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7205: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7206: } /* state */
7207: } /* vpopbased */
1.264 brouard 7208: 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 7209: } /* end nres */
7210: } /* k1 end 2 eme*/
7211:
7212:
7213: /*3eme*/
7214: for (k1=1; k1<= m ; k1 ++){
7215: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7216: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7217: continue;
7218:
7219: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7220: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7221: strcpy(gplotlabel,"(");
1.238 brouard 7222: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7223: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7224: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7225: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7226: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7227: vlv= nbcode[Tvaraff[k]][lv];
7228: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7229: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7230: }
7231: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7232: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7233: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7234: }
1.264 brouard 7235: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7236: fprintf(ficgp,"\n#\n");
7237: if(invalidvarcomb[k1]){
7238: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7239: continue;
7240: }
7241:
7242: /* k=2+nlstate*(2*cpt-2); */
7243: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7244: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7245: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7246: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7247: 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 7248: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7249: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7250: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7251: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7252: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7253: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7254:
1.238 brouard 7255: */
7256: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7257: 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 7258: /* 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 7259:
1.238 brouard 7260: }
1.261 brouard 7261: 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 7262: }
1.264 brouard 7263: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7264: } /* end nres */
7265: } /* end kl 3eme */
1.126 brouard 7266:
1.223 brouard 7267: /* 4eme */
1.201 brouard 7268: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7269: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7270: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7271: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7272: continue;
1.238 brouard 7273: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7274: strcpy(gplotlabel,"(");
1.238 brouard 7275: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7276: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7277: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7278: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7279: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7280: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7281: vlv= nbcode[Tvaraff[k]][lv];
7282: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7283: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7284: }
7285: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7286: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7287: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7288: }
1.264 brouard 7289: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7290: fprintf(ficgp,"\n#\n");
7291: if(invalidvarcomb[k1]){
7292: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7293: continue;
1.223 brouard 7294: }
1.238 brouard 7295:
1.241 brouard 7296: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7297: 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 7298: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7299: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7300: k=3;
7301: for (i=1; i<= nlstate ; i ++){
7302: if(i==1){
7303: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7304: }else{
7305: fprintf(ficgp,", '' ");
7306: }
7307: l=(nlstate+ndeath)*(i-1)+1;
7308: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7309: for (j=2; j<= nlstate+ndeath ; j ++)
7310: fprintf(ficgp,"+$%d",k+l+j-1);
7311: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7312: } /* nlstate */
1.264 brouard 7313: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7314: } /* end cpt state*/
7315: } /* end nres */
7316: } /* end covariate k1 */
7317:
1.220 brouard 7318: /* 5eme */
1.201 brouard 7319: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7320: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7321: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7322: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7323: continue;
1.238 brouard 7324: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7325: strcpy(gplotlabel,"(");
1.238 brouard 7326: 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);
7327: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7328: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7329: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7330: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7331: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7332: vlv= nbcode[Tvaraff[k]][lv];
7333: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7334: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7335: }
7336: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7337: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7338: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7339: }
1.264 brouard 7340: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7341: fprintf(ficgp,"\n#\n");
7342: if(invalidvarcomb[k1]){
7343: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7344: continue;
7345: }
1.227 brouard 7346:
1.241 brouard 7347: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7348: 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 7349: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7350: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7351: k=3;
7352: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7353: if(j==1)
7354: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7355: else
7356: fprintf(ficgp,", '' ");
7357: l=(nlstate+ndeath)*(cpt-1) +j;
7358: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7359: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7360: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7361: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7362: } /* nlstate */
7363: fprintf(ficgp,", '' ");
7364: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7365: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7366: l=(nlstate+ndeath)*(cpt-1) +j;
7367: if(j < nlstate)
7368: fprintf(ficgp,"$%d +",k+l);
7369: else
7370: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7371: }
1.264 brouard 7372: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7373: } /* end cpt state*/
7374: } /* end covariate */
7375: } /* end nres */
1.227 brouard 7376:
1.220 brouard 7377: /* 6eme */
1.202 brouard 7378: /* CV preval stable (period) for each covariate */
1.237 brouard 7379: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7380: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7381: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7382: continue;
1.255 brouard 7383: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7384: strcpy(gplotlabel,"(");
1.211 brouard 7385: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7386: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7387: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7388: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7389: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7390: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7391: vlv= nbcode[Tvaraff[k]][lv];
7392: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7393: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7394: }
1.237 brouard 7395: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7396: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7397: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7398: }
1.264 brouard 7399: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7400: fprintf(ficgp,"\n#\n");
1.223 brouard 7401: if(invalidvarcomb[k1]){
1.227 brouard 7402: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7403: continue;
1.223 brouard 7404: }
1.227 brouard 7405:
1.241 brouard 7406: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7407: 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 7408: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7409: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7410: k=3; /* Offset */
1.255 brouard 7411: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7412: if(i==1)
7413: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7414: else
7415: fprintf(ficgp,", '' ");
1.255 brouard 7416: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7417: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7418: for (j=2; j<= nlstate ; j ++)
7419: fprintf(ficgp,"+$%d",k+l+j-1);
7420: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7421: } /* nlstate */
1.264 brouard 7422: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7423: } /* end cpt state*/
7424: } /* end covariate */
1.227 brouard 7425:
7426:
1.220 brouard 7427: /* 7eme */
1.218 brouard 7428: if(backcast == 1){
1.217 brouard 7429: /* CV back preval stable (period) for each covariate */
1.237 brouard 7430: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7431: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7432: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7433: continue;
1.268 brouard 7434: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7435: strcpy(gplotlabel,"(");
7436: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7437: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7438: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7439: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7440: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7441: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7442: vlv= nbcode[Tvaraff[k]][lv];
7443: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7444: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7445: }
1.237 brouard 7446: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7447: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7448: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7449: }
1.264 brouard 7450: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7451: fprintf(ficgp,"\n#\n");
7452: if(invalidvarcomb[k1]){
7453: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7454: continue;
7455: }
7456:
1.241 brouard 7457: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7458: 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 7459: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7460: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7461: k=3; /* Offset */
1.268 brouard 7462: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7463: if(i==1)
7464: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7465: else
7466: fprintf(ficgp,", '' ");
7467: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7468: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7469: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7470: /* 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 7471: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7472: /* for (j=2; j<= nlstate ; j ++) */
7473: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7474: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7475: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7476: } /* nlstate */
1.264 brouard 7477: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7478: } /* end cpt state*/
7479: } /* end covariate */
7480: } /* End if backcast */
7481:
1.223 brouard 7482: /* 8eme */
1.218 brouard 7483: if(prevfcast==1){
7484: /* Projection from cross-sectional to stable (period) for each covariate */
7485:
1.237 brouard 7486: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7487: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7488: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7489: continue;
1.211 brouard 7490: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7491: strcpy(gplotlabel,"(");
1.227 brouard 7492: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7493: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7494: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7495: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7496: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7497: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7498: vlv= nbcode[Tvaraff[k]][lv];
7499: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7500: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7501: }
1.237 brouard 7502: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7503: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7504: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7505: }
1.264 brouard 7506: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7507: fprintf(ficgp,"\n#\n");
7508: if(invalidvarcomb[k1]){
7509: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7510: continue;
7511: }
7512:
7513: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7514: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7515: 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 7516: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7517: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7518:
7519: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7520: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7521: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7522: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7523: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7524: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7525: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7526: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7527: if(i==istart){
1.227 brouard 7528: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7529: }else{
7530: fprintf(ficgp,",\\\n '' ");
7531: }
7532: if(cptcoveff ==0){ /* No covariate */
7533: ioffset=2; /* Age is in 2 */
7534: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7535: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7536: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7537: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7538: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7539: if(i==nlstate+1){
1.270 brouard 7540: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7541: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7542: fprintf(ficgp,",\\\n '' ");
7543: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7544: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7545: offyear, \
1.268 brouard 7546: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7547: }else
1.227 brouard 7548: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7549: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7550: }else{ /* more than 2 covariates */
1.270 brouard 7551: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7552: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7553: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7554: iyearc=ioffset-1;
7555: iagec=ioffset;
1.227 brouard 7556: fprintf(ficgp," u %d:(",ioffset);
7557: kl=0;
7558: strcpy(gplotcondition,"(");
7559: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7560: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7561: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7562: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7563: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7564: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7565: kl++;
7566: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7567: kl++;
7568: if(k <cptcoveff && cptcoveff>1)
7569: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7570: }
7571: strcpy(gplotcondition+strlen(gplotcondition),")");
7572: /* 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 *\/ */
7573: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7574: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7575: /* '' 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*/
7576: if(i==nlstate+1){
1.270 brouard 7577: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7578: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7579: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7580: fprintf(ficgp," u %d:(",iagec);
7581: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7582: iyearc, iagec, offyear, \
7583: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7584: /* '' 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 7585: }else{
7586: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7587: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7588: }
7589: } /* end if covariate */
7590: } /* nlstate */
1.264 brouard 7591: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7592: } /* end cpt state*/
7593: } /* end covariate */
7594: } /* End if prevfcast */
1.227 brouard 7595:
1.268 brouard 7596: if(backcast==1){
7597: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7598:
7599: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7600: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7601: if(m != 1 && TKresult[nres]!= k1)
7602: continue;
7603: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7604: strcpy(gplotlabel,"(");
7605: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7606: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7607: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7608: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7609: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7610: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7611: vlv= nbcode[Tvaraff[k]][lv];
7612: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7613: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7614: }
7615: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7616: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7617: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7618: }
7619: strcpy(gplotlabel+strlen(gplotlabel),")");
7620: fprintf(ficgp,"\n#\n");
7621: if(invalidvarcomb[k1]){
7622: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7623: continue;
7624: }
7625:
7626: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7627: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7628: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7629: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7630: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7631:
7632: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7633: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7634: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7635: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7636: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7637: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7638: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7639: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7640: if(i==istart){
7641: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7642: }else{
7643: fprintf(ficgp,",\\\n '' ");
7644: }
7645: if(cptcoveff ==0){ /* No covariate */
7646: ioffset=2; /* Age is in 2 */
7647: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7648: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7649: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7650: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7651: fprintf(ficgp," u %d:(", ioffset);
7652: if(i==nlstate+1){
1.270 brouard 7653: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7654: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7655: fprintf(ficgp,",\\\n '' ");
7656: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7657: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7658: offbyear, \
7659: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7660: }else
7661: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7662: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7663: }else{ /* more than 2 covariates */
1.270 brouard 7664: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7665: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7666: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7667: iyearc=ioffset-1;
7668: iagec=ioffset;
1.268 brouard 7669: fprintf(ficgp," u %d:(",ioffset);
7670: kl=0;
7671: strcpy(gplotcondition,"(");
7672: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7673: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7674: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7675: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7676: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7677: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7678: kl++;
7679: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7680: kl++;
7681: if(k <cptcoveff && cptcoveff>1)
7682: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7683: }
7684: strcpy(gplotcondition+strlen(gplotcondition),")");
7685: /* 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 *\/ */
7686: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7687: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7688: /* '' 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*/
7689: if(i==nlstate+1){
1.270 brouard 7690: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7691: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7692: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7693: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7694: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7695: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7696: iyearc,iagec,offbyear, \
7697: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7698: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7699: }else{
7700: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7701: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7702: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7703: }
7704: } /* end if covariate */
7705: } /* nlstate */
7706: fprintf(ficgp,"\nset out; unset label;\n");
7707: } /* end cpt state*/
7708: } /* end covariate */
7709: } /* End if backcast */
7710:
1.227 brouard 7711:
1.238 brouard 7712: /* 9eme writing MLE parameters */
7713: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7714: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7715: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7716: for(k=1; k <=(nlstate+ndeath); k++){
7717: if (k != i) {
1.227 brouard 7718: fprintf(ficgp,"# current state %d\n",k);
7719: for(j=1; j <=ncovmodel; j++){
7720: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7721: jk++;
7722: }
7723: fprintf(ficgp,"\n");
1.126 brouard 7724: }
7725: }
1.223 brouard 7726: }
1.187 brouard 7727: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7728:
1.145 brouard 7729: /*goto avoid;*/
1.238 brouard 7730: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7731: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7732: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7733: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7734: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7735: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7736: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7737: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7738: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7739: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7740: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7741: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7742: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7743: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7744: fprintf(ficgp,"#\n");
1.223 brouard 7745: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7746: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7747: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7748: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7749: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7750: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7751: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7752: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7753: continue;
1.264 brouard 7754: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7755: strcpy(gplotlabel,"(");
7756: sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);
7757: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7758: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7759: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7760: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7761: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7762: vlv= nbcode[Tvaraff[k]][lv];
7763: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7764: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7765: }
1.237 brouard 7766: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7767: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7768: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7769: }
1.264 brouard 7770: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7771: fprintf(ficgp,"\n#\n");
1.264 brouard 7772: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
7773: fprintf(ficgp,"\nset label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7774: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7775: if (ng==1){
7776: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7777: fprintf(ficgp,"\nunset log y");
7778: }else if (ng==2){
7779: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7780: fprintf(ficgp,"\nset log y");
7781: }else if (ng==3){
7782: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7783: fprintf(ficgp,"\nset log y");
7784: }else
7785: fprintf(ficgp,"\nunset title ");
7786: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7787: i=1;
7788: for(k2=1; k2<=nlstate; k2++) {
7789: k3=i;
7790: for(k=1; k<=(nlstate+ndeath); k++) {
7791: if (k != k2){
7792: switch( ng) {
7793: case 1:
7794: if(nagesqr==0)
7795: fprintf(ficgp," p%d+p%d*x",i,i+1);
7796: else /* nagesqr =1 */
7797: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7798: break;
7799: case 2: /* ng=2 */
7800: if(nagesqr==0)
7801: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7802: else /* nagesqr =1 */
7803: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7804: break;
7805: case 3:
7806: if(nagesqr==0)
7807: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7808: else /* nagesqr =1 */
7809: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7810: break;
7811: }
7812: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7813: ijp=1; /* product no age */
7814: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7815: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7816: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 7817: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7818: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
7819: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
7820: if(DummyV[j]==0){
7821: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7822: }else{ /* quantitative */
7823: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7824: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7825: }
7826: ij++;
1.237 brouard 7827: }
1.268 brouard 7828: }
7829: }else if(cptcovprod >0){
7830: if(j==Tprod[ijp]) { /* */
7831: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7832: if(ijp <=cptcovprod) { /* Product */
7833: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7834: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
7835: /* 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)]); */
7836: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7837: }else{ /* Vn is dummy and Vm is quanti */
7838: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7839: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7840: }
7841: }else{ /* Vn*Vm Vn is quanti */
7842: if(DummyV[Tvard[ijp][2]]==0){
7843: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7844: }else{ /* Both quanti */
7845: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7846: }
1.237 brouard 7847: }
1.268 brouard 7848: ijp++;
1.237 brouard 7849: }
1.268 brouard 7850: } /* end Tprod */
1.237 brouard 7851: } else{ /* simple covariate */
1.264 brouard 7852: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7853: if(Dummy[j]==0){
7854: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7855: }else{ /* quantitative */
7856: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7857: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7858: }
1.237 brouard 7859: } /* end simple */
7860: } /* end j */
1.223 brouard 7861: }else{
7862: i=i-ncovmodel;
7863: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7864: fprintf(ficgp," (1.");
7865: }
1.227 brouard 7866:
1.223 brouard 7867: if(ng != 1){
7868: fprintf(ficgp,")/(1");
1.227 brouard 7869:
1.264 brouard 7870: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7871: if(nagesqr==0)
1.264 brouard 7872: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7873: else /* nagesqr =1 */
1.264 brouard 7874: 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 7875:
1.223 brouard 7876: ij=1;
7877: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 7878: if(cptcovage >0){
7879: if((j-2)==Tage[ij]) { /* Bug valgrind */
7880: if(ij <=cptcovage) { /* Bug valgrind */
7881: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7882: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
7883: ij++;
7884: }
7885: }
7886: }else
7887: 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 7888: }
7889: fprintf(ficgp,")");
7890: }
7891: fprintf(ficgp,")");
7892: if(ng ==2)
7893: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7894: else /* ng= 3 */
7895: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7896: }else{ /* end ng <> 1 */
7897: if( k !=k2) /* logit p11 is hard to draw */
7898: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7899: }
7900: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7901: fprintf(ficgp,",");
7902: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7903: fprintf(ficgp,",");
7904: i=i+ncovmodel;
7905: } /* end k */
7906: } /* end k2 */
1.264 brouard 7907: fprintf(ficgp,"\n set out; unset label;\n");
7908: } /* end k1 */
1.223 brouard 7909: } /* end ng */
7910: /* avoid: */
7911: fflush(ficgp);
1.126 brouard 7912: } /* end gnuplot */
7913:
7914:
7915: /*************** Moving average **************/
1.219 brouard 7916: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7917: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7918:
1.222 brouard 7919: int i, cpt, cptcod;
7920: int modcovmax =1;
7921: int mobilavrange, mob;
7922: int iage=0;
7923:
1.266 brouard 7924: double sum=0., sumr=0.;
1.222 brouard 7925: double age;
1.266 brouard 7926: double *sumnewp, *sumnewm, *sumnewmr;
7927: double *agemingood, *agemaxgood;
7928: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7929:
7930:
1.225 brouard 7931: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7932: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7933:
7934: sumnewp = vector(1,ncovcombmax);
7935: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7936: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7937: agemingood = vector(1,ncovcombmax);
1.266 brouard 7938: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7939: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7940: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7941:
7942: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7943: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7944: sumnewp[cptcod]=0.;
1.266 brouard 7945: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
7946: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7947: }
7948: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7949:
1.266 brouard 7950: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7951: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7952: else mobilavrange=mobilav;
7953: for (age=bage; age<=fage; age++)
7954: for (i=1; i<=nlstate;i++)
7955: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7956: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7957: /* We keep the original values on the extreme ages bage, fage and for
7958: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7959: we use a 5 terms etc. until the borders are no more concerned.
7960: */
7961: for (mob=3;mob <=mobilavrange;mob=mob+2){
7962: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 7963: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7964: sumnewm[cptcod]=0.;
7965: for (i=1; i<=nlstate;i++){
1.222 brouard 7966: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7967: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7968: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7969: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7970: }
7971: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 7972: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7973: } /* end i */
7974: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
7975: } /* end cptcod */
1.222 brouard 7976: }/* end age */
7977: }/* end mob */
1.266 brouard 7978: }else{
7979: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 7980: return -1;
1.266 brouard 7981: }
7982:
7983: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 7984: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7985: if(invalidvarcomb[cptcod]){
7986: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7987: continue;
7988: }
1.219 brouard 7989:
1.266 brouard 7990: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
7991: sumnewm[cptcod]=0.;
7992: sumnewmr[cptcod]=0.;
7993: for (i=1; i<=nlstate;i++){
7994: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7995: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
7996: }
7997: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
7998: agemingoodr[cptcod]=age;
7999: }
8000: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8001: agemingood[cptcod]=age;
8002: }
8003: } /* age */
8004: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8005: sumnewm[cptcod]=0.;
1.266 brouard 8006: sumnewmr[cptcod]=0.;
1.222 brouard 8007: for (i=1; i<=nlstate;i++){
8008: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8009: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8010: }
8011: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8012: agemaxgoodr[cptcod]=age;
1.222 brouard 8013: }
8014: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8015: agemaxgood[cptcod]=age;
8016: }
8017: } /* age */
8018: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8019: /* but they will change */
8020: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8021: sumnewm[cptcod]=0.;
8022: sumnewmr[cptcod]=0.;
8023: for (i=1; i<=nlstate;i++){
8024: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8025: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8026: }
8027: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8028: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8029: agemaxgoodr[cptcod]=age; /* age min */
8030: for (i=1; i<=nlstate;i++)
8031: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8032: }else{ /* bad we change the value with the values of good ages */
8033: for (i=1; i<=nlstate;i++){
8034: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8035: } /* i */
8036: } /* end bad */
8037: }else{
8038: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8039: agemaxgood[cptcod]=age;
8040: }else{ /* bad we change the value with the values of good ages */
8041: for (i=1; i<=nlstate;i++){
8042: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8043: } /* i */
8044: } /* end bad */
8045: }/* end else */
8046: sum=0.;sumr=0.;
8047: for (i=1; i<=nlstate;i++){
8048: sum+=mobaverage[(int)age][i][cptcod];
8049: sumr+=probs[(int)age][i][cptcod];
8050: }
8051: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8052: 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 8053: } /* end bad */
8054: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8055: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8056: 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 8057: } /* end bad */
8058: }/* age */
1.266 brouard 8059:
8060: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8061: sumnewm[cptcod]=0.;
1.266 brouard 8062: sumnewmr[cptcod]=0.;
1.222 brouard 8063: for (i=1; i<=nlstate;i++){
8064: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8065: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8066: }
8067: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8068: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8069: agemingoodr[cptcod]=age;
8070: for (i=1; i<=nlstate;i++)
8071: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8072: }else{ /* bad we change the value with the values of good ages */
8073: for (i=1; i<=nlstate;i++){
8074: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8075: } /* i */
8076: } /* end bad */
8077: }else{
8078: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8079: agemingood[cptcod]=age;
8080: }else{ /* bad */
8081: for (i=1; i<=nlstate;i++){
8082: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8083: } /* i */
8084: } /* end bad */
8085: }/* end else */
8086: sum=0.;sumr=0.;
8087: for (i=1; i<=nlstate;i++){
8088: sum+=mobaverage[(int)age][i][cptcod];
8089: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8090: }
1.266 brouard 8091: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8092: 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 8093: } /* end bad */
8094: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8095: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8096: 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 8097: } /* end bad */
8098: }/* age */
1.266 brouard 8099:
1.222 brouard 8100:
8101: for (age=bage; age<=fage; age++){
1.235 brouard 8102: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8103: sumnewp[cptcod]=0.;
8104: sumnewm[cptcod]=0.;
8105: for (i=1; i<=nlstate;i++){
8106: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8107: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8108: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8109: }
8110: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8111: }
8112: /* printf("\n"); */
8113: /* } */
1.266 brouard 8114:
1.222 brouard 8115: /* brutal averaging */
1.266 brouard 8116: /* for (i=1; i<=nlstate;i++){ */
8117: /* for (age=1; age<=bage; age++){ */
8118: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8119: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8120: /* } */
8121: /* for (age=fage; age<=AGESUP; age++){ */
8122: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8123: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8124: /* } */
8125: /* } /\* end i status *\/ */
8126: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8127: /* for (age=1; age<=AGESUP; age++){ */
8128: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8129: /* mobaverage[(int)age][i][cptcod]=0.; */
8130: /* } */
8131: /* } */
1.222 brouard 8132: }/* end cptcod */
1.266 brouard 8133: free_vector(agemaxgoodr,1, ncovcombmax);
8134: free_vector(agemaxgood,1, ncovcombmax);
8135: free_vector(agemingood,1, ncovcombmax);
8136: free_vector(agemingoodr,1, ncovcombmax);
8137: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8138: free_vector(sumnewm,1, ncovcombmax);
8139: free_vector(sumnewp,1, ncovcombmax);
8140: return 0;
8141: }/* End movingaverage */
1.218 brouard 8142:
1.126 brouard 8143:
8144: /************** Forecasting ******************/
1.269 brouard 8145: 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 8146: /* proj1, year, month, day of starting projection
8147: agemin, agemax range of age
8148: dateprev1 dateprev2 range of dates during which prevalence is computed
8149: anproj2 year of en of projection (same day and month as proj1).
8150: */
1.267 brouard 8151: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8152: double agec; /* generic age */
8153: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
8154: double *popeffectif,*popcount;
8155: double ***p3mat;
1.218 brouard 8156: /* double ***mobaverage; */
1.126 brouard 8157: char fileresf[FILENAMELENGTH];
8158:
8159: agelim=AGESUP;
1.211 brouard 8160: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8161: in each health status at the date of interview (if between dateprev1 and dateprev2).
8162: We still use firstpass and lastpass as another selection.
8163: */
1.214 brouard 8164: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8165: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8166:
1.201 brouard 8167: strcpy(fileresf,"F_");
8168: strcat(fileresf,fileresu);
1.126 brouard 8169: if((ficresf=fopen(fileresf,"w"))==NULL) {
8170: printf("Problem with forecast resultfile: %s\n", fileresf);
8171: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8172: }
1.235 brouard 8173: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8174: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8175:
1.225 brouard 8176: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8177:
8178:
8179: stepsize=(int) (stepm+YEARM-1)/YEARM;
8180: if (stepm<=12) stepsize=1;
8181: if(estepm < stepm){
8182: printf ("Problem %d lower than %d\n",estepm, stepm);
8183: }
1.270 brouard 8184: else{
8185: hstepm=estepm;
8186: }
8187: if(estepm > stepm){ /* Yes every two year */
8188: stepsize=2;
8189: }
1.126 brouard 8190:
8191: hstepm=hstepm/stepm;
8192: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8193: fractional in yp1 */
8194: anprojmean=yp;
8195: yp2=modf((yp1*12),&yp);
8196: mprojmean=yp;
8197: yp1=modf((yp2*30.5),&yp);
8198: jprojmean=yp;
8199: if(jprojmean==0) jprojmean=1;
8200: if(mprojmean==0) jprojmean=1;
8201:
1.227 brouard 8202: i1=pow(2,cptcoveff);
1.126 brouard 8203: if (cptcovn < 1){i1=1;}
8204:
8205: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
8206:
8207: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8208:
1.126 brouard 8209: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8210: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8211: for(k=1; k<=i1;k++){
1.253 brouard 8212: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8213: continue;
1.227 brouard 8214: if(invalidvarcomb[k]){
8215: printf("\nCombination (%d) projection ignored because no cases \n",k);
8216: continue;
8217: }
8218: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8219: for(j=1;j<=cptcoveff;j++) {
8220: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8221: }
1.235 brouard 8222: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8223: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8224: }
1.227 brouard 8225: fprintf(ficresf," yearproj age");
8226: for(j=1; j<=nlstate+ndeath;j++){
8227: for(i=1; i<=nlstate;i++)
8228: fprintf(ficresf," p%d%d",i,j);
8229: fprintf(ficresf," wp.%d",j);
8230: }
8231: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
8232: fprintf(ficresf,"\n");
8233: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
1.270 brouard 8234: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8235: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8236: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8237: nhstepm = nhstepm/hstepm;
8238: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8239: oldm=oldms;savm=savms;
1.268 brouard 8240: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8241: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8242: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8243: for (h=0; h<=nhstepm; h++){
8244: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8245: break;
8246: }
8247: }
8248: fprintf(ficresf,"\n");
8249: for(j=1;j<=cptcoveff;j++)
8250: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8251: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
8252:
8253: for(j=1; j<=nlstate+ndeath;j++) {
8254: ppij=0.;
8255: for(i=1; i<=nlstate;i++) {
8256: /* if (mobilav>=1) */
1.269 brouard 8257: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
1.268 brouard 8258: /* else { */ /* even if mobilav==-1 we use mobaverage */
8259: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8260: /* } */
8261: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8262: } /* end i */
8263: fprintf(ficresf," %.3f", ppij);
8264: }/* end j */
1.227 brouard 8265: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8266: } /* end agec */
1.266 brouard 8267: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8268: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8269: } /* end yearp */
8270: } /* end k */
1.219 brouard 8271:
1.126 brouard 8272: fclose(ficresf);
1.215 brouard 8273: printf("End of Computing forecasting \n");
8274: fprintf(ficlog,"End of Computing forecasting\n");
8275:
1.126 brouard 8276: }
8277:
1.269 brouard 8278: /************** Back Forecasting ******************/
8279: 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 8280: /* back1, year, month, day of starting backection
8281: agemin, agemax range of age
8282: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8283: anback2 year of end of backprojection (same day and month as back1).
8284: prevacurrent and prev are prevalences.
1.267 brouard 8285: */
8286: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8287: double agec; /* generic age */
1.268 brouard 8288: double agelim, ppij, ppi, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
1.267 brouard 8289: double *popeffectif,*popcount;
8290: double ***p3mat;
8291: /* double ***mobaverage; */
8292: char fileresfb[FILENAMELENGTH];
8293:
1.268 brouard 8294: agelim=AGEINF;
1.267 brouard 8295: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8296: in each health status at the date of interview (if between dateprev1 and dateprev2).
8297: We still use firstpass and lastpass as another selection.
8298: */
8299: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8300: /* firstpass, lastpass, stepm, weightopt, model); */
8301:
8302: /*Do we need to compute prevalence again?*/
8303:
8304: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8305:
8306: strcpy(fileresfb,"FB_");
8307: strcat(fileresfb,fileresu);
8308: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8309: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8310: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8311: }
8312: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8313: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8314:
8315: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8316:
8317:
8318: stepsize=(int) (stepm+YEARM-1)/YEARM;
8319: if (stepm<=12) stepsize=1;
8320: if(estepm < stepm){
8321: printf ("Problem %d lower than %d\n",estepm, stepm);
8322: }
1.270 brouard 8323: else{
8324: hstepm=estepm;
8325: }
8326: if(estepm >= stepm){ /* Yes every two year */
8327: stepsize=2;
8328: }
1.267 brouard 8329:
8330: hstepm=hstepm/stepm;
8331: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
8332: fractional in yp1 */
8333: anprojmean=yp;
8334: yp2=modf((yp1*12),&yp);
8335: mprojmean=yp;
8336: yp1=modf((yp2*30.5),&yp);
8337: jprojmean=yp;
8338: if(jprojmean==0) jprojmean=1;
8339: if(mprojmean==0) jprojmean=1;
8340:
8341: i1=pow(2,cptcoveff);
8342: if (cptcovn < 1){i1=1;}
8343:
8344: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.268 brouard 8345: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8346:
8347: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8348:
8349: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8350: for(k=1; k<=i1;k++){
8351: if(i1 != 1 && TKresult[nres]!= k)
8352: continue;
8353: if(invalidvarcomb[k]){
8354: printf("\nCombination (%d) projection ignored because no cases \n",k);
8355: continue;
8356: }
1.268 brouard 8357: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8358: for(j=1;j<=cptcoveff;j++) {
8359: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8360: }
8361: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8362: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8363: }
8364: fprintf(ficresfb," yearbproj age");
8365: for(j=1; j<=nlstate+ndeath;j++){
8366: for(i=1; i<=nlstate;i++)
1.268 brouard 8367: fprintf(ficresfb," b%d%d",i,j);
8368: fprintf(ficresfb," b.%d",j);
1.267 brouard 8369: }
8370: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
8371: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8372: fprintf(ficresfb,"\n");
8373: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
1.273 brouard 8374: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8375: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8376: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8377: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8378: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8379: nhstepm = nhstepm/hstepm;
8380: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8381: oldm=oldms;savm=savms;
1.268 brouard 8382: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8383: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8384: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8385: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8386: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8387: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8388: for (h=0; h<=nhstepm; h++){
1.268 brouard 8389: if (h*hstepm/YEARM*stepm ==-yearp) {
8390: break;
8391: }
8392: }
8393: fprintf(ficresfb,"\n");
8394: for(j=1;j<=cptcoveff;j++)
8395: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8396: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec-h*hstepm/YEARM*stepm);
8397: for(i=1; i<=nlstate+ndeath;i++) {
8398: ppij=0.;ppi=0.;
8399: for(j=1; j<=nlstate;j++) {
8400: /* if (mobilav==1) */
1.269 brouard 8401: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8402: ppi=ppi+prevacurrent[(int)agec][j][k];
8403: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8404: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8405: /* else { */
8406: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8407: /* } */
1.268 brouard 8408: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8409: } /* end j */
8410: if(ppi <0.99){
8411: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8412: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8413: }
8414: fprintf(ficresfb," %.3f", ppij);
8415: }/* end j */
1.267 brouard 8416: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8417: } /* end agec */
8418: } /* end yearp */
8419: } /* end k */
1.217 brouard 8420:
1.267 brouard 8421: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8422:
1.267 brouard 8423: fclose(ficresfb);
8424: printf("End of Computing Back forecasting \n");
8425: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8426:
1.267 brouard 8427: }
1.217 brouard 8428:
1.269 brouard 8429: /* Variance of prevalence limit: varprlim */
8430: 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){
8431: /*------- Variance of period (stable) prevalence------*/
8432:
8433: char fileresvpl[FILENAMELENGTH];
8434: FILE *ficresvpl;
8435: double **oldm, **savm;
8436: double **varpl; /* Variances of prevalence limits by age */
8437: int i1, k, nres, j ;
8438:
8439: strcpy(fileresvpl,"VPL_");
8440: strcat(fileresvpl,fileresu);
8441: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
8442: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
8443: exit(0);
8444: }
8445: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8446: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
8447:
8448: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8449: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8450:
8451: i1=pow(2,cptcoveff);
8452: if (cptcovn < 1){i1=1;}
8453:
8454: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8455: for(k=1; k<=i1;k++){
8456: if(i1 != 1 && TKresult[nres]!= k)
8457: continue;
8458: fprintf(ficresvpl,"\n#****** ");
8459: printf("\n#****** ");
8460: fprintf(ficlog,"\n#****** ");
8461: for(j=1;j<=cptcoveff;j++) {
8462: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8463: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8464: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8465: }
8466: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8467: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8468: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8469: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8470: }
8471: fprintf(ficresvpl,"******\n");
8472: printf("******\n");
8473: fprintf(ficlog,"******\n");
8474:
8475: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8476: oldm=oldms;savm=savms;
8477: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8478: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8479: /*}*/
8480: }
8481:
8482: fclose(ficresvpl);
8483: printf("done variance-covariance of period prevalence\n");fflush(stdout);
8484: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
8485:
8486: }
8487: /* Variance of back prevalence: varbprlim */
8488: 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){
8489: /*------- Variance of back (stable) prevalence------*/
8490:
8491: char fileresvbl[FILENAMELENGTH];
8492: FILE *ficresvbl;
8493:
8494: double **oldm, **savm;
8495: double **varbpl; /* Variances of back prevalence limits by age */
8496: int i1, k, nres, j ;
8497:
8498: strcpy(fileresvbl,"VBL_");
8499: strcat(fileresvbl,fileresu);
8500: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8501: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8502: exit(0);
8503: }
8504: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8505: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8506:
8507:
8508: i1=pow(2,cptcoveff);
8509: if (cptcovn < 1){i1=1;}
8510:
8511: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8512: for(k=1; k<=i1;k++){
8513: if(i1 != 1 && TKresult[nres]!= k)
8514: continue;
8515: fprintf(ficresvbl,"\n#****** ");
8516: printf("\n#****** ");
8517: fprintf(ficlog,"\n#****** ");
8518: for(j=1;j<=cptcoveff;j++) {
8519: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8520: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8521: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8522: }
8523: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8524: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8525: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8526: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8527: }
8528: fprintf(ficresvbl,"******\n");
8529: printf("******\n");
8530: fprintf(ficlog,"******\n");
8531:
8532: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8533: oldm=oldms;savm=savms;
8534:
8535: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8536: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8537: /*}*/
8538: }
8539:
8540: fclose(ficresvbl);
8541: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8542: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8543:
8544: } /* End of varbprlim */
8545:
1.126 brouard 8546: /************** Forecasting *****not tested NB*************/
1.227 brouard 8547: /* 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 8548:
1.227 brouard 8549: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8550: /* int *popage; */
8551: /* double calagedatem, agelim, kk1, kk2; */
8552: /* double *popeffectif,*popcount; */
8553: /* double ***p3mat,***tabpop,***tabpopprev; */
8554: /* /\* double ***mobaverage; *\/ */
8555: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8556:
1.227 brouard 8557: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8558: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8559: /* agelim=AGESUP; */
8560: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8561:
1.227 brouard 8562: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8563:
8564:
1.227 brouard 8565: /* strcpy(filerespop,"POP_"); */
8566: /* strcat(filerespop,fileresu); */
8567: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8568: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8569: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8570: /* } */
8571: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8572: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8573:
1.227 brouard 8574: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8575:
1.227 brouard 8576: /* /\* if (mobilav!=0) { *\/ */
8577: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8578: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8579: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8580: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8581: /* /\* } *\/ */
8582: /* /\* } *\/ */
1.126 brouard 8583:
1.227 brouard 8584: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8585: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8586:
1.227 brouard 8587: /* agelim=AGESUP; */
1.126 brouard 8588:
1.227 brouard 8589: /* hstepm=1; */
8590: /* hstepm=hstepm/stepm; */
1.218 brouard 8591:
1.227 brouard 8592: /* if (popforecast==1) { */
8593: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8594: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8595: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8596: /* } */
8597: /* popage=ivector(0,AGESUP); */
8598: /* popeffectif=vector(0,AGESUP); */
8599: /* popcount=vector(0,AGESUP); */
1.126 brouard 8600:
1.227 brouard 8601: /* i=1; */
8602: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8603:
1.227 brouard 8604: /* imx=i; */
8605: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8606: /* } */
1.218 brouard 8607:
1.227 brouard 8608: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8609: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8610: /* k=k+1; */
8611: /* fprintf(ficrespop,"\n#******"); */
8612: /* for(j=1;j<=cptcoveff;j++) { */
8613: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8614: /* } */
8615: /* fprintf(ficrespop,"******\n"); */
8616: /* fprintf(ficrespop,"# Age"); */
8617: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8618: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8619:
1.227 brouard 8620: /* for (cpt=0; cpt<=0;cpt++) { */
8621: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8622:
1.227 brouard 8623: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8624: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8625: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8626:
1.227 brouard 8627: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8628: /* oldm=oldms;savm=savms; */
8629: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8630:
1.227 brouard 8631: /* for (h=0; h<=nhstepm; h++){ */
8632: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8633: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8634: /* } */
8635: /* for(j=1; j<=nlstate+ndeath;j++) { */
8636: /* kk1=0.;kk2=0; */
8637: /* for(i=1; i<=nlstate;i++) { */
8638: /* if (mobilav==1) */
8639: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8640: /* else { */
8641: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8642: /* } */
8643: /* } */
8644: /* if (h==(int)(calagedatem+12*cpt)){ */
8645: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8646: /* /\*fprintf(ficrespop," %.3f", kk1); */
8647: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8648: /* } */
8649: /* } */
8650: /* for(i=1; i<=nlstate;i++){ */
8651: /* kk1=0.; */
8652: /* for(j=1; j<=nlstate;j++){ */
8653: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8654: /* } */
8655: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8656: /* } */
1.218 brouard 8657:
1.227 brouard 8658: /* if (h==(int)(calagedatem+12*cpt)) */
8659: /* for(j=1; j<=nlstate;j++) */
8660: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8661: /* } */
8662: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8663: /* } */
8664: /* } */
1.218 brouard 8665:
1.227 brouard 8666: /* /\******\/ */
1.218 brouard 8667:
1.227 brouard 8668: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8669: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8670: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8671: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8672: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8673:
1.227 brouard 8674: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8675: /* oldm=oldms;savm=savms; */
8676: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8677: /* for (h=0; h<=nhstepm; h++){ */
8678: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8679: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8680: /* } */
8681: /* for(j=1; j<=nlstate+ndeath;j++) { */
8682: /* kk1=0.;kk2=0; */
8683: /* for(i=1; i<=nlstate;i++) { */
8684: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8685: /* } */
8686: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8687: /* } */
8688: /* } */
8689: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8690: /* } */
8691: /* } */
8692: /* } */
8693: /* } */
1.218 brouard 8694:
1.227 brouard 8695: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8696:
1.227 brouard 8697: /* if (popforecast==1) { */
8698: /* free_ivector(popage,0,AGESUP); */
8699: /* free_vector(popeffectif,0,AGESUP); */
8700: /* free_vector(popcount,0,AGESUP); */
8701: /* } */
8702: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8703: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8704: /* fclose(ficrespop); */
8705: /* } /\* End of popforecast *\/ */
1.218 brouard 8706:
1.126 brouard 8707: int fileappend(FILE *fichier, char *optionfich)
8708: {
8709: if((fichier=fopen(optionfich,"a"))==NULL) {
8710: printf("Problem with file: %s\n", optionfich);
8711: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8712: return (0);
8713: }
8714: fflush(fichier);
8715: return (1);
8716: }
8717:
8718:
8719: /**************** function prwizard **********************/
8720: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8721: {
8722:
8723: /* Wizard to print covariance matrix template */
8724:
1.164 brouard 8725: char ca[32], cb[32];
8726: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8727: int numlinepar;
8728:
8729: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8730: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8731: for(i=1; i <=nlstate; i++){
8732: jj=0;
8733: for(j=1; j <=nlstate+ndeath; j++){
8734: if(j==i) continue;
8735: jj++;
8736: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8737: printf("%1d%1d",i,j);
8738: fprintf(ficparo,"%1d%1d",i,j);
8739: for(k=1; k<=ncovmodel;k++){
8740: /* printf(" %lf",param[i][j][k]); */
8741: /* fprintf(ficparo," %lf",param[i][j][k]); */
8742: printf(" 0.");
8743: fprintf(ficparo," 0.");
8744: }
8745: printf("\n");
8746: fprintf(ficparo,"\n");
8747: }
8748: }
8749: printf("# Scales (for hessian or gradient estimation)\n");
8750: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8751: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8752: for(i=1; i <=nlstate; i++){
8753: jj=0;
8754: for(j=1; j <=nlstate+ndeath; j++){
8755: if(j==i) continue;
8756: jj++;
8757: fprintf(ficparo,"%1d%1d",i,j);
8758: printf("%1d%1d",i,j);
8759: fflush(stdout);
8760: for(k=1; k<=ncovmodel;k++){
8761: /* printf(" %le",delti3[i][j][k]); */
8762: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8763: printf(" 0.");
8764: fprintf(ficparo," 0.");
8765: }
8766: numlinepar++;
8767: printf("\n");
8768: fprintf(ficparo,"\n");
8769: }
8770: }
8771: printf("# Covariance matrix\n");
8772: /* # 121 Var(a12)\n\ */
8773: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8774: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8775: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8776: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8777: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8778: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8779: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8780: fflush(stdout);
8781: fprintf(ficparo,"# Covariance matrix\n");
8782: /* # 121 Var(a12)\n\ */
8783: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8784: /* # ...\n\ */
8785: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8786:
8787: for(itimes=1;itimes<=2;itimes++){
8788: jj=0;
8789: for(i=1; i <=nlstate; i++){
8790: for(j=1; j <=nlstate+ndeath; j++){
8791: if(j==i) continue;
8792: for(k=1; k<=ncovmodel;k++){
8793: jj++;
8794: ca[0]= k+'a'-1;ca[1]='\0';
8795: if(itimes==1){
8796: printf("#%1d%1d%d",i,j,k);
8797: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8798: }else{
8799: printf("%1d%1d%d",i,j,k);
8800: fprintf(ficparo,"%1d%1d%d",i,j,k);
8801: /* printf(" %.5le",matcov[i][j]); */
8802: }
8803: ll=0;
8804: for(li=1;li <=nlstate; li++){
8805: for(lj=1;lj <=nlstate+ndeath; lj++){
8806: if(lj==li) continue;
8807: for(lk=1;lk<=ncovmodel;lk++){
8808: ll++;
8809: if(ll<=jj){
8810: cb[0]= lk +'a'-1;cb[1]='\0';
8811: if(ll<jj){
8812: if(itimes==1){
8813: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8814: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8815: }else{
8816: printf(" 0.");
8817: fprintf(ficparo," 0.");
8818: }
8819: }else{
8820: if(itimes==1){
8821: printf(" Var(%s%1d%1d)",ca,i,j);
8822: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8823: }else{
8824: printf(" 0.");
8825: fprintf(ficparo," 0.");
8826: }
8827: }
8828: }
8829: } /* end lk */
8830: } /* end lj */
8831: } /* end li */
8832: printf("\n");
8833: fprintf(ficparo,"\n");
8834: numlinepar++;
8835: } /* end k*/
8836: } /*end j */
8837: } /* end i */
8838: } /* end itimes */
8839:
8840: } /* end of prwizard */
8841: /******************* Gompertz Likelihood ******************************/
8842: double gompertz(double x[])
8843: {
8844: double A,B,L=0.0,sump=0.,num=0.;
8845: int i,n=0; /* n is the size of the sample */
8846:
1.220 brouard 8847: for (i=1;i<=imx ; i++) {
1.126 brouard 8848: sump=sump+weight[i];
8849: /* sump=sump+1;*/
8850: num=num+1;
8851: }
8852:
8853:
8854: /* for (i=0; i<=imx; i++)
8855: 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]);*/
8856:
8857: for (i=1;i<=imx ; i++)
8858: {
8859: if (cens[i] == 1 && wav[i]>1)
8860: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8861:
8862: if (cens[i] == 0 && wav[i]>1)
8863: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8864: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8865:
8866: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8867: if (wav[i] > 1 ) { /* ??? */
8868: L=L+A*weight[i];
8869: /* 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]);*/
8870: }
8871: }
8872:
8873: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8874:
8875: return -2*L*num/sump;
8876: }
8877:
1.136 brouard 8878: #ifdef GSL
8879: /******************* Gompertz_f Likelihood ******************************/
8880: double gompertz_f(const gsl_vector *v, void *params)
8881: {
8882: double A,B,LL=0.0,sump=0.,num=0.;
8883: double *x= (double *) v->data;
8884: int i,n=0; /* n is the size of the sample */
8885:
8886: for (i=0;i<=imx-1 ; i++) {
8887: sump=sump+weight[i];
8888: /* sump=sump+1;*/
8889: num=num+1;
8890: }
8891:
8892:
8893: /* for (i=0; i<=imx; i++)
8894: 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]);*/
8895: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8896: for (i=1;i<=imx ; i++)
8897: {
8898: if (cens[i] == 1 && wav[i]>1)
8899: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8900:
8901: if (cens[i] == 0 && wav[i]>1)
8902: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8903: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8904:
8905: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8906: if (wav[i] > 1 ) { /* ??? */
8907: LL=LL+A*weight[i];
8908: /* 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]);*/
8909: }
8910: }
8911:
8912: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8913: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8914:
8915: return -2*LL*num/sump;
8916: }
8917: #endif
8918:
1.126 brouard 8919: /******************* Printing html file ***********/
1.201 brouard 8920: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8921: int lastpass, int stepm, int weightopt, char model[],\
8922: int imx, double p[],double **matcov,double agemortsup){
8923: int i,k;
8924:
8925: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8926: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8927: for (i=1;i<=2;i++)
8928: 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 8929: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8930: fprintf(fichtm,"</ul>");
8931:
8932: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8933:
8934: 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>");
8935:
8936: for (k=agegomp;k<(agemortsup-2);k++)
8937: 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]);
8938:
8939:
8940: fflush(fichtm);
8941: }
8942:
8943: /******************* Gnuplot file **************/
1.201 brouard 8944: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8945:
8946: char dirfileres[132],optfileres[132];
1.164 brouard 8947:
1.126 brouard 8948: int ng;
8949:
8950:
8951: /*#ifdef windows */
8952: fprintf(ficgp,"cd \"%s\" \n",pathc);
8953: /*#endif */
8954:
8955:
8956: strcpy(dirfileres,optionfilefiname);
8957: strcpy(optfileres,"vpl");
1.199 brouard 8958: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8959: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8960: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8961: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8962: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8963:
8964: }
8965:
1.136 brouard 8966: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8967: {
1.126 brouard 8968:
1.136 brouard 8969: /*-------- data file ----------*/
8970: FILE *fic;
8971: char dummy[]=" ";
1.240 brouard 8972: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8973: int lstra;
1.136 brouard 8974: int linei, month, year,iout;
8975: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8976: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8977: char *stratrunc;
1.223 brouard 8978:
1.240 brouard 8979: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8980: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8981:
1.240 brouard 8982: for(v=1; v <=ncovcol;v++){
8983: DummyV[v]=0;
8984: FixedV[v]=0;
8985: }
8986: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8987: DummyV[v]=1;
8988: FixedV[v]=0;
8989: }
8990: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8991: DummyV[v]=0;
8992: FixedV[v]=1;
8993: }
8994: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8995: DummyV[v]=1;
8996: FixedV[v]=1;
8997: }
8998: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8999: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9000: 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]);
9001: }
1.126 brouard 9002:
1.136 brouard 9003: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9004: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9005: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9006: }
1.126 brouard 9007:
1.136 brouard 9008: i=1;
9009: linei=0;
9010: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9011: linei=linei+1;
9012: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9013: if(line[j] == '\t')
9014: line[j] = ' ';
9015: }
9016: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9017: ;
9018: };
9019: line[j+1]=0; /* Trims blanks at end of line */
9020: if(line[0]=='#'){
9021: fprintf(ficlog,"Comment line\n%s\n",line);
9022: printf("Comment line\n%s\n",line);
9023: continue;
9024: }
9025: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9026: strcpy(line, linetmp);
1.223 brouard 9027:
9028: /* Loops on waves */
9029: for (j=maxwav;j>=1;j--){
9030: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9031: cutv(stra, strb, line, ' ');
9032: if(strb[0]=='.') { /* Missing value */
9033: lval=-1;
9034: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9035: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9036: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9037: 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);
9038: 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);
9039: return 1;
9040: }
9041: }else{
9042: errno=0;
9043: /* what_kind_of_number(strb); */
9044: dval=strtod(strb,&endptr);
9045: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9046: /* if(strb != endptr && *endptr == '\0') */
9047: /* dval=dlval; */
9048: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9049: if( strb[0]=='\0' || (*endptr != '\0')){
9050: 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);
9051: 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);
9052: return 1;
9053: }
9054: cotqvar[j][iv][i]=dval;
9055: cotvar[j][ntv+iv][i]=dval;
9056: }
9057: strcpy(line,stra);
1.223 brouard 9058: }/* end loop ntqv */
1.225 brouard 9059:
1.223 brouard 9060: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9061: cutv(stra, strb, line, ' ');
9062: if(strb[0]=='.') { /* Missing value */
9063: lval=-1;
9064: }else{
9065: errno=0;
9066: lval=strtol(strb,&endptr,10);
9067: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9068: if( strb[0]=='\0' || (*endptr != '\0')){
9069: 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);
9070: 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);
9071: return 1;
9072: }
9073: }
9074: if(lval <-1 || lval >1){
9075: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9076: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9077: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9078: For example, for multinomial values like 1, 2 and 3,\n \
9079: build V1=0 V2=0 for the reference value (1),\n \
9080: V1=1 V2=0 for (2) \n \
1.223 brouard 9081: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9082: output of IMaCh is often meaningless.\n \
1.223 brouard 9083: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9084: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9085: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9086: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9087: For example, for multinomial values like 1, 2 and 3,\n \
9088: build V1=0 V2=0 for the reference value (1),\n \
9089: V1=1 V2=0 for (2) \n \
1.223 brouard 9090: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9091: output of IMaCh is often meaningless.\n \
1.223 brouard 9092: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9093: return 1;
9094: }
9095: cotvar[j][iv][i]=(double)(lval);
9096: strcpy(line,stra);
1.223 brouard 9097: }/* end loop ntv */
1.225 brouard 9098:
1.223 brouard 9099: /* Statuses at wave */
1.137 brouard 9100: cutv(stra, strb, line, ' ');
1.223 brouard 9101: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9102: lval=-1;
1.136 brouard 9103: }else{
1.238 brouard 9104: errno=0;
9105: lval=strtol(strb,&endptr,10);
9106: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9107: if( strb[0]=='\0' || (*endptr != '\0')){
9108: 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);
9109: 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);
9110: return 1;
9111: }
1.136 brouard 9112: }
1.225 brouard 9113:
1.136 brouard 9114: s[j][i]=lval;
1.225 brouard 9115:
1.223 brouard 9116: /* Date of Interview */
1.136 brouard 9117: strcpy(line,stra);
9118: cutv(stra, strb,line,' ');
1.169 brouard 9119: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9120: }
1.169 brouard 9121: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9122: month=99;
9123: year=9999;
1.136 brouard 9124: }else{
1.225 brouard 9125: 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);
9126: 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);
9127: return 1;
1.136 brouard 9128: }
9129: anint[j][i]= (double) year;
9130: mint[j][i]= (double)month;
9131: strcpy(line,stra);
1.223 brouard 9132: } /* End loop on waves */
1.225 brouard 9133:
1.223 brouard 9134: /* Date of death */
1.136 brouard 9135: cutv(stra, strb,line,' ');
1.169 brouard 9136: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9137: }
1.169 brouard 9138: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9139: month=99;
9140: year=9999;
9141: }else{
1.141 brouard 9142: 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 9143: 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);
9144: return 1;
1.136 brouard 9145: }
9146: andc[i]=(double) year;
9147: moisdc[i]=(double) month;
9148: strcpy(line,stra);
9149:
1.223 brouard 9150: /* Date of birth */
1.136 brouard 9151: cutv(stra, strb,line,' ');
1.169 brouard 9152: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9153: }
1.169 brouard 9154: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9155: month=99;
9156: year=9999;
9157: }else{
1.141 brouard 9158: 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);
9159: 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 9160: return 1;
1.136 brouard 9161: }
9162: if (year==9999) {
1.141 brouard 9163: 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);
9164: 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 9165: return 1;
9166:
1.136 brouard 9167: }
9168: annais[i]=(double)(year);
9169: moisnais[i]=(double)(month);
9170: strcpy(line,stra);
1.225 brouard 9171:
1.223 brouard 9172: /* Sample weight */
1.136 brouard 9173: cutv(stra, strb,line,' ');
9174: errno=0;
9175: dval=strtod(strb,&endptr);
9176: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9177: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9178: 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 9179: fflush(ficlog);
9180: return 1;
9181: }
9182: weight[i]=dval;
9183: strcpy(line,stra);
1.225 brouard 9184:
1.223 brouard 9185: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9186: cutv(stra, strb, line, ' ');
9187: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9188: lval=-1;
1.223 brouard 9189: }else{
1.225 brouard 9190: errno=0;
9191: /* what_kind_of_number(strb); */
9192: dval=strtod(strb,&endptr);
9193: /* if(strb != endptr && *endptr == '\0') */
9194: /* dval=dlval; */
9195: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9196: if( strb[0]=='\0' || (*endptr != '\0')){
9197: 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);
9198: 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);
9199: return 1;
9200: }
9201: coqvar[iv][i]=dval;
1.226 brouard 9202: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9203: }
9204: strcpy(line,stra);
9205: }/* end loop nqv */
1.136 brouard 9206:
1.223 brouard 9207: /* Covariate values */
1.136 brouard 9208: for (j=ncovcol;j>=1;j--){
9209: cutv(stra, strb,line,' ');
1.223 brouard 9210: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9211: lval=-1;
1.136 brouard 9212: }else{
1.225 brouard 9213: errno=0;
9214: lval=strtol(strb,&endptr,10);
9215: if( strb[0]=='\0' || (*endptr != '\0')){
9216: 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);
9217: 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);
9218: return 1;
9219: }
1.136 brouard 9220: }
9221: if(lval <-1 || lval >1){
1.225 brouard 9222: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9223: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9224: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9225: For example, for multinomial values like 1, 2 and 3,\n \
9226: build V1=0 V2=0 for the reference value (1),\n \
9227: V1=1 V2=0 for (2) \n \
1.136 brouard 9228: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9229: output of IMaCh is often meaningless.\n \
1.136 brouard 9230: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9231: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9232: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9233: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9234: For example, for multinomial values like 1, 2 and 3,\n \
9235: build V1=0 V2=0 for the reference value (1),\n \
9236: V1=1 V2=0 for (2) \n \
1.136 brouard 9237: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9238: output of IMaCh is often meaningless.\n \
1.136 brouard 9239: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9240: return 1;
1.136 brouard 9241: }
9242: covar[j][i]=(double)(lval);
9243: strcpy(line,stra);
9244: }
9245: lstra=strlen(stra);
1.225 brouard 9246:
1.136 brouard 9247: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9248: stratrunc = &(stra[lstra-9]);
9249: num[i]=atol(stratrunc);
9250: }
9251: else
9252: num[i]=atol(stra);
9253: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9254: 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;}*/
9255:
9256: i=i+1;
9257: } /* End loop reading data */
1.225 brouard 9258:
1.136 brouard 9259: *imax=i-1; /* Number of individuals */
9260: fclose(fic);
1.225 brouard 9261:
1.136 brouard 9262: return (0);
1.164 brouard 9263: /* endread: */
1.225 brouard 9264: printf("Exiting readdata: ");
9265: fclose(fic);
9266: return (1);
1.223 brouard 9267: }
1.126 brouard 9268:
1.234 brouard 9269: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9270: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9271: while (*p2 == ' ')
1.234 brouard 9272: p2++;
9273: /* while ((*p1++ = *p2++) !=0) */
9274: /* ; */
9275: /* do */
9276: /* while (*p2 == ' ') */
9277: /* p2++; */
9278: /* while (*p1++ == *p2++); */
9279: *stri=p2;
1.145 brouard 9280: }
9281:
1.235 brouard 9282: int decoderesult ( char resultline[], int nres)
1.230 brouard 9283: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9284: {
1.235 brouard 9285: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9286: char resultsav[MAXLINE];
1.234 brouard 9287: int resultmodel[MAXLINE];
9288: int modelresult[MAXLINE];
1.230 brouard 9289: char stra[80], strb[80], strc[80], strd[80],stre[80];
9290:
1.234 brouard 9291: removefirstspace(&resultline);
1.233 brouard 9292: printf("decoderesult:%s\n",resultline);
1.230 brouard 9293:
9294: if (strstr(resultline,"v") !=0){
9295: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9296: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9297: return 1;
9298: }
9299: trimbb(resultsav, resultline);
9300: if (strlen(resultsav) >1){
9301: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9302: }
1.253 brouard 9303: if(j == 0){ /* Resultline but no = */
9304: TKresult[nres]=0; /* Combination for the nresult and the model */
9305: return (0);
9306: }
9307:
1.234 brouard 9308: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9309: 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);
9310: 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);
9311: }
9312: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9313: if(nbocc(resultsav,'=') >1){
9314: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9315: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9316: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9317: }else
9318: cutl(strc,strd,resultsav,'=');
1.230 brouard 9319: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9320:
1.230 brouard 9321: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9322: Tvarsel[k]=atoi(strc);
9323: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9324: /* cptcovsel++; */
9325: if (nbocc(stra,'=') >0)
9326: strcpy(resultsav,stra); /* and analyzes it */
9327: }
1.235 brouard 9328: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9329: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9330: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9331: match=0;
1.236 brouard 9332: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9333: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9334: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9335: match=1;
9336: break;
9337: }
9338: }
9339: if(match == 0){
9340: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9341: }
9342: }
9343: }
1.235 brouard 9344: /* Checking for missing or useless values in comparison of current model needs */
9345: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9346: match=0;
1.235 brouard 9347: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9348: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9349: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9350: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9351: ++match;
9352: }
9353: }
9354: }
9355: if(match == 0){
9356: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9357: }else if(match > 1){
9358: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9359: }
9360: }
1.235 brouard 9361:
1.234 brouard 9362: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9363: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9364: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9365: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9366: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9367: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9368: /* 1 0 0 0 */
9369: /* 2 1 0 0 */
9370: /* 3 0 1 0 */
9371: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9372: /* 5 0 0 1 */
9373: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9374: /* 7 0 1 1 */
9375: /* 8 1 1 1 */
1.237 brouard 9376: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9377: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9378: /* V5*age V5 known which value for nres? */
9379: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9380: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9381: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9382: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9383: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9384: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9385: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9386: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9387: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9388: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9389: k4++;;
9390: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9391: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9392: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9393: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9394: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9395: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9396: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9397: k4q++;;
9398: }
9399: }
1.234 brouard 9400:
1.235 brouard 9401: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9402: return (0);
9403: }
1.235 brouard 9404:
1.230 brouard 9405: int decodemodel( char model[], int lastobs)
9406: /**< This routine decodes the model and returns:
1.224 brouard 9407: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9408: * - nagesqr = 1 if age*age in the model, otherwise 0.
9409: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9410: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9411: * - cptcovage number of covariates with age*products =2
9412: * - cptcovs number of simple covariates
9413: * - 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
9414: * which is a new column after the 9 (ncovcol) variables.
9415: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9416: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9417: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9418: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9419: */
1.136 brouard 9420: {
1.238 brouard 9421: int i, j, k, ks, v;
1.227 brouard 9422: int j1, k1, k2, k3, k4;
1.136 brouard 9423: char modelsav[80];
1.145 brouard 9424: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9425: char *strpt;
1.136 brouard 9426:
1.145 brouard 9427: /*removespace(model);*/
1.136 brouard 9428: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9429: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9430: if (strstr(model,"AGE") !=0){
1.192 brouard 9431: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9432: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9433: return 1;
9434: }
1.141 brouard 9435: if (strstr(model,"v") !=0){
9436: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9437: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9438: return 1;
9439: }
1.187 brouard 9440: strcpy(modelsav,model);
9441: if ((strpt=strstr(model,"age*age")) !=0){
9442: printf(" strpt=%s, model=%s\n",strpt, model);
9443: if(strpt != model){
1.234 brouard 9444: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9445: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9446: corresponding column of parameters.\n",model);
1.234 brouard 9447: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9448: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9449: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9450: return 1;
1.225 brouard 9451: }
1.187 brouard 9452: nagesqr=1;
9453: if (strstr(model,"+age*age") !=0)
1.234 brouard 9454: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9455: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9456: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9457: else
1.234 brouard 9458: substrchaine(modelsav, model, "age*age");
1.187 brouard 9459: }else
9460: nagesqr=0;
9461: if (strlen(modelsav) >1){
9462: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9463: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9464: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9465: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9466: * cst, age and age*age
9467: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9468: /* including age products which are counted in cptcovage.
9469: * but the covariates which are products must be treated
9470: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9471: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9472: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9473:
9474:
1.187 brouard 9475: /* Design
9476: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9477: * < ncovcol=8 >
9478: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9479: * k= 1 2 3 4 5 6 7 8
9480: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9481: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9482: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9483: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9484: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9485: * Tage[++cptcovage]=k
9486: * if products, new covar are created after ncovcol with k1
9487: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9488: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9489: * 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
9490: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9491: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9492: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9493: * < ncovcol=8 >
9494: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9495: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9496: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9497: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9498: * p Tprod[1]@2={ 6, 5}
9499: *p Tvard[1][1]@4= {7, 8, 5, 6}
9500: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9501: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9502: *How to reorganize?
9503: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9504: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9505: * {2, 1, 4, 8, 5, 6, 3, 7}
9506: * Struct []
9507: */
1.225 brouard 9508:
1.187 brouard 9509: /* This loop fills the array Tvar from the string 'model'.*/
9510: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9511: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9512: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9513: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9514: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9515: /* k=1 Tvar[1]=2 (from V2) */
9516: /* k=5 Tvar[5] */
9517: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9518: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9519: /* } */
1.198 brouard 9520: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9521: /*
9522: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9523: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9524: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9525: }
1.187 brouard 9526: cptcovage=0;
9527: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9528: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9529: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9530: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9531: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9532: /*scanf("%d",i);*/
9533: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9534: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9535: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9536: /* covar is not filled and then is empty */
9537: cptcovprod--;
9538: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9539: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9540: Typevar[k]=1; /* 1 for age product */
9541: cptcovage++; /* Sums the number of covariates which include age as a product */
9542: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9543: /*printf("stre=%s ", stre);*/
9544: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9545: cptcovprod--;
9546: cutl(stre,strb,strc,'V');
9547: Tvar[k]=atoi(stre);
9548: Typevar[k]=1; /* 1 for age product */
9549: cptcovage++;
9550: Tage[cptcovage]=k;
9551: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9552: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9553: cptcovn++;
9554: cptcovprodnoage++;k1++;
9555: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9556: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9557: because this model-covariate is a construction we invent a new column
9558: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9559: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9560: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9561: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9562: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9563: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9564: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9565: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9566: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9567: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9568: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9569: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9570: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9571: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9572: for (i=1; i<=lastobs;i++){
9573: /* Computes the new covariate which is a product of
9574: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9575: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9576: }
9577: } /* End age is not in the model */
9578: } /* End if model includes a product */
9579: else { /* no more sum */
9580: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9581: /* scanf("%d",i);*/
9582: cutl(strd,strc,strb,'V');
9583: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9584: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9585: Tvar[k]=atoi(strd);
9586: Typevar[k]=0; /* 0 for simple covariates */
9587: }
9588: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9589: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9590: scanf("%d",i);*/
1.187 brouard 9591: } /* end of loop + on total covariates */
9592: } /* end if strlen(modelsave == 0) age*age might exist */
9593: } /* end if strlen(model == 0) */
1.136 brouard 9594:
9595: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9596: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9597:
1.136 brouard 9598: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9599: printf("cptcovprod=%d ", cptcovprod);
9600: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9601: scanf("%d ",i);*/
9602:
9603:
1.230 brouard 9604: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9605: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9606: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9607: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9608: k = 1 2 3 4 5 6 7 8 9
9609: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9610: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9611: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9612: Dummy[k] 1 0 0 0 3 1 1 2 3
9613: Tmodelind[combination of covar]=k;
1.225 brouard 9614: */
9615: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9616: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9617: /* 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 9618: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9619: printf("Model=%s\n\
9620: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9621: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9622: 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);
9623: fprintf(ficlog,"Model=%s\n\
9624: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9625: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9626: 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 9627: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9628: 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 */
9629: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9630: Fixed[k]= 0;
9631: Dummy[k]= 0;
1.225 brouard 9632: ncoveff++;
1.232 brouard 9633: ncovf++;
1.234 brouard 9634: nsd++;
9635: modell[k].maintype= FTYPE;
9636: TvarsD[nsd]=Tvar[k];
9637: TvarsDind[nsd]=k;
9638: TvarF[ncovf]=Tvar[k];
9639: TvarFind[ncovf]=k;
9640: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9641: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9642: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9643: Fixed[k]= 0;
9644: Dummy[k]= 0;
9645: ncoveff++;
9646: ncovf++;
9647: modell[k].maintype= FTYPE;
9648: TvarF[ncovf]=Tvar[k];
9649: TvarFind[ncovf]=k;
1.230 brouard 9650: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9651: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9652: }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 9653: Fixed[k]= 0;
9654: Dummy[k]= 1;
1.230 brouard 9655: nqfveff++;
1.234 brouard 9656: modell[k].maintype= FTYPE;
9657: modell[k].subtype= FQ;
9658: nsq++;
9659: TvarsQ[nsq]=Tvar[k];
9660: TvarsQind[nsq]=k;
1.232 brouard 9661: ncovf++;
1.234 brouard 9662: TvarF[ncovf]=Tvar[k];
9663: TvarFind[ncovf]=k;
1.231 brouard 9664: 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 9665: 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 9666: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9667: Fixed[k]= 1;
9668: Dummy[k]= 0;
1.225 brouard 9669: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9670: modell[k].maintype= VTYPE;
9671: modell[k].subtype= VD;
9672: nsd++;
9673: TvarsD[nsd]=Tvar[k];
9674: TvarsDind[nsd]=k;
9675: ncovv++; /* Only simple time varying variables */
9676: TvarV[ncovv]=Tvar[k];
1.242 brouard 9677: 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 9678: 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 */
9679: 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 9680: 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);
9681: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9682: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9683: Fixed[k]= 1;
9684: Dummy[k]= 1;
9685: nqtveff++;
9686: modell[k].maintype= VTYPE;
9687: modell[k].subtype= VQ;
9688: ncovv++; /* Only simple time varying variables */
9689: nsq++;
9690: TvarsQ[nsq]=Tvar[k];
9691: TvarsQind[nsq]=k;
9692: TvarV[ncovv]=Tvar[k];
1.242 brouard 9693: 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 9694: 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 */
9695: 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 9696: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9697: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9698: 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 9699: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9700: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9701: ncova++;
9702: TvarA[ncova]=Tvar[k];
9703: TvarAind[ncova]=k;
1.231 brouard 9704: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9705: Fixed[k]= 2;
9706: Dummy[k]= 2;
9707: modell[k].maintype= ATYPE;
9708: modell[k].subtype= APFD;
9709: /* ncoveff++; */
1.227 brouard 9710: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9711: Fixed[k]= 2;
9712: Dummy[k]= 3;
9713: modell[k].maintype= ATYPE;
9714: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9715: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9716: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9717: Fixed[k]= 3;
9718: Dummy[k]= 2;
9719: modell[k].maintype= ATYPE;
9720: modell[k].subtype= APVD; /* Product age * varying dummy */
9721: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9722: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9723: Fixed[k]= 3;
9724: Dummy[k]= 3;
9725: modell[k].maintype= ATYPE;
9726: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9727: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9728: }
9729: }else if (Typevar[k] == 2) { /* product without age */
9730: k1=Tposprod[k];
9731: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9732: if(Tvard[k1][2] <=ncovcol){
9733: Fixed[k]= 1;
9734: Dummy[k]= 0;
9735: modell[k].maintype= FTYPE;
9736: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9737: ncovf++; /* Fixed variables without age */
9738: TvarF[ncovf]=Tvar[k];
9739: TvarFind[ncovf]=k;
9740: }else if(Tvard[k1][2] <=ncovcol+nqv){
9741: Fixed[k]= 0; /* or 2 ?*/
9742: Dummy[k]= 1;
9743: modell[k].maintype= FTYPE;
9744: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9745: ncovf++; /* Varying variables without age */
9746: TvarF[ncovf]=Tvar[k];
9747: TvarFind[ncovf]=k;
9748: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9749: Fixed[k]= 1;
9750: Dummy[k]= 0;
9751: modell[k].maintype= VTYPE;
9752: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9753: ncovv++; /* Varying variables without age */
9754: TvarV[ncovv]=Tvar[k];
9755: TvarVind[ncovv]=k;
9756: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9757: Fixed[k]= 1;
9758: Dummy[k]= 1;
9759: modell[k].maintype= VTYPE;
9760: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9761: ncovv++; /* Varying variables without age */
9762: TvarV[ncovv]=Tvar[k];
9763: TvarVind[ncovv]=k;
9764: }
1.227 brouard 9765: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9766: if(Tvard[k1][2] <=ncovcol){
9767: Fixed[k]= 0; /* or 2 ?*/
9768: Dummy[k]= 1;
9769: modell[k].maintype= FTYPE;
9770: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9771: ncovf++; /* Fixed variables without age */
9772: TvarF[ncovf]=Tvar[k];
9773: TvarFind[ncovf]=k;
9774: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9775: Fixed[k]= 1;
9776: Dummy[k]= 1;
9777: modell[k].maintype= VTYPE;
9778: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9779: ncovv++; /* Varying variables without age */
9780: TvarV[ncovv]=Tvar[k];
9781: TvarVind[ncovv]=k;
9782: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9783: Fixed[k]= 1;
9784: Dummy[k]= 1;
9785: modell[k].maintype= VTYPE;
9786: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9787: ncovv++; /* Varying variables without age */
9788: TvarV[ncovv]=Tvar[k];
9789: TvarVind[ncovv]=k;
9790: ncovv++; /* Varying variables without age */
9791: TvarV[ncovv]=Tvar[k];
9792: TvarVind[ncovv]=k;
9793: }
1.227 brouard 9794: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9795: if(Tvard[k1][2] <=ncovcol){
9796: Fixed[k]= 1;
9797: Dummy[k]= 1;
9798: modell[k].maintype= VTYPE;
9799: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9800: ncovv++; /* Varying variables without age */
9801: TvarV[ncovv]=Tvar[k];
9802: TvarVind[ncovv]=k;
9803: }else if(Tvard[k1][2] <=ncovcol+nqv){
9804: Fixed[k]= 1;
9805: Dummy[k]= 1;
9806: modell[k].maintype= VTYPE;
9807: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9808: ncovv++; /* Varying variables without age */
9809: TvarV[ncovv]=Tvar[k];
9810: TvarVind[ncovv]=k;
9811: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9812: Fixed[k]= 1;
9813: Dummy[k]= 0;
9814: modell[k].maintype= VTYPE;
9815: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9816: ncovv++; /* Varying variables without age */
9817: TvarV[ncovv]=Tvar[k];
9818: TvarVind[ncovv]=k;
9819: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9820: Fixed[k]= 1;
9821: Dummy[k]= 1;
9822: modell[k].maintype= VTYPE;
9823: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9824: ncovv++; /* Varying variables without age */
9825: TvarV[ncovv]=Tvar[k];
9826: TvarVind[ncovv]=k;
9827: }
1.227 brouard 9828: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9829: if(Tvard[k1][2] <=ncovcol){
9830: Fixed[k]= 1;
9831: Dummy[k]= 1;
9832: modell[k].maintype= VTYPE;
9833: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9834: ncovv++; /* Varying variables without age */
9835: TvarV[ncovv]=Tvar[k];
9836: TvarVind[ncovv]=k;
9837: }else if(Tvard[k1][2] <=ncovcol+nqv){
9838: Fixed[k]= 1;
9839: Dummy[k]= 1;
9840: modell[k].maintype= VTYPE;
9841: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9842: ncovv++; /* Varying variables without age */
9843: TvarV[ncovv]=Tvar[k];
9844: TvarVind[ncovv]=k;
9845: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9846: Fixed[k]= 1;
9847: Dummy[k]= 1;
9848: modell[k].maintype= VTYPE;
9849: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9850: ncovv++; /* Varying variables without age */
9851: TvarV[ncovv]=Tvar[k];
9852: TvarVind[ncovv]=k;
9853: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9854: Fixed[k]= 1;
9855: Dummy[k]= 1;
9856: modell[k].maintype= VTYPE;
9857: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9858: ncovv++; /* Varying variables without age */
9859: TvarV[ncovv]=Tvar[k];
9860: TvarVind[ncovv]=k;
9861: }
1.227 brouard 9862: }else{
1.240 brouard 9863: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9864: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9865: } /*end k1*/
1.225 brouard 9866: }else{
1.226 brouard 9867: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9868: 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 9869: }
1.227 brouard 9870: 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 9871: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9872: 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]);
9873: }
9874: /* Searching for doublons in the model */
9875: for(k1=1; k1<= cptcovt;k1++){
9876: for(k2=1; k2 <k1;k2++){
9877: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9878: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9879: if(Tvar[k1]==Tvar[k2]){
9880: 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]]);
9881: 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);
9882: return(1);
9883: }
9884: }else if (Typevar[k1] ==2){
9885: k3=Tposprod[k1];
9886: k4=Tposprod[k2];
9887: 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])) ){
9888: 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]]);
9889: 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);
9890: return(1);
9891: }
9892: }
1.227 brouard 9893: }
9894: }
1.225 brouard 9895: }
9896: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9897: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9898: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9899: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9900: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9901: /*endread:*/
1.225 brouard 9902: printf("Exiting decodemodel: ");
9903: return (1);
1.136 brouard 9904: }
9905:
1.169 brouard 9906: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9907: {/* Check ages at death */
1.136 brouard 9908: int i, m;
1.218 brouard 9909: int firstone=0;
9910:
1.136 brouard 9911: for (i=1; i<=imx; i++) {
9912: for(m=2; (m<= maxwav); m++) {
9913: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9914: anint[m][i]=9999;
1.216 brouard 9915: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9916: s[m][i]=-1;
1.136 brouard 9917: }
9918: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9919: *nberr = *nberr + 1;
1.218 brouard 9920: if(firstone == 0){
9921: firstone=1;
1.260 brouard 9922: 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 9923: }
1.262 brouard 9924: 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 9925: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9926: }
9927: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9928: (*nberr)++;
1.259 brouard 9929: 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 9930: 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 9931: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9932: }
9933: }
9934: }
9935:
9936: for (i=1; i<=imx; i++) {
9937: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9938: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9939: 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 9940: if (s[m][i] >= nlstate+1) {
1.169 brouard 9941: if(agedc[i]>0){
9942: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9943: agev[m][i]=agedc[i];
1.214 brouard 9944: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9945: }else {
1.136 brouard 9946: if ((int)andc[i]!=9999){
9947: nbwarn++;
9948: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9949: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9950: agev[m][i]=-1;
9951: }
9952: }
1.169 brouard 9953: } /* agedc > 0 */
1.214 brouard 9954: } /* end if */
1.136 brouard 9955: else if(s[m][i] !=9){ /* Standard case, age in fractional
9956: years but with the precision of a month */
9957: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9958: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9959: agev[m][i]=1;
9960: else if(agev[m][i] < *agemin){
9961: *agemin=agev[m][i];
9962: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9963: }
9964: else if(agev[m][i] >*agemax){
9965: *agemax=agev[m][i];
1.156 brouard 9966: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9967: }
9968: /*agev[m][i]=anint[m][i]-annais[i];*/
9969: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9970: } /* en if 9*/
1.136 brouard 9971: else { /* =9 */
1.214 brouard 9972: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9973: agev[m][i]=1;
9974: s[m][i]=-1;
9975: }
9976: }
1.214 brouard 9977: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9978: agev[m][i]=1;
1.214 brouard 9979: else{
9980: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9981: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9982: agev[m][i]=0;
9983: }
9984: } /* End for lastpass */
9985: }
1.136 brouard 9986:
9987: for (i=1; i<=imx; i++) {
9988: for(m=firstpass; (m<=lastpass); m++){
9989: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9990: (*nberr)++;
1.136 brouard 9991: 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);
9992: 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);
9993: return 1;
9994: }
9995: }
9996: }
9997:
9998: /*for (i=1; i<=imx; i++){
9999: for (m=firstpass; (m<lastpass); m++){
10000: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10001: }
10002:
10003: }*/
10004:
10005:
1.139 brouard 10006: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10007: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10008:
10009: return (0);
1.164 brouard 10010: /* endread:*/
1.136 brouard 10011: printf("Exiting calandcheckages: ");
10012: return (1);
10013: }
10014:
1.172 brouard 10015: #if defined(_MSC_VER)
10016: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10017: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10018: //#include "stdafx.h"
10019: //#include <stdio.h>
10020: //#include <tchar.h>
10021: //#include <windows.h>
10022: //#include <iostream>
10023: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10024:
10025: LPFN_ISWOW64PROCESS fnIsWow64Process;
10026:
10027: BOOL IsWow64()
10028: {
10029: BOOL bIsWow64 = FALSE;
10030:
10031: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10032: // (HANDLE, PBOOL);
10033:
10034: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10035:
10036: HMODULE module = GetModuleHandle(_T("kernel32"));
10037: const char funcName[] = "IsWow64Process";
10038: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10039: GetProcAddress(module, funcName);
10040:
10041: if (NULL != fnIsWow64Process)
10042: {
10043: if (!fnIsWow64Process(GetCurrentProcess(),
10044: &bIsWow64))
10045: //throw std::exception("Unknown error");
10046: printf("Unknown error\n");
10047: }
10048: return bIsWow64 != FALSE;
10049: }
10050: #endif
1.177 brouard 10051:
1.191 brouard 10052: void syscompilerinfo(int logged)
1.167 brouard 10053: {
10054: /* #include "syscompilerinfo.h"*/
1.185 brouard 10055: /* command line Intel compiler 32bit windows, XP compatible:*/
10056: /* /GS /W3 /Gy
10057: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10058: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10059: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10060: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10061: */
10062: /* 64 bits */
1.185 brouard 10063: /*
10064: /GS /W3 /Gy
10065: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10066: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10067: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10068: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10069: /* Optimization are useless and O3 is slower than O2 */
10070: /*
10071: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10072: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10073: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10074: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10075: */
1.186 brouard 10076: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10077: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10078: /PDB:"visual studio
10079: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10080: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10081: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10082: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10083: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10084: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10085: uiAccess='false'"
10086: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10087: /NOLOGO /TLBID:1
10088: */
1.177 brouard 10089: #if defined __INTEL_COMPILER
1.178 brouard 10090: #if defined(__GNUC__)
10091: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10092: #endif
1.177 brouard 10093: #elif defined(__GNUC__)
1.179 brouard 10094: #ifndef __APPLE__
1.174 brouard 10095: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10096: #endif
1.177 brouard 10097: struct utsname sysInfo;
1.178 brouard 10098: int cross = CROSS;
10099: if (cross){
10100: printf("Cross-");
1.191 brouard 10101: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10102: }
1.174 brouard 10103: #endif
10104:
1.171 brouard 10105: #include <stdint.h>
1.178 brouard 10106:
1.191 brouard 10107: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10108: #if defined(__clang__)
1.191 brouard 10109: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10110: #endif
10111: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10112: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10113: #endif
10114: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10115: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10116: #endif
10117: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10118: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10119: #endif
10120: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10121: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10122: #endif
10123: #if defined(_MSC_VER)
1.191 brouard 10124: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10125: #endif
10126: #if defined(__PGI)
1.191 brouard 10127: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10128: #endif
10129: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10130: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10131: #endif
1.191 brouard 10132: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10133:
1.167 brouard 10134: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10135: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10136: // Windows (x64 and x86)
1.191 brouard 10137: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10138: #elif __unix__ // all unices, not all compilers
10139: // Unix
1.191 brouard 10140: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10141: #elif __linux__
10142: // linux
1.191 brouard 10143: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10144: #elif __APPLE__
1.174 brouard 10145: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10146: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10147: #endif
10148:
10149: /* __MINGW32__ */
10150: /* __CYGWIN__ */
10151: /* __MINGW64__ */
10152: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10153: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10154: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10155: /* _WIN64 // Defined for applications for Win64. */
10156: /* _M_X64 // Defined for compilations that target x64 processors. */
10157: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10158:
1.167 brouard 10159: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10160: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10161: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10162: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10163: #else
1.191 brouard 10164: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10165: #endif
10166:
1.169 brouard 10167: #if defined(__GNUC__)
10168: # if defined(__GNUC_PATCHLEVEL__)
10169: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10170: + __GNUC_MINOR__ * 100 \
10171: + __GNUC_PATCHLEVEL__)
10172: # else
10173: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10174: + __GNUC_MINOR__ * 100)
10175: # endif
1.174 brouard 10176: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10177: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10178:
10179: if (uname(&sysInfo) != -1) {
10180: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10181: 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 10182: }
10183: else
10184: perror("uname() error");
1.179 brouard 10185: //#ifndef __INTEL_COMPILER
10186: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10187: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10188: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10189: #endif
1.169 brouard 10190: #endif
1.172 brouard 10191:
10192: // void main()
10193: // {
1.169 brouard 10194: #if defined(_MSC_VER)
1.174 brouard 10195: if (IsWow64()){
1.191 brouard 10196: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10197: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10198: }
10199: else{
1.191 brouard 10200: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10201: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10202: }
1.172 brouard 10203: // printf("\nPress Enter to continue...");
10204: // getchar();
10205: // }
10206:
1.169 brouard 10207: #endif
10208:
1.167 brouard 10209:
1.219 brouard 10210: }
1.136 brouard 10211:
1.219 brouard 10212: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 10213: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 10214: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10215: /* double ftolpl = 1.e-10; */
1.180 brouard 10216: double age, agebase, agelim;
1.203 brouard 10217: double tot;
1.180 brouard 10218:
1.202 brouard 10219: strcpy(filerespl,"PL_");
10220: strcat(filerespl,fileresu);
10221: if((ficrespl=fopen(filerespl,"w"))==NULL) {
10222: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10223: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10224: }
1.227 brouard 10225: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
10226: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10227: pstamp(ficrespl);
1.203 brouard 10228: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10229: fprintf(ficrespl,"#Age ");
10230: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10231: fprintf(ficrespl,"\n");
1.180 brouard 10232:
1.219 brouard 10233: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10234:
1.219 brouard 10235: agebase=ageminpar;
10236: agelim=agemaxpar;
1.180 brouard 10237:
1.227 brouard 10238: /* i1=pow(2,ncoveff); */
1.234 brouard 10239: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10240: if (cptcovn < 1){i1=1;}
1.180 brouard 10241:
1.238 brouard 10242: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10243: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10244: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10245: continue;
1.235 brouard 10246:
1.238 brouard 10247: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10248: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10249: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10250: /* k=k+1; */
10251: /* to clean */
10252: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10253: fprintf(ficrespl,"#******");
10254: printf("#******");
10255: fprintf(ficlog,"#******");
10256: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10257: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10258: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10259: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10260: }
10261: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10262: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10263: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10264: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10265: }
10266: fprintf(ficrespl,"******\n");
10267: printf("******\n");
10268: fprintf(ficlog,"******\n");
10269: if(invalidvarcomb[k]){
10270: printf("\nCombination (%d) ignored because no case \n",k);
10271: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10272: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10273: continue;
10274: }
1.219 brouard 10275:
1.238 brouard 10276: fprintf(ficrespl,"#Age ");
10277: for(j=1;j<=cptcoveff;j++) {
10278: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10279: }
10280: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10281: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10282:
1.238 brouard 10283: for (age=agebase; age<=agelim; age++){
10284: /* for (age=agebase; age<=agebase; age++){ */
10285: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10286: fprintf(ficrespl,"%.0f ",age );
10287: for(j=1;j<=cptcoveff;j++)
10288: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10289: tot=0.;
10290: for(i=1; i<=nlstate;i++){
10291: tot += prlim[i][i];
10292: fprintf(ficrespl," %.5f", prlim[i][i]);
10293: }
10294: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10295: } /* Age */
10296: /* was end of cptcod */
10297: } /* cptcov */
10298: } /* nres */
1.219 brouard 10299: return 0;
1.180 brouard 10300: }
10301:
1.218 brouard 10302: 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){
10303: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
10304:
10305: /* Computes the back prevalence limit for any combination of covariate values
10306: * at any age between ageminpar and agemaxpar
10307: */
1.235 brouard 10308: int i, j, k, i1, nres=0 ;
1.217 brouard 10309: /* double ftolpl = 1.e-10; */
10310: double age, agebase, agelim;
10311: double tot;
1.218 brouard 10312: /* double ***mobaverage; */
10313: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10314:
10315: strcpy(fileresplb,"PLB_");
10316: strcat(fileresplb,fileresu);
10317: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
10318: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10319: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
10320: }
10321: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10322: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
10323: pstamp(ficresplb);
10324: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
10325: fprintf(ficresplb,"#Age ");
10326: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10327: fprintf(ficresplb,"\n");
10328:
1.218 brouard 10329:
10330: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10331:
10332: agebase=ageminpar;
10333: agelim=agemaxpar;
10334:
10335:
1.227 brouard 10336: i1=pow(2,cptcoveff);
1.218 brouard 10337: if (cptcovn < 1){i1=1;}
1.227 brouard 10338:
1.238 brouard 10339: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10340: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10341: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10342: continue;
10343: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10344: fprintf(ficresplb,"#******");
10345: printf("#******");
10346: fprintf(ficlog,"#******");
10347: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10348: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10349: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10350: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10351: }
10352: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10353: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10354: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10355: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10356: }
10357: fprintf(ficresplb,"******\n");
10358: printf("******\n");
10359: fprintf(ficlog,"******\n");
10360: if(invalidvarcomb[k]){
10361: printf("\nCombination (%d) ignored because no cases \n",k);
10362: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10363: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10364: continue;
10365: }
1.218 brouard 10366:
1.238 brouard 10367: fprintf(ficresplb,"#Age ");
10368: for(j=1;j<=cptcoveff;j++) {
10369: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10370: }
10371: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10372: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10373:
10374:
1.238 brouard 10375: for (age=agebase; age<=agelim; age++){
10376: /* for (age=agebase; age<=agebase; age++){ */
10377: if(mobilavproj > 0){
10378: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10379: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10380: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10381: }else if (mobilavproj == 0){
10382: 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);
10383: 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);
10384: exit(1);
10385: }else{
10386: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10387: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10388: /* printf("TOTOT\n"); */
10389: /* exit(1); */
1.238 brouard 10390: }
10391: fprintf(ficresplb,"%.0f ",age );
10392: for(j=1;j<=cptcoveff;j++)
10393: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10394: tot=0.;
10395: for(i=1; i<=nlstate;i++){
10396: tot += bprlim[i][i];
10397: fprintf(ficresplb," %.5f", bprlim[i][i]);
10398: }
10399: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10400: } /* Age */
10401: /* was end of cptcod */
1.255 brouard 10402: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10403: } /* end of any combination */
10404: } /* end of nres */
1.218 brouard 10405: /* hBijx(p, bage, fage); */
10406: /* fclose(ficrespijb); */
10407:
10408: return 0;
1.217 brouard 10409: }
1.218 brouard 10410:
1.180 brouard 10411: int hPijx(double *p, int bage, int fage){
10412: /*------------- h Pij x at various ages ------------*/
10413:
10414: int stepsize;
10415: int agelim;
10416: int hstepm;
10417: int nhstepm;
1.235 brouard 10418: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10419:
10420: double agedeb;
10421: double ***p3mat;
10422:
1.201 brouard 10423: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10424: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10425: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10426: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10427: }
10428: printf("Computing pij: result on file '%s' \n", filerespij);
10429: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10430:
10431: stepsize=(int) (stepm+YEARM-1)/YEARM;
10432: /*if (stepm<=24) stepsize=2;*/
10433:
10434: agelim=AGESUP;
10435: hstepm=stepsize*YEARM; /* Every year of age */
10436: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10437:
1.180 brouard 10438: /* hstepm=1; aff par mois*/
10439: pstamp(ficrespij);
10440: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10441: i1= pow(2,cptcoveff);
1.218 brouard 10442: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10443: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10444: /* k=k+1; */
1.235 brouard 10445: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10446: for(k=1; k<=i1;k++){
1.253 brouard 10447: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10448: continue;
1.183 brouard 10449: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10450: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10451: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10452: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10453: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10454: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10455: }
1.183 brouard 10456: fprintf(ficrespij,"******\n");
10457:
10458: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10459: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10460: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10461:
10462: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10463:
1.183 brouard 10464: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10465: oldm=oldms;savm=savms;
1.235 brouard 10466: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10467: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10468: for(i=1; i<=nlstate;i++)
10469: for(j=1; j<=nlstate+ndeath;j++)
10470: fprintf(ficrespij," %1d-%1d",i,j);
10471: fprintf(ficrespij,"\n");
10472: for (h=0; h<=nhstepm; h++){
10473: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10474: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10475: for(i=1; i<=nlstate;i++)
10476: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10477: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10478: fprintf(ficrespij,"\n");
10479: }
1.183 brouard 10480: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10481: fprintf(ficrespij,"\n");
10482: }
1.180 brouard 10483: /*}*/
10484: }
1.218 brouard 10485: return 0;
1.180 brouard 10486: }
1.218 brouard 10487:
10488: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10489: /*------------- h Bij x at various ages ------------*/
10490:
10491: int stepsize;
1.218 brouard 10492: /* int agelim; */
10493: int ageminl;
1.217 brouard 10494: int hstepm;
10495: int nhstepm;
1.238 brouard 10496: int h, i, i1, j, k, nres;
1.218 brouard 10497:
1.217 brouard 10498: double agedeb;
10499: double ***p3mat;
1.218 brouard 10500:
10501: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10502: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10503: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10504: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10505: }
10506: printf("Computing pij back: result on file '%s' \n", filerespijb);
10507: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10508:
10509: stepsize=(int) (stepm+YEARM-1)/YEARM;
10510: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10511:
1.218 brouard 10512: /* agelim=AGESUP; */
10513: ageminl=30;
10514: hstepm=stepsize*YEARM; /* Every year of age */
10515: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10516:
10517: /* hstepm=1; aff par mois*/
10518: pstamp(ficrespijb);
1.255 brouard 10519: 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 10520: i1= pow(2,cptcoveff);
1.218 brouard 10521: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10522: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10523: /* k=k+1; */
1.238 brouard 10524: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10525: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10526: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10527: continue;
10528: fprintf(ficrespijb,"\n#****** ");
10529: for(j=1;j<=cptcoveff;j++)
10530: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10531: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10532: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10533: }
10534: fprintf(ficrespijb,"******\n");
1.264 brouard 10535: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10536: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10537: continue;
10538: }
10539:
10540: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10541: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10542: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10543: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10544: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10545:
10546: /* nhstepm=nhstepm*YEARM; aff par mois*/
10547:
1.266 brouard 10548: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10549: /* and memory limitations if stepm is small */
10550:
1.238 brouard 10551: /* oldm=oldms;savm=savms; */
10552: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10553: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10554: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10555: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10556: for(i=1; i<=nlstate;i++)
10557: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10558: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10559: fprintf(ficrespijb,"\n");
1.238 brouard 10560: for (h=0; h<=nhstepm; h++){
10561: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10562: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10563: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10564: for(i=1; i<=nlstate;i++)
10565: for(j=1; j<=nlstate+ndeath;j++)
10566: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10567: fprintf(ficrespijb,"\n");
10568: }
10569: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10570: fprintf(ficrespijb,"\n");
10571: } /* end age deb */
10572: } /* end combination */
10573: } /* end nres */
1.218 brouard 10574: return 0;
10575: } /* hBijx */
1.217 brouard 10576:
1.180 brouard 10577:
1.136 brouard 10578: /***********************************************/
10579: /**************** Main Program *****************/
10580: /***********************************************/
10581:
10582: int main(int argc, char *argv[])
10583: {
10584: #ifdef GSL
10585: const gsl_multimin_fminimizer_type *T;
10586: size_t iteri = 0, it;
10587: int rval = GSL_CONTINUE;
10588: int status = GSL_SUCCESS;
10589: double ssval;
10590: #endif
10591: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10592: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10593: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10594: int jj, ll, li, lj, lk;
1.136 brouard 10595: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10596: int num_filled;
1.136 brouard 10597: int itimes;
10598: int NDIM=2;
10599: int vpopbased=0;
1.235 brouard 10600: int nres=0;
1.258 brouard 10601: int endishere=0;
1.136 brouard 10602:
1.274 ! brouard 10603: int ncurrv=0; /* Temporary variable */
! 10604:
1.164 brouard 10605: char ca[32], cb[32];
1.136 brouard 10606: /* FILE *fichtm; *//* Html File */
10607: /* FILE *ficgp;*/ /*Gnuplot File */
10608: struct stat info;
1.191 brouard 10609: double agedeb=0.;
1.194 brouard 10610:
10611: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10612: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10613:
1.165 brouard 10614: double fret;
1.191 brouard 10615: double dum=0.; /* Dummy variable */
1.136 brouard 10616: double ***p3mat;
1.218 brouard 10617: /* double ***mobaverage; */
1.164 brouard 10618:
10619: char line[MAXLINE];
1.197 brouard 10620: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10621:
1.234 brouard 10622: char modeltemp[MAXLINE];
1.230 brouard 10623: char resultline[MAXLINE];
10624:
1.136 brouard 10625: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10626: char *tok, *val; /* pathtot */
1.136 brouard 10627: int firstobs=1, lastobs=10;
1.195 brouard 10628: int c, h , cpt, c2;
1.191 brouard 10629: int jl=0;
10630: int i1, j1, jk, stepsize=0;
1.194 brouard 10631: int count=0;
10632:
1.164 brouard 10633: int *tab;
1.136 brouard 10634: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10635: int backcast=0;
1.136 brouard 10636: int mobilav=0,popforecast=0;
1.191 brouard 10637: int hstepm=0, nhstepm=0;
1.136 brouard 10638: int agemortsup;
10639: float sumlpop=0.;
10640: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10641: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10642:
1.191 brouard 10643: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10644: double ftolpl=FTOL;
10645: double **prlim;
1.217 brouard 10646: double **bprlim;
1.136 brouard 10647: double ***param; /* Matrix of parameters */
1.251 brouard 10648: double ***paramstart; /* Matrix of starting parameter values */
10649: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10650: double **matcov; /* Matrix of covariance */
1.203 brouard 10651: double **hess; /* Hessian matrix */
1.136 brouard 10652: double ***delti3; /* Scale */
10653: double *delti; /* Scale */
10654: double ***eij, ***vareij;
10655: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10656:
1.136 brouard 10657: double *epj, vepp;
1.164 brouard 10658:
1.273 brouard 10659: double dateprev1, dateprev2;
10660: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0;
10661: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0;
1.217 brouard 10662:
1.136 brouard 10663: double **ximort;
1.145 brouard 10664: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10665: int *dcwave;
10666:
1.164 brouard 10667: char z[1]="c";
1.136 brouard 10668:
10669: /*char *strt;*/
10670: char strtend[80];
1.126 brouard 10671:
1.164 brouard 10672:
1.126 brouard 10673: /* setlocale (LC_ALL, ""); */
10674: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10675: /* textdomain (PACKAGE); */
10676: /* setlocale (LC_CTYPE, ""); */
10677: /* setlocale (LC_MESSAGES, ""); */
10678:
10679: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10680: rstart_time = time(NULL);
10681: /* (void) gettimeofday(&start_time,&tzp);*/
10682: start_time = *localtime(&rstart_time);
1.126 brouard 10683: curr_time=start_time;
1.157 brouard 10684: /*tml = *localtime(&start_time.tm_sec);*/
10685: /* strcpy(strstart,asctime(&tml)); */
10686: strcpy(strstart,asctime(&start_time));
1.126 brouard 10687:
10688: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10689: /* tp.tm_sec = tp.tm_sec +86400; */
10690: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10691: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10692: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10693: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10694: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10695: /* strt=asctime(&tmg); */
10696: /* printf("Time(after) =%s",strstart); */
10697: /* (void) time (&time_value);
10698: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10699: * tm = *localtime(&time_value);
10700: * strstart=asctime(&tm);
10701: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10702: */
10703:
10704: nberr=0; /* Number of errors and warnings */
10705: nbwarn=0;
1.184 brouard 10706: #ifdef WIN32
10707: _getcwd(pathcd, size);
10708: #else
1.126 brouard 10709: getcwd(pathcd, size);
1.184 brouard 10710: #endif
1.191 brouard 10711: syscompilerinfo(0);
1.196 brouard 10712: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10713: if(argc <=1){
10714: printf("\nEnter the parameter file name: ");
1.205 brouard 10715: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10716: printf("ERROR Empty parameter file name\n");
10717: goto end;
10718: }
1.126 brouard 10719: i=strlen(pathr);
10720: if(pathr[i-1]=='\n')
10721: pathr[i-1]='\0';
1.156 brouard 10722: i=strlen(pathr);
1.205 brouard 10723: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10724: pathr[i-1]='\0';
1.205 brouard 10725: }
10726: i=strlen(pathr);
10727: if( i==0 ){
10728: printf("ERROR Empty parameter file name\n");
10729: goto end;
10730: }
10731: for (tok = pathr; tok != NULL; ){
1.126 brouard 10732: printf("Pathr |%s|\n",pathr);
10733: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10734: printf("val= |%s| pathr=%s\n",val,pathr);
10735: strcpy (pathtot, val);
10736: if(pathr[0] == '\0') break; /* Dirty */
10737: }
10738: }
10739: else{
10740: strcpy(pathtot,argv[1]);
10741: }
10742: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10743: /*cygwin_split_path(pathtot,path,optionfile);
10744: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10745: /* cutv(path,optionfile,pathtot,'\\');*/
10746:
10747: /* Split argv[0], imach program to get pathimach */
10748: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10749: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10750: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10751: /* strcpy(pathimach,argv[0]); */
10752: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10753: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10754: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10755: #ifdef WIN32
10756: _chdir(path); /* Can be a relative path */
10757: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10758: #else
1.126 brouard 10759: chdir(path); /* Can be a relative path */
1.184 brouard 10760: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10761: #endif
10762: printf("Current directory %s!\n",pathcd);
1.126 brouard 10763: strcpy(command,"mkdir ");
10764: strcat(command,optionfilefiname);
10765: if((outcmd=system(command)) != 0){
1.169 brouard 10766: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10767: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10768: /* fclose(ficlog); */
10769: /* exit(1); */
10770: }
10771: /* if((imk=mkdir(optionfilefiname))<0){ */
10772: /* perror("mkdir"); */
10773: /* } */
10774:
10775: /*-------- arguments in the command line --------*/
10776:
1.186 brouard 10777: /* Main Log file */
1.126 brouard 10778: strcat(filelog, optionfilefiname);
10779: strcat(filelog,".log"); /* */
10780: if((ficlog=fopen(filelog,"w"))==NULL) {
10781: printf("Problem with logfile %s\n",filelog);
10782: goto end;
10783: }
10784: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10785: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10786: fprintf(ficlog,"\nEnter the parameter file name: \n");
10787: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10788: path=%s \n\
10789: optionfile=%s\n\
10790: optionfilext=%s\n\
1.156 brouard 10791: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10792:
1.197 brouard 10793: syscompilerinfo(1);
1.167 brouard 10794:
1.126 brouard 10795: printf("Local time (at start):%s",strstart);
10796: fprintf(ficlog,"Local time (at start): %s",strstart);
10797: fflush(ficlog);
10798: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10799: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10800:
10801: /* */
10802: strcpy(fileres,"r");
10803: strcat(fileres, optionfilefiname);
1.201 brouard 10804: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10805: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10806: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10807:
1.186 brouard 10808: /* Main ---------arguments file --------*/
1.126 brouard 10809:
10810: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10811: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10812: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10813: fflush(ficlog);
1.149 brouard 10814: /* goto end; */
10815: exit(70);
1.126 brouard 10816: }
10817:
10818:
10819:
10820: strcpy(filereso,"o");
1.201 brouard 10821: strcat(filereso,fileresu);
1.126 brouard 10822: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10823: printf("Problem with Output resultfile: %s\n", filereso);
10824: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10825: fflush(ficlog);
10826: goto end;
10827: }
10828:
10829: /* Reads comments: lines beginning with '#' */
10830: numlinepar=0;
1.197 brouard 10831:
10832: /* First parameter line */
10833: while(fgets(line, MAXLINE, ficpar)) {
10834: /* If line starts with a # it is a comment */
10835: if (line[0] == '#') {
10836: numlinepar++;
10837: fputs(line,stdout);
10838: fputs(line,ficparo);
10839: fputs(line,ficlog);
10840: continue;
10841: }else
10842: break;
10843: }
10844: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10845: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10846: if (num_filled != 5) {
10847: printf("Should be 5 parameters\n");
10848: }
1.126 brouard 10849: numlinepar++;
1.197 brouard 10850: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10851: }
10852: /* Second parameter line */
10853: while(fgets(line, MAXLINE, ficpar)) {
10854: /* If line starts with a # it is a comment */
10855: if (line[0] == '#') {
10856: numlinepar++;
10857: fputs(line,stdout);
10858: fputs(line,ficparo);
10859: fputs(line,ficlog);
10860: continue;
10861: }else
10862: break;
10863: }
1.223 brouard 10864: 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", \
10865: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10866: if (num_filled != 11) {
10867: 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 10868: printf("but line=%s\n",line);
1.197 brouard 10869: }
1.223 brouard 10870: 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 10871: }
1.203 brouard 10872: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10873: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10874: /* Third parameter line */
10875: while(fgets(line, MAXLINE, ficpar)) {
10876: /* If line starts with a # it is a comment */
10877: if (line[0] == '#') {
10878: numlinepar++;
10879: fputs(line,stdout);
10880: fputs(line,ficparo);
10881: fputs(line,ficlog);
10882: continue;
10883: }else
10884: break;
10885: }
1.201 brouard 10886: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 brouard 10887: if (num_filled == 0){
10888: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10889: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10890: model[0]='\0';
10891: goto end;
10892: } else if (num_filled != 1){
1.197 brouard 10893: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10894: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10895: model[0]='\0';
10896: goto end;
10897: }
10898: else{
10899: if (model[0]=='+'){
10900: for(i=1; i<=strlen(model);i++)
10901: modeltemp[i-1]=model[i];
1.201 brouard 10902: strcpy(model,modeltemp);
1.197 brouard 10903: }
10904: }
1.199 brouard 10905: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10906: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10907: }
10908: /* 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); */
10909: /* numlinepar=numlinepar+3; /\* In general *\/ */
10910: /* 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 10911: 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);
10912: 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 10913: fflush(ficlog);
1.190 brouard 10914: /* if(model[0]=='#'|| model[0]== '\0'){ */
10915: if(model[0]=='#'){
1.187 brouard 10916: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10917: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10918: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10919: if(mle != -1){
10920: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10921: exit(1);
10922: }
10923: }
1.126 brouard 10924: while((c=getc(ficpar))=='#' && c!= EOF){
10925: ungetc(c,ficpar);
10926: fgets(line, MAXLINE, ficpar);
10927: numlinepar++;
1.195 brouard 10928: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10929: z[0]=line[1];
10930: }
10931: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10932: fputs(line, stdout);
10933: //puts(line);
1.126 brouard 10934: fputs(line,ficparo);
10935: fputs(line,ficlog);
10936: }
10937: ungetc(c,ficpar);
10938:
10939:
1.145 brouard 10940: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.268 brouard 10941: if(nqv>=1)coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
10942: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
10943: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 10944: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10945: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10946: v1+v2*age+v2*v3 makes cptcovn = 3
10947: */
10948: if (strlen(model)>1)
1.187 brouard 10949: 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 10950: else
1.187 brouard 10951: ncovmodel=2; /* Constant and age */
1.133 brouard 10952: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10953: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10954: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10955: 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);
10956: 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);
10957: fflush(stdout);
10958: fclose (ficlog);
10959: goto end;
10960: }
1.126 brouard 10961: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10962: delti=delti3[1][1];
10963: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10964: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10965: /* We could also provide initial parameters values giving by simple logistic regression
10966: * only one way, that is without matrix product. We will have nlstate maximizations */
10967: /* for(i=1;i<nlstate;i++){ */
10968: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10969: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10970: /* } */
1.126 brouard 10971: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10972: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10973: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10974: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10975: fclose (ficparo);
10976: fclose (ficlog);
10977: goto end;
10978: exit(0);
1.220 brouard 10979: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10980: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10981: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10982: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10983: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10984: matcov=matrix(1,npar,1,npar);
1.203 brouard 10985: hess=matrix(1,npar,1,npar);
1.220 brouard 10986: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10987: /* Read guessed parameters */
1.126 brouard 10988: /* Reads comments: lines beginning with '#' */
10989: while((c=getc(ficpar))=='#' && c!= EOF){
10990: ungetc(c,ficpar);
10991: fgets(line, MAXLINE, ficpar);
10992: numlinepar++;
1.141 brouard 10993: fputs(line,stdout);
1.126 brouard 10994: fputs(line,ficparo);
10995: fputs(line,ficlog);
10996: }
10997: ungetc(c,ficpar);
10998:
10999: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11000: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11001: for(i=1; i <=nlstate; i++){
1.234 brouard 11002: j=0;
1.126 brouard 11003: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11004: if(jj==i) continue;
11005: j++;
11006: fscanf(ficpar,"%1d%1d",&i1,&j1);
11007: if ((i1 != i) || (j1 != jj)){
11008: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11009: It might be a problem of design; if ncovcol and the model are correct\n \
11010: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11011: exit(1);
11012: }
11013: fprintf(ficparo,"%1d%1d",i1,j1);
11014: if(mle==1)
11015: printf("%1d%1d",i,jj);
11016: fprintf(ficlog,"%1d%1d",i,jj);
11017: for(k=1; k<=ncovmodel;k++){
11018: fscanf(ficpar," %lf",¶m[i][j][k]);
11019: if(mle==1){
11020: printf(" %lf",param[i][j][k]);
11021: fprintf(ficlog," %lf",param[i][j][k]);
11022: }
11023: else
11024: fprintf(ficlog," %lf",param[i][j][k]);
11025: fprintf(ficparo," %lf",param[i][j][k]);
11026: }
11027: fscanf(ficpar,"\n");
11028: numlinepar++;
11029: if(mle==1)
11030: printf("\n");
11031: fprintf(ficlog,"\n");
11032: fprintf(ficparo,"\n");
1.126 brouard 11033: }
11034: }
11035: fflush(ficlog);
1.234 brouard 11036:
1.251 brouard 11037: /* Reads parameters values */
1.126 brouard 11038: p=param[1][1];
1.251 brouard 11039: pstart=paramstart[1][1];
1.126 brouard 11040:
11041: /* Reads comments: lines beginning with '#' */
11042: while((c=getc(ficpar))=='#' && c!= EOF){
11043: ungetc(c,ficpar);
11044: fgets(line, MAXLINE, ficpar);
11045: numlinepar++;
1.141 brouard 11046: fputs(line,stdout);
1.126 brouard 11047: fputs(line,ficparo);
11048: fputs(line,ficlog);
11049: }
11050: ungetc(c,ficpar);
11051:
11052: for(i=1; i <=nlstate; i++){
11053: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11054: fscanf(ficpar,"%1d%1d",&i1,&j1);
11055: if ( (i1-i) * (j1-j) != 0){
11056: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11057: exit(1);
11058: }
11059: printf("%1d%1d",i,j);
11060: fprintf(ficparo,"%1d%1d",i1,j1);
11061: fprintf(ficlog,"%1d%1d",i1,j1);
11062: for(k=1; k<=ncovmodel;k++){
11063: fscanf(ficpar,"%le",&delti3[i][j][k]);
11064: printf(" %le",delti3[i][j][k]);
11065: fprintf(ficparo," %le",delti3[i][j][k]);
11066: fprintf(ficlog," %le",delti3[i][j][k]);
11067: }
11068: fscanf(ficpar,"\n");
11069: numlinepar++;
11070: printf("\n");
11071: fprintf(ficparo,"\n");
11072: fprintf(ficlog,"\n");
1.126 brouard 11073: }
11074: }
11075: fflush(ficlog);
1.234 brouard 11076:
1.145 brouard 11077: /* Reads covariance matrix */
1.126 brouard 11078: delti=delti3[1][1];
1.220 brouard 11079:
11080:
1.126 brouard 11081: /* 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 11082:
1.126 brouard 11083: /* Reads comments: lines beginning with '#' */
11084: while((c=getc(ficpar))=='#' && c!= EOF){
11085: ungetc(c,ficpar);
11086: fgets(line, MAXLINE, ficpar);
11087: numlinepar++;
1.141 brouard 11088: fputs(line,stdout);
1.126 brouard 11089: fputs(line,ficparo);
11090: fputs(line,ficlog);
11091: }
11092: ungetc(c,ficpar);
1.220 brouard 11093:
1.126 brouard 11094: matcov=matrix(1,npar,1,npar);
1.203 brouard 11095: hess=matrix(1,npar,1,npar);
1.131 brouard 11096: for(i=1; i <=npar; i++)
11097: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11098:
1.194 brouard 11099: /* Scans npar lines */
1.126 brouard 11100: for(i=1; i <=npar; i++){
1.226 brouard 11101: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11102: if(count != 3){
1.226 brouard 11103: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11104: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11105: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11106: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11107: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11108: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11109: exit(1);
1.220 brouard 11110: }else{
1.226 brouard 11111: if(mle==1)
11112: printf("%1d%1d%d",i1,j1,jk);
11113: }
11114: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11115: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11116: for(j=1; j <=i; j++){
1.226 brouard 11117: fscanf(ficpar," %le",&matcov[i][j]);
11118: if(mle==1){
11119: printf(" %.5le",matcov[i][j]);
11120: }
11121: fprintf(ficlog," %.5le",matcov[i][j]);
11122: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11123: }
11124: fscanf(ficpar,"\n");
11125: numlinepar++;
11126: if(mle==1)
1.220 brouard 11127: printf("\n");
1.126 brouard 11128: fprintf(ficlog,"\n");
11129: fprintf(ficparo,"\n");
11130: }
1.194 brouard 11131: /* End of read covariance matrix npar lines */
1.126 brouard 11132: for(i=1; i <=npar; i++)
11133: for(j=i+1;j<=npar;j++)
1.226 brouard 11134: matcov[i][j]=matcov[j][i];
1.126 brouard 11135:
11136: if(mle==1)
11137: printf("\n");
11138: fprintf(ficlog,"\n");
11139:
11140: fflush(ficlog);
11141:
11142: /*-------- Rewriting parameter file ----------*/
11143: strcpy(rfileres,"r"); /* "Rparameterfile */
11144: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
11145: strcat(rfileres,"."); /* */
11146: strcat(rfileres,optionfilext); /* Other files have txt extension */
11147: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 11148: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11149: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 11150: }
11151: fprintf(ficres,"#%s\n",version);
11152: } /* End of mle != -3 */
1.218 brouard 11153:
1.186 brouard 11154: /* Main data
11155: */
1.126 brouard 11156: n= lastobs;
11157: num=lvector(1,n);
11158: moisnais=vector(1,n);
11159: annais=vector(1,n);
11160: moisdc=vector(1,n);
11161: andc=vector(1,n);
1.220 brouard 11162: weight=vector(1,n);
1.126 brouard 11163: agedc=vector(1,n);
11164: cod=ivector(1,n);
1.220 brouard 11165: for(i=1;i<=n;i++){
1.234 brouard 11166: num[i]=0;
11167: moisnais[i]=0;
11168: annais[i]=0;
11169: moisdc[i]=0;
11170: andc[i]=0;
11171: agedc[i]=0;
11172: cod[i]=0;
11173: weight[i]=1.0; /* Equal weights, 1 by default */
11174: }
1.126 brouard 11175: mint=matrix(1,maxwav,1,n);
11176: anint=matrix(1,maxwav,1,n);
1.131 brouard 11177: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11178: tab=ivector(1,NCOVMAX);
1.144 brouard 11179: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11180: 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 11181:
1.136 brouard 11182: /* Reads data from file datafile */
11183: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11184: goto end;
11185:
11186: /* Calculation of the number of parameters from char model */
1.234 brouard 11187: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11188: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11189: k=3 V4 Tvar[k=3]= 4 (from V4)
11190: k=2 V1 Tvar[k=2]= 1 (from V1)
11191: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11192: */
11193:
11194: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11195: TvarsDind=ivector(1,NCOVMAX); /* */
11196: TvarsD=ivector(1,NCOVMAX); /* */
11197: TvarsQind=ivector(1,NCOVMAX); /* */
11198: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11199: TvarF=ivector(1,NCOVMAX); /* */
11200: TvarFind=ivector(1,NCOVMAX); /* */
11201: TvarV=ivector(1,NCOVMAX); /* */
11202: TvarVind=ivector(1,NCOVMAX); /* */
11203: TvarA=ivector(1,NCOVMAX); /* */
11204: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11205: TvarFD=ivector(1,NCOVMAX); /* */
11206: TvarFDind=ivector(1,NCOVMAX); /* */
11207: TvarFQ=ivector(1,NCOVMAX); /* */
11208: TvarFQind=ivector(1,NCOVMAX); /* */
11209: TvarVD=ivector(1,NCOVMAX); /* */
11210: TvarVDind=ivector(1,NCOVMAX); /* */
11211: TvarVQ=ivector(1,NCOVMAX); /* */
11212: TvarVQind=ivector(1,NCOVMAX); /* */
11213:
1.230 brouard 11214: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11215: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11216: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11217: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11218: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11219: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11220: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11221: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11222: */
11223: /* For model-covariate k tells which data-covariate to use but
11224: because this model-covariate is a construction we invent a new column
11225: ncovcol + k1
11226: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11227: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11228: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11229: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11230: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11231: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11232: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11233: */
1.145 brouard 11234: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11235: 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 11236: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11237: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11238: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11239: 4 covariates (3 plus signs)
11240: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11241: */
1.230 brouard 11242: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11243: * individual dummy, fixed or varying:
11244: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11245: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11246: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11247: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11248: * Tmodelind[1]@9={9,0,3,2,}*/
11249: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11250: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11251: * individual quantitative, fixed or varying:
11252: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11253: * 3, 1, 0, 0, 0, 0, 0, 0},
11254: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11255: /* Main decodemodel */
11256:
1.187 brouard 11257:
1.223 brouard 11258: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11259: goto end;
11260:
1.137 brouard 11261: if((double)(lastobs-imx)/(double)imx > 1.10){
11262: nbwarn++;
11263: 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);
11264: 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);
11265: }
1.136 brouard 11266: /* if(mle==1){*/
1.137 brouard 11267: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11268: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11269: }
11270:
11271: /*-calculation of age at interview from date of interview and age at death -*/
11272: agev=matrix(1,maxwav,1,imx);
11273:
11274: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11275: goto end;
11276:
1.126 brouard 11277:
1.136 brouard 11278: agegomp=(int)agemin;
11279: free_vector(moisnais,1,n);
11280: free_vector(annais,1,n);
1.126 brouard 11281: /* free_matrix(mint,1,maxwav,1,n);
11282: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11283: /* free_vector(moisdc,1,n); */
11284: /* free_vector(andc,1,n); */
1.145 brouard 11285: /* */
11286:
1.126 brouard 11287: wav=ivector(1,imx);
1.214 brouard 11288: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11289: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11290: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11291: 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.*/
11292: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11293: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11294:
11295: /* Concatenates waves */
1.214 brouard 11296: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11297: Death is a valid wave (if date is known).
11298: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11299: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11300: and mw[mi+1][i]. dh depends on stepm.
11301: */
11302:
1.126 brouard 11303: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11304: /* Concatenates waves */
1.145 brouard 11305:
1.215 brouard 11306: free_vector(moisdc,1,n);
11307: free_vector(andc,1,n);
11308:
1.126 brouard 11309: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11310: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11311: ncodemax[1]=1;
1.145 brouard 11312: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11313: cptcoveff=0;
1.220 brouard 11314: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11315: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11316: }
11317:
11318: ncovcombmax=pow(2,cptcoveff);
11319: invalidvarcomb=ivector(1, ncovcombmax);
11320: for(i=1;i<ncovcombmax;i++)
11321: invalidvarcomb[i]=0;
11322:
1.211 brouard 11323: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11324: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11325: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11326:
1.200 brouard 11327: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11328: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11329: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11330: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11331: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11332: * (currently 0 or 1) in the data.
11333: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11334: * corresponding modality (h,j).
11335: */
11336:
1.145 brouard 11337: h=0;
11338: /*if (cptcovn > 0) */
1.126 brouard 11339: m=pow(2,cptcoveff);
11340:
1.144 brouard 11341: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11342: * For k=4 covariates, h goes from 1 to m=2**k
11343: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11344: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11345: * h\k 1 2 3 4
1.143 brouard 11346: *______________________________
11347: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11348: * 2 2 1 1 1
11349: * 3 i=2 1 2 1 1
11350: * 4 2 2 1 1
11351: * 5 i=3 1 i=2 1 2 1
11352: * 6 2 1 2 1
11353: * 7 i=4 1 2 2 1
11354: * 8 2 2 2 1
1.197 brouard 11355: * 9 i=5 1 i=3 1 i=2 1 2
11356: * 10 2 1 1 2
11357: * 11 i=6 1 2 1 2
11358: * 12 2 2 1 2
11359: * 13 i=7 1 i=4 1 2 2
11360: * 14 2 1 2 2
11361: * 15 i=8 1 2 2 2
11362: * 16 2 2 2 2
1.143 brouard 11363: */
1.212 brouard 11364: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11365: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11366: * and the value of each covariate?
11367: * V1=1, V2=1, V3=2, V4=1 ?
11368: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11369: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11370: * In order to get the real value in the data, we use nbcode
11371: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11372: * We are keeping this crazy system in order to be able (in the future?)
11373: * to have more than 2 values (0 or 1) for a covariate.
11374: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11375: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11376: * bbbbbbbb
11377: * 76543210
11378: * h-1 00000101 (6-1=5)
1.219 brouard 11379: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11380: * &
11381: * 1 00000001 (1)
1.219 brouard 11382: * 00000000 = 1 & ((h-1) >> (k-1))
11383: * +1= 00000001 =1
1.211 brouard 11384: *
11385: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11386: * h' 1101 =2^3+2^2+0x2^1+2^0
11387: * >>k' 11
11388: * & 00000001
11389: * = 00000001
11390: * +1 = 00000010=2 = codtabm(14,3)
11391: * Reverse h=6 and m=16?
11392: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11393: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11394: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11395: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11396: * V3=decodtabm(14,3,2**4)=2
11397: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11398: *(h-1) >> (j-1) 0011 =13 >> 2
11399: * &1 000000001
11400: * = 000000001
11401: * +1= 000000010 =2
11402: * 2211
11403: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11404: * V3=2
1.220 brouard 11405: * codtabm and decodtabm are identical
1.211 brouard 11406: */
11407:
1.145 brouard 11408:
11409: free_ivector(Ndum,-1,NCOVMAX);
11410:
11411:
1.126 brouard 11412:
1.186 brouard 11413: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11414: strcpy(optionfilegnuplot,optionfilefiname);
11415: if(mle==-3)
1.201 brouard 11416: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11417: strcat(optionfilegnuplot,".gp");
11418:
11419: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11420: printf("Problem with file %s",optionfilegnuplot);
11421: }
11422: else{
1.204 brouard 11423: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11424: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11425: //fprintf(ficgp,"set missing 'NaNq'\n");
11426: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11427: }
11428: /* fclose(ficgp);*/
1.186 brouard 11429:
11430:
11431: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11432:
11433: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11434: if(mle==-3)
1.201 brouard 11435: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11436: strcat(optionfilehtm,".htm");
11437: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11438: printf("Problem with %s \n",optionfilehtm);
11439: exit(0);
1.126 brouard 11440: }
11441:
11442: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11443: strcat(optionfilehtmcov,"-cov.htm");
11444: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11445: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11446: }
11447: else{
11448: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11449: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11450: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11451: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11452: }
11453:
1.213 brouard 11454: 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 11455: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11456: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11457: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11458: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11459: \n\
11460: <hr size=\"2\" color=\"#EC5E5E\">\
11461: <ul><li><h4>Parameter files</h4>\n\
11462: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11463: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11464: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11465: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11466: - Date and time at start: %s</ul>\n",\
11467: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11468: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11469: fileres,fileres,\
11470: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11471: fflush(fichtm);
11472:
11473: strcpy(pathr,path);
11474: strcat(pathr,optionfilefiname);
1.184 brouard 11475: #ifdef WIN32
11476: _chdir(optionfilefiname); /* Move to directory named optionfile */
11477: #else
1.126 brouard 11478: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11479: #endif
11480:
1.126 brouard 11481:
1.220 brouard 11482: /* Calculates basic frequencies. Computes observed prevalence at single age
11483: and for any valid combination of covariates
1.126 brouard 11484: and prints on file fileres'p'. */
1.251 brouard 11485: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11486: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11487:
11488: fprintf(fichtm,"\n");
1.274 ! brouard 11489: 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",\
! 11490: ftol, stepm);
! 11491: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
! 11492: ncurrv=1;
! 11493: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
! 11494: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
! 11495: ncurrv=i;
! 11496: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
! 11497: fprintf(fichtm,"\n<li> Number of time varying (wave varying) covariates: ntv=%d ", ntv);
! 11498: ncurrv=i;
! 11499: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
! 11500: fprintf(fichtm,"\n<li>Number of quantitative time varying covariates: nqtv=%d ", nqtv);
! 11501: ncurrv=i;
! 11502: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
! 11503: 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", \
! 11504: nlstate, ndeath, maxwav, mle, weightopt);
! 11505:
! 11506: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
! 11507: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
! 11508:
! 11509:
! 11510: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11511: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11512: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 ! brouard 11513: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11514: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11515: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11516: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11517: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11518: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11519:
1.126 brouard 11520: /* For Powell, parameters are in a vector p[] starting at p[1]
11521: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11522: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11523:
11524: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11525: /* For mortality only */
1.126 brouard 11526: if (mle==-3){
1.136 brouard 11527: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11528: for(i=1;i<=NDIM;i++)
11529: for(j=1;j<=NDIM;j++)
11530: ximort[i][j]=0.;
1.186 brouard 11531: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11532: cens=ivector(1,n);
11533: ageexmed=vector(1,n);
11534: agecens=vector(1,n);
11535: dcwave=ivector(1,n);
1.223 brouard 11536:
1.126 brouard 11537: for (i=1; i<=imx; i++){
11538: dcwave[i]=-1;
11539: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11540: if (s[m][i]>nlstate) {
11541: dcwave[i]=m;
11542: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11543: break;
11544: }
1.126 brouard 11545: }
1.226 brouard 11546:
1.126 brouard 11547: for (i=1; i<=imx; i++) {
11548: if (wav[i]>0){
1.226 brouard 11549: ageexmed[i]=agev[mw[1][i]][i];
11550: j=wav[i];
11551: agecens[i]=1.;
11552:
11553: if (ageexmed[i]> 1 && wav[i] > 0){
11554: agecens[i]=agev[mw[j][i]][i];
11555: cens[i]= 1;
11556: }else if (ageexmed[i]< 1)
11557: cens[i]= -1;
11558: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11559: cens[i]=0 ;
1.126 brouard 11560: }
11561: else cens[i]=-1;
11562: }
11563:
11564: for (i=1;i<=NDIM;i++) {
11565: for (j=1;j<=NDIM;j++)
1.226 brouard 11566: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11567: }
11568:
1.145 brouard 11569: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11570: /*printf("%lf %lf", p[1], p[2]);*/
11571:
11572:
1.136 brouard 11573: #ifdef GSL
11574: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11575: #else
1.126 brouard 11576: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11577: #endif
1.201 brouard 11578: strcpy(filerespow,"POW-MORT_");
11579: strcat(filerespow,fileresu);
1.126 brouard 11580: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11581: printf("Problem with resultfile: %s\n", filerespow);
11582: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11583: }
1.136 brouard 11584: #ifdef GSL
11585: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11586: #else
1.126 brouard 11587: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11588: #endif
1.126 brouard 11589: /* for (i=1;i<=nlstate;i++)
11590: for(j=1;j<=nlstate+ndeath;j++)
11591: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11592: */
11593: fprintf(ficrespow,"\n");
1.136 brouard 11594: #ifdef GSL
11595: /* gsl starts here */
11596: T = gsl_multimin_fminimizer_nmsimplex;
11597: gsl_multimin_fminimizer *sfm = NULL;
11598: gsl_vector *ss, *x;
11599: gsl_multimin_function minex_func;
11600:
11601: /* Initial vertex size vector */
11602: ss = gsl_vector_alloc (NDIM);
11603:
11604: if (ss == NULL){
11605: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11606: }
11607: /* Set all step sizes to 1 */
11608: gsl_vector_set_all (ss, 0.001);
11609:
11610: /* Starting point */
1.126 brouard 11611:
1.136 brouard 11612: x = gsl_vector_alloc (NDIM);
11613:
11614: if (x == NULL){
11615: gsl_vector_free(ss);
11616: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11617: }
11618:
11619: /* Initialize method and iterate */
11620: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11621: /* gsl_vector_set(x, 0, 0.0268); */
11622: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11623: gsl_vector_set(x, 0, p[1]);
11624: gsl_vector_set(x, 1, p[2]);
11625:
11626: minex_func.f = &gompertz_f;
11627: minex_func.n = NDIM;
11628: minex_func.params = (void *)&p; /* ??? */
11629:
11630: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11631: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11632:
11633: printf("Iterations beginning .....\n\n");
11634: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11635:
11636: iteri=0;
11637: while (rval == GSL_CONTINUE){
11638: iteri++;
11639: status = gsl_multimin_fminimizer_iterate(sfm);
11640:
11641: if (status) printf("error: %s\n", gsl_strerror (status));
11642: fflush(0);
11643:
11644: if (status)
11645: break;
11646:
11647: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11648: ssval = gsl_multimin_fminimizer_size (sfm);
11649:
11650: if (rval == GSL_SUCCESS)
11651: printf ("converged to a local maximum at\n");
11652:
11653: printf("%5d ", iteri);
11654: for (it = 0; it < NDIM; it++){
11655: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11656: }
11657: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11658: }
11659:
11660: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11661:
11662: gsl_vector_free(x); /* initial values */
11663: gsl_vector_free(ss); /* inital step size */
11664: for (it=0; it<NDIM; it++){
11665: p[it+1]=gsl_vector_get(sfm->x,it);
11666: fprintf(ficrespow," %.12lf", p[it]);
11667: }
11668: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11669: #endif
11670: #ifdef POWELL
11671: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11672: #endif
1.126 brouard 11673: fclose(ficrespow);
11674:
1.203 brouard 11675: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11676:
11677: for(i=1; i <=NDIM; i++)
11678: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11679: matcov[i][j]=matcov[j][i];
1.126 brouard 11680:
11681: printf("\nCovariance matrix\n ");
1.203 brouard 11682: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11683: for(i=1; i <=NDIM; i++) {
11684: for(j=1;j<=NDIM;j++){
1.220 brouard 11685: printf("%f ",matcov[i][j]);
11686: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11687: }
1.203 brouard 11688: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11689: }
11690:
11691: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11692: for (i=1;i<=NDIM;i++) {
1.126 brouard 11693: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11694: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11695: }
1.126 brouard 11696: lsurv=vector(1,AGESUP);
11697: lpop=vector(1,AGESUP);
11698: tpop=vector(1,AGESUP);
11699: lsurv[agegomp]=100000;
11700:
11701: for (k=agegomp;k<=AGESUP;k++) {
11702: agemortsup=k;
11703: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11704: }
11705:
11706: for (k=agegomp;k<agemortsup;k++)
11707: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11708:
11709: for (k=agegomp;k<agemortsup;k++){
11710: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11711: sumlpop=sumlpop+lpop[k];
11712: }
11713:
11714: tpop[agegomp]=sumlpop;
11715: for (k=agegomp;k<(agemortsup-3);k++){
11716: /* tpop[k+1]=2;*/
11717: tpop[k+1]=tpop[k]-lpop[k];
11718: }
11719:
11720:
11721: printf("\nAge lx qx dx Lx Tx e(x)\n");
11722: for (k=agegomp;k<(agemortsup-2);k++)
11723: 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]);
11724:
11725:
11726: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11727: ageminpar=50;
11728: agemaxpar=100;
1.194 brouard 11729: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11730: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11731: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11732: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11733: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11734: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11735: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11736: }else{
11737: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11738: 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 11739: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11740: }
1.201 brouard 11741: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11742: stepm, weightopt,\
11743: model,imx,p,matcov,agemortsup);
11744:
11745: free_vector(lsurv,1,AGESUP);
11746: free_vector(lpop,1,AGESUP);
11747: free_vector(tpop,1,AGESUP);
1.220 brouard 11748: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11749: free_ivector(cens,1,n);
11750: free_vector(agecens,1,n);
11751: free_ivector(dcwave,1,n);
1.220 brouard 11752: #ifdef GSL
1.136 brouard 11753: #endif
1.186 brouard 11754: } /* Endof if mle==-3 mortality only */
1.205 brouard 11755: /* Standard */
11756: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11757: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11758: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11759: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11760: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11761: for (k=1; k<=npar;k++)
11762: printf(" %d %8.5f",k,p[k]);
11763: printf("\n");
1.205 brouard 11764: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11765: /* mlikeli uses func not funcone */
1.247 brouard 11766: /* for(i=1;i<nlstate;i++){ */
11767: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11768: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11769: /* } */
1.205 brouard 11770: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11771: }
11772: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11773: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11774: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11775: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11776: }
11777: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11778: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11779: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11780: for (k=1; k<=npar;k++)
11781: printf(" %d %8.5f",k,p[k]);
11782: printf("\n");
11783:
11784: /*--------- results files --------------*/
1.224 brouard 11785: 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 11786:
11787:
11788: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11789: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11790: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11791: for(i=1,jk=1; i <=nlstate; i++){
11792: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11793: if (k != i) {
11794: printf("%d%d ",i,k);
11795: fprintf(ficlog,"%d%d ",i,k);
11796: fprintf(ficres,"%1d%1d ",i,k);
11797: for(j=1; j <=ncovmodel; j++){
11798: printf("%12.7f ",p[jk]);
11799: fprintf(ficlog,"%12.7f ",p[jk]);
11800: fprintf(ficres,"%12.7f ",p[jk]);
11801: jk++;
11802: }
11803: printf("\n");
11804: fprintf(ficlog,"\n");
11805: fprintf(ficres,"\n");
11806: }
1.126 brouard 11807: }
11808: }
1.203 brouard 11809: if(mle != 0){
11810: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11811: ftolhess=ftol; /* Usually correct */
1.203 brouard 11812: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11813: 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");
11814: 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");
11815: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11816: for(k=1; k <=(nlstate+ndeath); k++){
11817: if (k != i) {
11818: printf("%d%d ",i,k);
11819: fprintf(ficlog,"%d%d ",i,k);
11820: for(j=1; j <=ncovmodel; j++){
11821: 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]));
11822: 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]));
11823: jk++;
11824: }
11825: printf("\n");
11826: fprintf(ficlog,"\n");
11827: }
11828: }
1.193 brouard 11829: }
1.203 brouard 11830: } /* end of hesscov and Wald tests */
1.225 brouard 11831:
1.203 brouard 11832: /* */
1.126 brouard 11833: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11834: printf("# Scales (for hessian or gradient estimation)\n");
11835: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11836: for(i=1,jk=1; i <=nlstate; i++){
11837: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11838: if (j!=i) {
11839: fprintf(ficres,"%1d%1d",i,j);
11840: printf("%1d%1d",i,j);
11841: fprintf(ficlog,"%1d%1d",i,j);
11842: for(k=1; k<=ncovmodel;k++){
11843: printf(" %.5e",delti[jk]);
11844: fprintf(ficlog," %.5e",delti[jk]);
11845: fprintf(ficres," %.5e",delti[jk]);
11846: jk++;
11847: }
11848: printf("\n");
11849: fprintf(ficlog,"\n");
11850: fprintf(ficres,"\n");
11851: }
1.126 brouard 11852: }
11853: }
11854:
11855: 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 11856: if(mle >= 1) /* To big for the screen */
1.126 brouard 11857: 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");
11858: 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");
11859: /* # 121 Var(a12)\n\ */
11860: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11861: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11862: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11863: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11864: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11865: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11866: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11867:
11868:
11869: /* Just to have a covariance matrix which will be more understandable
11870: even is we still don't want to manage dictionary of variables
11871: */
11872: for(itimes=1;itimes<=2;itimes++){
11873: jj=0;
11874: for(i=1; i <=nlstate; i++){
1.225 brouard 11875: for(j=1; j <=nlstate+ndeath; j++){
11876: if(j==i) continue;
11877: for(k=1; k<=ncovmodel;k++){
11878: jj++;
11879: ca[0]= k+'a'-1;ca[1]='\0';
11880: if(itimes==1){
11881: if(mle>=1)
11882: printf("#%1d%1d%d",i,j,k);
11883: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11884: fprintf(ficres,"#%1d%1d%d",i,j,k);
11885: }else{
11886: if(mle>=1)
11887: printf("%1d%1d%d",i,j,k);
11888: fprintf(ficlog,"%1d%1d%d",i,j,k);
11889: fprintf(ficres,"%1d%1d%d",i,j,k);
11890: }
11891: ll=0;
11892: for(li=1;li <=nlstate; li++){
11893: for(lj=1;lj <=nlstate+ndeath; lj++){
11894: if(lj==li) continue;
11895: for(lk=1;lk<=ncovmodel;lk++){
11896: ll++;
11897: if(ll<=jj){
11898: cb[0]= lk +'a'-1;cb[1]='\0';
11899: if(ll<jj){
11900: if(itimes==1){
11901: if(mle>=1)
11902: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11903: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11904: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11905: }else{
11906: if(mle>=1)
11907: printf(" %.5e",matcov[jj][ll]);
11908: fprintf(ficlog," %.5e",matcov[jj][ll]);
11909: fprintf(ficres," %.5e",matcov[jj][ll]);
11910: }
11911: }else{
11912: if(itimes==1){
11913: if(mle>=1)
11914: printf(" Var(%s%1d%1d)",ca,i,j);
11915: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11916: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11917: }else{
11918: if(mle>=1)
11919: printf(" %.7e",matcov[jj][ll]);
11920: fprintf(ficlog," %.7e",matcov[jj][ll]);
11921: fprintf(ficres," %.7e",matcov[jj][ll]);
11922: }
11923: }
11924: }
11925: } /* end lk */
11926: } /* end lj */
11927: } /* end li */
11928: if(mle>=1)
11929: printf("\n");
11930: fprintf(ficlog,"\n");
11931: fprintf(ficres,"\n");
11932: numlinepar++;
11933: } /* end k*/
11934: } /*end j */
1.126 brouard 11935: } /* end i */
11936: } /* end itimes */
11937:
11938: fflush(ficlog);
11939: fflush(ficres);
1.225 brouard 11940: while(fgets(line, MAXLINE, ficpar)) {
11941: /* If line starts with a # it is a comment */
11942: if (line[0] == '#') {
11943: numlinepar++;
11944: fputs(line,stdout);
11945: fputs(line,ficparo);
11946: fputs(line,ficlog);
11947: continue;
11948: }else
11949: break;
11950: }
11951:
1.209 brouard 11952: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11953: /* ungetc(c,ficpar); */
11954: /* fgets(line, MAXLINE, ficpar); */
11955: /* fputs(line,stdout); */
11956: /* fputs(line,ficparo); */
11957: /* } */
11958: /* ungetc(c,ficpar); */
1.126 brouard 11959:
11960: estepm=0;
1.209 brouard 11961: 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 11962:
11963: if (num_filled != 6) {
11964: 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);
11965: 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);
11966: goto end;
11967: }
11968: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11969: }
11970: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11971: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11972:
1.209 brouard 11973: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11974: if (estepm==0 || estepm < stepm) estepm=stepm;
11975: if (fage <= 2) {
11976: bage = ageminpar;
11977: fage = agemaxpar;
11978: }
11979:
11980: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11981: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11982: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11983:
1.186 brouard 11984: /* Other stuffs, more or less useful */
1.254 brouard 11985: while(fgets(line, MAXLINE, ficpar)) {
11986: /* If line starts with a # it is a comment */
11987: if (line[0] == '#') {
11988: numlinepar++;
11989: fputs(line,stdout);
11990: fputs(line,ficparo);
11991: fputs(line,ficlog);
11992: continue;
11993: }else
11994: break;
11995: }
11996:
11997: 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){
11998:
11999: if (num_filled != 7) {
12000: 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);
12001: 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);
12002: goto end;
12003: }
12004: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12005: 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);
12006: 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);
12007: 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 12008: }
1.254 brouard 12009:
12010: while(fgets(line, MAXLINE, ficpar)) {
12011: /* If line starts with a # it is a comment */
12012: if (line[0] == '#') {
12013: numlinepar++;
12014: fputs(line,stdout);
12015: fputs(line,ficparo);
12016: fputs(line,ficlog);
12017: continue;
12018: }else
12019: break;
1.126 brouard 12020: }
12021:
12022:
12023: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12024: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12025:
1.254 brouard 12026: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12027: if (num_filled != 1) {
12028: 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);
12029: 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);
12030: goto end;
12031: }
12032: printf("pop_based=%d\n",popbased);
12033: fprintf(ficlog,"pop_based=%d\n",popbased);
12034: fprintf(ficparo,"pop_based=%d\n",popbased);
12035: fprintf(ficres,"pop_based=%d\n",popbased);
12036: }
12037:
1.258 brouard 12038: /* Results */
12039: nresult=0;
12040: do{
12041: if(!fgets(line, MAXLINE, ficpar)){
12042: endishere=1;
12043: parameterline=14;
12044: }else if (line[0] == '#') {
12045: /* If line starts with a # it is a comment */
1.254 brouard 12046: numlinepar++;
12047: fputs(line,stdout);
12048: fputs(line,ficparo);
12049: fputs(line,ficlog);
12050: continue;
1.258 brouard 12051: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12052: parameterline=11;
12053: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
12054: parameterline=12;
12055: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12056: parameterline=13;
12057: else{
12058: parameterline=14;
1.254 brouard 12059: }
1.258 brouard 12060: switch (parameterline){
12061: case 11:
12062: 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){
12063: if (num_filled != 8) {
12064: 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);
12065: 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);
12066: goto end;
12067: }
12068: 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);
12069: 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);
12070: 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);
12071: 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);
12072: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12073: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12074: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
12075:
1.258 brouard 12076: }
1.254 brouard 12077: break;
1.258 brouard 12078: case 12:
12079: /*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);*/
12080: 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){
12081: if (num_filled != 8) {
1.262 brouard 12082: 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);
12083: 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 12084: goto end;
12085: }
12086: 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);
12087: 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);
12088: 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);
12089: 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);
12090: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12091: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12092: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.258 brouard 12093: }
1.230 brouard 12094: break;
1.258 brouard 12095: case 13:
12096: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12097: if (num_filled == 0){
12098: resultline[0]='\0';
12099: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12100: 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);
12101: break;
12102: } else if (num_filled != 1){
12103: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12104: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12105: }
12106: nresult++; /* Sum of resultlines */
12107: printf("Result %d: result=%s\n",nresult, resultline);
12108: if(nresult > MAXRESULTLINES){
12109: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12110: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12111: goto end;
12112: }
12113: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12114: fprintf(ficparo,"result: %s\n",resultline);
12115: fprintf(ficres,"result: %s\n",resultline);
12116: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12117: break;
1.258 brouard 12118: case 14:
1.259 brouard 12119: if(ncovmodel >2 && nresult==0 ){
12120: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12121: goto end;
12122: }
1.259 brouard 12123: break;
1.258 brouard 12124: default:
12125: nresult=1;
12126: decoderesult(".",nresult ); /* No covariate */
12127: }
12128: } /* End switch parameterline */
12129: }while(endishere==0); /* End do */
1.126 brouard 12130:
1.230 brouard 12131: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12132: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12133:
12134: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12135: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12136: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12137: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12138: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12139: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12140: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12141: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12142: }else{
1.270 brouard 12143: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
12144: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, backcast, pathc,p, (int)anproj1-bage, (int)anback1-fage);
1.220 brouard 12145: }
12146: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 12147: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.273 brouard 12148: jprev1,mprev1,anprev1,dateprev1, dateproj1, dateback1,jprev2,mprev2,anprev2,dateprev2,dateproj2, dateback2);
1.220 brouard 12149:
1.225 brouard 12150: /*------------ free_vector -------------*/
12151: /* chdir(path); */
1.220 brouard 12152:
1.215 brouard 12153: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12154: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12155: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12156: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 12157: free_lvector(num,1,n);
12158: free_vector(agedc,1,n);
12159: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12160: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12161: fclose(ficparo);
12162: fclose(ficres);
1.220 brouard 12163:
12164:
1.186 brouard 12165: /* Other results (useful)*/
1.220 brouard 12166:
12167:
1.126 brouard 12168: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12169: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12170: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12171: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12172: fclose(ficrespl);
12173:
12174: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12175: /*#include "hpijx.h"*/
12176: hPijx(p, bage, fage);
1.145 brouard 12177: fclose(ficrespij);
1.227 brouard 12178:
1.220 brouard 12179: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12180: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12181: k=1;
1.126 brouard 12182: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12183:
1.269 brouard 12184: /* Prevalence for each covariate combination in probs[age][status][cov] */
12185: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12186: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12187: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12188: for(k=1;k<=ncovcombmax;k++)
12189: probs[i][j][k]=0.;
1.269 brouard 12190: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12191: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12192: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12193: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12194: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12195: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12196: for(k=1;k<=ncovcombmax;k++)
12197: mobaverages[i][j][k]=0.;
1.219 brouard 12198: mobaverage=mobaverages;
12199: if (mobilav!=0) {
1.235 brouard 12200: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12201: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12202: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12203: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12204: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12205: }
1.269 brouard 12206: } else if (mobilavproj !=0) {
1.235 brouard 12207: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12208: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12209: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12210: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12211: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12212: }
1.269 brouard 12213: }else{
12214: printf("Internal error moving average\n");
12215: fflush(stdout);
12216: exit(1);
1.219 brouard 12217: }
12218: }/* end if moving average */
1.227 brouard 12219:
1.126 brouard 12220: /*---------- Forecasting ------------------*/
12221: if(prevfcast==1){
12222: /* if(stepm ==1){*/
1.269 brouard 12223: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 12224: }
1.269 brouard 12225:
12226: /* Backcasting */
1.217 brouard 12227: if(backcast==1){
1.219 brouard 12228: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12229: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12230: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12231:
12232: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12233:
12234: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12235:
1.219 brouard 12236: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12237: fclose(ficresplb);
12238:
1.222 brouard 12239: hBijx(p, bage, fage, mobaverage);
12240: fclose(ficrespijb);
1.219 brouard 12241:
1.269 brouard 12242: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2,
12243: mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
12244: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12245:
12246:
1.269 brouard 12247: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12248: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12249: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12250: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.269 brouard 12251: } /* end Backcasting */
1.268 brouard 12252:
1.186 brouard 12253:
12254: /* ------ Other prevalence ratios------------ */
1.126 brouard 12255:
1.215 brouard 12256: free_ivector(wav,1,imx);
12257: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12258: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12259: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12260:
12261:
1.127 brouard 12262: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12263:
1.201 brouard 12264: strcpy(filerese,"E_");
12265: strcat(filerese,fileresu);
1.126 brouard 12266: if((ficreseij=fopen(filerese,"w"))==NULL) {
12267: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12268: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12269: }
1.208 brouard 12270: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12271: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12272:
12273: pstamp(ficreseij);
1.219 brouard 12274:
1.235 brouard 12275: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12276: if (cptcovn < 1){i1=1;}
12277:
12278: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12279: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12280: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12281: continue;
1.219 brouard 12282: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12283: printf("\n#****** ");
1.225 brouard 12284: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12285: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12286: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12287: }
12288: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12289: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12290: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12291: }
12292: fprintf(ficreseij,"******\n");
1.235 brouard 12293: printf("******\n");
1.219 brouard 12294:
12295: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12296: oldm=oldms;savm=savms;
1.235 brouard 12297: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12298:
1.219 brouard 12299: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12300: }
12301: fclose(ficreseij);
1.208 brouard 12302: printf("done evsij\n");fflush(stdout);
12303: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12304:
1.218 brouard 12305:
1.227 brouard 12306: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12307:
1.201 brouard 12308: strcpy(filerest,"T_");
12309: strcat(filerest,fileresu);
1.127 brouard 12310: if((ficrest=fopen(filerest,"w"))==NULL) {
12311: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12312: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12313: }
1.208 brouard 12314: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12315: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12316: strcpy(fileresstde,"STDE_");
12317: strcat(fileresstde,fileresu);
1.126 brouard 12318: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12319: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12320: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12321: }
1.227 brouard 12322: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12323: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12324:
1.201 brouard 12325: strcpy(filerescve,"CVE_");
12326: strcat(filerescve,fileresu);
1.126 brouard 12327: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12328: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12329: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12330: }
1.227 brouard 12331: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12332: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12333:
1.201 brouard 12334: strcpy(fileresv,"V_");
12335: strcat(fileresv,fileresu);
1.126 brouard 12336: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12337: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12338: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12339: }
1.227 brouard 12340: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12341: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12342:
1.235 brouard 12343: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12344: if (cptcovn < 1){i1=1;}
12345:
12346: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12347: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12348: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12349: continue;
1.242 brouard 12350: printf("\n#****** Result for:");
12351: fprintf(ficrest,"\n#****** Result for:");
12352: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12353: for(j=1;j<=cptcoveff;j++){
12354: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12355: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12356: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12357: }
1.235 brouard 12358: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12359: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12360: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12361: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12362: }
1.208 brouard 12363: fprintf(ficrest,"******\n");
1.227 brouard 12364: fprintf(ficlog,"******\n");
12365: printf("******\n");
1.208 brouard 12366:
12367: fprintf(ficresstdeij,"\n#****** ");
12368: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12369: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12370: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12371: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12372: }
1.235 brouard 12373: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12374: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12375: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12376: }
1.208 brouard 12377: fprintf(ficresstdeij,"******\n");
12378: fprintf(ficrescveij,"******\n");
12379:
12380: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12381: /* pstamp(ficresvij); */
1.225 brouard 12382: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12383: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12384: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12385: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12386: }
1.208 brouard 12387: fprintf(ficresvij,"******\n");
12388:
12389: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12390: oldm=oldms;savm=savms;
1.235 brouard 12391: printf(" cvevsij ");
12392: fprintf(ficlog, " cvevsij ");
12393: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12394: printf(" end cvevsij \n ");
12395: fprintf(ficlog, " end cvevsij \n ");
12396:
12397: /*
12398: */
12399: /* goto endfree; */
12400:
12401: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12402: pstamp(ficrest);
12403:
1.269 brouard 12404: epj=vector(1,nlstate+1);
1.208 brouard 12405: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12406: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12407: cptcod= 0; /* To be deleted */
12408: printf("varevsij vpopbased=%d \n",vpopbased);
12409: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12410: 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 12411: 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 ");
12412: if(vpopbased==1)
12413: 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);
12414: else
12415: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
12416: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12417: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12418: fprintf(ficrest,"\n");
12419: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
12420: printf("Computing age specific period (stable) prevalences in each health state \n");
12421: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
12422: for(age=bage; age <=fage ;age++){
1.235 brouard 12423: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12424: if (vpopbased==1) {
12425: if(mobilav ==0){
12426: for(i=1; i<=nlstate;i++)
12427: prlim[i][i]=probs[(int)age][i][k];
12428: }else{ /* mobilav */
12429: for(i=1; i<=nlstate;i++)
12430: prlim[i][i]=mobaverage[(int)age][i][k];
12431: }
12432: }
1.219 brouard 12433:
1.227 brouard 12434: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12435: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12436: /* printf(" age %4.0f ",age); */
12437: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12438: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12439: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12440: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12441: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12442: }
12443: epj[nlstate+1] +=epj[j];
12444: }
12445: /* printf(" age %4.0f \n",age); */
1.219 brouard 12446:
1.227 brouard 12447: for(i=1, vepp=0.;i <=nlstate;i++)
12448: for(j=1;j <=nlstate;j++)
12449: vepp += vareij[i][j][(int)age];
12450: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12451: for(j=1;j <=nlstate;j++){
12452: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12453: }
12454: fprintf(ficrest,"\n");
12455: }
1.208 brouard 12456: } /* End vpopbased */
1.269 brouard 12457: free_vector(epj,1,nlstate+1);
1.208 brouard 12458: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12459: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12460: printf("done selection\n");fflush(stdout);
12461: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12462:
1.235 brouard 12463: } /* End k selection */
1.227 brouard 12464:
12465: printf("done State-specific expectancies\n");fflush(stdout);
12466: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12467:
1.269 brouard 12468: /* variance-covariance of period prevalence*/
12469: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12470:
1.227 brouard 12471:
12472: free_vector(weight,1,n);
12473: free_imatrix(Tvard,1,NCOVMAX,1,2);
12474: free_imatrix(s,1,maxwav+1,1,n);
12475: free_matrix(anint,1,maxwav,1,n);
12476: free_matrix(mint,1,maxwav,1,n);
12477: free_ivector(cod,1,n);
12478: free_ivector(tab,1,NCOVMAX);
12479: fclose(ficresstdeij);
12480: fclose(ficrescveij);
12481: fclose(ficresvij);
12482: fclose(ficrest);
12483: fclose(ficpar);
12484:
12485:
1.126 brouard 12486: /*---------- End : free ----------------*/
1.219 brouard 12487: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12488: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12489: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12490: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12491: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12492: } /* mle==-3 arrives here for freeing */
1.227 brouard 12493: /* endfree:*/
12494: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12495: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12496: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.268 brouard 12497: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
12498: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
12499: if(nqv>=1)free_matrix(coqvar,1,nqv,1,n);
1.227 brouard 12500: free_matrix(covar,0,NCOVMAX,1,n);
12501: free_matrix(matcov,1,npar,1,npar);
12502: free_matrix(hess,1,npar,1,npar);
12503: /*free_vector(delti,1,npar);*/
12504: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12505: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12506: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12507: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12508:
12509: free_ivector(ncodemax,1,NCOVMAX);
12510: free_ivector(ncodemaxwundef,1,NCOVMAX);
12511: free_ivector(Dummy,-1,NCOVMAX);
12512: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12513: free_ivector(DummyV,1,NCOVMAX);
12514: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12515: free_ivector(Typevar,-1,NCOVMAX);
12516: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12517: free_ivector(TvarsQ,1,NCOVMAX);
12518: free_ivector(TvarsQind,1,NCOVMAX);
12519: free_ivector(TvarsD,1,NCOVMAX);
12520: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12521: free_ivector(TvarFD,1,NCOVMAX);
12522: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12523: free_ivector(TvarF,1,NCOVMAX);
12524: free_ivector(TvarFind,1,NCOVMAX);
12525: free_ivector(TvarV,1,NCOVMAX);
12526: free_ivector(TvarVind,1,NCOVMAX);
12527: free_ivector(TvarA,1,NCOVMAX);
12528: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12529: free_ivector(TvarFQ,1,NCOVMAX);
12530: free_ivector(TvarFQind,1,NCOVMAX);
12531: free_ivector(TvarVD,1,NCOVMAX);
12532: free_ivector(TvarVDind,1,NCOVMAX);
12533: free_ivector(TvarVQ,1,NCOVMAX);
12534: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12535: free_ivector(Tvarsel,1,NCOVMAX);
12536: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12537: free_ivector(Tposprod,1,NCOVMAX);
12538: free_ivector(Tprod,1,NCOVMAX);
12539: free_ivector(Tvaraff,1,NCOVMAX);
12540: free_ivector(invalidvarcomb,1,ncovcombmax);
12541: free_ivector(Tage,1,NCOVMAX);
12542: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12543: free_ivector(TmodelInvind,1,NCOVMAX);
12544: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12545:
12546: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12547: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12548: fflush(fichtm);
12549: fflush(ficgp);
12550:
1.227 brouard 12551:
1.126 brouard 12552: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12553: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12554: 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 12555: }else{
12556: printf("End of Imach\n");
12557: fprintf(ficlog,"End of Imach\n");
12558: }
12559: printf("See log file on %s\n",filelog);
12560: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12561: /*(void) gettimeofday(&end_time,&tzp);*/
12562: rend_time = time(NULL);
12563: end_time = *localtime(&rend_time);
12564: /* tml = *localtime(&end_time.tm_sec); */
12565: strcpy(strtend,asctime(&end_time));
1.126 brouard 12566: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12567: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12568: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12569:
1.157 brouard 12570: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12571: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12572: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12573: /* printf("Total time was %d uSec.\n", total_usecs);*/
12574: /* if(fileappend(fichtm,optionfilehtm)){ */
12575: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12576: fclose(fichtm);
12577: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12578: fclose(fichtmcov);
12579: fclose(ficgp);
12580: fclose(ficlog);
12581: /*------ End -----------*/
1.227 brouard 12582:
12583:
12584: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12585: #ifdef WIN32
1.227 brouard 12586: if (_chdir(pathcd) != 0)
12587: printf("Can't move to directory %s!\n",path);
12588: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12589: #else
1.227 brouard 12590: if(chdir(pathcd) != 0)
12591: printf("Can't move to directory %s!\n", path);
12592: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12593: #endif
1.126 brouard 12594: printf("Current directory %s!\n",pathcd);
12595: /*strcat(plotcmd,CHARSEPARATOR);*/
12596: sprintf(plotcmd,"gnuplot");
1.157 brouard 12597: #ifdef _WIN32
1.126 brouard 12598: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12599: #endif
12600: if(!stat(plotcmd,&info)){
1.158 brouard 12601: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12602: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12603: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12604: }else
12605: strcpy(pplotcmd,plotcmd);
1.157 brouard 12606: #ifdef __unix
1.126 brouard 12607: strcpy(plotcmd,GNUPLOTPROGRAM);
12608: if(!stat(plotcmd,&info)){
1.158 brouard 12609: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12610: }else
12611: strcpy(pplotcmd,plotcmd);
12612: #endif
12613: }else
12614: strcpy(pplotcmd,plotcmd);
12615:
12616: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12617: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12618:
1.126 brouard 12619: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12620: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12621: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12622: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12623: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12624: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12625: }
1.158 brouard 12626: printf(" Successful, please wait...");
1.126 brouard 12627: while (z[0] != 'q') {
12628: /* chdir(path); */
1.154 brouard 12629: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12630: scanf("%s",z);
12631: /* if (z[0] == 'c') system("./imach"); */
12632: if (z[0] == 'e') {
1.158 brouard 12633: #ifdef __APPLE__
1.152 brouard 12634: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12635: #elif __linux
12636: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12637: #else
1.152 brouard 12638: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12639: #endif
12640: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12641: system(pplotcmd);
1.126 brouard 12642: }
12643: else if (z[0] == 'g') system(plotcmd);
12644: else if (z[0] == 'q') exit(0);
12645: }
1.227 brouard 12646: end:
1.126 brouard 12647: while (z[0] != 'q') {
1.195 brouard 12648: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12649: scanf("%s",z);
12650: }
12651: }
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