Annotation of imach/src/imach.c, revision 1.256
1.256 ! brouard 1: /* $Id: imach.c,v 1.255 2017/03/08 16:02:28 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.256 ! brouard 4: Revision 1.255 2017/03/08 16:02:28 brouard
! 5: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
! 6:
1.255 brouard 7: Revision 1.254 2017/03/08 07:13:00 brouard
8: Summary: Fixing data parameter line
9:
1.254 brouard 10: Revision 1.253 2016/12/15 11:59:41 brouard
11: Summary: 0.99 in progress
12:
1.253 brouard 13: Revision 1.252 2016/09/15 21:15:37 brouard
14: *** empty log message ***
15:
1.252 brouard 16: Revision 1.251 2016/09/15 15:01:13 brouard
17: Summary: not working
18:
1.251 brouard 19: Revision 1.250 2016/09/08 16:07:27 brouard
20: Summary: continue
21:
1.250 brouard 22: Revision 1.249 2016/09/07 17:14:18 brouard
23: Summary: Starting values from frequencies
24:
1.249 brouard 25: Revision 1.248 2016/09/07 14:10:18 brouard
26: *** empty log message ***
27:
1.248 brouard 28: Revision 1.247 2016/09/02 11:11:21 brouard
29: *** empty log message ***
30:
1.247 brouard 31: Revision 1.246 2016/09/02 08:49:22 brouard
32: *** empty log message ***
33:
1.246 brouard 34: Revision 1.245 2016/09/02 07:25:01 brouard
35: *** empty log message ***
36:
1.245 brouard 37: Revision 1.244 2016/09/02 07:17:34 brouard
38: *** empty log message ***
39:
1.244 brouard 40: Revision 1.243 2016/09/02 06:45:35 brouard
41: *** empty log message ***
42:
1.243 brouard 43: Revision 1.242 2016/08/30 15:01:20 brouard
44: Summary: Fixing a lots
45:
1.242 brouard 46: Revision 1.241 2016/08/29 17:17:25 brouard
47: Summary: gnuplot problem in Back projection to fix
48:
1.241 brouard 49: Revision 1.240 2016/08/29 07:53:18 brouard
50: Summary: Better
51:
1.240 brouard 52: Revision 1.239 2016/08/26 15:51:03 brouard
53: Summary: Improvement in Powell output in order to copy and paste
54:
55: Author:
56:
1.239 brouard 57: Revision 1.238 2016/08/26 14:23:35 brouard
58: Summary: Starting tests of 0.99
59:
1.238 brouard 60: Revision 1.237 2016/08/26 09:20:19 brouard
61: Summary: to valgrind
62:
1.237 brouard 63: Revision 1.236 2016/08/25 10:50:18 brouard
64: *** empty log message ***
65:
1.236 brouard 66: Revision 1.235 2016/08/25 06:59:23 brouard
67: *** empty log message ***
68:
1.235 brouard 69: Revision 1.234 2016/08/23 16:51:20 brouard
70: *** empty log message ***
71:
1.234 brouard 72: Revision 1.233 2016/08/23 07:40:50 brouard
73: Summary: not working
74:
1.233 brouard 75: Revision 1.232 2016/08/22 14:20:21 brouard
76: Summary: not working
77:
1.232 brouard 78: Revision 1.231 2016/08/22 07:17:15 brouard
79: Summary: not working
80:
1.231 brouard 81: Revision 1.230 2016/08/22 06:55:53 brouard
82: Summary: Not working
83:
1.230 brouard 84: Revision 1.229 2016/07/23 09:45:53 brouard
85: Summary: Completing for func too
86:
1.229 brouard 87: Revision 1.228 2016/07/22 17:45:30 brouard
88: Summary: Fixing some arrays, still debugging
89:
1.227 brouard 90: Revision 1.226 2016/07/12 18:42:34 brouard
91: Summary: temp
92:
1.226 brouard 93: Revision 1.225 2016/07/12 08:40:03 brouard
94: Summary: saving but not running
95:
1.225 brouard 96: Revision 1.224 2016/07/01 13:16:01 brouard
97: Summary: Fixes
98:
1.224 brouard 99: Revision 1.223 2016/02/19 09:23:35 brouard
100: Summary: temporary
101:
1.223 brouard 102: Revision 1.222 2016/02/17 08:14:50 brouard
103: Summary: Probably last 0.98 stable version 0.98r6
104:
1.222 brouard 105: Revision 1.221 2016/02/15 23:35:36 brouard
106: Summary: minor bug
107:
1.220 brouard 108: Revision 1.219 2016/02/15 00:48:12 brouard
109: *** empty log message ***
110:
1.219 brouard 111: Revision 1.218 2016/02/12 11:29:23 brouard
112: Summary: 0.99 Back projections
113:
1.218 brouard 114: Revision 1.217 2015/12/23 17:18:31 brouard
115: Summary: Experimental backcast
116:
1.217 brouard 117: Revision 1.216 2015/12/18 17:32:11 brouard
118: Summary: 0.98r4 Warning and status=-2
119:
120: Version 0.98r4 is now:
121: - displaying an error when status is -1, date of interview unknown and date of death known;
122: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
123: Older changes concerning s=-2, dating from 2005 have been supersed.
124:
1.216 brouard 125: Revision 1.215 2015/12/16 08:52:24 brouard
126: Summary: 0.98r4 working
127:
1.215 brouard 128: Revision 1.214 2015/12/16 06:57:54 brouard
129: Summary: temporary not working
130:
1.214 brouard 131: Revision 1.213 2015/12/11 18:22:17 brouard
132: Summary: 0.98r4
133:
1.213 brouard 134: Revision 1.212 2015/11/21 12:47:24 brouard
135: Summary: minor typo
136:
1.212 brouard 137: Revision 1.211 2015/11/21 12:41:11 brouard
138: Summary: 0.98r3 with some graph of projected cross-sectional
139:
140: Author: Nicolas Brouard
141:
1.211 brouard 142: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 143: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 144: Summary: Adding ftolpl parameter
145: Author: N Brouard
146:
147: We had difficulties to get smoothed confidence intervals. It was due
148: to the period prevalence which wasn't computed accurately. The inner
149: parameter ftolpl is now an outer parameter of the .imach parameter
150: file after estepm. If ftolpl is small 1.e-4 and estepm too,
151: computation are long.
152:
1.209 brouard 153: Revision 1.208 2015/11/17 14:31:57 brouard
154: Summary: temporary
155:
1.208 brouard 156: Revision 1.207 2015/10/27 17:36:57 brouard
157: *** empty log message ***
158:
1.207 brouard 159: Revision 1.206 2015/10/24 07:14:11 brouard
160: *** empty log message ***
161:
1.206 brouard 162: Revision 1.205 2015/10/23 15:50:53 brouard
163: Summary: 0.98r3 some clarification for graphs on likelihood contributions
164:
1.205 brouard 165: Revision 1.204 2015/10/01 16:20:26 brouard
166: Summary: Some new graphs of contribution to likelihood
167:
1.204 brouard 168: Revision 1.203 2015/09/30 17:45:14 brouard
169: Summary: looking at better estimation of the hessian
170:
171: Also a better criteria for convergence to the period prevalence And
172: therefore adding the number of years needed to converge. (The
173: prevalence in any alive state shold sum to one
174:
1.203 brouard 175: Revision 1.202 2015/09/22 19:45:16 brouard
176: Summary: Adding some overall graph on contribution to likelihood. Might change
177:
1.202 brouard 178: Revision 1.201 2015/09/15 17:34:58 brouard
179: Summary: 0.98r0
180:
181: - Some new graphs like suvival functions
182: - Some bugs fixed like model=1+age+V2.
183:
1.201 brouard 184: Revision 1.200 2015/09/09 16:53:55 brouard
185: Summary: Big bug thanks to Flavia
186:
187: Even model=1+age+V2. did not work anymore
188:
1.200 brouard 189: Revision 1.199 2015/09/07 14:09:23 brouard
190: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
191:
1.199 brouard 192: Revision 1.198 2015/09/03 07:14:39 brouard
193: Summary: 0.98q5 Flavia
194:
1.198 brouard 195: Revision 1.197 2015/09/01 18:24:39 brouard
196: *** empty log message ***
197:
1.197 brouard 198: Revision 1.196 2015/08/18 23:17:52 brouard
199: Summary: 0.98q5
200:
1.196 brouard 201: Revision 1.195 2015/08/18 16:28:39 brouard
202: Summary: Adding a hack for testing purpose
203:
204: After reading the title, ftol and model lines, if the comment line has
205: a q, starting with #q, the answer at the end of the run is quit. It
206: permits to run test files in batch with ctest. The former workaround was
207: $ echo q | imach foo.imach
208:
1.195 brouard 209: Revision 1.194 2015/08/18 13:32:00 brouard
210: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
211:
1.194 brouard 212: Revision 1.193 2015/08/04 07:17:42 brouard
213: Summary: 0.98q4
214:
1.193 brouard 215: Revision 1.192 2015/07/16 16:49:02 brouard
216: Summary: Fixing some outputs
217:
1.192 brouard 218: Revision 1.191 2015/07/14 10:00:33 brouard
219: Summary: Some fixes
220:
1.191 brouard 221: Revision 1.190 2015/05/05 08:51:13 brouard
222: Summary: Adding digits in output parameters (7 digits instead of 6)
223:
224: Fix 1+age+.
225:
1.190 brouard 226: Revision 1.189 2015/04/30 14:45:16 brouard
227: Summary: 0.98q2
228:
1.189 brouard 229: Revision 1.188 2015/04/30 08:27:53 brouard
230: *** empty log message ***
231:
1.188 brouard 232: Revision 1.187 2015/04/29 09:11:15 brouard
233: *** empty log message ***
234:
1.187 brouard 235: Revision 1.186 2015/04/23 12:01:52 brouard
236: Summary: V1*age is working now, version 0.98q1
237:
238: Some codes had been disabled in order to simplify and Vn*age was
239: working in the optimization phase, ie, giving correct MLE parameters,
240: but, as usual, outputs were not correct and program core dumped.
241:
1.186 brouard 242: Revision 1.185 2015/03/11 13:26:42 brouard
243: Summary: Inclusion of compile and links command line for Intel Compiler
244:
1.185 brouard 245: Revision 1.184 2015/03/11 11:52:39 brouard
246: Summary: Back from Windows 8. Intel Compiler
247:
1.184 brouard 248: Revision 1.183 2015/03/10 20:34:32 brouard
249: Summary: 0.98q0, trying with directest, mnbrak fixed
250:
251: We use directest instead of original Powell test; probably no
252: incidence on the results, but better justifications;
253: We fixed Numerical Recipes mnbrak routine which was wrong and gave
254: wrong results.
255:
1.183 brouard 256: Revision 1.182 2015/02/12 08:19:57 brouard
257: Summary: Trying to keep directest which seems simpler and more general
258: Author: Nicolas Brouard
259:
1.182 brouard 260: Revision 1.181 2015/02/11 23:22:24 brouard
261: Summary: Comments on Powell added
262:
263: Author:
264:
1.181 brouard 265: Revision 1.180 2015/02/11 17:33:45 brouard
266: Summary: Finishing move from main to function (hpijx and prevalence_limit)
267:
1.180 brouard 268: Revision 1.179 2015/01/04 09:57:06 brouard
269: Summary: back to OS/X
270:
1.179 brouard 271: Revision 1.178 2015/01/04 09:35:48 brouard
272: *** empty log message ***
273:
1.178 brouard 274: Revision 1.177 2015/01/03 18:40:56 brouard
275: Summary: Still testing ilc32 on OSX
276:
1.177 brouard 277: Revision 1.176 2015/01/03 16:45:04 brouard
278: *** empty log message ***
279:
1.176 brouard 280: Revision 1.175 2015/01/03 16:33:42 brouard
281: *** empty log message ***
282:
1.175 brouard 283: Revision 1.174 2015/01/03 16:15:49 brouard
284: Summary: Still in cross-compilation
285:
1.174 brouard 286: Revision 1.173 2015/01/03 12:06:26 brouard
287: Summary: trying to detect cross-compilation
288:
1.173 brouard 289: Revision 1.172 2014/12/27 12:07:47 brouard
290: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
291:
1.172 brouard 292: Revision 1.171 2014/12/23 13:26:59 brouard
293: Summary: Back from Visual C
294:
295: Still problem with utsname.h on Windows
296:
1.171 brouard 297: Revision 1.170 2014/12/23 11:17:12 brouard
298: Summary: Cleaning some \%% back to %%
299:
300: The escape was mandatory for a specific compiler (which one?), but too many warnings.
301:
1.170 brouard 302: Revision 1.169 2014/12/22 23:08:31 brouard
303: Summary: 0.98p
304:
305: Outputs some informations on compiler used, OS etc. Testing on different platforms.
306:
1.169 brouard 307: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 308: Summary: update
1.169 brouard 309:
1.168 brouard 310: Revision 1.167 2014/12/22 13:50:56 brouard
311: Summary: Testing uname and compiler version and if compiled 32 or 64
312:
313: Testing on Linux 64
314:
1.167 brouard 315: Revision 1.166 2014/12/22 11:40:47 brouard
316: *** empty log message ***
317:
1.166 brouard 318: Revision 1.165 2014/12/16 11:20:36 brouard
319: Summary: After compiling on Visual C
320:
321: * imach.c (Module): Merging 1.61 to 1.162
322:
1.165 brouard 323: Revision 1.164 2014/12/16 10:52:11 brouard
324: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
325:
326: * imach.c (Module): Merging 1.61 to 1.162
327:
1.164 brouard 328: Revision 1.163 2014/12/16 10:30:11 brouard
329: * imach.c (Module): Merging 1.61 to 1.162
330:
1.163 brouard 331: Revision 1.162 2014/09/25 11:43:39 brouard
332: Summary: temporary backup 0.99!
333:
1.162 brouard 334: Revision 1.1 2014/09/16 11:06:58 brouard
335: Summary: With some code (wrong) for nlopt
336:
337: Author:
338:
339: Revision 1.161 2014/09/15 20:41:41 brouard
340: Summary: Problem with macro SQR on Intel compiler
341:
1.161 brouard 342: Revision 1.160 2014/09/02 09:24:05 brouard
343: *** empty log message ***
344:
1.160 brouard 345: Revision 1.159 2014/09/01 10:34:10 brouard
346: Summary: WIN32
347: Author: Brouard
348:
1.159 brouard 349: Revision 1.158 2014/08/27 17:11:51 brouard
350: *** empty log message ***
351:
1.158 brouard 352: Revision 1.157 2014/08/27 16:26:55 brouard
353: Summary: Preparing windows Visual studio version
354: Author: Brouard
355:
356: In order to compile on Visual studio, time.h is now correct and time_t
357: and tm struct should be used. difftime should be used but sometimes I
358: just make the differences in raw time format (time(&now).
359: Trying to suppress #ifdef LINUX
360: Add xdg-open for __linux in order to open default browser.
361:
1.157 brouard 362: Revision 1.156 2014/08/25 20:10:10 brouard
363: *** empty log message ***
364:
1.156 brouard 365: Revision 1.155 2014/08/25 18:32:34 brouard
366: Summary: New compile, minor changes
367: Author: Brouard
368:
1.155 brouard 369: Revision 1.154 2014/06/20 17:32:08 brouard
370: Summary: Outputs now all graphs of convergence to period prevalence
371:
1.154 brouard 372: Revision 1.153 2014/06/20 16:45:46 brouard
373: Summary: If 3 live state, convergence to period prevalence on same graph
374: Author: Brouard
375:
1.153 brouard 376: Revision 1.152 2014/06/18 17:54:09 brouard
377: Summary: open browser, use gnuplot on same dir than imach if not found in the path
378:
1.152 brouard 379: Revision 1.151 2014/06/18 16:43:30 brouard
380: *** empty log message ***
381:
1.151 brouard 382: Revision 1.150 2014/06/18 16:42:35 brouard
383: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
384: Author: brouard
385:
1.150 brouard 386: Revision 1.149 2014/06/18 15:51:14 brouard
387: Summary: Some fixes in parameter files errors
388: Author: Nicolas Brouard
389:
1.149 brouard 390: Revision 1.148 2014/06/17 17:38:48 brouard
391: Summary: Nothing new
392: Author: Brouard
393:
394: Just a new packaging for OS/X version 0.98nS
395:
1.148 brouard 396: Revision 1.147 2014/06/16 10:33:11 brouard
397: *** empty log message ***
398:
1.147 brouard 399: Revision 1.146 2014/06/16 10:20:28 brouard
400: Summary: Merge
401: Author: Brouard
402:
403: Merge, before building revised version.
404:
1.146 brouard 405: Revision 1.145 2014/06/10 21:23:15 brouard
406: Summary: Debugging with valgrind
407: Author: Nicolas Brouard
408:
409: Lot of changes in order to output the results with some covariates
410: After the Edimburgh REVES conference 2014, it seems mandatory to
411: improve the code.
412: No more memory valgrind error but a lot has to be done in order to
413: continue the work of splitting the code into subroutines.
414: Also, decodemodel has been improved. Tricode is still not
415: optimal. nbcode should be improved. Documentation has been added in
416: the source code.
417:
1.144 brouard 418: Revision 1.143 2014/01/26 09:45:38 brouard
419: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
420:
421: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
422: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
423:
1.143 brouard 424: Revision 1.142 2014/01/26 03:57:36 brouard
425: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
426:
427: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
428:
1.142 brouard 429: Revision 1.141 2014/01/26 02:42:01 brouard
430: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
431:
1.141 brouard 432: Revision 1.140 2011/09/02 10:37:54 brouard
433: Summary: times.h is ok with mingw32 now.
434:
1.140 brouard 435: Revision 1.139 2010/06/14 07:50:17 brouard
436: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
437: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
438:
1.139 brouard 439: Revision 1.138 2010/04/30 18:19:40 brouard
440: *** empty log message ***
441:
1.138 brouard 442: Revision 1.137 2010/04/29 18:11:38 brouard
443: (Module): Checking covariates for more complex models
444: than V1+V2. A lot of change to be done. Unstable.
445:
1.137 brouard 446: Revision 1.136 2010/04/26 20:30:53 brouard
447: (Module): merging some libgsl code. Fixing computation
448: of likelione (using inter/intrapolation if mle = 0) in order to
449: get same likelihood as if mle=1.
450: Some cleaning of code and comments added.
451:
1.136 brouard 452: Revision 1.135 2009/10/29 15:33:14 brouard
453: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
454:
1.135 brouard 455: Revision 1.134 2009/10/29 13:18:53 brouard
456: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
457:
1.134 brouard 458: Revision 1.133 2009/07/06 10:21:25 brouard
459: just nforces
460:
1.133 brouard 461: Revision 1.132 2009/07/06 08:22:05 brouard
462: Many tings
463:
1.132 brouard 464: Revision 1.131 2009/06/20 16:22:47 brouard
465: Some dimensions resccaled
466:
1.131 brouard 467: Revision 1.130 2009/05/26 06:44:34 brouard
468: (Module): Max Covariate is now set to 20 instead of 8. A
469: lot of cleaning with variables initialized to 0. Trying to make
470: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
471:
1.130 brouard 472: Revision 1.129 2007/08/31 13:49:27 lievre
473: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
474:
1.129 lievre 475: Revision 1.128 2006/06/30 13:02:05 brouard
476: (Module): Clarifications on computing e.j
477:
1.128 brouard 478: Revision 1.127 2006/04/28 18:11:50 brouard
479: (Module): Yes the sum of survivors was wrong since
480: imach-114 because nhstepm was no more computed in the age
481: loop. Now we define nhstepma in the age loop.
482: (Module): In order to speed up (in case of numerous covariates) we
483: compute health expectancies (without variances) in a first step
484: and then all the health expectancies with variances or standard
485: deviation (needs data from the Hessian matrices) which slows the
486: computation.
487: In the future we should be able to stop the program is only health
488: expectancies and graph are needed without standard deviations.
489:
1.127 brouard 490: Revision 1.126 2006/04/28 17:23:28 brouard
491: (Module): Yes the sum of survivors was wrong since
492: imach-114 because nhstepm was no more computed in the age
493: loop. Now we define nhstepma in the age loop.
494: Version 0.98h
495:
1.126 brouard 496: Revision 1.125 2006/04/04 15:20:31 lievre
497: Errors in calculation of health expectancies. Age was not initialized.
498: Forecasting file added.
499:
500: Revision 1.124 2006/03/22 17:13:53 lievre
501: Parameters are printed with %lf instead of %f (more numbers after the comma).
502: The log-likelihood is printed in the log file
503:
504: Revision 1.123 2006/03/20 10:52:43 brouard
505: * imach.c (Module): <title> changed, corresponds to .htm file
506: name. <head> headers where missing.
507:
508: * imach.c (Module): Weights can have a decimal point as for
509: English (a comma might work with a correct LC_NUMERIC environment,
510: otherwise the weight is truncated).
511: Modification of warning when the covariates values are not 0 or
512: 1.
513: Version 0.98g
514:
515: Revision 1.122 2006/03/20 09:45:41 brouard
516: (Module): Weights can have a decimal point as for
517: English (a comma might work with a correct LC_NUMERIC environment,
518: otherwise the weight is truncated).
519: Modification of warning when the covariates values are not 0 or
520: 1.
521: Version 0.98g
522:
523: Revision 1.121 2006/03/16 17:45:01 lievre
524: * imach.c (Module): Comments concerning covariates added
525:
526: * imach.c (Module): refinements in the computation of lli if
527: status=-2 in order to have more reliable computation if stepm is
528: not 1 month. Version 0.98f
529:
530: Revision 1.120 2006/03/16 15:10:38 lievre
531: (Module): refinements in the computation of lli if
532: status=-2 in order to have more reliable computation if stepm is
533: not 1 month. Version 0.98f
534:
535: Revision 1.119 2006/03/15 17:42:26 brouard
536: (Module): Bug if status = -2, the loglikelihood was
537: computed as likelihood omitting the logarithm. Version O.98e
538:
539: Revision 1.118 2006/03/14 18:20:07 brouard
540: (Module): varevsij Comments added explaining the second
541: table of variances if popbased=1 .
542: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
543: (Module): Function pstamp added
544: (Module): Version 0.98d
545:
546: Revision 1.117 2006/03/14 17:16:22 brouard
547: (Module): varevsij Comments added explaining the second
548: table of variances if popbased=1 .
549: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
550: (Module): Function pstamp added
551: (Module): Version 0.98d
552:
553: Revision 1.116 2006/03/06 10:29:27 brouard
554: (Module): Variance-covariance wrong links and
555: varian-covariance of ej. is needed (Saito).
556:
557: Revision 1.115 2006/02/27 12:17:45 brouard
558: (Module): One freematrix added in mlikeli! 0.98c
559:
560: Revision 1.114 2006/02/26 12:57:58 brouard
561: (Module): Some improvements in processing parameter
562: filename with strsep.
563:
564: Revision 1.113 2006/02/24 14:20:24 brouard
565: (Module): Memory leaks checks with valgrind and:
566: datafile was not closed, some imatrix were not freed and on matrix
567: allocation too.
568:
569: Revision 1.112 2006/01/30 09:55:26 brouard
570: (Module): Back to gnuplot.exe instead of wgnuplot.exe
571:
572: Revision 1.111 2006/01/25 20:38:18 brouard
573: (Module): Lots of cleaning and bugs added (Gompertz)
574: (Module): Comments can be added in data file. Missing date values
575: can be a simple dot '.'.
576:
577: Revision 1.110 2006/01/25 00:51:50 brouard
578: (Module): Lots of cleaning and bugs added (Gompertz)
579:
580: Revision 1.109 2006/01/24 19:37:15 brouard
581: (Module): Comments (lines starting with a #) are allowed in data.
582:
583: Revision 1.108 2006/01/19 18:05:42 lievre
584: Gnuplot problem appeared...
585: To be fixed
586:
587: Revision 1.107 2006/01/19 16:20:37 brouard
588: Test existence of gnuplot in imach path
589:
590: Revision 1.106 2006/01/19 13:24:36 brouard
591: Some cleaning and links added in html output
592:
593: Revision 1.105 2006/01/05 20:23:19 lievre
594: *** empty log message ***
595:
596: Revision 1.104 2005/09/30 16:11:43 lievre
597: (Module): sump fixed, loop imx fixed, and simplifications.
598: (Module): If the status is missing at the last wave but we know
599: that the person is alive, then we can code his/her status as -2
600: (instead of missing=-1 in earlier versions) and his/her
601: contributions to the likelihood is 1 - Prob of dying from last
602: health status (= 1-p13= p11+p12 in the easiest case of somebody in
603: the healthy state at last known wave). Version is 0.98
604:
605: Revision 1.103 2005/09/30 15:54:49 lievre
606: (Module): sump fixed, loop imx fixed, and simplifications.
607:
608: Revision 1.102 2004/09/15 17:31:30 brouard
609: Add the possibility to read data file including tab characters.
610:
611: Revision 1.101 2004/09/15 10:38:38 brouard
612: Fix on curr_time
613:
614: Revision 1.100 2004/07/12 18:29:06 brouard
615: Add version for Mac OS X. Just define UNIX in Makefile
616:
617: Revision 1.99 2004/06/05 08:57:40 brouard
618: *** empty log message ***
619:
620: Revision 1.98 2004/05/16 15:05:56 brouard
621: New version 0.97 . First attempt to estimate force of mortality
622: directly from the data i.e. without the need of knowing the health
623: state at each age, but using a Gompertz model: log u =a + b*age .
624: This is the basic analysis of mortality and should be done before any
625: other analysis, in order to test if the mortality estimated from the
626: cross-longitudinal survey is different from the mortality estimated
627: from other sources like vital statistic data.
628:
629: The same imach parameter file can be used but the option for mle should be -3.
630:
1.133 brouard 631: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 632: former routines in order to include the new code within the former code.
633:
634: The output is very simple: only an estimate of the intercept and of
635: the slope with 95% confident intervals.
636:
637: Current limitations:
638: A) Even if you enter covariates, i.e. with the
639: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
640: B) There is no computation of Life Expectancy nor Life Table.
641:
642: Revision 1.97 2004/02/20 13:25:42 lievre
643: Version 0.96d. Population forecasting command line is (temporarily)
644: suppressed.
645:
646: Revision 1.96 2003/07/15 15:38:55 brouard
647: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
648: rewritten within the same printf. Workaround: many printfs.
649:
650: Revision 1.95 2003/07/08 07:54:34 brouard
651: * imach.c (Repository):
652: (Repository): Using imachwizard code to output a more meaningful covariance
653: matrix (cov(a12,c31) instead of numbers.
654:
655: Revision 1.94 2003/06/27 13:00:02 brouard
656: Just cleaning
657:
658: Revision 1.93 2003/06/25 16:33:55 brouard
659: (Module): On windows (cygwin) function asctime_r doesn't
660: exist so I changed back to asctime which exists.
661: (Module): Version 0.96b
662:
663: Revision 1.92 2003/06/25 16:30:45 brouard
664: (Module): On windows (cygwin) function asctime_r doesn't
665: exist so I changed back to asctime which exists.
666:
667: Revision 1.91 2003/06/25 15:30:29 brouard
668: * imach.c (Repository): Duplicated warning errors corrected.
669: (Repository): Elapsed time after each iteration is now output. It
670: helps to forecast when convergence will be reached. Elapsed time
671: is stamped in powell. We created a new html file for the graphs
672: concerning matrix of covariance. It has extension -cov.htm.
673:
674: Revision 1.90 2003/06/24 12:34:15 brouard
675: (Module): Some bugs corrected for windows. Also, when
676: mle=-1 a template is output in file "or"mypar.txt with the design
677: of the covariance matrix to be input.
678:
679: Revision 1.89 2003/06/24 12:30:52 brouard
680: (Module): Some bugs corrected for windows. Also, when
681: mle=-1 a template is output in file "or"mypar.txt with the design
682: of the covariance matrix to be input.
683:
684: Revision 1.88 2003/06/23 17:54:56 brouard
685: * 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.
686:
687: Revision 1.87 2003/06/18 12:26:01 brouard
688: Version 0.96
689:
690: Revision 1.86 2003/06/17 20:04:08 brouard
691: (Module): Change position of html and gnuplot routines and added
692: routine fileappend.
693:
694: Revision 1.85 2003/06/17 13:12:43 brouard
695: * imach.c (Repository): Check when date of death was earlier that
696: current date of interview. It may happen when the death was just
697: prior to the death. In this case, dh was negative and likelihood
698: was wrong (infinity). We still send an "Error" but patch by
699: assuming that the date of death was just one stepm after the
700: interview.
701: (Repository): Because some people have very long ID (first column)
702: we changed int to long in num[] and we added a new lvector for
703: memory allocation. But we also truncated to 8 characters (left
704: truncation)
705: (Repository): No more line truncation errors.
706:
707: Revision 1.84 2003/06/13 21:44:43 brouard
708: * imach.c (Repository): Replace "freqsummary" at a correct
709: place. It differs from routine "prevalence" which may be called
710: many times. Probs is memory consuming and must be used with
711: parcimony.
712: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
713:
714: Revision 1.83 2003/06/10 13:39:11 lievre
715: *** empty log message ***
716:
717: Revision 1.82 2003/06/05 15:57:20 brouard
718: Add log in imach.c and fullversion number is now printed.
719:
720: */
721: /*
722: Interpolated Markov Chain
723:
724: Short summary of the programme:
725:
1.227 brouard 726: This program computes Healthy Life Expectancies or State-specific
727: (if states aren't health statuses) Expectancies from
728: cross-longitudinal data. Cross-longitudinal data consist in:
729:
730: -1- a first survey ("cross") where individuals from different ages
731: are interviewed on their health status or degree of disability (in
732: the case of a health survey which is our main interest)
733:
734: -2- at least a second wave of interviews ("longitudinal") which
735: measure each change (if any) in individual health status. Health
736: expectancies are computed from the time spent in each health state
737: according to a model. More health states you consider, more time is
738: necessary to reach the Maximum Likelihood of the parameters involved
739: in the model. The simplest model is the multinomial logistic model
740: where pij is the probability to be observed in state j at the second
741: wave conditional to be observed in state i at the first
742: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
743: etc , where 'age' is age and 'sex' is a covariate. If you want to
744: have a more complex model than "constant and age", you should modify
745: the program where the markup *Covariates have to be included here
746: again* invites you to do it. More covariates you add, slower the
1.126 brouard 747: convergence.
748:
749: The advantage of this computer programme, compared to a simple
750: multinomial logistic model, is clear when the delay between waves is not
751: identical for each individual. Also, if a individual missed an
752: intermediate interview, the information is lost, but taken into
753: account using an interpolation or extrapolation.
754:
755: hPijx is the probability to be observed in state i at age x+h
756: conditional to the observed state i at age x. The delay 'h' can be
757: split into an exact number (nh*stepm) of unobserved intermediate
758: states. This elementary transition (by month, quarter,
759: semester or year) is modelled as a multinomial logistic. The hPx
760: matrix is simply the matrix product of nh*stepm elementary matrices
761: and the contribution of each individual to the likelihood is simply
762: hPijx.
763:
764: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 765: of the life expectancies. It also computes the period (stable) prevalence.
766:
767: Back prevalence and projections:
1.227 brouard 768:
769: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
770: double agemaxpar, double ftolpl, int *ncvyearp, double
771: dateprev1,double dateprev2, int firstpass, int lastpass, int
772: mobilavproj)
773:
774: Computes the back prevalence limit for any combination of
775: covariate values k at any age between ageminpar and agemaxpar and
776: returns it in **bprlim. In the loops,
777:
778: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
779: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
780:
781: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 782: Computes for any combination of covariates k and any age between bage and fage
783: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
784: oldm=oldms;savm=savms;
1.227 brouard 785:
786: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 787: Computes the transition matrix starting at age 'age' over
788: 'nhstepm*hstepm*stepm' months (i.e. until
789: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 790: nhstepm*hstepm matrices.
791:
792: Returns p3mat[i][j][h] after calling
793: p3mat[i][j][h]=matprod2(newm,
794: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
795: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
796: oldm);
1.226 brouard 797:
798: Important routines
799:
800: - func (or funcone), computes logit (pij) distinguishing
801: o fixed variables (single or product dummies or quantitative);
802: o varying variables by:
803: (1) wave (single, product dummies, quantitative),
804: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
805: % fixed dummy (treated) or quantitative (not done because time-consuming);
806: % varying dummy (not done) or quantitative (not done);
807: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
808: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
809: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
810: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
811: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 812:
1.226 brouard 813:
814:
1.133 brouard 815: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
816: Institut national d'études démographiques, Paris.
1.126 brouard 817: This software have been partly granted by Euro-REVES, a concerted action
818: from the European Union.
819: It is copyrighted identically to a GNU software product, ie programme and
820: software can be distributed freely for non commercial use. Latest version
821: can be accessed at http://euroreves.ined.fr/imach .
822:
823: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
824: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
825:
826: **********************************************************************/
827: /*
828: main
829: read parameterfile
830: read datafile
831: concatwav
832: freqsummary
833: if (mle >= 1)
834: mlikeli
835: print results files
836: if mle==1
837: computes hessian
838: read end of parameter file: agemin, agemax, bage, fage, estepm
839: begin-prev-date,...
840: open gnuplot file
841: open html file
1.145 brouard 842: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
843: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
844: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
845: freexexit2 possible for memory heap.
846:
847: h Pij x | pij_nom ficrestpij
848: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
849: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
850: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
851:
852: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
853: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
854: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
855: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
856: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
857:
1.126 brouard 858: forecasting if prevfcast==1 prevforecast call prevalence()
859: health expectancies
860: Variance-covariance of DFLE
861: prevalence()
862: movingaverage()
863: varevsij()
864: if popbased==1 varevsij(,popbased)
865: total life expectancies
866: Variance of period (stable) prevalence
867: end
868: */
869:
1.187 brouard 870: /* #define DEBUG */
871: /* #define DEBUGBRENT */
1.203 brouard 872: /* #define DEBUGLINMIN */
873: /* #define DEBUGHESS */
874: #define DEBUGHESSIJ
1.224 brouard 875: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 876: #define POWELL /* Instead of NLOPT */
1.224 brouard 877: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 878: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
879: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 880:
881: #include <math.h>
882: #include <stdio.h>
883: #include <stdlib.h>
884: #include <string.h>
1.226 brouard 885: #include <ctype.h>
1.159 brouard 886:
887: #ifdef _WIN32
888: #include <io.h>
1.172 brouard 889: #include <windows.h>
890: #include <tchar.h>
1.159 brouard 891: #else
1.126 brouard 892: #include <unistd.h>
1.159 brouard 893: #endif
1.126 brouard 894:
895: #include <limits.h>
896: #include <sys/types.h>
1.171 brouard 897:
898: #if defined(__GNUC__)
899: #include <sys/utsname.h> /* Doesn't work on Windows */
900: #endif
901:
1.126 brouard 902: #include <sys/stat.h>
903: #include <errno.h>
1.159 brouard 904: /* extern int errno; */
1.126 brouard 905:
1.157 brouard 906: /* #ifdef LINUX */
907: /* #include <time.h> */
908: /* #include "timeval.h" */
909: /* #else */
910: /* #include <sys/time.h> */
911: /* #endif */
912:
1.126 brouard 913: #include <time.h>
914:
1.136 brouard 915: #ifdef GSL
916: #include <gsl/gsl_errno.h>
917: #include <gsl/gsl_multimin.h>
918: #endif
919:
1.167 brouard 920:
1.162 brouard 921: #ifdef NLOPT
922: #include <nlopt.h>
923: typedef struct {
924: double (* function)(double [] );
925: } myfunc_data ;
926: #endif
927:
1.126 brouard 928: /* #include <libintl.h> */
929: /* #define _(String) gettext (String) */
930:
1.251 brouard 931: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 932:
933: #define GNUPLOTPROGRAM "gnuplot"
934: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
935: #define FILENAMELENGTH 132
936:
937: #define GLOCK_ERROR_NOPATH -1 /* empty path */
938: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
939:
1.144 brouard 940: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
941: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 942:
943: #define NINTERVMAX 8
1.144 brouard 944: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
945: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
946: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 947: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 948: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
949: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 950: #define MAXN 20000
1.144 brouard 951: #define YEARM 12. /**< Number of months per year */
1.218 brouard 952: /* #define AGESUP 130 */
953: #define AGESUP 150
954: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 955: #define AGEBASE 40
1.194 brouard 956: #define AGEOVERFLOW 1.e20
1.164 brouard 957: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 958: #ifdef _WIN32
959: #define DIRSEPARATOR '\\'
960: #define CHARSEPARATOR "\\"
961: #define ODIRSEPARATOR '/'
962: #else
1.126 brouard 963: #define DIRSEPARATOR '/'
964: #define CHARSEPARATOR "/"
965: #define ODIRSEPARATOR '\\'
966: #endif
967:
1.256 ! brouard 968: /* $Id: imach.c,v 1.255 2017/03/08 16:02:28 brouard Exp $ */
1.126 brouard 969: /* $State: Exp $ */
1.196 brouard 970: #include "version.h"
971: char version[]=__IMACH_VERSION__;
1.224 brouard 972: 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";
1.256 ! brouard 973: char fullversion[]="$Revision: 1.255 $ $Date: 2017/03/08 16:02:28 $";
1.126 brouard 974: char strstart[80];
975: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 976: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 977: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 978: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
979: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
980: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 981: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
982: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 983: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
984: int cptcovprodnoage=0; /**< Number of covariate products without age */
985: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 986: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
987: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 988: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 989: int nsd=0; /**< Total number of single dummy variables (output) */
990: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 991: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 992: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 993: int ntveff=0; /**< ntveff number of effective time varying variables */
994: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 995: int cptcov=0; /* Working variable */
1.218 brouard 996: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 997: int npar=NPARMAX;
998: int nlstate=2; /* Number of live states */
999: int ndeath=1; /* Number of dead states */
1.130 brouard 1000: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1001: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1002: int popbased=0;
1003:
1004: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1005: int maxwav=0; /* Maxim number of waves */
1006: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1007: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1008: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1009: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1010: int mle=1, weightopt=0;
1.126 brouard 1011: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1012: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1013: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1014: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1015: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1016: int selected(int kvar); /* Is covariate kvar selected for printing results */
1017:
1.130 brouard 1018: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1019: double **matprod2(); /* test */
1.126 brouard 1020: double **oldm, **newm, **savm; /* Working pointers to matrices */
1021: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1022: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1023:
1.136 brouard 1024: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1025: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1026: FILE *ficlog, *ficrespow;
1.130 brouard 1027: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1028: double fretone; /* Only one call to likelihood */
1.130 brouard 1029: long ipmx=0; /* Number of contributions */
1.126 brouard 1030: double sw; /* Sum of weights */
1031: char filerespow[FILENAMELENGTH];
1032: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1033: FILE *ficresilk;
1034: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1035: FILE *ficresprobmorprev;
1036: FILE *fichtm, *fichtmcov; /* Html File */
1037: FILE *ficreseij;
1038: char filerese[FILENAMELENGTH];
1039: FILE *ficresstdeij;
1040: char fileresstde[FILENAMELENGTH];
1041: FILE *ficrescveij;
1042: char filerescve[FILENAMELENGTH];
1043: FILE *ficresvij;
1044: char fileresv[FILENAMELENGTH];
1045: FILE *ficresvpl;
1046: char fileresvpl[FILENAMELENGTH];
1047: char title[MAXLINE];
1.234 brouard 1048: char model[MAXLINE]; /**< The model line */
1.217 brouard 1049: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1050: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1051: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1052: char command[FILENAMELENGTH];
1053: int outcmd=0;
1054:
1.217 brouard 1055: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1056: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1057: char filelog[FILENAMELENGTH]; /* Log file */
1058: char filerest[FILENAMELENGTH];
1059: char fileregp[FILENAMELENGTH];
1060: char popfile[FILENAMELENGTH];
1061:
1062: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1063:
1.157 brouard 1064: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1065: /* struct timezone tzp; */
1066: /* extern int gettimeofday(); */
1067: struct tm tml, *gmtime(), *localtime();
1068:
1069: extern time_t time();
1070:
1071: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1072: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1073: struct tm tm;
1074:
1.126 brouard 1075: char strcurr[80], strfor[80];
1076:
1077: char *endptr;
1078: long lval;
1079: double dval;
1080:
1081: #define NR_END 1
1082: #define FREE_ARG char*
1083: #define FTOL 1.0e-10
1084:
1085: #define NRANSI
1.240 brouard 1086: #define ITMAX 200
1087: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1088:
1089: #define TOL 2.0e-4
1090:
1091: #define CGOLD 0.3819660
1092: #define ZEPS 1.0e-10
1093: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1094:
1095: #define GOLD 1.618034
1096: #define GLIMIT 100.0
1097: #define TINY 1.0e-20
1098:
1099: static double maxarg1,maxarg2;
1100: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1101: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1102:
1103: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1104: #define rint(a) floor(a+0.5)
1.166 brouard 1105: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1106: #define mytinydouble 1.0e-16
1.166 brouard 1107: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1108: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1109: /* static double dsqrarg; */
1110: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1111: static double sqrarg;
1112: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1113: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1114: int agegomp= AGEGOMP;
1115:
1116: int imx;
1117: int stepm=1;
1118: /* Stepm, step in month: minimum step interpolation*/
1119:
1120: int estepm;
1121: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1122:
1123: int m,nb;
1124: long *num;
1.197 brouard 1125: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1126: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1127: covariate for which somebody answered excluding
1128: undefined. Usually 2: 0 and 1. */
1129: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1130: covariate for which somebody answered including
1131: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1132: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1133: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1134: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1135: double *ageexmed,*agecens;
1136: double dateintmean=0;
1137:
1138: double *weight;
1139: int **s; /* Status */
1.141 brouard 1140: double *agedc;
1.145 brouard 1141: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1142: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1143: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1144: double **coqvar; /* Fixed quantitative covariate iqv */
1145: double ***cotvar; /* Time varying covariate itv */
1146: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1147: double idx;
1148: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1149: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1150: /*k 1 2 3 4 5 6 7 8 9 */
1151: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1152: /* Tndvar[k] 1 2 3 4 5 */
1153: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1154: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1155: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1156: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1157: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1158: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1159: /* Tprod[i]=k 4 7 */
1160: /* Tage[i]=k 5 8 */
1161: /* */
1162: /* Type */
1163: /* V 1 2 3 4 5 */
1164: /* F F V V V */
1165: /* D Q D D Q */
1166: /* */
1167: int *TvarsD;
1168: int *TvarsDind;
1169: int *TvarsQ;
1170: int *TvarsQind;
1171:
1.235 brouard 1172: #define MAXRESULTLINES 10
1173: int nresult=0;
1174: int TKresult[MAXRESULTLINES];
1.237 brouard 1175: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1176: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1177: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1178: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1179: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1180: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1181:
1.234 brouard 1182: /* 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 1183: 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 */
1184: 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 */
1185: 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 */
1186: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1187: 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 */
1188: 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 1189: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1190: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1191: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1192: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1193: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1194: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1195: 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 */
1196: 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 */
1197:
1.230 brouard 1198: int *Tvarsel; /**< Selected covariates for output */
1199: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1200: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1201: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1202: 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 1203: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1204: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1205: int *Tage;
1.227 brouard 1206: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1207: 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 1208: 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*/
1209: 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 1210: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1211: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1212: int **Tvard;
1213: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1214: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1215: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1216: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1217: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1218: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1219: double *lsurv, *lpop, *tpop;
1220:
1.231 brouard 1221: #define FD 1; /* Fixed dummy covariate */
1222: #define FQ 2; /* Fixed quantitative covariate */
1223: #define FP 3; /* Fixed product covariate */
1224: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1225: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1226: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1227: #define VD 10; /* Varying dummy covariate */
1228: #define VQ 11; /* Varying quantitative covariate */
1229: #define VP 12; /* Varying product covariate */
1230: #define VPDD 13; /* Varying product dummy*dummy covariate */
1231: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1232: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1233: #define APFD 16; /* Age product * fixed dummy covariate */
1234: #define APFQ 17; /* Age product * fixed quantitative covariate */
1235: #define APVD 18; /* Age product * varying dummy covariate */
1236: #define APVQ 19; /* Age product * varying quantitative covariate */
1237:
1238: #define FTYPE 1; /* Fixed covariate */
1239: #define VTYPE 2; /* Varying covariate (loop in wave) */
1240: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1241:
1242: struct kmodel{
1243: int maintype; /* main type */
1244: int subtype; /* subtype */
1245: };
1246: struct kmodel modell[NCOVMAX];
1247:
1.143 brouard 1248: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1249: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1250:
1251: /**************** split *************************/
1252: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1253: {
1254: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1255: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1256: */
1257: char *ss; /* pointer */
1.186 brouard 1258: int l1=0, l2=0; /* length counters */
1.126 brouard 1259:
1260: l1 = strlen(path ); /* length of path */
1261: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1262: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1263: if ( ss == NULL ) { /* no directory, so determine current directory */
1264: strcpy( name, path ); /* we got the fullname name because no directory */
1265: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1266: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1267: /* get current working directory */
1268: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1269: #ifdef WIN32
1270: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1271: #else
1272: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1273: #endif
1.126 brouard 1274: return( GLOCK_ERROR_GETCWD );
1275: }
1276: /* got dirc from getcwd*/
1277: printf(" DIRC = %s \n",dirc);
1.205 brouard 1278: } else { /* strip directory from path */
1.126 brouard 1279: ss++; /* after this, the filename */
1280: l2 = strlen( ss ); /* length of filename */
1281: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1282: strcpy( name, ss ); /* save file name */
1283: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1284: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1285: printf(" DIRC2 = %s \n",dirc);
1286: }
1287: /* We add a separator at the end of dirc if not exists */
1288: l1 = strlen( dirc ); /* length of directory */
1289: if( dirc[l1-1] != DIRSEPARATOR ){
1290: dirc[l1] = DIRSEPARATOR;
1291: dirc[l1+1] = 0;
1292: printf(" DIRC3 = %s \n",dirc);
1293: }
1294: ss = strrchr( name, '.' ); /* find last / */
1295: if (ss >0){
1296: ss++;
1297: strcpy(ext,ss); /* save extension */
1298: l1= strlen( name);
1299: l2= strlen(ss)+1;
1300: strncpy( finame, name, l1-l2);
1301: finame[l1-l2]= 0;
1302: }
1303:
1304: return( 0 ); /* we're done */
1305: }
1306:
1307:
1308: /******************************************/
1309:
1310: void replace_back_to_slash(char *s, char*t)
1311: {
1312: int i;
1313: int lg=0;
1314: i=0;
1315: lg=strlen(t);
1316: for(i=0; i<= lg; i++) {
1317: (s[i] = t[i]);
1318: if (t[i]== '\\') s[i]='/';
1319: }
1320: }
1321:
1.132 brouard 1322: char *trimbb(char *out, char *in)
1.137 brouard 1323: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1324: char *s;
1325: s=out;
1326: while (*in != '\0'){
1.137 brouard 1327: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1328: in++;
1329: }
1330: *out++ = *in++;
1331: }
1332: *out='\0';
1333: return s;
1334: }
1335:
1.187 brouard 1336: /* char *substrchaine(char *out, char *in, char *chain) */
1337: /* { */
1338: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1339: /* char *s, *t; */
1340: /* t=in;s=out; */
1341: /* while ((*in != *chain) && (*in != '\0')){ */
1342: /* *out++ = *in++; */
1343: /* } */
1344:
1345: /* /\* *in matches *chain *\/ */
1346: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1347: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1348: /* } */
1349: /* in--; chain--; */
1350: /* while ( (*in != '\0')){ */
1351: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1352: /* *out++ = *in++; */
1353: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1354: /* } */
1355: /* *out='\0'; */
1356: /* out=s; */
1357: /* return out; */
1358: /* } */
1359: char *substrchaine(char *out, char *in, char *chain)
1360: {
1361: /* Substract chain 'chain' from 'in', return and output 'out' */
1362: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1363:
1364: char *strloc;
1365:
1366: strcpy (out, in);
1367: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1368: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1369: if(strloc != NULL){
1370: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1371: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1372: /* strcpy (strloc, strloc +strlen(chain));*/
1373: }
1374: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1375: return out;
1376: }
1377:
1378:
1.145 brouard 1379: char *cutl(char *blocc, char *alocc, char *in, char occ)
1380: {
1.187 brouard 1381: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1382: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1383: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1384: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1385: */
1.160 brouard 1386: char *s, *t;
1.145 brouard 1387: t=in;s=in;
1388: while ((*in != occ) && (*in != '\0')){
1389: *alocc++ = *in++;
1390: }
1391: if( *in == occ){
1392: *(alocc)='\0';
1393: s=++in;
1394: }
1395:
1396: if (s == t) {/* occ not found */
1397: *(alocc-(in-s))='\0';
1398: in=s;
1399: }
1400: while ( *in != '\0'){
1401: *blocc++ = *in++;
1402: }
1403:
1404: *blocc='\0';
1405: return t;
1406: }
1.137 brouard 1407: char *cutv(char *blocc, char *alocc, char *in, char occ)
1408: {
1.187 brouard 1409: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1410: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1411: gives blocc="abcdef2ghi" and alocc="j".
1412: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1413: */
1414: char *s, *t;
1415: t=in;s=in;
1416: while (*in != '\0'){
1417: while( *in == occ){
1418: *blocc++ = *in++;
1419: s=in;
1420: }
1421: *blocc++ = *in++;
1422: }
1423: if (s == t) /* occ not found */
1424: *(blocc-(in-s))='\0';
1425: else
1426: *(blocc-(in-s)-1)='\0';
1427: in=s;
1428: while ( *in != '\0'){
1429: *alocc++ = *in++;
1430: }
1431:
1432: *alocc='\0';
1433: return s;
1434: }
1435:
1.126 brouard 1436: int nbocc(char *s, char occ)
1437: {
1438: int i,j=0;
1439: int lg=20;
1440: i=0;
1441: lg=strlen(s);
1442: for(i=0; i<= lg; i++) {
1.234 brouard 1443: if (s[i] == occ ) j++;
1.126 brouard 1444: }
1445: return j;
1446: }
1447:
1.137 brouard 1448: /* void cutv(char *u,char *v, char*t, char occ) */
1449: /* { */
1450: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1451: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1452: /* gives u="abcdef2ghi" and v="j" *\/ */
1453: /* int i,lg,j,p=0; */
1454: /* i=0; */
1455: /* lg=strlen(t); */
1456: /* for(j=0; j<=lg-1; j++) { */
1457: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1458: /* } */
1.126 brouard 1459:
1.137 brouard 1460: /* for(j=0; j<p; j++) { */
1461: /* (u[j] = t[j]); */
1462: /* } */
1463: /* u[p]='\0'; */
1.126 brouard 1464:
1.137 brouard 1465: /* for(j=0; j<= lg; j++) { */
1466: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1467: /* } */
1468: /* } */
1.126 brouard 1469:
1.160 brouard 1470: #ifdef _WIN32
1471: char * strsep(char **pp, const char *delim)
1472: {
1473: char *p, *q;
1474:
1475: if ((p = *pp) == NULL)
1476: return 0;
1477: if ((q = strpbrk (p, delim)) != NULL)
1478: {
1479: *pp = q + 1;
1480: *q = '\0';
1481: }
1482: else
1483: *pp = 0;
1484: return p;
1485: }
1486: #endif
1487:
1.126 brouard 1488: /********************** nrerror ********************/
1489:
1490: void nrerror(char error_text[])
1491: {
1492: fprintf(stderr,"ERREUR ...\n");
1493: fprintf(stderr,"%s\n",error_text);
1494: exit(EXIT_FAILURE);
1495: }
1496: /*********************** vector *******************/
1497: double *vector(int nl, int nh)
1498: {
1499: double *v;
1500: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1501: if (!v) nrerror("allocation failure in vector");
1502: return v-nl+NR_END;
1503: }
1504:
1505: /************************ free vector ******************/
1506: void free_vector(double*v, int nl, int nh)
1507: {
1508: free((FREE_ARG)(v+nl-NR_END));
1509: }
1510:
1511: /************************ivector *******************************/
1512: int *ivector(long nl,long nh)
1513: {
1514: int *v;
1515: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1516: if (!v) nrerror("allocation failure in ivector");
1517: return v-nl+NR_END;
1518: }
1519:
1520: /******************free ivector **************************/
1521: void free_ivector(int *v, long nl, long nh)
1522: {
1523: free((FREE_ARG)(v+nl-NR_END));
1524: }
1525:
1526: /************************lvector *******************************/
1527: long *lvector(long nl,long nh)
1528: {
1529: long *v;
1530: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1531: if (!v) nrerror("allocation failure in ivector");
1532: return v-nl+NR_END;
1533: }
1534:
1535: /******************free lvector **************************/
1536: void free_lvector(long *v, long nl, long nh)
1537: {
1538: free((FREE_ARG)(v+nl-NR_END));
1539: }
1540:
1541: /******************* imatrix *******************************/
1542: int **imatrix(long nrl, long nrh, long ncl, long nch)
1543: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1544: {
1545: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1546: int **m;
1547:
1548: /* allocate pointers to rows */
1549: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1550: if (!m) nrerror("allocation failure 1 in matrix()");
1551: m += NR_END;
1552: m -= nrl;
1553:
1554:
1555: /* allocate rows and set pointers to them */
1556: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1557: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1558: m[nrl] += NR_END;
1559: m[nrl] -= ncl;
1560:
1561: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1562:
1563: /* return pointer to array of pointers to rows */
1564: return m;
1565: }
1566:
1567: /****************** free_imatrix *************************/
1568: void free_imatrix(m,nrl,nrh,ncl,nch)
1569: int **m;
1570: long nch,ncl,nrh,nrl;
1571: /* free an int matrix allocated by imatrix() */
1572: {
1573: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1574: free((FREE_ARG) (m+nrl-NR_END));
1575: }
1576:
1577: /******************* matrix *******************************/
1578: double **matrix(long nrl, long nrh, long ncl, long nch)
1579: {
1580: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1581: double **m;
1582:
1583: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1584: if (!m) nrerror("allocation failure 1 in matrix()");
1585: m += NR_END;
1586: m -= nrl;
1587:
1588: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1589: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1590: m[nrl] += NR_END;
1591: m[nrl] -= ncl;
1592:
1593: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1594: return m;
1.145 brouard 1595: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1596: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1597: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1598: */
1599: }
1600:
1601: /*************************free matrix ************************/
1602: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1603: {
1604: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1605: free((FREE_ARG)(m+nrl-NR_END));
1606: }
1607:
1608: /******************* ma3x *******************************/
1609: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1610: {
1611: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1612: double ***m;
1613:
1614: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1615: if (!m) nrerror("allocation failure 1 in matrix()");
1616: m += NR_END;
1617: m -= nrl;
1618:
1619: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1620: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1621: m[nrl] += NR_END;
1622: m[nrl] -= ncl;
1623:
1624: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1625:
1626: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1627: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1628: m[nrl][ncl] += NR_END;
1629: m[nrl][ncl] -= nll;
1630: for (j=ncl+1; j<=nch; j++)
1631: m[nrl][j]=m[nrl][j-1]+nlay;
1632:
1633: for (i=nrl+1; i<=nrh; i++) {
1634: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1635: for (j=ncl+1; j<=nch; j++)
1636: m[i][j]=m[i][j-1]+nlay;
1637: }
1638: return m;
1639: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1640: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1641: */
1642: }
1643:
1644: /*************************free ma3x ************************/
1645: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1646: {
1647: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1648: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1649: free((FREE_ARG)(m+nrl-NR_END));
1650: }
1651:
1652: /*************** function subdirf ***********/
1653: char *subdirf(char fileres[])
1654: {
1655: /* Caution optionfilefiname is hidden */
1656: strcpy(tmpout,optionfilefiname);
1657: strcat(tmpout,"/"); /* Add to the right */
1658: strcat(tmpout,fileres);
1659: return tmpout;
1660: }
1661:
1662: /*************** function subdirf2 ***********/
1663: char *subdirf2(char fileres[], char *preop)
1664: {
1665:
1666: /* Caution optionfilefiname is hidden */
1667: strcpy(tmpout,optionfilefiname);
1668: strcat(tmpout,"/");
1669: strcat(tmpout,preop);
1670: strcat(tmpout,fileres);
1671: return tmpout;
1672: }
1673:
1674: /*************** function subdirf3 ***********/
1675: char *subdirf3(char fileres[], char *preop, char *preop2)
1676: {
1677:
1678: /* Caution optionfilefiname is hidden */
1679: strcpy(tmpout,optionfilefiname);
1680: strcat(tmpout,"/");
1681: strcat(tmpout,preop);
1682: strcat(tmpout,preop2);
1683: strcat(tmpout,fileres);
1684: return tmpout;
1685: }
1.213 brouard 1686:
1687: /*************** function subdirfext ***********/
1688: char *subdirfext(char fileres[], char *preop, char *postop)
1689: {
1690:
1691: strcpy(tmpout,preop);
1692: strcat(tmpout,fileres);
1693: strcat(tmpout,postop);
1694: return tmpout;
1695: }
1.126 brouard 1696:
1.213 brouard 1697: /*************** function subdirfext3 ***********/
1698: char *subdirfext3(char fileres[], char *preop, char *postop)
1699: {
1700:
1701: /* Caution optionfilefiname is hidden */
1702: strcpy(tmpout,optionfilefiname);
1703: strcat(tmpout,"/");
1704: strcat(tmpout,preop);
1705: strcat(tmpout,fileres);
1706: strcat(tmpout,postop);
1707: return tmpout;
1708: }
1709:
1.162 brouard 1710: char *asc_diff_time(long time_sec, char ascdiff[])
1711: {
1712: long sec_left, days, hours, minutes;
1713: days = (time_sec) / (60*60*24);
1714: sec_left = (time_sec) % (60*60*24);
1715: hours = (sec_left) / (60*60) ;
1716: sec_left = (sec_left) %(60*60);
1717: minutes = (sec_left) /60;
1718: sec_left = (sec_left) % (60);
1719: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1720: return ascdiff;
1721: }
1722:
1.126 brouard 1723: /***************** f1dim *************************/
1724: extern int ncom;
1725: extern double *pcom,*xicom;
1726: extern double (*nrfunc)(double []);
1727:
1728: double f1dim(double x)
1729: {
1730: int j;
1731: double f;
1732: double *xt;
1733:
1734: xt=vector(1,ncom);
1735: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1736: f=(*nrfunc)(xt);
1737: free_vector(xt,1,ncom);
1738: return f;
1739: }
1740:
1741: /*****************brent *************************/
1742: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1743: {
1744: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1745: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1746: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1747: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1748: * returned function value.
1749: */
1.126 brouard 1750: int iter;
1751: double a,b,d,etemp;
1.159 brouard 1752: double fu=0,fv,fw,fx;
1.164 brouard 1753: double ftemp=0.;
1.126 brouard 1754: double p,q,r,tol1,tol2,u,v,w,x,xm;
1755: double e=0.0;
1756:
1757: a=(ax < cx ? ax : cx);
1758: b=(ax > cx ? ax : cx);
1759: x=w=v=bx;
1760: fw=fv=fx=(*f)(x);
1761: for (iter=1;iter<=ITMAX;iter++) {
1762: xm=0.5*(a+b);
1763: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1764: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1765: printf(".");fflush(stdout);
1766: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1767: #ifdef DEBUGBRENT
1.126 brouard 1768: 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);
1769: 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);
1770: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1771: #endif
1772: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1773: *xmin=x;
1774: return fx;
1775: }
1776: ftemp=fu;
1777: if (fabs(e) > tol1) {
1778: r=(x-w)*(fx-fv);
1779: q=(x-v)*(fx-fw);
1780: p=(x-v)*q-(x-w)*r;
1781: q=2.0*(q-r);
1782: if (q > 0.0) p = -p;
1783: q=fabs(q);
1784: etemp=e;
1785: e=d;
1786: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1787: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1788: else {
1.224 brouard 1789: d=p/q;
1790: u=x+d;
1791: if (u-a < tol2 || b-u < tol2)
1792: d=SIGN(tol1,xm-x);
1.126 brouard 1793: }
1794: } else {
1795: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1796: }
1797: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1798: fu=(*f)(u);
1799: if (fu <= fx) {
1800: if (u >= x) a=x; else b=x;
1801: SHFT(v,w,x,u)
1.183 brouard 1802: SHFT(fv,fw,fx,fu)
1803: } else {
1804: if (u < x) a=u; else b=u;
1805: if (fu <= fw || w == x) {
1.224 brouard 1806: v=w;
1807: w=u;
1808: fv=fw;
1809: fw=fu;
1.183 brouard 1810: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1811: v=u;
1812: fv=fu;
1.183 brouard 1813: }
1814: }
1.126 brouard 1815: }
1816: nrerror("Too many iterations in brent");
1817: *xmin=x;
1818: return fx;
1819: }
1820:
1821: /****************** mnbrak ***********************/
1822:
1823: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1824: double (*func)(double))
1.183 brouard 1825: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1826: the downhill direction (defined by the function as evaluated at the initial points) and returns
1827: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1828: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1829: */
1.126 brouard 1830: double ulim,u,r,q, dum;
1831: double fu;
1.187 brouard 1832:
1833: double scale=10.;
1834: int iterscale=0;
1835:
1836: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1837: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1838:
1839:
1840: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1841: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1842: /* *bx = *ax - (*ax - *bx)/scale; */
1843: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1844: /* } */
1845:
1.126 brouard 1846: if (*fb > *fa) {
1847: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1848: SHFT(dum,*fb,*fa,dum)
1849: }
1.126 brouard 1850: *cx=(*bx)+GOLD*(*bx-*ax);
1851: *fc=(*func)(*cx);
1.183 brouard 1852: #ifdef DEBUG
1.224 brouard 1853: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1854: 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 1855: #endif
1.224 brouard 1856: 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 1857: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1858: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1859: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1860: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1861: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1862: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1863: fu=(*func)(u);
1.163 brouard 1864: #ifdef DEBUG
1865: /* f(x)=A(x-u)**2+f(u) */
1866: double A, fparabu;
1867: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1868: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1869: 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);
1870: 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 1871: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1872: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1873: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1874: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1875: #endif
1.184 brouard 1876: #ifdef MNBRAKORIGINAL
1.183 brouard 1877: #else
1.191 brouard 1878: /* if (fu > *fc) { */
1879: /* #ifdef DEBUG */
1880: /* printf("mnbrak4 fu > fc \n"); */
1881: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1882: /* #endif */
1883: /* /\* 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 *\\/ *\/ */
1884: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1885: /* dum=u; /\* Shifting c and u *\/ */
1886: /* u = *cx; */
1887: /* *cx = dum; */
1888: /* dum = fu; */
1889: /* fu = *fc; */
1890: /* *fc =dum; */
1891: /* } else { /\* end *\/ */
1892: /* #ifdef DEBUG */
1893: /* printf("mnbrak3 fu < fc \n"); */
1894: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1895: /* #endif */
1896: /* dum=u; /\* Shifting c and u *\/ */
1897: /* u = *cx; */
1898: /* *cx = dum; */
1899: /* dum = fu; */
1900: /* fu = *fc; */
1901: /* *fc =dum; */
1902: /* } */
1.224 brouard 1903: #ifdef DEBUGMNBRAK
1904: double A, fparabu;
1905: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1906: fparabu= *fa - A*(*ax-u)*(*ax-u);
1907: 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);
1908: 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 1909: #endif
1.191 brouard 1910: dum=u; /* Shifting c and u */
1911: u = *cx;
1912: *cx = dum;
1913: dum = fu;
1914: fu = *fc;
1915: *fc =dum;
1.183 brouard 1916: #endif
1.162 brouard 1917: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1918: #ifdef DEBUG
1.224 brouard 1919: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1920: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1921: #endif
1.126 brouard 1922: fu=(*func)(u);
1923: if (fu < *fc) {
1.183 brouard 1924: #ifdef DEBUG
1.224 brouard 1925: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1926: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1927: #endif
1928: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1929: SHFT(*fb,*fc,fu,(*func)(u))
1930: #ifdef DEBUG
1931: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1932: #endif
1933: }
1.162 brouard 1934: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1935: #ifdef DEBUG
1.224 brouard 1936: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1937: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1938: #endif
1.126 brouard 1939: u=ulim;
1940: fu=(*func)(u);
1.183 brouard 1941: } else { /* u could be left to b (if r > q parabola has a maximum) */
1942: #ifdef DEBUG
1.224 brouard 1943: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1944: 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 1945: #endif
1.126 brouard 1946: u=(*cx)+GOLD*(*cx-*bx);
1947: fu=(*func)(u);
1.224 brouard 1948: #ifdef DEBUG
1949: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1950: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1951: #endif
1.183 brouard 1952: } /* end tests */
1.126 brouard 1953: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1954: SHFT(*fa,*fb,*fc,fu)
1955: #ifdef DEBUG
1.224 brouard 1956: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1957: 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 1958: #endif
1959: } /* 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 1960: }
1961:
1962: /*************** linmin ************************/
1.162 brouard 1963: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1964: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1965: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1966: the value of func at the returned location p . This is actually all accomplished by calling the
1967: routines mnbrak and brent .*/
1.126 brouard 1968: int ncom;
1969: double *pcom,*xicom;
1970: double (*nrfunc)(double []);
1971:
1.224 brouard 1972: #ifdef LINMINORIGINAL
1.126 brouard 1973: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1974: #else
1975: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1976: #endif
1.126 brouard 1977: {
1978: double brent(double ax, double bx, double cx,
1979: double (*f)(double), double tol, double *xmin);
1980: double f1dim(double x);
1981: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1982: double *fc, double (*func)(double));
1983: int j;
1984: double xx,xmin,bx,ax;
1985: double fx,fb,fa;
1.187 brouard 1986:
1.203 brouard 1987: #ifdef LINMINORIGINAL
1988: #else
1989: double scale=10., axs, xxs; /* Scale added for infinity */
1990: #endif
1991:
1.126 brouard 1992: ncom=n;
1993: pcom=vector(1,n);
1994: xicom=vector(1,n);
1995: nrfunc=func;
1996: for (j=1;j<=n;j++) {
1997: pcom[j]=p[j];
1.202 brouard 1998: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1999: }
1.187 brouard 2000:
1.203 brouard 2001: #ifdef LINMINORIGINAL
2002: xx=1.;
2003: #else
2004: axs=0.0;
2005: xxs=1.;
2006: do{
2007: xx= xxs;
2008: #endif
1.187 brouard 2009: ax=0.;
2010: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2011: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2012: /* 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)) */
2013: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2014: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2015: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2016: /* 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 2017: #ifdef LINMINORIGINAL
2018: #else
2019: if (fx != fx){
1.224 brouard 2020: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2021: printf("|");
2022: fprintf(ficlog,"|");
1.203 brouard 2023: #ifdef DEBUGLINMIN
1.224 brouard 2024: 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 2025: #endif
2026: }
1.224 brouard 2027: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2028: #endif
2029:
1.191 brouard 2030: #ifdef DEBUGLINMIN
2031: 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 2032: 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 2033: #endif
1.224 brouard 2034: #ifdef LINMINORIGINAL
2035: #else
2036: if(fb == fx){ /* Flat function in the direction */
2037: xmin=xx;
2038: *flat=1;
2039: }else{
2040: *flat=0;
2041: #endif
2042: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2043: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2044: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2045: /* fmin = f(p[j] + xmin * xi[j]) */
2046: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2047: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2048: #ifdef DEBUG
1.224 brouard 2049: 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);
2050: 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);
2051: #endif
2052: #ifdef LINMINORIGINAL
2053: #else
2054: }
1.126 brouard 2055: #endif
1.191 brouard 2056: #ifdef DEBUGLINMIN
2057: printf("linmin end ");
1.202 brouard 2058: fprintf(ficlog,"linmin end ");
1.191 brouard 2059: #endif
1.126 brouard 2060: for (j=1;j<=n;j++) {
1.203 brouard 2061: #ifdef LINMINORIGINAL
2062: xi[j] *= xmin;
2063: #else
2064: #ifdef DEBUGLINMIN
2065: if(xxs <1.0)
2066: printf(" before xi[%d]=%12.8f", j,xi[j]);
2067: #endif
2068: 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) */
2069: #ifdef DEBUGLINMIN
2070: if(xxs <1.0)
2071: 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 );
2072: #endif
2073: #endif
1.187 brouard 2074: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2075: }
1.191 brouard 2076: #ifdef DEBUGLINMIN
1.203 brouard 2077: printf("\n");
1.191 brouard 2078: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2079: 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 2080: for (j=1;j<=n;j++) {
1.202 brouard 2081: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2082: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2083: if(j % ncovmodel == 0){
1.191 brouard 2084: printf("\n");
1.202 brouard 2085: fprintf(ficlog,"\n");
2086: }
1.191 brouard 2087: }
1.203 brouard 2088: #else
1.191 brouard 2089: #endif
1.126 brouard 2090: free_vector(xicom,1,n);
2091: free_vector(pcom,1,n);
2092: }
2093:
2094:
2095: /*************** powell ************************/
1.162 brouard 2096: /*
2097: Minimization of a function func of n variables. Input consists of an initial starting point
2098: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2099: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2100: such that failure to decrease by more than this amount on one iteration signals doneness. On
2101: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2102: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2103: */
1.224 brouard 2104: #ifdef LINMINORIGINAL
2105: #else
2106: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2107: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2108: #endif
1.126 brouard 2109: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2110: double (*func)(double []))
2111: {
1.224 brouard 2112: #ifdef LINMINORIGINAL
2113: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2114: double (*func)(double []));
1.224 brouard 2115: #else
1.241 brouard 2116: void linmin(double p[], double xi[], int n, double *fret,
2117: double (*func)(double []),int *flat);
1.224 brouard 2118: #endif
1.239 brouard 2119: int i,ibig,j,jk,k;
1.126 brouard 2120: double del,t,*pt,*ptt,*xit;
1.181 brouard 2121: double directest;
1.126 brouard 2122: double fp,fptt;
2123: double *xits;
2124: int niterf, itmp;
1.224 brouard 2125: #ifdef LINMINORIGINAL
2126: #else
2127:
2128: flatdir=ivector(1,n);
2129: for (j=1;j<=n;j++) flatdir[j]=0;
2130: #endif
1.126 brouard 2131:
2132: pt=vector(1,n);
2133: ptt=vector(1,n);
2134: xit=vector(1,n);
2135: xits=vector(1,n);
2136: *fret=(*func)(p);
2137: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2138: rcurr_time = time(NULL);
1.126 brouard 2139: for (*iter=1;;++(*iter)) {
1.187 brouard 2140: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2141: ibig=0;
2142: del=0.0;
1.157 brouard 2143: rlast_time=rcurr_time;
2144: /* (void) gettimeofday(&curr_time,&tzp); */
2145: rcurr_time = time(NULL);
2146: curr_time = *localtime(&rcurr_time);
2147: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2148: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2149: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2150: for (i=1;i<=n;i++) {
1.126 brouard 2151: fprintf(ficrespow," %.12lf", p[i]);
2152: }
1.239 brouard 2153: fprintf(ficrespow,"\n");fflush(ficrespow);
2154: printf("\n#model= 1 + age ");
2155: fprintf(ficlog,"\n#model= 1 + age ");
2156: if(nagesqr==1){
1.241 brouard 2157: printf(" + age*age ");
2158: fprintf(ficlog," + age*age ");
1.239 brouard 2159: }
2160: for(j=1;j <=ncovmodel-2;j++){
2161: if(Typevar[j]==0) {
2162: printf(" + V%d ",Tvar[j]);
2163: fprintf(ficlog," + V%d ",Tvar[j]);
2164: }else if(Typevar[j]==1) {
2165: printf(" + V%d*age ",Tvar[j]);
2166: fprintf(ficlog," + V%d*age ",Tvar[j]);
2167: }else if(Typevar[j]==2) {
2168: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2169: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2170: }
2171: }
1.126 brouard 2172: printf("\n");
1.239 brouard 2173: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2174: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2175: fprintf(ficlog,"\n");
1.239 brouard 2176: for(i=1,jk=1; i <=nlstate; i++){
2177: for(k=1; k <=(nlstate+ndeath); k++){
2178: if (k != i) {
2179: printf("%d%d ",i,k);
2180: fprintf(ficlog,"%d%d ",i,k);
2181: for(j=1; j <=ncovmodel; j++){
2182: printf("%12.7f ",p[jk]);
2183: fprintf(ficlog,"%12.7f ",p[jk]);
2184: jk++;
2185: }
2186: printf("\n");
2187: fprintf(ficlog,"\n");
2188: }
2189: }
2190: }
1.241 brouard 2191: if(*iter <=3 && *iter >1){
1.157 brouard 2192: tml = *localtime(&rcurr_time);
2193: strcpy(strcurr,asctime(&tml));
2194: rforecast_time=rcurr_time;
1.126 brouard 2195: itmp = strlen(strcurr);
2196: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2197: strcurr[itmp-1]='\0';
1.162 brouard 2198: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2199: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2200: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2201: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2202: forecast_time = *localtime(&rforecast_time);
2203: strcpy(strfor,asctime(&forecast_time));
2204: itmp = strlen(strfor);
2205: if(strfor[itmp-1]=='\n')
2206: strfor[itmp-1]='\0';
2207: 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);
2208: 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 2209: }
2210: }
1.187 brouard 2211: for (i=1;i<=n;i++) { /* For each direction i */
2212: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2213: fptt=(*fret);
2214: #ifdef DEBUG
1.203 brouard 2215: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2216: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2217: #endif
1.203 brouard 2218: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2219: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2220: #ifdef LINMINORIGINAL
1.188 brouard 2221: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2222: #else
2223: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2224: flatdir[i]=flat; /* Function is vanishing in that direction i */
2225: #endif
2226: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2227: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2228: /* because that direction will be replaced unless the gain del is small */
2229: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2230: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2231: /* with the new direction. */
2232: del=fabs(fptt-(*fret));
2233: ibig=i;
1.126 brouard 2234: }
2235: #ifdef DEBUG
2236: printf("%d %.12e",i,(*fret));
2237: fprintf(ficlog,"%d %.12e",i,(*fret));
2238: for (j=1;j<=n;j++) {
1.224 brouard 2239: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2240: printf(" x(%d)=%.12e",j,xit[j]);
2241: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2242: }
2243: for(j=1;j<=n;j++) {
1.225 brouard 2244: printf(" p(%d)=%.12e",j,p[j]);
2245: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2246: }
2247: printf("\n");
2248: fprintf(ficlog,"\n");
2249: #endif
1.187 brouard 2250: } /* end loop on each direction i */
2251: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2252: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2253: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2254: for(j=1;j<=n;j++) {
1.225 brouard 2255: if(flatdir[j] >0){
2256: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2257: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2258: }
2259: /* printf("\n"); */
2260: /* fprintf(ficlog,"\n"); */
2261: }
1.243 brouard 2262: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2263: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2264: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2265: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2266: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2267: /* decreased of more than 3.84 */
2268: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2269: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2270: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2271:
1.188 brouard 2272: /* Starting the program with initial values given by a former maximization will simply change */
2273: /* the scales of the directions and the directions, because the are reset to canonical directions */
2274: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2275: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2276: #ifdef DEBUG
2277: int k[2],l;
2278: k[0]=1;
2279: k[1]=-1;
2280: printf("Max: %.12e",(*func)(p));
2281: fprintf(ficlog,"Max: %.12e",(*func)(p));
2282: for (j=1;j<=n;j++) {
2283: printf(" %.12e",p[j]);
2284: fprintf(ficlog," %.12e",p[j]);
2285: }
2286: printf("\n");
2287: fprintf(ficlog,"\n");
2288: for(l=0;l<=1;l++) {
2289: for (j=1;j<=n;j++) {
2290: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2291: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2292: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2293: }
2294: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2295: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2296: }
2297: #endif
2298:
1.224 brouard 2299: #ifdef LINMINORIGINAL
2300: #else
2301: free_ivector(flatdir,1,n);
2302: #endif
1.126 brouard 2303: free_vector(xit,1,n);
2304: free_vector(xits,1,n);
2305: free_vector(ptt,1,n);
2306: free_vector(pt,1,n);
2307: return;
1.192 brouard 2308: } /* enough precision */
1.240 brouard 2309: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2310: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2311: ptt[j]=2.0*p[j]-pt[j];
2312: xit[j]=p[j]-pt[j];
2313: pt[j]=p[j];
2314: }
1.181 brouard 2315: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2316: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2317: if (*iter <=4) {
1.225 brouard 2318: #else
2319: #endif
1.224 brouard 2320: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2321: #else
1.161 brouard 2322: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2323: #endif
1.162 brouard 2324: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2325: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2326: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2327: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2328: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2329: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2330: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2331: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2332: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2333: /* Even if f3 <f1, directest can be negative and t >0 */
2334: /* mu² and del² are equal when f3=f1 */
2335: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2336: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2337: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2338: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2339: #ifdef NRCORIGINAL
2340: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2341: #else
2342: 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 2343: t= t- del*SQR(fp-fptt);
1.183 brouard 2344: #endif
1.202 brouard 2345: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2346: #ifdef DEBUG
1.181 brouard 2347: 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);
2348: 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 2349: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2350: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2351: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2352: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2353: 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);
2354: 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);
2355: #endif
1.183 brouard 2356: #ifdef POWELLORIGINAL
2357: if (t < 0.0) { /* Then we use it for new direction */
2358: #else
1.182 brouard 2359: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2360: 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 2361: 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 2362: 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 2363: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2364: }
1.181 brouard 2365: if (directest < 0.0) { /* Then we use it for new direction */
2366: #endif
1.191 brouard 2367: #ifdef DEBUGLINMIN
1.234 brouard 2368: printf("Before linmin in direction P%d-P0\n",n);
2369: for (j=1;j<=n;j++) {
2370: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2371: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2372: if(j % ncovmodel == 0){
2373: printf("\n");
2374: fprintf(ficlog,"\n");
2375: }
2376: }
1.224 brouard 2377: #endif
2378: #ifdef LINMINORIGINAL
1.234 brouard 2379: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2380: #else
1.234 brouard 2381: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2382: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2383: #endif
1.234 brouard 2384:
1.191 brouard 2385: #ifdef DEBUGLINMIN
1.234 brouard 2386: for (j=1;j<=n;j++) {
2387: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2388: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2389: if(j % ncovmodel == 0){
2390: printf("\n");
2391: fprintf(ficlog,"\n");
2392: }
2393: }
1.224 brouard 2394: #endif
1.234 brouard 2395: for (j=1;j<=n;j++) {
2396: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2397: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2398: }
1.224 brouard 2399: #ifdef LINMINORIGINAL
2400: #else
1.234 brouard 2401: for (j=1, flatd=0;j<=n;j++) {
2402: if(flatdir[j]>0)
2403: flatd++;
2404: }
2405: if(flatd >0){
1.255 brouard 2406: printf("%d flat directions: ",flatd);
2407: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2408: for (j=1;j<=n;j++) {
2409: if(flatdir[j]>0){
2410: printf("%d ",j);
2411: fprintf(ficlog,"%d ",j);
2412: }
2413: }
2414: printf("\n");
2415: fprintf(ficlog,"\n");
2416: }
1.191 brouard 2417: #endif
1.234 brouard 2418: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2419: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2420:
1.126 brouard 2421: #ifdef DEBUG
1.234 brouard 2422: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2423: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2424: for(j=1;j<=n;j++){
2425: printf(" %lf",xit[j]);
2426: fprintf(ficlog," %lf",xit[j]);
2427: }
2428: printf("\n");
2429: fprintf(ficlog,"\n");
1.126 brouard 2430: #endif
1.192 brouard 2431: } /* end of t or directest negative */
1.224 brouard 2432: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2433: #else
1.234 brouard 2434: } /* end if (fptt < fp) */
1.192 brouard 2435: #endif
1.225 brouard 2436: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2437: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2438: #else
1.224 brouard 2439: #endif
1.234 brouard 2440: } /* loop iteration */
1.126 brouard 2441: }
1.234 brouard 2442:
1.126 brouard 2443: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2444:
1.235 brouard 2445: 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 2446: {
1.235 brouard 2447: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2448: (and selected quantitative values in nres)
2449: by left multiplying the unit
1.234 brouard 2450: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2451: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2452: /* Wx is row vector: population in state 1, population in state 2, population dead */
2453: /* or prevalence in state 1, prevalence in state 2, 0 */
2454: /* newm is the matrix after multiplications, its rows are identical at a factor */
2455: /* Initial matrix pimij */
2456: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2457: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2458: /* 0, 0 , 1} */
2459: /*
2460: * and after some iteration: */
2461: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2462: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2463: /* 0, 0 , 1} */
2464: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2465: /* {0.51571254859325999, 0.4842874514067399, */
2466: /* 0.51326036147820708, 0.48673963852179264} */
2467: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2468:
1.126 brouard 2469: int i, ii,j,k;
1.209 brouard 2470: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2471: /* double **matprod2(); */ /* test */
1.218 brouard 2472: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2473: double **newm;
1.209 brouard 2474: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2475: int ncvloop=0;
1.169 brouard 2476:
1.209 brouard 2477: min=vector(1,nlstate);
2478: max=vector(1,nlstate);
2479: meandiff=vector(1,nlstate);
2480:
1.218 brouard 2481: /* Starting with matrix unity */
1.126 brouard 2482: for (ii=1;ii<=nlstate+ndeath;ii++)
2483: for (j=1;j<=nlstate+ndeath;j++){
2484: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2485: }
1.169 brouard 2486:
2487: cov[1]=1.;
2488:
2489: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2490: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2491: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2492: ncvloop++;
1.126 brouard 2493: newm=savm;
2494: /* Covariates have to be included here again */
1.138 brouard 2495: cov[2]=agefin;
1.187 brouard 2496: if(nagesqr==1)
2497: cov[3]= agefin*agefin;;
1.234 brouard 2498: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2499: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2500: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2501: /* 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 2502: }
2503: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2504: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2505: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2506: /* 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 2507: }
1.237 brouard 2508: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2509: if(Dummy[Tvar[Tage[k]]]){
2510: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2511: } else{
1.235 brouard 2512: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2513: }
1.235 brouard 2514: /* 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 2515: }
1.237 brouard 2516: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2517: /* 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 2518: if(Dummy[Tvard[k][1]==0]){
2519: if(Dummy[Tvard[k][2]==0]){
2520: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2521: }else{
2522: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2523: }
2524: }else{
2525: if(Dummy[Tvard[k][2]==0]){
2526: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2527: }else{
2528: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2529: }
2530: }
1.234 brouard 2531: }
1.138 brouard 2532: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2533: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2534: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2535: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2536: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2537: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2538: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2539:
1.126 brouard 2540: savm=oldm;
2541: oldm=newm;
1.209 brouard 2542:
2543: for(j=1; j<=nlstate; j++){
2544: max[j]=0.;
2545: min[j]=1.;
2546: }
2547: for(i=1;i<=nlstate;i++){
2548: sumnew=0;
2549: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2550: for(j=1; j<=nlstate; j++){
2551: prlim[i][j]= newm[i][j]/(1-sumnew);
2552: max[j]=FMAX(max[j],prlim[i][j]);
2553: min[j]=FMIN(min[j],prlim[i][j]);
2554: }
2555: }
2556:
1.126 brouard 2557: maxmax=0.;
1.209 brouard 2558: for(j=1; j<=nlstate; j++){
2559: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2560: maxmax=FMAX(maxmax,meandiff[j]);
2561: /* 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 2562: } /* j loop */
1.203 brouard 2563: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2564: /* 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 2565: if(maxmax < ftolpl){
1.209 brouard 2566: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2567: free_vector(min,1,nlstate);
2568: free_vector(max,1,nlstate);
2569: free_vector(meandiff,1,nlstate);
1.126 brouard 2570: return prlim;
2571: }
1.169 brouard 2572: } /* age loop */
1.208 brouard 2573: /* After some age loop it doesn't converge */
1.209 brouard 2574: 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 2575: 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 2576: /* 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); */
2577: free_vector(min,1,nlstate);
2578: free_vector(max,1,nlstate);
2579: free_vector(meandiff,1,nlstate);
1.208 brouard 2580:
1.169 brouard 2581: return prlim; /* should not reach here */
1.126 brouard 2582: }
2583:
1.217 brouard 2584:
2585: /**** Back Prevalence limit (stable or period prevalence) ****************/
2586:
1.218 brouard 2587: /* 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) */
2588: /* 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 2589: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2590: {
1.218 brouard 2591: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2592: matrix by transitions matrix until convergence is reached with precision ftolpl */
2593: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2594: /* Wx is row vector: population in state 1, population in state 2, population dead */
2595: /* or prevalence in state 1, prevalence in state 2, 0 */
2596: /* newm is the matrix after multiplications, its rows are identical at a factor */
2597: /* Initial matrix pimij */
2598: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2599: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2600: /* 0, 0 , 1} */
2601: /*
2602: * and after some iteration: */
2603: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2604: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2605: /* 0, 0 , 1} */
2606: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2607: /* {0.51571254859325999, 0.4842874514067399, */
2608: /* 0.51326036147820708, 0.48673963852179264} */
2609: /* If we start from prlim again, prlim tends to a constant matrix */
2610:
2611: int i, ii,j,k;
1.247 brouard 2612: int first=0;
1.217 brouard 2613: double *min, *max, *meandiff, maxmax,sumnew=0.;
2614: /* double **matprod2(); */ /* test */
2615: double **out, cov[NCOVMAX+1], **bmij();
2616: double **newm;
1.218 brouard 2617: double **dnewm, **doldm, **dsavm; /* for use */
2618: double **oldm, **savm; /* for use */
2619:
1.217 brouard 2620: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2621: int ncvloop=0;
2622:
2623: min=vector(1,nlstate);
2624: max=vector(1,nlstate);
2625: meandiff=vector(1,nlstate);
2626:
1.218 brouard 2627: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2628: oldm=oldms; savm=savms;
2629:
2630: /* Starting with matrix unity */
2631: for (ii=1;ii<=nlstate+ndeath;ii++)
2632: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2633: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2634: }
2635:
2636: cov[1]=1.;
2637:
2638: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2639: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2640: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2641: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2642: ncvloop++;
1.218 brouard 2643: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2644: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2645: /* Covariates have to be included here again */
2646: cov[2]=agefin;
2647: if(nagesqr==1)
2648: cov[3]= agefin*agefin;;
1.242 brouard 2649: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2650: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2651: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2652: /* printf("bprevalim 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)); */
2653: }
2654: /* for (k=1; k<=cptcovn;k++) { */
2655: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2656: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2657: /* /\* 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])]); *\/ */
2658: /* } */
2659: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2660: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2661: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2662: /* 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]); */
2663: }
2664: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2665: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2666: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2667: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2668: for (k=1; k<=cptcovage;k++){ /* For product with age */
2669: if(Dummy[Tvar[Tage[k]]]){
2670: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2671: } else{
2672: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2673: }
2674: /* 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]); */
2675: }
2676: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2677: /* 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]); */
2678: if(Dummy[Tvard[k][1]==0]){
2679: if(Dummy[Tvard[k][2]==0]){
2680: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2681: }else{
2682: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2683: }
2684: }else{
2685: if(Dummy[Tvard[k][2]==0]){
2686: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2687: }else{
2688: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2689: }
2690: }
1.217 brouard 2691: }
2692:
2693: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2694: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2695: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2696: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2697: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2698: /* ij should be linked to the correct index of cov */
2699: /* age and covariate values ij are in 'cov', but we need to pass
2700: * ij for the observed prevalence at age and status and covariate
2701: * number: prevacurrent[(int)agefin][ii][ij]
2702: */
2703: /* 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 *\/ */
2704: /* 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 *\/ */
2705: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.217 brouard 2706: savm=oldm;
2707: oldm=newm;
2708: for(j=1; j<=nlstate; j++){
2709: max[j]=0.;
2710: min[j]=1.;
2711: }
2712: for(j=1; j<=nlstate; j++){
2713: for(i=1;i<=nlstate;i++){
1.234 brouard 2714: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2715: bprlim[i][j]= newm[i][j];
2716: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2717: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2718: }
2719: }
1.218 brouard 2720:
1.217 brouard 2721: maxmax=0.;
2722: for(i=1; i<=nlstate; i++){
2723: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2724: maxmax=FMAX(maxmax,meandiff[i]);
2725: /* 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); */
2726: } /* j loop */
2727: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2728: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2729: if(maxmax < ftolpl){
1.220 brouard 2730: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2731: free_vector(min,1,nlstate);
2732: free_vector(max,1,nlstate);
2733: free_vector(meandiff,1,nlstate);
2734: return bprlim;
2735: }
2736: } /* age loop */
2737: /* After some age loop it doesn't converge */
1.247 brouard 2738: if(first){
2739: first=1;
2740: 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\
2741: 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);
2742: }
2743: 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 2744: 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);
2745: /* 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); */
2746: free_vector(min,1,nlstate);
2747: free_vector(max,1,nlstate);
2748: free_vector(meandiff,1,nlstate);
2749:
2750: return bprlim; /* should not reach here */
2751: }
2752:
1.126 brouard 2753: /*************** transition probabilities ***************/
2754:
2755: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2756: {
1.138 brouard 2757: /* According to parameters values stored in x and the covariate's values stored in cov,
2758: computes the probability to be observed in state j being in state i by appying the
2759: model to the ncovmodel covariates (including constant and age).
2760: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2761: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2762: ncth covariate in the global vector x is given by the formula:
2763: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2764: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2765: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2766: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2767: Outputs ps[i][j] the probability to be observed in j being in j according to
2768: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2769: */
2770: double s1, lnpijopii;
1.126 brouard 2771: /*double t34;*/
1.164 brouard 2772: int i,j, nc, ii, jj;
1.126 brouard 2773:
1.223 brouard 2774: for(i=1; i<= nlstate; i++){
2775: for(j=1; j<i;j++){
2776: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2777: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2778: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2779: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2780: }
2781: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2782: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2783: }
2784: for(j=i+1; j<=nlstate+ndeath;j++){
2785: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2786: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2787: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2788: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2789: }
2790: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2791: }
2792: }
1.218 brouard 2793:
1.223 brouard 2794: for(i=1; i<= nlstate; i++){
2795: s1=0;
2796: for(j=1; j<i; j++){
2797: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2798: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2799: }
2800: for(j=i+1; j<=nlstate+ndeath; j++){
2801: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2802: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2803: }
2804: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2805: ps[i][i]=1./(s1+1.);
2806: /* Computing other pijs */
2807: for(j=1; j<i; j++)
2808: ps[i][j]= exp(ps[i][j])*ps[i][i];
2809: for(j=i+1; j<=nlstate+ndeath; j++)
2810: ps[i][j]= exp(ps[i][j])*ps[i][i];
2811: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2812: } /* end i */
1.218 brouard 2813:
1.223 brouard 2814: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2815: for(jj=1; jj<= nlstate+ndeath; jj++){
2816: ps[ii][jj]=0;
2817: ps[ii][ii]=1;
2818: }
2819: }
1.218 brouard 2820:
2821:
1.223 brouard 2822: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2823: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2824: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2825: /* } */
2826: /* printf("\n "); */
2827: /* } */
2828: /* printf("\n ");printf("%lf ",cov[2]);*/
2829: /*
2830: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2831: goto end;*/
1.223 brouard 2832: return ps;
1.126 brouard 2833: }
2834:
1.218 brouard 2835: /*************** backward transition probabilities ***************/
2836:
2837: /* 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 ) */
2838: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2839: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2840: {
1.222 brouard 2841: /* Computes the backward probability at age agefin and covariate ij
2842: * and returns in **ps as well as **bmij.
2843: */
1.218 brouard 2844: int i, ii, j,k;
1.222 brouard 2845:
2846: double **out, **pmij();
2847: double sumnew=0.;
1.218 brouard 2848: double agefin;
1.222 brouard 2849:
2850: double **dnewm, **dsavm, **doldm;
2851: double **bbmij;
2852:
1.218 brouard 2853: doldm=ddoldms; /* global pointers */
1.222 brouard 2854: dnewm=ddnewms;
2855: dsavm=ddsavms;
2856:
2857: agefin=cov[2];
2858: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2859: the observed prevalence (with this covariate ij) */
2860: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2861: /* We do have the matrix Px in savm and we need pij */
2862: for (j=1;j<=nlstate+ndeath;j++){
2863: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2864: for (ii=1;ii<=nlstate;ii++){
2865: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2866: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2867: for (ii=1;ii<=nlstate+ndeath;ii++){
2868: if(sumnew >= 1.e-10){
2869: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2870: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2871: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2872: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2873: /* }else */
2874: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2875: }else{
1.242 brouard 2876: ;
2877: /* printf("ii=%d, i=%d, doldm=%lf dsavm=%lf, probs=%lf, sumnew=%lf,agefin=%d\n",ii,j,doldm[ii][j],dsavm[ii][j],prevacurrent[(int)agefin][ii][ij],sumnew, (int)agefin); */
1.222 brouard 2878: }
2879: } /*End ii */
2880: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2881: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2882: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2883: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2884: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2885: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2886: /* left Product of this matrix by diag matrix of prevalences (savm) */
2887: for (j=1;j<=nlstate+ndeath;j++){
2888: for (ii=1;ii<=nlstate+ndeath;ii++){
2889: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2890: }
2891: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2892: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2893: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2894: /* end bmij */
2895: return ps;
1.218 brouard 2896: }
1.217 brouard 2897: /*************** transition probabilities ***************/
2898:
1.218 brouard 2899: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2900: {
2901: /* According to parameters values stored in x and the covariate's values stored in cov,
2902: computes the probability to be observed in state j being in state i by appying the
2903: model to the ncovmodel covariates (including constant and age).
2904: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2905: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2906: ncth covariate in the global vector x is given by the formula:
2907: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2908: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2909: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2910: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2911: Outputs ps[i][j] the probability to be observed in j being in j according to
2912: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2913: */
2914: double s1, lnpijopii;
2915: /*double t34;*/
2916: int i,j, nc, ii, jj;
2917:
1.234 brouard 2918: for(i=1; i<= nlstate; i++){
2919: for(j=1; j<i;j++){
2920: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2921: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2922: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2923: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2924: }
2925: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2926: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2927: }
2928: for(j=i+1; j<=nlstate+ndeath;j++){
2929: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2930: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2931: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2932: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2933: }
2934: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2935: }
2936: }
2937:
2938: for(i=1; i<= nlstate; i++){
2939: s1=0;
2940: for(j=1; j<i; j++){
2941: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2942: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2943: }
2944: for(j=i+1; j<=nlstate+ndeath; j++){
2945: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2946: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2947: }
2948: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2949: ps[i][i]=1./(s1+1.);
2950: /* Computing other pijs */
2951: for(j=1; j<i; j++)
2952: ps[i][j]= exp(ps[i][j])*ps[i][i];
2953: for(j=i+1; j<=nlstate+ndeath; j++)
2954: ps[i][j]= exp(ps[i][j])*ps[i][i];
2955: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2956: } /* end i */
2957:
2958: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2959: for(jj=1; jj<= nlstate+ndeath; jj++){
2960: ps[ii][jj]=0;
2961: ps[ii][ii]=1;
2962: }
2963: }
2964: /* Added for backcast */ /* Transposed matrix too */
2965: for(jj=1; jj<= nlstate+ndeath; jj++){
2966: s1=0.;
2967: for(ii=1; ii<= nlstate+ndeath; ii++){
2968: s1+=ps[ii][jj];
2969: }
2970: for(ii=1; ii<= nlstate; ii++){
2971: ps[ii][jj]=ps[ii][jj]/s1;
2972: }
2973: }
2974: /* Transposition */
2975: for(jj=1; jj<= nlstate+ndeath; jj++){
2976: for(ii=jj; ii<= nlstate+ndeath; ii++){
2977: s1=ps[ii][jj];
2978: ps[ii][jj]=ps[jj][ii];
2979: ps[jj][ii]=s1;
2980: }
2981: }
2982: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2983: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2984: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2985: /* } */
2986: /* printf("\n "); */
2987: /* } */
2988: /* printf("\n ");printf("%lf ",cov[2]);*/
2989: /*
2990: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2991: goto end;*/
2992: return ps;
1.217 brouard 2993: }
2994:
2995:
1.126 brouard 2996: /**************** Product of 2 matrices ******************/
2997:
1.145 brouard 2998: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2999: {
3000: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3001: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3002: /* in, b, out are matrice of pointers which should have been initialized
3003: before: only the contents of out is modified. The function returns
3004: a pointer to pointers identical to out */
1.145 brouard 3005: int i, j, k;
1.126 brouard 3006: for(i=nrl; i<= nrh; i++)
1.145 brouard 3007: for(k=ncolol; k<=ncoloh; k++){
3008: out[i][k]=0.;
3009: for(j=ncl; j<=nch; j++)
3010: out[i][k] +=in[i][j]*b[j][k];
3011: }
1.126 brouard 3012: return out;
3013: }
3014:
3015:
3016: /************* Higher Matrix Product ***************/
3017:
1.235 brouard 3018: 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 3019: {
1.218 brouard 3020: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3021: 'nhstepm*hstepm*stepm' months (i.e. until
3022: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3023: nhstepm*hstepm matrices.
3024: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3025: (typically every 2 years instead of every month which is too big
3026: for the memory).
3027: Model is determined by parameters x and covariates have to be
3028: included manually here.
3029:
3030: */
3031:
3032: int i, j, d, h, k;
1.131 brouard 3033: double **out, cov[NCOVMAX+1];
1.126 brouard 3034: double **newm;
1.187 brouard 3035: double agexact;
1.214 brouard 3036: double agebegin, ageend;
1.126 brouard 3037:
3038: /* Hstepm could be zero and should return the unit matrix */
3039: for (i=1;i<=nlstate+ndeath;i++)
3040: for (j=1;j<=nlstate+ndeath;j++){
3041: oldm[i][j]=(i==j ? 1.0 : 0.0);
3042: po[i][j][0]=(i==j ? 1.0 : 0.0);
3043: }
3044: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3045: for(h=1; h <=nhstepm; h++){
3046: for(d=1; d <=hstepm; d++){
3047: newm=savm;
3048: /* Covariates have to be included here again */
3049: cov[1]=1.;
1.214 brouard 3050: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3051: cov[2]=agexact;
3052: if(nagesqr==1)
1.227 brouard 3053: cov[3]= agexact*agexact;
1.235 brouard 3054: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3055: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3056: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3057: /* 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)); */
3058: }
3059: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3060: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3061: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3062: /* 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]); */
3063: }
3064: for (k=1; k<=cptcovage;k++){
3065: if(Dummy[Tvar[Tage[k]]]){
3066: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3067: } else{
3068: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3069: }
3070: /* 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]); */
3071: }
3072: for (k=1; k<=cptcovprod;k++){ /* */
3073: /* 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]); */
3074: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3075: }
3076: /* for (k=1; k<=cptcovn;k++) */
3077: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3078: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3079: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3080: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3081: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3082:
3083:
1.126 brouard 3084: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3085: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3086: /* right multiplication of oldm by the current matrix */
1.126 brouard 3087: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3088: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3089: /* if((int)age == 70){ */
3090: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3091: /* for(i=1; i<=nlstate+ndeath; i++) { */
3092: /* printf("%d pmmij ",i); */
3093: /* for(j=1;j<=nlstate+ndeath;j++) { */
3094: /* printf("%f ",pmmij[i][j]); */
3095: /* } */
3096: /* printf(" oldm "); */
3097: /* for(j=1;j<=nlstate+ndeath;j++) { */
3098: /* printf("%f ",oldm[i][j]); */
3099: /* } */
3100: /* printf("\n"); */
3101: /* } */
3102: /* } */
1.126 brouard 3103: savm=oldm;
3104: oldm=newm;
3105: }
3106: for(i=1; i<=nlstate+ndeath; i++)
3107: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3108: po[i][j][h]=newm[i][j];
3109: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3110: }
1.128 brouard 3111: /*printf("h=%d ",h);*/
1.126 brouard 3112: } /* end h */
1.218 brouard 3113: /* printf("\n H=%d \n",h); */
1.126 brouard 3114: return po;
3115: }
3116:
1.217 brouard 3117: /************* Higher Back Matrix Product ***************/
1.218 brouard 3118: /* double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, int ij ) */
1.222 brouard 3119: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3120: {
1.218 brouard 3121: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3122: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 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
1.217 brouard 3127: for the memory).
1.218 brouard 3128: Model is determined by parameters x and covariates have to be
3129: included manually here.
1.217 brouard 3130:
1.222 brouard 3131: */
1.217 brouard 3132:
3133: int i, j, d, h, k;
3134: double **out, cov[NCOVMAX+1];
3135: double **newm;
3136: double agexact;
3137: double agebegin, ageend;
1.222 brouard 3138: double **oldm, **savm;
1.217 brouard 3139:
1.222 brouard 3140: oldm=oldms;savm=savms;
1.217 brouard 3141: /* Hstepm could be zero and should return the unit matrix */
3142: for (i=1;i<=nlstate+ndeath;i++)
3143: for (j=1;j<=nlstate+ndeath;j++){
3144: oldm[i][j]=(i==j ? 1.0 : 0.0);
3145: po[i][j][0]=(i==j ? 1.0 : 0.0);
3146: }
3147: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3148: for(h=1; h <=nhstepm; h++){
3149: for(d=1; d <=hstepm; d++){
3150: newm=savm;
3151: /* Covariates have to be included here again */
3152: cov[1]=1.;
3153: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3154: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3155: cov[2]=agexact;
3156: if(nagesqr==1)
1.222 brouard 3157: cov[3]= agexact*agexact;
1.218 brouard 3158: for (k=1; k<=cptcovn;k++)
1.222 brouard 3159: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3160: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3161: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3162: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3163: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3164: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3165: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3166: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3167: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
1.218 brouard 3168:
3169:
1.217 brouard 3170: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3171: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3172: /* Careful transposed matrix */
1.222 brouard 3173: /* age is in cov[2] */
1.218 brouard 3174: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3175: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3176: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3177: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3178: /* if((int)age == 70){ */
3179: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3180: /* for(i=1; i<=nlstate+ndeath; i++) { */
3181: /* printf("%d pmmij ",i); */
3182: /* for(j=1;j<=nlstate+ndeath;j++) { */
3183: /* printf("%f ",pmmij[i][j]); */
3184: /* } */
3185: /* printf(" oldm "); */
3186: /* for(j=1;j<=nlstate+ndeath;j++) { */
3187: /* printf("%f ",oldm[i][j]); */
3188: /* } */
3189: /* printf("\n"); */
3190: /* } */
3191: /* } */
3192: savm=oldm;
3193: oldm=newm;
3194: }
3195: for(i=1; i<=nlstate+ndeath; i++)
3196: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3197: po[i][j][h]=newm[i][j];
3198: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3199: }
3200: /*printf("h=%d ",h);*/
3201: } /* end h */
1.222 brouard 3202: /* printf("\n H=%d \n",h); */
1.217 brouard 3203: return po;
3204: }
3205:
3206:
1.162 brouard 3207: #ifdef NLOPT
3208: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3209: double fret;
3210: double *xt;
3211: int j;
3212: myfunc_data *d2 = (myfunc_data *) pd;
3213: /* xt = (p1-1); */
3214: xt=vector(1,n);
3215: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3216:
3217: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3218: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3219: printf("Function = %.12lf ",fret);
3220: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3221: printf("\n");
3222: free_vector(xt,1,n);
3223: return fret;
3224: }
3225: #endif
1.126 brouard 3226:
3227: /*************** log-likelihood *************/
3228: double func( double *x)
3229: {
1.226 brouard 3230: int i, ii, j, k, mi, d, kk;
3231: int ioffset=0;
3232: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3233: double **out;
3234: double lli; /* Individual log likelihood */
3235: int s1, s2;
1.228 brouard 3236: 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 3237: double bbh, survp;
3238: long ipmx;
3239: double agexact;
3240: /*extern weight */
3241: /* We are differentiating ll according to initial status */
3242: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3243: /*for(i=1;i<imx;i++)
3244: printf(" %d\n",s[4][i]);
3245: */
1.162 brouard 3246:
1.226 brouard 3247: ++countcallfunc;
1.162 brouard 3248:
1.226 brouard 3249: cov[1]=1.;
1.126 brouard 3250:
1.226 brouard 3251: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3252: ioffset=0;
1.226 brouard 3253: if(mle==1){
3254: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3255: /* Computes the values of the ncovmodel covariates of the model
3256: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3257: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3258: to be observed in j being in i according to the model.
3259: */
1.243 brouard 3260: ioffset=2+nagesqr ;
1.233 brouard 3261: /* Fixed */
1.234 brouard 3262: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3263: 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)*/
3264: }
1.226 brouard 3265: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3266: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3267: has been calculated etc */
3268: /* For an individual i, wav[i] gives the number of effective waves */
3269: /* We compute the contribution to Likelihood of each effective transition
3270: mw[mi][i] is real wave of the mi th effectve wave */
3271: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3272: s2=s[mw[mi+1][i]][i];
3273: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3274: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3275: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3276: */
3277: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3278: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3279: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3280: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3281: }
3282: for (ii=1;ii<=nlstate+ndeath;ii++)
3283: for (j=1;j<=nlstate+ndeath;j++){
3284: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3285: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3286: }
3287: for(d=0; d<dh[mi][i]; d++){
3288: newm=savm;
3289: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3290: cov[2]=agexact;
3291: if(nagesqr==1)
3292: cov[3]= agexact*agexact; /* Should be changed here */
3293: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3294: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3295: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3296: else
3297: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3298: }
3299: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3300: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3301: savm=oldm;
3302: oldm=newm;
3303: } /* end mult */
3304:
3305: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3306: /* But now since version 0.9 we anticipate for bias at large stepm.
3307: * If stepm is larger than one month (smallest stepm) and if the exact delay
3308: * (in months) between two waves is not a multiple of stepm, we rounded to
3309: * the nearest (and in case of equal distance, to the lowest) interval but now
3310: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3311: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3312: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3313: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3314: * -stepm/2 to stepm/2 .
3315: * For stepm=1 the results are the same as for previous versions of Imach.
3316: * For stepm > 1 the results are less biased than in previous versions.
3317: */
1.234 brouard 3318: s1=s[mw[mi][i]][i];
3319: s2=s[mw[mi+1][i]][i];
3320: bbh=(double)bh[mi][i]/(double)stepm;
3321: /* bias bh is positive if real duration
3322: * is higher than the multiple of stepm and negative otherwise.
3323: */
3324: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3325: if( s2 > nlstate){
3326: /* i.e. if s2 is a death state and if the date of death is known
3327: then the contribution to the likelihood is the probability to
3328: die between last step unit time and current step unit time,
3329: which is also equal to probability to die before dh
3330: minus probability to die before dh-stepm .
3331: In version up to 0.92 likelihood was computed
3332: as if date of death was unknown. Death was treated as any other
3333: health state: the date of the interview describes the actual state
3334: and not the date of a change in health state. The former idea was
3335: to consider that at each interview the state was recorded
3336: (healthy, disable or death) and IMaCh was corrected; but when we
3337: introduced the exact date of death then we should have modified
3338: the contribution of an exact death to the likelihood. This new
3339: contribution is smaller and very dependent of the step unit
3340: stepm. It is no more the probability to die between last interview
3341: and month of death but the probability to survive from last
3342: interview up to one month before death multiplied by the
3343: probability to die within a month. Thanks to Chris
3344: Jackson for correcting this bug. Former versions increased
3345: mortality artificially. The bad side is that we add another loop
3346: which slows down the processing. The difference can be up to 10%
3347: lower mortality.
3348: */
3349: /* If, at the beginning of the maximization mostly, the
3350: cumulative probability or probability to be dead is
3351: constant (ie = 1) over time d, the difference is equal to
3352: 0. out[s1][3] = savm[s1][3]: probability, being at state
3353: s1 at precedent wave, to be dead a month before current
3354: wave is equal to probability, being at state s1 at
3355: precedent wave, to be dead at mont of the current
3356: wave. Then the observed probability (that this person died)
3357: is null according to current estimated parameter. In fact,
3358: it should be very low but not zero otherwise the log go to
3359: infinity.
3360: */
1.183 brouard 3361: /* #ifdef INFINITYORIGINAL */
3362: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3363: /* #else */
3364: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3365: /* lli=log(mytinydouble); */
3366: /* else */
3367: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3368: /* #endif */
1.226 brouard 3369: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3370:
1.226 brouard 3371: } else if ( s2==-1 ) { /* alive */
3372: for (j=1,survp=0. ; j<=nlstate; j++)
3373: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3374: /*survp += out[s1][j]; */
3375: lli= log(survp);
3376: }
3377: else if (s2==-4) {
3378: for (j=3,survp=0. ; j<=nlstate; j++)
3379: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3380: lli= log(survp);
3381: }
3382: else if (s2==-5) {
3383: for (j=1,survp=0. ; j<=2; j++)
3384: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3385: lli= log(survp);
3386: }
3387: else{
3388: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3389: /* 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 */
3390: }
3391: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3392: /*if(lli ==000.0)*/
3393: /*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); */
3394: ipmx +=1;
3395: sw += weight[i];
3396: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3397: /* if (lli < log(mytinydouble)){ */
3398: /* 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); */
3399: /* 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]); */
3400: /* } */
3401: } /* end of wave */
3402: } /* end of individual */
3403: } else if(mle==2){
3404: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3405: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3406: for(mi=1; mi<= wav[i]-1; mi++){
3407: for (ii=1;ii<=nlstate+ndeath;ii++)
3408: for (j=1;j<=nlstate+ndeath;j++){
3409: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3410: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3411: }
3412: for(d=0; d<=dh[mi][i]; d++){
3413: newm=savm;
3414: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3415: cov[2]=agexact;
3416: if(nagesqr==1)
3417: cov[3]= agexact*agexact;
3418: for (kk=1; kk<=cptcovage;kk++) {
3419: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3420: }
3421: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3422: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3423: savm=oldm;
3424: oldm=newm;
3425: } /* end mult */
3426:
3427: s1=s[mw[mi][i]][i];
3428: s2=s[mw[mi+1][i]][i];
3429: bbh=(double)bh[mi][i]/(double)stepm;
3430: 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 */
3431: ipmx +=1;
3432: sw += weight[i];
3433: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3434: } /* end of wave */
3435: } /* end of individual */
3436: } else if(mle==3){ /* exponential inter-extrapolation */
3437: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3438: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3439: for(mi=1; mi<= wav[i]-1; mi++){
3440: for (ii=1;ii<=nlstate+ndeath;ii++)
3441: for (j=1;j<=nlstate+ndeath;j++){
3442: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3443: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3444: }
3445: for(d=0; d<dh[mi][i]; d++){
3446: newm=savm;
3447: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3448: cov[2]=agexact;
3449: if(nagesqr==1)
3450: cov[3]= agexact*agexact;
3451: for (kk=1; kk<=cptcovage;kk++) {
3452: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3453: }
3454: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3455: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3456: savm=oldm;
3457: oldm=newm;
3458: } /* end mult */
3459:
3460: s1=s[mw[mi][i]][i];
3461: s2=s[mw[mi+1][i]][i];
3462: bbh=(double)bh[mi][i]/(double)stepm;
3463: 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 */
3464: ipmx +=1;
3465: sw += weight[i];
3466: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3467: } /* end of wave */
3468: } /* end of individual */
3469: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3470: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3471: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3472: for(mi=1; mi<= wav[i]-1; mi++){
3473: for (ii=1;ii<=nlstate+ndeath;ii++)
3474: for (j=1;j<=nlstate+ndeath;j++){
3475: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3476: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3477: }
3478: for(d=0; d<dh[mi][i]; d++){
3479: newm=savm;
3480: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3481: cov[2]=agexact;
3482: if(nagesqr==1)
3483: cov[3]= agexact*agexact;
3484: for (kk=1; kk<=cptcovage;kk++) {
3485: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3486: }
1.126 brouard 3487:
1.226 brouard 3488: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3489: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3490: savm=oldm;
3491: oldm=newm;
3492: } /* end mult */
3493:
3494: s1=s[mw[mi][i]][i];
3495: s2=s[mw[mi+1][i]][i];
3496: if( s2 > nlstate){
3497: lli=log(out[s1][s2] - savm[s1][s2]);
3498: } else if ( s2==-1 ) { /* alive */
3499: for (j=1,survp=0. ; j<=nlstate; j++)
3500: survp += out[s1][j];
3501: lli= log(survp);
3502: }else{
3503: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3504: }
3505: ipmx +=1;
3506: sw += weight[i];
3507: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3508: /* 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 3509: } /* end of wave */
3510: } /* end of individual */
3511: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3512: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3513: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3514: for(mi=1; mi<= wav[i]-1; mi++){
3515: for (ii=1;ii<=nlstate+ndeath;ii++)
3516: for (j=1;j<=nlstate+ndeath;j++){
3517: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3518: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3519: }
3520: for(d=0; d<dh[mi][i]; d++){
3521: newm=savm;
3522: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3523: cov[2]=agexact;
3524: if(nagesqr==1)
3525: cov[3]= agexact*agexact;
3526: for (kk=1; kk<=cptcovage;kk++) {
3527: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3528: }
1.126 brouard 3529:
1.226 brouard 3530: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3531: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3532: savm=oldm;
3533: oldm=newm;
3534: } /* end mult */
3535:
3536: s1=s[mw[mi][i]][i];
3537: s2=s[mw[mi+1][i]][i];
3538: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3539: ipmx +=1;
3540: sw += weight[i];
3541: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3542: /*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]);*/
3543: } /* end of wave */
3544: } /* end of individual */
3545: } /* End of if */
3546: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3547: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3548: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3549: return -l;
1.126 brouard 3550: }
3551:
3552: /*************** log-likelihood *************/
3553: double funcone( double *x)
3554: {
1.228 brouard 3555: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3556: int i, ii, j, k, mi, d, kk;
1.228 brouard 3557: int ioffset=0;
1.131 brouard 3558: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3559: double **out;
3560: double lli; /* Individual log likelihood */
3561: double llt;
3562: int s1, s2;
1.228 brouard 3563: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3564:
1.126 brouard 3565: double bbh, survp;
1.187 brouard 3566: double agexact;
1.214 brouard 3567: double agebegin, ageend;
1.126 brouard 3568: /*extern weight */
3569: /* We are differentiating ll according to initial status */
3570: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3571: /*for(i=1;i<imx;i++)
3572: printf(" %d\n",s[4][i]);
3573: */
3574: cov[1]=1.;
3575:
3576: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3577: ioffset=0;
3578: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3579: /* ioffset=2+nagesqr+cptcovage; */
3580: ioffset=2+nagesqr;
1.232 brouard 3581: /* Fixed */
1.224 brouard 3582: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3583: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3584: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3585: 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)*/
3586: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3587: /* cov[2+6]=covar[Tvar[6]][i]; */
3588: /* cov[2+6]=covar[2][i]; V2 */
3589: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3590: /* cov[2+7]=covar[Tvar[7]][i]; */
3591: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3592: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3593: /* cov[2+9]=covar[Tvar[9]][i]; */
3594: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3595: }
1.232 brouard 3596: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3597: /* 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?)*\/ */
3598: /* } */
1.231 brouard 3599: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3600: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3601: /* } */
1.225 brouard 3602:
1.233 brouard 3603:
3604: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3605: /* Wave varying (but not age varying) */
3606: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3607: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3608: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3609: }
1.232 brouard 3610: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3611: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3612: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3613: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3614: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3615: /* 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 3616: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3617: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3618: /* /\* 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]); *\/ */
3619: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3620: /* } */
1.126 brouard 3621: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3622: for (j=1;j<=nlstate+ndeath;j++){
3623: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3624: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3625: }
1.214 brouard 3626:
3627: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3628: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3629: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3630: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3631: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3632: and mw[mi+1][i]. dh depends on stepm.*/
3633: newm=savm;
1.247 brouard 3634: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3635: cov[2]=agexact;
3636: if(nagesqr==1)
3637: cov[3]= agexact*agexact;
3638: for (kk=1; kk<=cptcovage;kk++) {
3639: if(!FixedV[Tvar[Tage[kk]]])
3640: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3641: else
3642: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3643: }
3644: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3645: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3646: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3647: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3648: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3649: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3650: savm=oldm;
3651: oldm=newm;
1.126 brouard 3652: } /* end mult */
3653:
3654: s1=s[mw[mi][i]][i];
3655: s2=s[mw[mi+1][i]][i];
1.217 brouard 3656: /* if(s2==-1){ */
3657: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3658: /* /\* exit(1); *\/ */
3659: /* } */
1.126 brouard 3660: bbh=(double)bh[mi][i]/(double)stepm;
3661: /* bias is positive if real duration
3662: * is higher than the multiple of stepm and negative otherwise.
3663: */
3664: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3665: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3666: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3667: for (j=1,survp=0. ; j<=nlstate; j++)
3668: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3669: lli= log(survp);
1.126 brouard 3670: }else if (mle==1){
1.242 brouard 3671: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3672: } else if(mle==2){
1.242 brouard 3673: 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 3674: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3675: 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 3676: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3677: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3678: } else{ /* mle=0 back to 1 */
1.242 brouard 3679: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3680: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3681: } /* End of if */
3682: ipmx +=1;
3683: sw += weight[i];
3684: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3685: /*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 3686: if(globpr){
1.246 brouard 3687: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3688: %11.6f %11.6f %11.6f ", \
1.242 brouard 3689: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3690: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3691: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3692: llt +=ll[k]*gipmx/gsw;
3693: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3694: }
3695: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3696: }
1.232 brouard 3697: } /* end of wave */
3698: } /* end of individual */
3699: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3700: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3701: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3702: if(globpr==0){ /* First time we count the contributions and weights */
3703: gipmx=ipmx;
3704: gsw=sw;
3705: }
3706: return -l;
1.126 brouard 3707: }
3708:
3709:
3710: /*************** function likelione ***********/
3711: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3712: {
3713: /* This routine should help understanding what is done with
3714: the selection of individuals/waves and
3715: to check the exact contribution to the likelihood.
3716: Plotting could be done.
3717: */
3718: int k;
3719:
3720: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3721: strcpy(fileresilk,"ILK_");
1.202 brouard 3722: strcat(fileresilk,fileresu);
1.126 brouard 3723: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3724: printf("Problem with resultfile: %s\n", fileresilk);
3725: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3726: }
1.214 brouard 3727: 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");
3728: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3729: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3730: for(k=1; k<=nlstate; k++)
3731: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3732: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3733: }
3734:
3735: *fretone=(*funcone)(p);
3736: if(*globpri !=0){
3737: fclose(ficresilk);
1.205 brouard 3738: if (mle ==0)
3739: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3740: else if(mle >=1)
3741: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3742: fprintf(fichtm," You should at least run with mle >= 1 to get starting values corresponding to the optimized parameters in order to visualize the real contribution of each individual/wave: <a href=\"%s\">%s</a><br>\n",subdirf(fileresilk),subdirf(fileresilk));
1.207 brouard 3743:
1.208 brouard 3744:
3745: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3746: 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 3747: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3748: }
1.207 brouard 3749: 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 3750: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3751: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3752: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3753: fflush(fichtm);
1.205 brouard 3754: }
1.126 brouard 3755: return;
3756: }
3757:
3758:
3759: /*********** Maximum Likelihood Estimation ***************/
3760:
3761: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3762: {
1.165 brouard 3763: int i,j, iter=0;
1.126 brouard 3764: double **xi;
3765: double fret;
3766: double fretone; /* Only one call to likelihood */
3767: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3768:
3769: #ifdef NLOPT
3770: int creturn;
3771: nlopt_opt opt;
3772: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3773: double *lb;
3774: double minf; /* the minimum objective value, upon return */
3775: double * p1; /* Shifted parameters from 0 instead of 1 */
3776: myfunc_data dinst, *d = &dinst;
3777: #endif
3778:
3779:
1.126 brouard 3780: xi=matrix(1,npar,1,npar);
3781: for (i=1;i<=npar;i++)
3782: for (j=1;j<=npar;j++)
3783: xi[i][j]=(i==j ? 1.0 : 0.0);
3784: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3785: strcpy(filerespow,"POW_");
1.126 brouard 3786: strcat(filerespow,fileres);
3787: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3788: printf("Problem with resultfile: %s\n", filerespow);
3789: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3790: }
3791: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3792: for (i=1;i<=nlstate;i++)
3793: for(j=1;j<=nlstate+ndeath;j++)
3794: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3795: fprintf(ficrespow,"\n");
1.162 brouard 3796: #ifdef POWELL
1.126 brouard 3797: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3798: #endif
1.126 brouard 3799:
1.162 brouard 3800: #ifdef NLOPT
3801: #ifdef NEWUOA
3802: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3803: #else
3804: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3805: #endif
3806: lb=vector(0,npar-1);
3807: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3808: nlopt_set_lower_bounds(opt, lb);
3809: nlopt_set_initial_step1(opt, 0.1);
3810:
3811: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3812: d->function = func;
3813: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3814: nlopt_set_min_objective(opt, myfunc, d);
3815: nlopt_set_xtol_rel(opt, ftol);
3816: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3817: printf("nlopt failed! %d\n",creturn);
3818: }
3819: else {
3820: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3821: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3822: iter=1; /* not equal */
3823: }
3824: nlopt_destroy(opt);
3825: #endif
1.126 brouard 3826: free_matrix(xi,1,npar,1,npar);
3827: fclose(ficrespow);
1.203 brouard 3828: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3829: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3830: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3831:
3832: }
3833:
3834: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3835: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3836: {
3837: double **a,**y,*x,pd;
1.203 brouard 3838: /* double **hess; */
1.164 brouard 3839: int i, j;
1.126 brouard 3840: int *indx;
3841:
3842: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3843: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3844: void lubksb(double **a, int npar, int *indx, double b[]) ;
3845: void ludcmp(double **a, int npar, int *indx, double *d) ;
3846: double gompertz(double p[]);
1.203 brouard 3847: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3848:
3849: printf("\nCalculation of the hessian matrix. Wait...\n");
3850: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3851: for (i=1;i<=npar;i++){
1.203 brouard 3852: printf("%d-",i);fflush(stdout);
3853: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3854:
3855: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3856:
3857: /* printf(" %f ",p[i]);
3858: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3859: }
3860:
3861: for (i=1;i<=npar;i++) {
3862: for (j=1;j<=npar;j++) {
3863: if (j>i) {
1.203 brouard 3864: printf(".%d-%d",i,j);fflush(stdout);
3865: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3866: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3867:
3868: hess[j][i]=hess[i][j];
3869: /*printf(" %lf ",hess[i][j]);*/
3870: }
3871: }
3872: }
3873: printf("\n");
3874: fprintf(ficlog,"\n");
3875:
3876: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3877: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3878:
3879: a=matrix(1,npar,1,npar);
3880: y=matrix(1,npar,1,npar);
3881: x=vector(1,npar);
3882: indx=ivector(1,npar);
3883: for (i=1;i<=npar;i++)
3884: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3885: ludcmp(a,npar,indx,&pd);
3886:
3887: for (j=1;j<=npar;j++) {
3888: for (i=1;i<=npar;i++) x[i]=0;
3889: x[j]=1;
3890: lubksb(a,npar,indx,x);
3891: for (i=1;i<=npar;i++){
3892: matcov[i][j]=x[i];
3893: }
3894: }
3895:
3896: printf("\n#Hessian matrix#\n");
3897: fprintf(ficlog,"\n#Hessian matrix#\n");
3898: for (i=1;i<=npar;i++) {
3899: for (j=1;j<=npar;j++) {
1.203 brouard 3900: printf("%.6e ",hess[i][j]);
3901: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3902: }
3903: printf("\n");
3904: fprintf(ficlog,"\n");
3905: }
3906:
1.203 brouard 3907: /* printf("\n#Covariance matrix#\n"); */
3908: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3909: /* for (i=1;i<=npar;i++) { */
3910: /* for (j=1;j<=npar;j++) { */
3911: /* printf("%.6e ",matcov[i][j]); */
3912: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3913: /* } */
3914: /* printf("\n"); */
3915: /* fprintf(ficlog,"\n"); */
3916: /* } */
3917:
1.126 brouard 3918: /* Recompute Inverse */
1.203 brouard 3919: /* for (i=1;i<=npar;i++) */
3920: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3921: /* ludcmp(a,npar,indx,&pd); */
3922:
3923: /* printf("\n#Hessian matrix recomputed#\n"); */
3924:
3925: /* for (j=1;j<=npar;j++) { */
3926: /* for (i=1;i<=npar;i++) x[i]=0; */
3927: /* x[j]=1; */
3928: /* lubksb(a,npar,indx,x); */
3929: /* for (i=1;i<=npar;i++){ */
3930: /* y[i][j]=x[i]; */
3931: /* printf("%.3e ",y[i][j]); */
3932: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3933: /* } */
3934: /* printf("\n"); */
3935: /* fprintf(ficlog,"\n"); */
3936: /* } */
3937:
3938: /* Verifying the inverse matrix */
3939: #ifdef DEBUGHESS
3940: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3941:
1.203 brouard 3942: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3943: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3944:
3945: for (j=1;j<=npar;j++) {
3946: for (i=1;i<=npar;i++){
1.203 brouard 3947: printf("%.2f ",y[i][j]);
3948: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3949: }
3950: printf("\n");
3951: fprintf(ficlog,"\n");
3952: }
1.203 brouard 3953: #endif
1.126 brouard 3954:
3955: free_matrix(a,1,npar,1,npar);
3956: free_matrix(y,1,npar,1,npar);
3957: free_vector(x,1,npar);
3958: free_ivector(indx,1,npar);
1.203 brouard 3959: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3960:
3961:
3962: }
3963:
3964: /*************** hessian matrix ****************/
3965: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3966: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3967: int i;
3968: int l=1, lmax=20;
1.203 brouard 3969: double k1,k2, res, fx;
1.132 brouard 3970: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3971: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3972: int k=0,kmax=10;
3973: double l1;
3974:
3975: fx=func(x);
3976: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3977: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3978: l1=pow(10,l);
3979: delts=delt;
3980: for(k=1 ; k <kmax; k=k+1){
3981: delt = delta*(l1*k);
3982: p2[theta]=x[theta] +delt;
1.145 brouard 3983: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3984: p2[theta]=x[theta]-delt;
3985: k2=func(p2)-fx;
3986: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3987: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3988:
1.203 brouard 3989: #ifdef DEBUGHESSII
1.126 brouard 3990: 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);
3991: 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);
3992: #endif
3993: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3994: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3995: k=kmax;
3996: }
3997: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3998: k=kmax; l=lmax*10;
1.126 brouard 3999: }
4000: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4001: delts=delt;
4002: }
1.203 brouard 4003: } /* End loop k */
1.126 brouard 4004: }
4005: delti[theta]=delts;
4006: return res;
4007:
4008: }
4009:
1.203 brouard 4010: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4011: {
4012: int i;
1.164 brouard 4013: int l=1, lmax=20;
1.126 brouard 4014: double k1,k2,k3,k4,res,fx;
1.132 brouard 4015: double p2[MAXPARM+1];
1.203 brouard 4016: int k, kmax=1;
4017: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4018:
4019: int firstime=0;
1.203 brouard 4020:
1.126 brouard 4021: fx=func(x);
1.203 brouard 4022: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4023: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4024: p2[thetai]=x[thetai]+delti[thetai]*k;
4025: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4026: k1=func(p2)-fx;
4027:
1.203 brouard 4028: p2[thetai]=x[thetai]+delti[thetai]*k;
4029: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4030: k2=func(p2)-fx;
4031:
1.203 brouard 4032: p2[thetai]=x[thetai]-delti[thetai]*k;
4033: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4034: k3=func(p2)-fx;
4035:
1.203 brouard 4036: p2[thetai]=x[thetai]-delti[thetai]*k;
4037: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4038: k4=func(p2)-fx;
1.203 brouard 4039: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4040: if(k1*k2*k3*k4 <0.){
1.208 brouard 4041: firstime=1;
1.203 brouard 4042: kmax=kmax+10;
1.208 brouard 4043: }
4044: if(kmax >=10 || firstime ==1){
1.246 brouard 4045: 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);
4046: 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 4047: 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);
4048: 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);
4049: }
4050: #ifdef DEBUGHESSIJ
4051: v1=hess[thetai][thetai];
4052: v2=hess[thetaj][thetaj];
4053: cv12=res;
4054: /* Computing eigen value of Hessian matrix */
4055: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4056: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4057: if ((lc2 <0) || (lc1 <0) ){
4058: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4059: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4060: 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);
4061: 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);
4062: }
1.126 brouard 4063: #endif
4064: }
4065: return res;
4066: }
4067:
1.203 brouard 4068: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4069: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4070: /* { */
4071: /* int i; */
4072: /* int l=1, lmax=20; */
4073: /* double k1,k2,k3,k4,res,fx; */
4074: /* double p2[MAXPARM+1]; */
4075: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4076: /* int k=0,kmax=10; */
4077: /* double l1; */
4078:
4079: /* fx=func(x); */
4080: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4081: /* l1=pow(10,l); */
4082: /* delts=delt; */
4083: /* for(k=1 ; k <kmax; k=k+1){ */
4084: /* delt = delti*(l1*k); */
4085: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4086: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4087: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4088: /* k1=func(p2)-fx; */
4089:
4090: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4091: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4092: /* k2=func(p2)-fx; */
4093:
4094: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4095: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4096: /* k3=func(p2)-fx; */
4097:
4098: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4099: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4100: /* k4=func(p2)-fx; */
4101: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4102: /* #ifdef DEBUGHESSIJ */
4103: /* 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); */
4104: /* 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); */
4105: /* #endif */
4106: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4107: /* k=kmax; */
4108: /* } */
4109: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4110: /* k=kmax; l=lmax*10; */
4111: /* } */
4112: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4113: /* delts=delt; */
4114: /* } */
4115: /* } /\* End loop k *\/ */
4116: /* } */
4117: /* delti[theta]=delts; */
4118: /* return res; */
4119: /* } */
4120:
4121:
1.126 brouard 4122: /************** Inverse of matrix **************/
4123: void ludcmp(double **a, int n, int *indx, double *d)
4124: {
4125: int i,imax,j,k;
4126: double big,dum,sum,temp;
4127: double *vv;
4128:
4129: vv=vector(1,n);
4130: *d=1.0;
4131: for (i=1;i<=n;i++) {
4132: big=0.0;
4133: for (j=1;j<=n;j++)
4134: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 ! brouard 4135: if (big == 0.0){
! 4136: printf(" Singular Hessian matrix at row %d:\n",i);
! 4137: for (j=1;j<=n;j++) {
! 4138: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
! 4139: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
! 4140: }
! 4141: fflush(ficlog);
! 4142: fclose(ficlog);
! 4143: nrerror("Singular matrix in routine ludcmp");
! 4144: }
1.126 brouard 4145: vv[i]=1.0/big;
4146: }
4147: for (j=1;j<=n;j++) {
4148: for (i=1;i<j;i++) {
4149: sum=a[i][j];
4150: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4151: a[i][j]=sum;
4152: }
4153: big=0.0;
4154: for (i=j;i<=n;i++) {
4155: sum=a[i][j];
4156: for (k=1;k<j;k++)
4157: sum -= a[i][k]*a[k][j];
4158: a[i][j]=sum;
4159: if ( (dum=vv[i]*fabs(sum)) >= big) {
4160: big=dum;
4161: imax=i;
4162: }
4163: }
4164: if (j != imax) {
4165: for (k=1;k<=n;k++) {
4166: dum=a[imax][k];
4167: a[imax][k]=a[j][k];
4168: a[j][k]=dum;
4169: }
4170: *d = -(*d);
4171: vv[imax]=vv[j];
4172: }
4173: indx[j]=imax;
4174: if (a[j][j] == 0.0) a[j][j]=TINY;
4175: if (j != n) {
4176: dum=1.0/(a[j][j]);
4177: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4178: }
4179: }
4180: free_vector(vv,1,n); /* Doesn't work */
4181: ;
4182: }
4183:
4184: void lubksb(double **a, int n, int *indx, double b[])
4185: {
4186: int i,ii=0,ip,j;
4187: double sum;
4188:
4189: for (i=1;i<=n;i++) {
4190: ip=indx[i];
4191: sum=b[ip];
4192: b[ip]=b[i];
4193: if (ii)
4194: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4195: else if (sum) ii=i;
4196: b[i]=sum;
4197: }
4198: for (i=n;i>=1;i--) {
4199: sum=b[i];
4200: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4201: b[i]=sum/a[i][i];
4202: }
4203: }
4204:
4205: void pstamp(FILE *fichier)
4206: {
1.196 brouard 4207: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4208: }
4209:
1.253 brouard 4210: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4211:
4212: /* y=a+bx regression */
4213: double sumx = 0.0; /* sum of x */
4214: double sumx2 = 0.0; /* sum of x**2 */
4215: double sumxy = 0.0; /* sum of x * y */
4216: double sumy = 0.0; /* sum of y */
4217: double sumy2 = 0.0; /* sum of y**2 */
4218: double sume2; /* sum of square or residuals */
4219: double yhat;
4220:
4221: double denom=0;
4222: int i;
4223: int ne=*no;
4224:
4225: for ( i=ifi, ne=0;i<=ila;i++) {
4226: if(!isfinite(x[i]) || !isfinite(y[i])){
4227: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4228: continue;
4229: }
4230: ne=ne+1;
4231: sumx += x[i];
4232: sumx2 += x[i]*x[i];
4233: sumxy += x[i] * y[i];
4234: sumy += y[i];
4235: sumy2 += y[i]*y[i];
4236: denom = (ne * sumx2 - sumx*sumx);
4237: /* 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); */
4238: }
4239:
4240: denom = (ne * sumx2 - sumx*sumx);
4241: if (denom == 0) {
4242: // vertical, slope m is infinity
4243: *b = INFINITY;
4244: *a = 0;
4245: if (r) *r = 0;
4246: return 1;
4247: }
4248:
4249: *b = (ne * sumxy - sumx * sumy) / denom;
4250: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4251: if (r!=NULL) {
4252: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4253: sqrt((sumx2 - sumx*sumx/ne) *
4254: (sumy2 - sumy*sumy/ne));
4255: }
4256: *no=ne;
4257: for ( i=ifi, ne=0;i<=ila;i++) {
4258: if(!isfinite(x[i]) || !isfinite(y[i])){
4259: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4260: continue;
4261: }
4262: ne=ne+1;
4263: yhat = y[i] - *a -*b* x[i];
4264: sume2 += yhat * yhat ;
4265:
4266: denom = (ne * sumx2 - sumx*sumx);
4267: /* 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); */
4268: }
4269: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4270: *sa= *sb * sqrt(sumx2/ne);
4271:
4272: return 0;
4273: }
4274:
1.126 brouard 4275: /************ Frequencies ********************/
1.251 brouard 4276: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4277: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4278: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4279: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4280:
1.253 brouard 4281: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0;
1.226 brouard 4282: int iind=0, iage=0;
4283: int mi; /* Effective wave */
4284: int first;
4285: double ***freq; /* Frequencies */
1.253 brouard 4286: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4287: int no;
1.226 brouard 4288: double *meanq;
4289: double **meanqt;
4290: double *pp, **prop, *posprop, *pospropt;
4291: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4292: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4293: double agebegin, ageend;
4294:
4295: pp=vector(1,nlstate);
1.251 brouard 4296: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4297: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4298: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4299: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4300: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4301: meanqt=matrix(1,lastpass,1,nqtveff);
4302: strcpy(fileresp,"P_");
4303: strcat(fileresp,fileresu);
4304: /*strcat(fileresphtm,fileresu);*/
4305: if((ficresp=fopen(fileresp,"w"))==NULL) {
4306: printf("Problem with prevalence resultfile: %s\n", fileresp);
4307: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4308: exit(0);
4309: }
1.240 brouard 4310:
1.226 brouard 4311: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4312: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4313: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4314: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4315: fflush(ficlog);
4316: exit(70);
4317: }
4318: else{
4319: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4320: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4321: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4322: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4323: }
1.237 brouard 4324: 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 4325:
1.226 brouard 4326: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4327: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4328: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4329: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4330: fflush(ficlog);
4331: exit(70);
1.240 brouard 4332: } else{
1.226 brouard 4333: 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 4334: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4335: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4336: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4337: }
1.240 brouard 4338: 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);
4339:
1.253 brouard 4340: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4341: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4342: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4343: j1=0;
1.126 brouard 4344:
1.227 brouard 4345: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4346: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4347: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4348:
4349:
1.226 brouard 4350: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4351: reference=low_education V1=0,V2=0
4352: med_educ V1=1 V2=0,
4353: high_educ V1=0 V2=1
4354: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4355: */
1.249 brouard 4356: dateintsum=0;
4357: k2cpt=0;
4358:
1.253 brouard 4359: if(cptcoveff == 0 )
4360: nl=1; /* Constant model only */
4361: else
4362: nl=2;
4363: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4364: if(nj==1)
4365: j=0; /* First pass for the constant */
4366: else
4367: j=cptcoveff; /* Other passes for the covariate values */
1.251 brouard 4368: first=1;
4369: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on covariates combination in order of model, excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
4370: posproptt=0.;
4371: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4372: scanf("%d", i);*/
4373: for (i=-5; i<=nlstate+ndeath; i++)
4374: for (jk=-5; jk<=nlstate+ndeath; jk++)
4375: for(m=iagemin; m <= iagemax+3; m++)
4376: freq[i][jk][m]=0;
4377:
4378: for (i=1; i<=nlstate; i++) {
1.240 brouard 4379: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4380: prop[i][m]=0;
4381: posprop[i]=0;
4382: pospropt[i]=0;
4383: }
4384: /* for (z1=1; z1<= nqfveff; z1++) { */
4385: /* meanq[z1]+=0.; */
4386: /* for(m=1;m<=lastpass;m++){ */
4387: /* meanqt[m][z1]=0.; */
4388: /* } */
4389: /* } */
4390:
4391: /* dateintsum=0; */
4392: /* k2cpt=0; */
4393:
4394: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4395: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4396: bool=1;
4397: if(j !=0){
4398: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4399: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4400: /* for (z1=1; z1<= nqfveff; z1++) { */
4401: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4402: /* } */
4403: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4404: /* if(Tvaraff[z1] ==-20){ */
4405: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4406: /* }else if(Tvaraff[z1] ==-10){ */
4407: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4408: /* }else */
4409: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4410: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4411: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4412: /* 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",
4413: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4414: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4415: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4416: } /* Onlyf fixed */
4417: } /* end z1 */
4418: } /* cptcovn > 0 */
4419: } /* end any */
4420: }/* end j==0 */
4421: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4422: /* for(m=firstpass; m<=lastpass; m++){ */
4423: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4424: m=mw[mi][iind];
4425: if(j!=0){
4426: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4427: for (z1=1; z1<=cptcoveff; z1++) {
4428: if( Fixed[Tmodelind[z1]]==1){
4429: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4430: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4431: value is -1, we don't select. It differs from the
4432: constant and age model which counts them. */
4433: bool=0; /* not selected */
4434: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4435: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4436: bool=0;
4437: }
4438: }
4439: }
4440: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4441: } /* end j==0 */
4442: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4443: if(bool==1){
4444: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4445: and mw[mi+1][iind]. dh depends on stepm. */
4446: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4447: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4448: if(m >=firstpass && m <=lastpass){
4449: k2=anint[m][iind]+(mint[m][iind]/12.);
4450: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4451: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4452: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4453: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4454: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4455: if (m<lastpass) {
4456: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4457: /* 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]); */
4458: if(s[m][iind]==-1)
4459: 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.));
4460: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4461: /* if((int)agev[m][iind] == 55) */
4462: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4463: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4464: 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 4465: }
1.251 brouard 4466: } /* end if between passes */
4467: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4468: dateintsum=dateintsum+k2; /* on all covariates ?*/
4469: k2cpt++;
4470: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4471: }
1.251 brouard 4472: }else{
4473: bool=1;
4474: }/* end bool 2 */
4475: } /* end m */
4476: } /* end bool */
4477: } /* end iind = 1 to imx */
4478: /* prop[s][age] is feeded for any initial and valid live state as well as
4479: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4480:
4481:
4482: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4483: pstamp(ficresp);
4484: if (cptcoveff>0 && j!=0){
4485: printf( "\n#********** Variable ");
4486: fprintf(ficresp, "\n#********** Variable ");
4487: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4488: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4489: fprintf(ficlog, "\n#********** Variable ");
4490: for (z1=1; z1<=cptcoveff; z1++){
4491: if(!FixedV[Tvaraff[z1]]){
4492: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4493: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4494: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4495: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4496: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4497: }else{
1.251 brouard 4498: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4499: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4500: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4501: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4502: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4503: }
4504: }
4505: printf( "**********\n#");
4506: fprintf(ficresp, "**********\n#");
4507: fprintf(ficresphtm, "**********</h3>\n");
4508: fprintf(ficresphtmfr, "**********</h3>\n");
4509: fprintf(ficlog, "**********\n");
4510: }
4511: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4512: for(i=1; i<=nlstate;i++) {
4513: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4514: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4515: }
4516: fprintf(ficresp, "\n");
4517: fprintf(ficresphtm, "\n");
4518:
4519: /* Header of frequency table by age */
4520: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4521: fprintf(ficresphtmfr,"<th>Age</th> ");
4522: for(jk=-1; jk <=nlstate+ndeath; jk++){
4523: for(m=-1; m <=nlstate+ndeath; m++){
4524: if(jk!=0 && m!=0)
4525: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4526: }
1.226 brouard 4527: }
1.251 brouard 4528: fprintf(ficresphtmfr, "\n");
4529:
4530: /* For each age */
4531: for(iage=iagemin; iage <= iagemax+3; iage++){
4532: fprintf(ficresphtm,"<tr>");
4533: if(iage==iagemax+1){
4534: fprintf(ficlog,"1");
4535: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4536: }else if(iage==iagemax+2){
4537: fprintf(ficlog,"0");
4538: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4539: }else if(iage==iagemax+3){
4540: fprintf(ficlog,"Total");
4541: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4542: }else{
1.240 brouard 4543: if(first==1){
1.251 brouard 4544: first=0;
4545: printf("See log file for details...\n");
4546: }
4547: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4548: fprintf(ficlog,"Age %d", iage);
4549: }
4550: for(jk=1; jk <=nlstate ; jk++){
4551: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4552: pp[jk] += freq[jk][m][iage];
4553: }
4554: for(jk=1; jk <=nlstate ; jk++){
4555: for(m=-1, pos=0; m <=0 ; m++)
4556: pos += freq[jk][m][iage];
4557: if(pp[jk]>=1.e-10){
4558: if(first==1){
4559: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4560: }
4561: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4562: }else{
4563: if(first==1)
4564: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4565: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4566: }
4567: }
4568:
1.251 brouard 4569: for(jk=1; jk <=nlstate ; jk++){
4570: /* posprop[jk]=0; */
4571: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4572: pp[jk] += freq[jk][m][iage];
4573: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4574:
4575: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4576: pos += pp[jk]; /* pos is the total number of transitions until this age */
4577: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4578: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4579: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4580: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4581: }
1.251 brouard 4582: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4583: if(pos>=1.e-5){
1.251 brouard 4584: if(first==1)
4585: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4586: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4587: }else{
4588: if(first==1)
4589: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4590: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4591: }
4592: if( iage <= iagemax){
4593: if(pos>=1.e-5){
4594: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4595: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4596: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4597: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4598: }
4599: else{
4600: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4601: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4602: }
1.240 brouard 4603: }
1.251 brouard 4604: pospropt[jk] +=posprop[jk];
4605: } /* end loop jk */
4606: /* pospropt=0.; */
4607: for(jk=-1; jk <=nlstate+ndeath; jk++){
4608: for(m=-1; m <=nlstate+ndeath; m++){
4609: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4610: if(first==1){
4611: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4612: }
1.253 brouard 4613: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]); */
1.251 brouard 4614: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4615: }
4616: if(jk!=0 && m!=0)
4617: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4618: }
1.251 brouard 4619: } /* end loop jk */
4620: posproptt=0.;
4621: for(jk=1; jk <=nlstate; jk++){
4622: posproptt += pospropt[jk];
4623: }
4624: fprintf(ficresphtmfr,"</tr>\n ");
4625: if(iage <= iagemax){
4626: fprintf(ficresp,"\n");
4627: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4628: }
1.251 brouard 4629: if(first==1)
4630: printf("Others in log...\n");
4631: fprintf(ficlog,"\n");
4632: } /* end loop age iage */
4633: fprintf(ficresphtm,"<tr><th>Tot</th>");
4634: for(jk=1; jk <=nlstate ; jk++){
4635: if(posproptt < 1.e-5){
4636: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4637: }else{
4638: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4639: }
1.226 brouard 4640: }
1.251 brouard 4641: fprintf(ficresphtm,"</tr>\n");
4642: fprintf(ficresphtm,"</table>\n");
4643: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4644: if(posproptt < 1.e-5){
1.251 brouard 4645: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4646: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4647: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4648: invalidvarcomb[j1]=1;
1.226 brouard 4649: }else{
1.251 brouard 4650: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4651: invalidvarcomb[j1]=0;
1.226 brouard 4652: }
1.251 brouard 4653: fprintf(ficresphtmfr,"</table>\n");
4654: fprintf(ficlog,"\n");
4655: if(j!=0){
4656: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4657: for(i=1,jk=1; i <=nlstate; i++){
4658: for(k=1; k <=(nlstate+ndeath); k++){
4659: if (k != i) {
4660: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
1.253 brouard 4661: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4662: if(j1==1){ /* All dummy covariates to zero */
4663: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4664: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4665: printf("%d%d ",i,k);
4666: fprintf(ficlog,"%d%d ",i,k);
4667: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[jk],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]));
4668: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4669: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4670: }
1.253 brouard 4671: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4672: for(iage=iagemin; iage <= iagemax+3; iage++){
4673: x[iage]= (double)iage;
4674: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
4675: /* printf("i=%d, k=%d, jk=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,jk,j1,jj, iage, y[iage]); */
4676: }
4677: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
4678: pstart[jk]=b;
4679: pstart[jk-1]=a;
1.252 brouard 4680: }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 */
4681: 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]);
4682: 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]);
1.251 brouard 4683: pstart[jk]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 brouard 4684: printf("%d%d ",i,k);
4685: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4686: printf("jk=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",jk,i,k,jk,p[jk],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]));
4687: }else{ /* Other cases, like quantitative fixed or varying covariates */
4688: ;
4689: }
4690: /* printf("%12.7f )", param[i][jj][k]); */
4691: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4692: jk++;
4693: } /* end jj */
4694: } /* end k!= i */
4695: } /* end k */
4696: } /* end i, jk */
4697: } /* end j !=0 */
4698: } /* end selected combination of covariate j1 */
4699: if(j==0){ /* We can estimate starting values from the occurences in each case */
4700: printf("#Freqsummary: Starting values for the constants:\n");
4701: fprintf(ficlog,"\n");
4702: for(i=1,jk=1; i <=nlstate; i++){
4703: for(k=1; k <=(nlstate+ndeath); k++){
4704: if (k != i) {
4705: printf("%d%d ",i,k);
4706: fprintf(ficlog,"%d%d ",i,k);
4707: for(jj=1; jj <=ncovmodel; jj++){
1.253 brouard 4708: pstart[jk]=p[jk]; /* Setting pstart to p values by default */
4709: if(jj==1){ /* Age has to be done */
4710: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4711: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4712: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4713: }
4714: /* printf("%12.7f )", param[i][jj][k]); */
4715: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4716: jk++;
1.250 brouard 4717: }
1.251 brouard 4718: printf("\n");
4719: fprintf(ficlog,"\n");
1.250 brouard 4720: }
4721: }
4722: }
1.251 brouard 4723: printf("#Freqsummary\n");
4724: fprintf(ficlog,"\n");
4725: for(jk=-1; jk <=nlstate+ndeath; jk++){
4726: for(m=-1; m <=nlstate+ndeath; m++){
4727: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4728: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4729: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4730: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4731: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4732: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4733: /* } */
4734: }
4735: } /* end loop jk */
4736:
4737: printf("\n");
4738: fprintf(ficlog,"\n");
4739: } /* end j=0 */
1.249 brouard 4740: } /* end j */
1.252 brouard 4741:
1.253 brouard 4742: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4743: for(i=1, jk=1; i <=nlstate; i++){
4744: for(j=1; j <=nlstate+ndeath; j++){
4745: if(j!=i){
4746: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4747: printf("%1d%1d",i,j);
4748: fprintf(ficparo,"%1d%1d",i,j);
4749: for(k=1; k<=ncovmodel;k++){
4750: /* printf(" %lf",param[i][j][k]); */
4751: /* fprintf(ficparo," %lf",param[i][j][k]); */
4752: p[jk]=pstart[jk];
4753: printf(" %f ",pstart[jk]);
4754: fprintf(ficparo," %f ",pstart[jk]);
4755: jk++;
4756: }
4757: printf("\n");
4758: fprintf(ficparo,"\n");
4759: }
4760: }
4761: }
4762: } /* end mle=-2 */
1.226 brouard 4763: dateintmean=dateintsum/k2cpt;
1.240 brouard 4764:
1.226 brouard 4765: fclose(ficresp);
4766: fclose(ficresphtm);
4767: fclose(ficresphtmfr);
4768: free_vector(meanq,1,nqfveff);
4769: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4770: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4771: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4772: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4773: free_vector(pospropt,1,nlstate);
4774: free_vector(posprop,1,nlstate);
1.251 brouard 4775: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4776: free_vector(pp,1,nlstate);
4777: /* End of freqsummary */
4778: }
1.126 brouard 4779:
4780: /************ Prevalence ********************/
1.227 brouard 4781: 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)
4782: {
4783: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4784: in each health status at the date of interview (if between dateprev1 and dateprev2).
4785: We still use firstpass and lastpass as another selection.
4786: */
1.126 brouard 4787:
1.227 brouard 4788: int i, m, jk, j1, bool, z1,j, iv;
4789: int mi; /* Effective wave */
4790: int iage;
4791: double agebegin, ageend;
4792:
4793: double **prop;
4794: double posprop;
4795: double y2; /* in fractional years */
4796: int iagemin, iagemax;
4797: int first; /** to stop verbosity which is redirected to log file */
4798:
4799: iagemin= (int) agemin;
4800: iagemax= (int) agemax;
4801: /*pp=vector(1,nlstate);*/
1.251 brouard 4802: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4803: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4804: j1=0;
1.222 brouard 4805:
1.227 brouard 4806: /*j=cptcoveff;*/
4807: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4808:
1.227 brouard 4809: first=1;
4810: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4811: for (i=1; i<=nlstate; i++)
1.251 brouard 4812: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4813: prop[i][iage]=0.0;
4814: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4815: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4816: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4817:
4818: for (i=1; i<=imx; i++) { /* Each individual */
4819: bool=1;
4820: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4821: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4822: m=mw[mi][i];
4823: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4824: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4825: for (z1=1; z1<=cptcoveff; z1++){
4826: if( Fixed[Tmodelind[z1]]==1){
4827: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4828: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4829: bool=0;
4830: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4831: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4832: bool=0;
4833: }
4834: }
4835: if(bool==1){ /* Otherwise we skip that wave/person */
4836: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4837: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4838: if(m >=firstpass && m <=lastpass){
4839: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4840: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4841: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4842: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4843: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4844: 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);
4845: exit(1);
4846: }
4847: if (s[m][i]>0 && s[m][i]<=nlstate) {
4848: /*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]]);*/
4849: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4850: prop[s[m][i]][iagemax+3] += weight[i];
4851: } /* end valid statuses */
4852: } /* end selection of dates */
4853: } /* end selection of waves */
4854: } /* end bool */
4855: } /* end wave */
4856: } /* end individual */
4857: for(i=iagemin; i <= iagemax+3; i++){
4858: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4859: posprop += prop[jk][i];
4860: }
4861:
4862: for(jk=1; jk <=nlstate ; jk++){
4863: if( i <= iagemax){
4864: if(posprop>=1.e-5){
4865: probs[i][jk][j1]= prop[jk][i]/posprop;
4866: } else{
4867: if(first==1){
4868: first=0;
4869: printf("Warning Observed prevalence probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,j1,probs[i][jk][j1]);
4870: }
4871: }
4872: }
4873: }/* end jk */
4874: }/* end i */
1.222 brouard 4875: /*} *//* end i1 */
1.227 brouard 4876: } /* end j1 */
1.222 brouard 4877:
1.227 brouard 4878: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4879: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4880: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4881: } /* End of prevalence */
1.126 brouard 4882:
4883: /************* Waves Concatenation ***************/
4884:
4885: 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)
4886: {
4887: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4888: Death is a valid wave (if date is known).
4889: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4890: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4891: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4892: */
1.126 brouard 4893:
1.224 brouard 4894: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4895: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4896: double sum=0., jmean=0.;*/
1.224 brouard 4897: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4898: int j, k=0,jk, ju, jl;
4899: double sum=0.;
4900: first=0;
1.214 brouard 4901: firstwo=0;
1.217 brouard 4902: firsthree=0;
1.218 brouard 4903: firstfour=0;
1.164 brouard 4904: jmin=100000;
1.126 brouard 4905: jmax=-1;
4906: jmean=0.;
1.224 brouard 4907:
4908: /* Treating live states */
1.214 brouard 4909: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4910: mi=0; /* First valid wave */
1.227 brouard 4911: mli=0; /* Last valid wave */
1.126 brouard 4912: m=firstpass;
1.214 brouard 4913: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4914: 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 */
4915: mli=m-1;/* mw[++mi][i]=m-1; */
4916: }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 */
4917: mw[++mi][i]=m;
4918: mli=m;
1.224 brouard 4919: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4920: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4921: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4922: }
1.227 brouard 4923: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4924: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4925: break;
1.224 brouard 4926: #else
1.227 brouard 4927: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4928: if(firsthree == 0){
4929: 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 pi. .\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);
4930: firsthree=1;
4931: }
4932: 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 pi. .\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);
4933: mw[++mi][i]=m;
4934: mli=m;
4935: }
4936: if(s[m][i]==-2){ /* Vital status is really unknown */
4937: nbwarn++;
4938: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4939: 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);
4940: 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);
4941: }
4942: break;
4943: }
4944: break;
1.224 brouard 4945: #endif
1.227 brouard 4946: }/* End m >= lastpass */
1.126 brouard 4947: }/* end while */
1.224 brouard 4948:
1.227 brouard 4949: /* 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 4950: /* After last pass */
1.224 brouard 4951: /* Treating death states */
1.214 brouard 4952: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4953: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4954: /* } */
1.126 brouard 4955: mi++; /* Death is another wave */
4956: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4957: /* Only death is a correct wave */
1.126 brouard 4958: mw[mi][i]=m;
1.224 brouard 4959: }
4960: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4961: else if ((int) andc[i] != 9999) { /* Status is negative. A death occured after lastpass, we can't take it into account because of potential bias */
1.216 brouard 4962: /* m++; */
4963: /* mi++; */
4964: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4965: /* mw[mi][i]=m; */
1.218 brouard 4966: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4967: 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 */
4968: nbwarn++;
4969: if(firstfiv==0){
4970: 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 );
4971: firstfiv=1;
4972: }else{
4973: 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 );
4974: }
4975: }else{ /* Death occured afer last wave potential bias */
4976: nberr++;
4977: if(firstwo==0){
4978: 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.\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 );
4979: firstwo=1;
4980: }
4981: 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.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
4982: }
1.218 brouard 4983: }else{ /* end date of interview is known */
1.227 brouard 4984: /* death is known but not confirmed by death status at any wave */
4985: if(firstfour==0){
4986: 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 );
4987: firstfour=1;
4988: }
4989: 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 4990: }
1.224 brouard 4991: } /* end if date of death is known */
4992: #endif
4993: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4994: /* wav[i]=mw[mi][i]; */
1.126 brouard 4995: if(mi==0){
4996: nbwarn++;
4997: if(first==0){
1.227 brouard 4998: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4999: first=1;
1.126 brouard 5000: }
5001: if(first==1){
1.227 brouard 5002: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5003: }
5004: } /* end mi==0 */
5005: } /* End individuals */
1.214 brouard 5006: /* wav and mw are no more changed */
1.223 brouard 5007:
1.214 brouard 5008:
1.126 brouard 5009: for(i=1; i<=imx; i++){
5010: for(mi=1; mi<wav[i];mi++){
5011: if (stepm <=0)
1.227 brouard 5012: dh[mi][i]=1;
1.126 brouard 5013: else{
1.227 brouard 5014: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
5015: if (agedc[i] < 2*AGESUP) {
5016: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5017: if(j==0) j=1; /* Survives at least one month after exam */
5018: else if(j<0){
5019: nberr++;
5020: 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]);
5021: j=1; /* Temporary Dangerous patch */
5022: 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);
5023: 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]);
5024: 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);
5025: }
5026: k=k+1;
5027: if (j >= jmax){
5028: jmax=j;
5029: ijmax=i;
5030: }
5031: if (j <= jmin){
5032: jmin=j;
5033: ijmin=i;
5034: }
5035: sum=sum+j;
5036: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5037: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5038: }
5039: }
5040: else{
5041: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5042: /* 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 5043:
1.227 brouard 5044: k=k+1;
5045: if (j >= jmax) {
5046: jmax=j;
5047: ijmax=i;
5048: }
5049: else if (j <= jmin){
5050: jmin=j;
5051: ijmin=i;
5052: }
5053: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5054: /*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]);*/
5055: if(j<0){
5056: nberr++;
5057: 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]);
5058: 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]);
5059: }
5060: sum=sum+j;
5061: }
5062: jk= j/stepm;
5063: jl= j -jk*stepm;
5064: ju= j -(jk+1)*stepm;
5065: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5066: if(jl==0){
5067: dh[mi][i]=jk;
5068: bh[mi][i]=0;
5069: }else{ /* We want a negative bias in order to only have interpolation ie
5070: * to avoid the price of an extra matrix product in likelihood */
5071: dh[mi][i]=jk+1;
5072: bh[mi][i]=ju;
5073: }
5074: }else{
5075: if(jl <= -ju){
5076: dh[mi][i]=jk;
5077: bh[mi][i]=jl; /* bias is positive if real duration
5078: * is higher than the multiple of stepm and negative otherwise.
5079: */
5080: }
5081: else{
5082: dh[mi][i]=jk+1;
5083: bh[mi][i]=ju;
5084: }
5085: if(dh[mi][i]==0){
5086: dh[mi][i]=1; /* At least one step */
5087: bh[mi][i]=ju; /* At least one step */
5088: /* 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);*/
5089: }
5090: } /* end if mle */
1.126 brouard 5091: }
5092: } /* end wave */
5093: }
5094: jmean=sum/k;
5095: 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 5096: 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 5097: }
1.126 brouard 5098:
5099: /*********** Tricode ****************************/
1.220 brouard 5100: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5101: {
5102: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5103: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5104: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5105: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5106: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5107: */
1.130 brouard 5108:
1.242 brouard 5109: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5110: int modmaxcovj=0; /* Modality max of covariates j */
5111: int cptcode=0; /* Modality max of covariates j */
5112: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5113:
5114:
1.242 brouard 5115: /* cptcoveff=0; */
5116: /* *cptcov=0; */
1.126 brouard 5117:
1.242 brouard 5118: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5119:
1.242 brouard 5120: /* Loop on covariates without age and products and no quantitative variable */
5121: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5122: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5123: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5124: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5125: switch(Fixed[k]) {
5126: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5127: 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*/
5128: ij=(int)(covar[Tvar[k]][i]);
5129: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5130: * If product of Vn*Vm, still boolean *:
5131: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5132: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5133: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5134: modality of the nth covariate of individual i. */
5135: if (ij > modmaxcovj)
5136: modmaxcovj=ij;
5137: else if (ij < modmincovj)
5138: modmincovj=ij;
5139: if ((ij < -1) && (ij > NCOVMAX)){
5140: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5141: exit(1);
5142: }else
5143: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5144: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5145: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5146: /* getting the maximum value of the modality of the covariate
5147: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5148: female ies 1, then modmaxcovj=1.
5149: */
5150: } /* end for loop on individuals i */
5151: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5152: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5153: cptcode=modmaxcovj;
5154: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5155: /*for (i=0; i<=cptcode; i++) {*/
5156: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5157: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5158: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5159: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5160: if( j != -1){
5161: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5162: covariate for which somebody answered excluding
5163: undefined. Usually 2: 0 and 1. */
5164: }
5165: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5166: covariate for which somebody answered including
5167: undefined. Usually 3: -1, 0 and 1. */
5168: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5169: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5170: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5171:
1.242 brouard 5172: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5173: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5174: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5175: /* modmincovj=3; modmaxcovj = 7; */
5176: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5177: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5178: /* defining two dummy variables: variables V1_1 and V1_2.*/
5179: /* nbcode[Tvar[j]][ij]=k; */
5180: /* nbcode[Tvar[j]][1]=0; */
5181: /* nbcode[Tvar[j]][2]=1; */
5182: /* nbcode[Tvar[j]][3]=2; */
5183: /* To be continued (not working yet). */
5184: ij=0; /* ij is similar to i but can jump over null modalities */
5185: 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*/
5186: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5187: break;
5188: }
5189: ij++;
5190: 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*/
5191: cptcode = ij; /* New max modality for covar j */
5192: } /* end of loop on modality i=-1 to 1 or more */
5193: break;
5194: case 1: /* Testing on varying covariate, could be simple and
5195: * should look at waves or product of fixed *
5196: * varying. No time to test -1, assuming 0 and 1 only */
5197: ij=0;
5198: for(i=0; i<=1;i++){
5199: nbcode[Tvar[k]][++ij]=i;
5200: }
5201: break;
5202: default:
5203: break;
5204: } /* end switch */
5205: } /* end dummy test */
5206:
5207: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5208: /* /\*recode from 0 *\/ */
5209: /* k is a modality. If we have model=V1+V1*sex */
5210: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5211: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5212: /* } */
5213: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5214: /* if (ij > ncodemax[j]) { */
5215: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5216: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5217: /* break; */
5218: /* } */
5219: /* } /\* end of loop on modality k *\/ */
5220: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5221:
5222: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5223: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5224: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5225: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5226: 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 */
5227: 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 */
5228: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5229: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5230:
5231: ij=0;
5232: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5233: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5234: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5235: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5236: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5237: /* If product not in single variable we don't print results */
5238: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5239: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5240: 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*/
5241: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5242: 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 */
5243: if(Fixed[k]!=0)
5244: anyvaryingduminmodel=1;
5245: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5246: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5247: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5248: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5249: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5250: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5251: }
5252: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5253: /* ij--; */
5254: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5255: *cptcov=ij; /*Number of total real effective covariates: effective
5256: * because they can be excluded from the model and real
5257: * if in the model but excluded because missing values, but how to get k from ij?*/
5258: for(j=ij+1; j<= cptcovt; j++){
5259: Tvaraff[j]=0;
5260: Tmodelind[j]=0;
5261: }
5262: for(j=ntveff+1; j<= cptcovt; j++){
5263: TmodelInvind[j]=0;
5264: }
5265: /* To be sorted */
5266: ;
5267: }
1.126 brouard 5268:
1.145 brouard 5269:
1.126 brouard 5270: /*********** Health Expectancies ****************/
5271:
1.235 brouard 5272: 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 5273:
5274: {
5275: /* Health expectancies, no variances */
1.164 brouard 5276: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5277: int nhstepma, nstepma; /* Decreasing with age */
5278: double age, agelim, hf;
5279: double ***p3mat;
5280: double eip;
5281:
1.238 brouard 5282: /* pstamp(ficreseij); */
1.126 brouard 5283: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5284: fprintf(ficreseij,"# Age");
5285: for(i=1; i<=nlstate;i++){
5286: for(j=1; j<=nlstate;j++){
5287: fprintf(ficreseij," e%1d%1d ",i,j);
5288: }
5289: fprintf(ficreseij," e%1d. ",i);
5290: }
5291: fprintf(ficreseij,"\n");
5292:
5293:
5294: if(estepm < stepm){
5295: printf ("Problem %d lower than %d\n",estepm, stepm);
5296: }
5297: else hstepm=estepm;
5298: /* We compute the life expectancy from trapezoids spaced every estepm months
5299: * This is mainly to measure the difference between two models: for example
5300: * if stepm=24 months pijx are given only every 2 years and by summing them
5301: * we are calculating an estimate of the Life Expectancy assuming a linear
5302: * progression in between and thus overestimating or underestimating according
5303: * to the curvature of the survival function. If, for the same date, we
5304: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5305: * to compare the new estimate of Life expectancy with the same linear
5306: * hypothesis. A more precise result, taking into account a more precise
5307: * curvature will be obtained if estepm is as small as stepm. */
5308:
5309: /* For example we decided to compute the life expectancy with the smallest unit */
5310: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5311: nhstepm is the number of hstepm from age to agelim
5312: nstepm is the number of stepm from age to agelin.
5313: Look at hpijx to understand the reason of that which relies in memory size
5314: and note for a fixed period like estepm months */
5315: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5316: survival function given by stepm (the optimization length). Unfortunately it
5317: means that if the survival funtion is printed only each two years of age and if
5318: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5319: results. So we changed our mind and took the option of the best precision.
5320: */
5321: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5322:
5323: agelim=AGESUP;
5324: /* If stepm=6 months */
5325: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5326: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5327:
5328: /* nhstepm age range expressed in number of stepm */
5329: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5330: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5331: /* if (stepm >= YEARM) hstepm=1;*/
5332: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5333: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5334:
5335: for (age=bage; age<=fage; age ++){
5336: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5337: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5338: /* if (stepm >= YEARM) hstepm=1;*/
5339: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5340:
5341: /* If stepm=6 months */
5342: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5343: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5344:
1.235 brouard 5345: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5346:
5347: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5348:
5349: printf("%d|",(int)age);fflush(stdout);
5350: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5351:
5352: /* Computing expectancies */
5353: for(i=1; i<=nlstate;i++)
5354: for(j=1; j<=nlstate;j++)
5355: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5356: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5357:
5358: /* 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]);*/
5359:
5360: }
5361:
5362: fprintf(ficreseij,"%3.0f",age );
5363: for(i=1; i<=nlstate;i++){
5364: eip=0;
5365: for(j=1; j<=nlstate;j++){
5366: eip +=eij[i][j][(int)age];
5367: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5368: }
5369: fprintf(ficreseij,"%9.4f", eip );
5370: }
5371: fprintf(ficreseij,"\n");
5372:
5373: }
5374: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5375: printf("\n");
5376: fprintf(ficlog,"\n");
5377:
5378: }
5379:
1.235 brouard 5380: 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 5381:
5382: {
5383: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5384: to initial status i, ei. .
1.126 brouard 5385: */
5386: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5387: int nhstepma, nstepma; /* Decreasing with age */
5388: double age, agelim, hf;
5389: double ***p3matp, ***p3matm, ***varhe;
5390: double **dnewm,**doldm;
5391: double *xp, *xm;
5392: double **gp, **gm;
5393: double ***gradg, ***trgradg;
5394: int theta;
5395:
5396: double eip, vip;
5397:
5398: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5399: xp=vector(1,npar);
5400: xm=vector(1,npar);
5401: dnewm=matrix(1,nlstate*nlstate,1,npar);
5402: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5403:
5404: pstamp(ficresstdeij);
5405: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5406: fprintf(ficresstdeij,"# Age");
5407: for(i=1; i<=nlstate;i++){
5408: for(j=1; j<=nlstate;j++)
5409: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5410: fprintf(ficresstdeij," e%1d. ",i);
5411: }
5412: fprintf(ficresstdeij,"\n");
5413:
5414: pstamp(ficrescveij);
5415: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5416: fprintf(ficrescveij,"# Age");
5417: for(i=1; i<=nlstate;i++)
5418: for(j=1; j<=nlstate;j++){
5419: cptj= (j-1)*nlstate+i;
5420: for(i2=1; i2<=nlstate;i2++)
5421: for(j2=1; j2<=nlstate;j2++){
5422: cptj2= (j2-1)*nlstate+i2;
5423: if(cptj2 <= cptj)
5424: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5425: }
5426: }
5427: fprintf(ficrescveij,"\n");
5428:
5429: if(estepm < stepm){
5430: printf ("Problem %d lower than %d\n",estepm, stepm);
5431: }
5432: else hstepm=estepm;
5433: /* We compute the life expectancy from trapezoids spaced every estepm months
5434: * This is mainly to measure the difference between two models: for example
5435: * if stepm=24 months pijx are given only every 2 years and by summing them
5436: * we are calculating an estimate of the Life Expectancy assuming a linear
5437: * progression in between and thus overestimating or underestimating according
5438: * to the curvature of the survival function. If, for the same date, we
5439: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5440: * to compare the new estimate of Life expectancy with the same linear
5441: * hypothesis. A more precise result, taking into account a more precise
5442: * curvature will be obtained if estepm is as small as stepm. */
5443:
5444: /* For example we decided to compute the life expectancy with the smallest unit */
5445: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5446: nhstepm is the number of hstepm from age to agelim
5447: nstepm is the number of stepm from age to agelin.
5448: Look at hpijx to understand the reason of that which relies in memory size
5449: and note for a fixed period like estepm months */
5450: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5451: survival function given by stepm (the optimization length). Unfortunately it
5452: means that if the survival funtion is printed only each two years of age and if
5453: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5454: results. So we changed our mind and took the option of the best precision.
5455: */
5456: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5457:
5458: /* If stepm=6 months */
5459: /* nhstepm age range expressed in number of stepm */
5460: agelim=AGESUP;
5461: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5462: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5463: /* if (stepm >= YEARM) hstepm=1;*/
5464: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5465:
5466: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5467: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5468: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5469: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5470: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5471: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5472:
5473: for (age=bage; age<=fage; age ++){
5474: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5475: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5476: /* if (stepm >= YEARM) hstepm=1;*/
5477: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5478:
1.126 brouard 5479: /* If stepm=6 months */
5480: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5481: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5482:
5483: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5484:
1.126 brouard 5485: /* Computing Variances of health expectancies */
5486: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5487: decrease memory allocation */
5488: for(theta=1; theta <=npar; theta++){
5489: for(i=1; i<=npar; i++){
1.222 brouard 5490: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5491: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5492: }
1.235 brouard 5493: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5494: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5495:
1.126 brouard 5496: for(j=1; j<= nlstate; j++){
1.222 brouard 5497: for(i=1; i<=nlstate; i++){
5498: for(h=0; h<=nhstepm-1; h++){
5499: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5500: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5501: }
5502: }
1.126 brouard 5503: }
1.218 brouard 5504:
1.126 brouard 5505: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5506: for(h=0; h<=nhstepm-1; h++){
5507: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5508: }
1.126 brouard 5509: }/* End theta */
5510:
5511:
5512: for(h=0; h<=nhstepm-1; h++)
5513: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5514: for(theta=1; theta <=npar; theta++)
5515: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5516:
1.218 brouard 5517:
1.222 brouard 5518: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5519: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5520: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5521:
1.222 brouard 5522: printf("%d|",(int)age);fflush(stdout);
5523: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5524: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5525: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5526: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5527: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5528: for(ij=1;ij<=nlstate*nlstate;ij++)
5529: for(ji=1;ji<=nlstate*nlstate;ji++)
5530: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5531: }
5532: }
1.218 brouard 5533:
1.126 brouard 5534: /* Computing expectancies */
1.235 brouard 5535: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5536: for(i=1; i<=nlstate;i++)
5537: for(j=1; j<=nlstate;j++)
1.222 brouard 5538: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5539: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5540:
1.222 brouard 5541: /* 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 5542:
1.222 brouard 5543: }
1.218 brouard 5544:
1.126 brouard 5545: fprintf(ficresstdeij,"%3.0f",age );
5546: for(i=1; i<=nlstate;i++){
5547: eip=0.;
5548: vip=0.;
5549: for(j=1; j<=nlstate;j++){
1.222 brouard 5550: eip += eij[i][j][(int)age];
5551: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5552: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5553: 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 5554: }
5555: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5556: }
5557: fprintf(ficresstdeij,"\n");
1.218 brouard 5558:
1.126 brouard 5559: fprintf(ficrescveij,"%3.0f",age );
5560: for(i=1; i<=nlstate;i++)
5561: for(j=1; j<=nlstate;j++){
1.222 brouard 5562: cptj= (j-1)*nlstate+i;
5563: for(i2=1; i2<=nlstate;i2++)
5564: for(j2=1; j2<=nlstate;j2++){
5565: cptj2= (j2-1)*nlstate+i2;
5566: if(cptj2 <= cptj)
5567: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5568: }
1.126 brouard 5569: }
5570: fprintf(ficrescveij,"\n");
1.218 brouard 5571:
1.126 brouard 5572: }
5573: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5574: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5575: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5576: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5577: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5578: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5579: printf("\n");
5580: fprintf(ficlog,"\n");
1.218 brouard 5581:
1.126 brouard 5582: free_vector(xm,1,npar);
5583: free_vector(xp,1,npar);
5584: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5585: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5586: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5587: }
1.218 brouard 5588:
1.126 brouard 5589: /************ Variance ******************/
1.235 brouard 5590: 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 5591: {
5592: /* Variance of health expectancies */
5593: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5594: /* double **newm;*/
5595: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5596:
5597: /* int movingaverage(); */
5598: double **dnewm,**doldm;
5599: double **dnewmp,**doldmp;
5600: int i, j, nhstepm, hstepm, h, nstepm ;
5601: int k;
5602: double *xp;
5603: double **gp, **gm; /* for var eij */
5604: double ***gradg, ***trgradg; /*for var eij */
5605: double **gradgp, **trgradgp; /* for var p point j */
5606: double *gpp, *gmp; /* for var p point j */
5607: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5608: double ***p3mat;
5609: double age,agelim, hf;
5610: /* double ***mobaverage; */
5611: int theta;
5612: char digit[4];
5613: char digitp[25];
5614:
5615: char fileresprobmorprev[FILENAMELENGTH];
5616:
5617: if(popbased==1){
5618: if(mobilav!=0)
5619: strcpy(digitp,"-POPULBASED-MOBILAV_");
5620: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5621: }
5622: else
5623: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5624:
1.218 brouard 5625: /* if (mobilav!=0) { */
5626: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5627: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5628: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5629: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5630: /* } */
5631: /* } */
5632:
5633: strcpy(fileresprobmorprev,"PRMORPREV-");
5634: sprintf(digit,"%-d",ij);
5635: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5636: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5637: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5638: strcat(fileresprobmorprev,fileresu);
5639: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5640: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5641: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5642: }
5643: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5644: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5645: pstamp(ficresprobmorprev);
5646: 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 5647: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5648: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5649: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5650: }
5651: for(j=1;j<=cptcoveff;j++)
5652: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5653: fprintf(ficresprobmorprev,"\n");
5654:
1.218 brouard 5655: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5656: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5657: fprintf(ficresprobmorprev," p.%-d SE",j);
5658: for(i=1; i<=nlstate;i++)
5659: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5660: }
5661: fprintf(ficresprobmorprev,"\n");
5662:
5663: fprintf(ficgp,"\n# Routine varevsij");
5664: fprintf(ficgp,"\nunset title \n");
5665: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5666: 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");
5667: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5668: /* } */
5669: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5670: pstamp(ficresvij);
5671: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5672: if(popbased==1)
5673: 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);
5674: else
5675: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5676: fprintf(ficresvij,"# Age");
5677: for(i=1; i<=nlstate;i++)
5678: for(j=1; j<=nlstate;j++)
5679: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5680: fprintf(ficresvij,"\n");
5681:
5682: xp=vector(1,npar);
5683: dnewm=matrix(1,nlstate,1,npar);
5684: doldm=matrix(1,nlstate,1,nlstate);
5685: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5686: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5687:
5688: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5689: gpp=vector(nlstate+1,nlstate+ndeath);
5690: gmp=vector(nlstate+1,nlstate+ndeath);
5691: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5692:
1.218 brouard 5693: if(estepm < stepm){
5694: printf ("Problem %d lower than %d\n",estepm, stepm);
5695: }
5696: else hstepm=estepm;
5697: /* For example we decided to compute the life expectancy with the smallest unit */
5698: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5699: nhstepm is the number of hstepm from age to agelim
5700: nstepm is the number of stepm from age to agelim.
5701: Look at function hpijx to understand why because of memory size limitations,
5702: we decided (b) to get a life expectancy respecting the most precise curvature of the
5703: survival function given by stepm (the optimization length). Unfortunately it
5704: means that if the survival funtion is printed every two years of age and if
5705: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5706: results. So we changed our mind and took the option of the best precision.
5707: */
5708: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5709: agelim = AGESUP;
5710: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5711: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5712: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5713: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5714: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5715: gp=matrix(0,nhstepm,1,nlstate);
5716: gm=matrix(0,nhstepm,1,nlstate);
5717:
5718:
5719: for(theta=1; theta <=npar; theta++){
5720: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5721: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5722: }
5723:
1.242 brouard 5724: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5725:
5726: if (popbased==1) {
5727: if(mobilav ==0){
5728: for(i=1; i<=nlstate;i++)
5729: prlim[i][i]=probs[(int)age][i][ij];
5730: }else{ /* mobilav */
5731: for(i=1; i<=nlstate;i++)
5732: prlim[i][i]=mobaverage[(int)age][i][ij];
5733: }
5734: }
5735:
1.235 brouard 5736: 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 5737: for(j=1; j<= nlstate; j++){
5738: for(h=0; h<=nhstepm; h++){
5739: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5740: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5741: }
5742: }
5743: /* Next for computing probability of death (h=1 means
5744: computed over hstepm matrices product = hstepm*stepm months)
5745: as a weighted average of prlim.
5746: */
5747: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5748: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5749: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5750: }
5751: /* end probability of death */
5752:
5753: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5754: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5755:
1.242 brouard 5756: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5757:
5758: if (popbased==1) {
5759: if(mobilav ==0){
5760: for(i=1; i<=nlstate;i++)
5761: prlim[i][i]=probs[(int)age][i][ij];
5762: }else{ /* mobilav */
5763: for(i=1; i<=nlstate;i++)
5764: prlim[i][i]=mobaverage[(int)age][i][ij];
5765: }
5766: }
5767:
1.235 brouard 5768: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5769:
5770: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5771: for(h=0; h<=nhstepm; h++){
5772: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5773: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5774: }
5775: }
5776: /* This for computing probability of death (h=1 means
5777: computed over hstepm matrices product = hstepm*stepm months)
5778: as a weighted average of prlim.
5779: */
5780: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5781: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5782: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5783: }
5784: /* end probability of death */
5785:
5786: for(j=1; j<= nlstate; j++) /* vareij */
5787: for(h=0; h<=nhstepm; h++){
5788: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5789: }
5790:
5791: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5792: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5793: }
5794:
5795: } /* End theta */
5796:
5797: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5798:
5799: for(h=0; h<=nhstepm; h++) /* veij */
5800: for(j=1; j<=nlstate;j++)
5801: for(theta=1; theta <=npar; theta++)
5802: trgradg[h][j][theta]=gradg[h][theta][j];
5803:
5804: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5805: for(theta=1; theta <=npar; theta++)
5806: trgradgp[j][theta]=gradgp[theta][j];
5807:
5808:
5809: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5810: for(i=1;i<=nlstate;i++)
5811: for(j=1;j<=nlstate;j++)
5812: vareij[i][j][(int)age] =0.;
5813:
5814: for(h=0;h<=nhstepm;h++){
5815: for(k=0;k<=nhstepm;k++){
5816: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5817: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5818: for(i=1;i<=nlstate;i++)
5819: for(j=1;j<=nlstate;j++)
5820: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5821: }
5822: }
5823:
5824: /* pptj */
5825: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5826: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5827: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5828: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5829: varppt[j][i]=doldmp[j][i];
5830: /* end ppptj */
5831: /* x centered again */
5832:
1.242 brouard 5833: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5834:
5835: if (popbased==1) {
5836: if(mobilav ==0){
5837: for(i=1; i<=nlstate;i++)
5838: prlim[i][i]=probs[(int)age][i][ij];
5839: }else{ /* mobilav */
5840: for(i=1; i<=nlstate;i++)
5841: prlim[i][i]=mobaverage[(int)age][i][ij];
5842: }
5843: }
5844:
5845: /* This for computing probability of death (h=1 means
5846: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5847: as a weighted average of prlim.
5848: */
1.235 brouard 5849: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5850: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5851: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5852: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5853: }
5854: /* end probability of death */
5855:
5856: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5857: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5858: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5859: for(i=1; i<=nlstate;i++){
5860: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5861: }
5862: }
5863: fprintf(ficresprobmorprev,"\n");
5864:
5865: fprintf(ficresvij,"%.0f ",age );
5866: for(i=1; i<=nlstate;i++)
5867: for(j=1; j<=nlstate;j++){
5868: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5869: }
5870: fprintf(ficresvij,"\n");
5871: free_matrix(gp,0,nhstepm,1,nlstate);
5872: free_matrix(gm,0,nhstepm,1,nlstate);
5873: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5874: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5875: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5876: } /* End age */
5877: free_vector(gpp,nlstate+1,nlstate+ndeath);
5878: free_vector(gmp,nlstate+1,nlstate+ndeath);
5879: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5880: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5881: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5882: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5883: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5884: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5885: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5886: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5887: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5888: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5889: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5890: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5891: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5892: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5893: 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);
5894: /* 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 5895: */
1.218 brouard 5896: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5897: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5898:
1.218 brouard 5899: free_vector(xp,1,npar);
5900: free_matrix(doldm,1,nlstate,1,nlstate);
5901: free_matrix(dnewm,1,nlstate,1,npar);
5902: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5903: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5904: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5905: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5906: fclose(ficresprobmorprev);
5907: fflush(ficgp);
5908: fflush(fichtm);
5909: } /* end varevsij */
1.126 brouard 5910:
5911: /************ Variance of prevlim ******************/
1.235 brouard 5912: void varprevlim(char fileres[], double **varpl, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **prlim, double ftolpl, int *ncvyearp, int ij, char strstart[], int nres)
1.126 brouard 5913: {
1.205 brouard 5914: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5915: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5916:
1.126 brouard 5917: double **dnewm,**doldm;
5918: int i, j, nhstepm, hstepm;
5919: double *xp;
5920: double *gp, *gm;
5921: double **gradg, **trgradg;
1.208 brouard 5922: double **mgm, **mgp;
1.126 brouard 5923: double age,agelim;
5924: int theta;
5925:
5926: pstamp(ficresvpl);
5927: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5928: fprintf(ficresvpl,"# Age ");
5929: if(nresult >=1)
5930: fprintf(ficresvpl," Result# ");
1.126 brouard 5931: for(i=1; i<=nlstate;i++)
5932: fprintf(ficresvpl," %1d-%1d",i,i);
5933: fprintf(ficresvpl,"\n");
5934:
5935: xp=vector(1,npar);
5936: dnewm=matrix(1,nlstate,1,npar);
5937: doldm=matrix(1,nlstate,1,nlstate);
5938:
5939: hstepm=1*YEARM; /* Every year of age */
5940: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5941: agelim = AGESUP;
5942: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5943: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5944: if (stepm >= YEARM) hstepm=1;
5945: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5946: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5947: mgp=matrix(1,npar,1,nlstate);
5948: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5949: gp=vector(1,nlstate);
5950: gm=vector(1,nlstate);
5951:
5952: for(theta=1; theta <=npar; theta++){
5953: for(i=1; i<=npar; i++){ /* Computes gradient */
5954: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5955: }
1.209 brouard 5956: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5957: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5958: else
1.235 brouard 5959: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5960: for(i=1;i<=nlstate;i++){
1.126 brouard 5961: gp[i] = prlim[i][i];
1.208 brouard 5962: mgp[theta][i] = prlim[i][i];
5963: }
1.126 brouard 5964: for(i=1; i<=npar; i++) /* Computes gradient */
5965: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5966: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5967: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5968: else
1.235 brouard 5969: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5970: for(i=1;i<=nlstate;i++){
1.126 brouard 5971: gm[i] = prlim[i][i];
1.208 brouard 5972: mgm[theta][i] = prlim[i][i];
5973: }
1.126 brouard 5974: for(i=1;i<=nlstate;i++)
5975: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5976: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5977: } /* End theta */
5978:
5979: trgradg =matrix(1,nlstate,1,npar);
5980:
5981: for(j=1; j<=nlstate;j++)
5982: for(theta=1; theta <=npar; theta++)
5983: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5984: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5985: /* printf("\nmgm mgp %d ",(int)age); */
5986: /* for(j=1; j<=nlstate;j++){ */
5987: /* printf(" %d ",j); */
5988: /* for(theta=1; theta <=npar; theta++) */
5989: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5990: /* printf("\n "); */
5991: /* } */
5992: /* } */
5993: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5994: /* printf("\n gradg %d ",(int)age); */
5995: /* for(j=1; j<=nlstate;j++){ */
5996: /* printf("%d ",j); */
5997: /* for(theta=1; theta <=npar; theta++) */
5998: /* printf("%d %lf ",theta,gradg[theta][j]); */
5999: /* printf("\n "); */
6000: /* } */
6001: /* } */
1.126 brouard 6002:
6003: for(i=1;i<=nlstate;i++)
6004: varpl[i][(int)age] =0.;
1.209 brouard 6005: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 6006: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6007: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
6008: }else{
1.126 brouard 6009: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6010: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6011: }
1.126 brouard 6012: for(i=1;i<=nlstate;i++)
6013: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6014:
6015: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6016: if(nresult >=1)
6017: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6018: for(i=1; i<=nlstate;i++)
6019: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6020: fprintf(ficresvpl,"\n");
6021: free_vector(gp,1,nlstate);
6022: free_vector(gm,1,nlstate);
1.208 brouard 6023: free_matrix(mgm,1,npar,1,nlstate);
6024: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6025: free_matrix(gradg,1,npar,1,nlstate);
6026: free_matrix(trgradg,1,nlstate,1,npar);
6027: } /* End age */
6028:
6029: free_vector(xp,1,npar);
6030: free_matrix(doldm,1,nlstate,1,npar);
6031: free_matrix(dnewm,1,nlstate,1,nlstate);
6032:
6033: }
6034:
6035: /************ Variance of one-step probabilities ******************/
6036: 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 6037: {
6038: int i, j=0, k1, l1, tj;
6039: int k2, l2, j1, z1;
6040: int k=0, l;
6041: int first=1, first1, first2;
6042: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6043: double **dnewm,**doldm;
6044: double *xp;
6045: double *gp, *gm;
6046: double **gradg, **trgradg;
6047: double **mu;
6048: double age, cov[NCOVMAX+1];
6049: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6050: int theta;
6051: char fileresprob[FILENAMELENGTH];
6052: char fileresprobcov[FILENAMELENGTH];
6053: char fileresprobcor[FILENAMELENGTH];
6054: double ***varpij;
6055:
6056: strcpy(fileresprob,"PROB_");
6057: strcat(fileresprob,fileres);
6058: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6059: printf("Problem with resultfile: %s\n", fileresprob);
6060: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6061: }
6062: strcpy(fileresprobcov,"PROBCOV_");
6063: strcat(fileresprobcov,fileresu);
6064: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6065: printf("Problem with resultfile: %s\n", fileresprobcov);
6066: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6067: }
6068: strcpy(fileresprobcor,"PROBCOR_");
6069: strcat(fileresprobcor,fileresu);
6070: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6071: printf("Problem with resultfile: %s\n", fileresprobcor);
6072: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6073: }
6074: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6075: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6076: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6077: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6078: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6079: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6080: pstamp(ficresprob);
6081: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6082: fprintf(ficresprob,"# Age");
6083: pstamp(ficresprobcov);
6084: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6085: fprintf(ficresprobcov,"# Age");
6086: pstamp(ficresprobcor);
6087: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6088: fprintf(ficresprobcor,"# Age");
1.126 brouard 6089:
6090:
1.222 brouard 6091: for(i=1; i<=nlstate;i++)
6092: for(j=1; j<=(nlstate+ndeath);j++){
6093: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6094: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6095: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6096: }
6097: /* fprintf(ficresprob,"\n");
6098: fprintf(ficresprobcov,"\n");
6099: fprintf(ficresprobcor,"\n");
6100: */
6101: xp=vector(1,npar);
6102: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6103: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6104: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6105: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6106: first=1;
6107: fprintf(ficgp,"\n# Routine varprob");
6108: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6109: fprintf(fichtm,"\n");
6110:
6111: 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.</li>\n",optionfilehtmcov);
6112: 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);
6113: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6114: and drawn. It helps understanding how is the covariance between two incidences.\
6115: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6116: 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 6117: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6118: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6119: standard deviations wide on each axis. <br>\
6120: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6121: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6122: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6123:
1.222 brouard 6124: cov[1]=1;
6125: /* tj=cptcoveff; */
1.225 brouard 6126: tj = (int) pow(2,cptcoveff);
1.222 brouard 6127: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6128: j1=0;
1.224 brouard 6129: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6130: if (cptcovn>0) {
6131: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6132: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6133: fprintf(ficresprob, "**********\n#\n");
6134: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6135: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6136: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6137:
1.222 brouard 6138: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6139: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6140: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6141:
6142:
1.222 brouard 6143: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6144: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6145: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6146:
1.222 brouard 6147: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6148: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6149: fprintf(ficresprobcor, "**********\n#");
6150: if(invalidvarcomb[j1]){
6151: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6152: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6153: continue;
6154: }
6155: }
6156: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6157: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6158: gp=vector(1,(nlstate)*(nlstate+ndeath));
6159: gm=vector(1,(nlstate)*(nlstate+ndeath));
6160: for (age=bage; age<=fage; age ++){
6161: cov[2]=age;
6162: if(nagesqr==1)
6163: cov[3]= age*age;
6164: for (k=1; k<=cptcovn;k++) {
6165: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6166: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6167: * 1 1 1 1 1
6168: * 2 2 1 1 1
6169: * 3 1 2 1 1
6170: */
6171: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6172: }
6173: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6174: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6175: for (k=1; k<=cptcovprod;k++)
6176: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6177:
6178:
1.222 brouard 6179: for(theta=1; theta <=npar; theta++){
6180: for(i=1; i<=npar; i++)
6181: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6182:
1.222 brouard 6183: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6184:
1.222 brouard 6185: k=0;
6186: for(i=1; i<= (nlstate); i++){
6187: for(j=1; j<=(nlstate+ndeath);j++){
6188: k=k+1;
6189: gp[k]=pmmij[i][j];
6190: }
6191: }
1.220 brouard 6192:
1.222 brouard 6193: for(i=1; i<=npar; i++)
6194: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6195:
1.222 brouard 6196: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6197: k=0;
6198: for(i=1; i<=(nlstate); i++){
6199: for(j=1; j<=(nlstate+ndeath);j++){
6200: k=k+1;
6201: gm[k]=pmmij[i][j];
6202: }
6203: }
1.220 brouard 6204:
1.222 brouard 6205: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6206: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6207: }
1.126 brouard 6208:
1.222 brouard 6209: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6210: for(theta=1; theta <=npar; theta++)
6211: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6212:
1.222 brouard 6213: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6214: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6215:
1.222 brouard 6216: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6217:
1.222 brouard 6218: k=0;
6219: for(i=1; i<=(nlstate); i++){
6220: for(j=1; j<=(nlstate+ndeath);j++){
6221: k=k+1;
6222: mu[k][(int) age]=pmmij[i][j];
6223: }
6224: }
6225: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6226: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6227: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6228:
1.222 brouard 6229: /*printf("\n%d ",(int)age);
6230: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6231: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6232: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6233: }*/
1.220 brouard 6234:
1.222 brouard 6235: fprintf(ficresprob,"\n%d ",(int)age);
6236: fprintf(ficresprobcov,"\n%d ",(int)age);
6237: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6238:
1.222 brouard 6239: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6240: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6241: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6242: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6243: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6244: }
6245: i=0;
6246: for (k=1; k<=(nlstate);k++){
6247: for (l=1; l<=(nlstate+ndeath);l++){
6248: i++;
6249: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6250: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6251: for (j=1; j<=i;j++){
6252: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6253: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6254: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6255: }
6256: }
6257: }/* end of loop for state */
6258: } /* end of loop for age */
6259: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6260: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6261: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6262: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6263:
6264: /* Confidence intervalle of pij */
6265: /*
6266: fprintf(ficgp,"\nunset parametric;unset label");
6267: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6268: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6269: 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);
6270: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6271: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6272: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6273: */
6274:
6275: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6276: first1=1;first2=2;
6277: for (k2=1; k2<=(nlstate);k2++){
6278: for (l2=1; l2<=(nlstate+ndeath);l2++){
6279: if(l2==k2) continue;
6280: j=(k2-1)*(nlstate+ndeath)+l2;
6281: for (k1=1; k1<=(nlstate);k1++){
6282: for (l1=1; l1<=(nlstate+ndeath);l1++){
6283: if(l1==k1) continue;
6284: i=(k1-1)*(nlstate+ndeath)+l1;
6285: if(i<=j) continue;
6286: for (age=bage; age<=fage; age ++){
6287: if ((int)age %5==0){
6288: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6289: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6290: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6291: mu1=mu[i][(int) age]/stepm*YEARM ;
6292: mu2=mu[j][(int) age]/stepm*YEARM;
6293: c12=cv12/sqrt(v1*v2);
6294: /* Computing eigen value of matrix of covariance */
6295: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6296: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6297: if ((lc2 <0) || (lc1 <0) ){
6298: if(first2==1){
6299: first1=0;
6300: 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);
6301: }
6302: 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);
6303: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6304: /* lc2=fabs(lc2); */
6305: }
1.220 brouard 6306:
1.222 brouard 6307: /* Eigen vectors */
6308: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6309: /*v21=sqrt(1.-v11*v11); *//* error */
6310: v21=(lc1-v1)/cv12*v11;
6311: v12=-v21;
6312: v22=v11;
6313: tnalp=v21/v11;
6314: if(first1==1){
6315: first1=0;
6316: 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);
6317: }
6318: 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);
6319: /*printf(fignu*/
6320: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6321: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6322: if(first==1){
6323: first=0;
6324: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6325: fprintf(ficgp,"\nset parametric;unset label");
6326: 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);
6327: fprintf(ficgp,"\nset ter svg size 640, 480");
6328: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6329: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6330: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6331: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6332: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6333: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6334: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6335: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6336: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6337: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6338: 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", \
6339: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6340: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6341: }else{
6342: first=0;
6343: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6344: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6345: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6346: 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", \
6347: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6348: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6349: }/* if first */
6350: } /* age mod 5 */
6351: } /* end loop age */
6352: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6353: first=1;
6354: } /*l12 */
6355: } /* k12 */
6356: } /*l1 */
6357: }/* k1 */
6358: } /* loop on combination of covariates j1 */
6359: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6360: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6361: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6362: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6363: free_vector(xp,1,npar);
6364: fclose(ficresprob);
6365: fclose(ficresprobcov);
6366: fclose(ficresprobcor);
6367: fflush(ficgp);
6368: fflush(fichtmcov);
6369: }
1.126 brouard 6370:
6371:
6372: /******************* Printing html file ***********/
1.201 brouard 6373: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6374: int lastpass, int stepm, int weightopt, char model[],\
6375: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6376: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6377: double jprev1, double mprev1,double anprev1, double dateprev1, \
6378: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6379: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6380:
6381: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6382: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6383: </ul>");
1.237 brouard 6384: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6385: </ul>", model);
1.214 brouard 6386: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6387: 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",
6388: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6389: 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 6390: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6391: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6392: fprintf(fichtm,"\
6393: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6394: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6395: fprintf(fichtm,"\
1.217 brouard 6396: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6397: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6398: fprintf(fichtm,"\
1.126 brouard 6399: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6400: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6401: fprintf(fichtm,"\
1.217 brouard 6402: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6403: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6404: fprintf(fichtm,"\
1.211 brouard 6405: - (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 6406: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6407: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6408: if(prevfcast==1){
6409: fprintf(fichtm,"\
6410: - Prevalence projections by age and states: \
1.201 brouard 6411: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6412: }
1.126 brouard 6413:
1.222 brouard 6414: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6415:
1.225 brouard 6416: m=pow(2,cptcoveff);
1.222 brouard 6417: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6418:
1.222 brouard 6419: jj1=0;
1.237 brouard 6420:
6421: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6422: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6423: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6424: continue;
1.220 brouard 6425:
1.222 brouard 6426: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6427: jj1++;
6428: if (cptcovn > 0) {
6429: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6430: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6431: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6432: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6433: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6434: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6435: }
1.237 brouard 6436: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6437: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6438: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6439: }
6440:
1.230 brouard 6441: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6442: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6443: if(invalidvarcomb[k1]){
6444: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6445: printf("\nCombination (%d) ignored because no cases \n",k1);
6446: continue;
6447: }
6448: }
6449: /* aij, bij */
1.241 brouard 6450: fprintf(fichtm,"<br>- Logit model (yours is: 1+age+%s), for example: logit(pij)=log(pij/pii)= aij+ bij age + V1 age + etc. as a function of age: <a href=\"%s_%d-1-%d.svg\">%s_%d-1-%d.svg</a><br> \
6451: <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 6452: /* Pij */
1.241 brouard 6453: 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> \
6454: <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 6455: /* Quasi-incidences */
6456: 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 6457: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6458: 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 6459: 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> \
6460: <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 6461: /* Survival functions (period) in state j */
6462: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6463: 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> \
6464: <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 6465: }
6466: /* State specific survival functions (period) */
6467: for(cpt=1; cpt<=nlstate;cpt++){
6468: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6469: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6470: <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 6471: }
6472: /* Period (stable) prevalence in each health state */
6473: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6474: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability to be in state %d some years earlier, knowing that we will be in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 6475: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 6476: }
6477: if(backcast==1){
6478: /* Period (stable) back prevalence in each health state */
6479: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6480: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability to be in state %d at a younger age, knowing that we will be 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 6481: <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 6482: }
1.217 brouard 6483: }
1.222 brouard 6484: if(prevfcast==1){
6485: /* Projection of prevalence up to period (stable) prevalence in each health state */
6486: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6487: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f) up to period (stable) prevalence in state %d. Or probability to be in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6488: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6489: }
6490: }
1.220 brouard 6491:
1.222 brouard 6492: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6493: 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> \
6494: <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 6495: }
6496: /* } /\* end i1 *\/ */
6497: }/* End k1 */
6498: fprintf(fichtm,"</ul>");
1.126 brouard 6499:
1.222 brouard 6500: fprintf(fichtm,"\
1.126 brouard 6501: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6502: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6503: - 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 6504: But because parameters are usually highly correlated (a higher incidence of disability \
6505: and a higher incidence of recovery can give very close observed transition) it might \
6506: be very useful to look not only at linear confidence intervals estimated from the \
6507: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6508: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6509: covariance matrix of the one-step probabilities. \
6510: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6511:
1.222 brouard 6512: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6513: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6514: fprintf(fichtm,"\
1.126 brouard 6515: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6516: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6517:
1.222 brouard 6518: fprintf(fichtm,"\
1.126 brouard 6519: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6520: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6521: fprintf(fichtm,"\
1.126 brouard 6522: - 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): \
6523: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6524: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6525: fprintf(fichtm,"\
1.126 brouard 6526: - (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): \
6527: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6528: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6529: fprintf(fichtm,"\
1.128 brouard 6530: - 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 6531: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6532: fprintf(fichtm,"\
1.128 brouard 6533: - 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 6534: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6535: fprintf(fichtm,"\
1.126 brouard 6536: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6537: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6538:
6539: /* if(popforecast==1) fprintf(fichtm,"\n */
6540: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6541: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6542: /* <br>",fileres,fileres,fileres,fileres); */
6543: /* else */
6544: /* 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 6545: fflush(fichtm);
6546: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6547:
1.225 brouard 6548: m=pow(2,cptcoveff);
1.222 brouard 6549: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6550:
1.222 brouard 6551: jj1=0;
1.237 brouard 6552:
1.241 brouard 6553: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6554: for(k1=1; k1<=m;k1++){
1.253 brouard 6555: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6556: continue;
1.222 brouard 6557: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6558: jj1++;
1.126 brouard 6559: if (cptcovn > 0) {
6560: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6561: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6562: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6563: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6564: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6565: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6566: }
6567:
1.126 brouard 6568: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6569:
1.222 brouard 6570: if(invalidvarcomb[k1]){
6571: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6572: continue;
6573: }
1.126 brouard 6574: }
6575: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6576: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6577: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6578: <img src=\"%s_%d-%d-%d.svg\">",cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
1.126 brouard 6579: }
6580: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6581: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6582: true period expectancies (those weighted with period prevalences are also\
6583: drawn in addition to the population based expectancies computed using\
1.241 brouard 6584: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6585: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6586: /* } /\* end i1 *\/ */
6587: }/* End k1 */
1.241 brouard 6588: }/* End nres */
1.222 brouard 6589: fprintf(fichtm,"</ul>");
6590: fflush(fichtm);
1.126 brouard 6591: }
6592:
6593: /******************* Gnuplot file **************/
1.223 brouard 6594: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6595:
6596: char dirfileres[132],optfileres[132];
1.223 brouard 6597: char gplotcondition[132];
1.237 brouard 6598: 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 6599: int lv=0, vlv=0, kl=0;
1.130 brouard 6600: int ng=0;
1.201 brouard 6601: int vpopbased;
1.223 brouard 6602: int ioffset; /* variable offset for columns */
1.235 brouard 6603: int nres=0; /* Index of resultline */
1.219 brouard 6604:
1.126 brouard 6605: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6606: /* printf("Problem with file %s",optionfilegnuplot); */
6607: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6608: /* } */
6609:
6610: /*#ifdef windows */
6611: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6612: /*#endif */
1.225 brouard 6613: m=pow(2,cptcoveff);
1.126 brouard 6614:
1.202 brouard 6615: /* Contribution to likelihood */
6616: /* Plot the probability implied in the likelihood */
1.223 brouard 6617: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6618: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6619: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6620: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6621: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6622: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6623: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6624: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6625: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6626: 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));
6627: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6628: 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));
6629: for (i=1; i<= nlstate ; i ++) {
6630: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6631: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6632: 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);
6633: for (j=2; j<= nlstate+ndeath ; j ++) {
6634: 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);
6635: }
6636: fprintf(ficgp,";\nset out; unset ylabel;\n");
6637: }
6638: /* 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 */
6639: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6640: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6641: fprintf(ficgp,"\nset out;unset log\n");
6642: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6643:
1.126 brouard 6644: strcpy(dirfileres,optionfilefiname);
6645: strcpy(optfileres,"vpl");
1.223 brouard 6646: /* 1eme*/
1.238 brouard 6647: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6648: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6649: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6650: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6651: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6652: continue;
6653: /* We are interested in selected combination by the resultline */
1.246 brouard 6654: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6655: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6656: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6657: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6658: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6659: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6660: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6661: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6662: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6663: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6664: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6665: }
6666: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6667: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6668: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6669: }
1.246 brouard 6670: /* printf("\n#\n"); */
1.238 brouard 6671: fprintf(ficgp,"\n#\n");
6672: if(invalidvarcomb[k1]){
6673: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6674: continue;
6675: }
1.235 brouard 6676:
1.241 brouard 6677: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6678: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6679: 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);
1.235 brouard 6680:
1.238 brouard 6681: for (i=1; i<= nlstate ; i ++) {
6682: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6683: else fprintf(ficgp," %%*lf (%%*lf)");
6684: }
1.242 brouard 6685: 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_"),k1-1,k1-1,nres);
1.238 brouard 6686: for (i=1; i<= nlstate ; i ++) {
6687: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6688: else fprintf(ficgp," %%*lf (%%*lf)");
6689: }
1.242 brouard 6690: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6691: for (i=1; i<= nlstate ; i ++) {
6692: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6693: else fprintf(ficgp," %%*lf (%%*lf)");
6694: }
6695: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6696: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6697: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6698: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6699: if(cptcoveff ==0){
1.245 brouard 6700: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6701: }else{
6702: kl=0;
6703: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6704: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6705: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6706: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6707: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6708: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6709: kl++;
1.238 brouard 6710: /* 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 *\/ */
6711: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6712: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6713: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6714: if(k==cptcoveff){
1.245 brouard 6715: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6716: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6717: }else{
6718: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6719: kl++;
6720: }
6721: } /* end covariate */
6722: } /* end if no covariate */
6723: } /* end if backcast */
6724: fprintf(ficgp,"\nset out \n");
6725: } /* nres */
1.201 brouard 6726: } /* k1 */
6727: } /* cpt */
1.235 brouard 6728:
6729:
1.126 brouard 6730: /*2 eme*/
1.238 brouard 6731: for (k1=1; k1<= m ; k1 ++){
6732: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6733: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6734: continue;
6735: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6736: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6737: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6738: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6739: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6740: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6741: vlv= nbcode[Tvaraff[k]][lv];
6742: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6743: }
1.237 brouard 6744: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6745: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6746: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6747: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6748: }
1.211 brouard 6749: fprintf(ficgp,"\n#\n");
1.223 brouard 6750: if(invalidvarcomb[k1]){
6751: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6752: continue;
6753: }
1.219 brouard 6754:
1.241 brouard 6755: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6756: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6757: if(vpopbased==0)
6758: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6759: else
6760: fprintf(ficgp,"\nreplot ");
6761: for (i=1; i<= nlstate+1 ; i ++) {
6762: k=2*i;
6763: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1, vpopbased);
6764: for (j=1; j<= nlstate+1 ; j ++) {
6765: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6766: else fprintf(ficgp," %%*lf (%%*lf)");
6767: }
6768: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6769: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6770: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1,vpopbased);
6771: for (j=1; j<= nlstate+1 ; j ++) {
6772: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6773: else fprintf(ficgp," %%*lf (%%*lf)");
6774: }
6775: fprintf(ficgp,"\" t\"\" w l lt 0,");
6776: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1,vpopbased);
6777: for (j=1; j<= nlstate+1 ; j ++) {
6778: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6779: else fprintf(ficgp," %%*lf (%%*lf)");
6780: }
6781: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6782: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6783: } /* state */
6784: } /* vpopbased */
1.244 brouard 6785: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6786: } /* end nres */
6787: } /* k1 end 2 eme*/
6788:
6789:
6790: /*3eme*/
6791: for (k1=1; k1<= m ; k1 ++){
6792: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6793: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6794: continue;
6795:
6796: for (cpt=1; cpt<= nlstate ; cpt ++) {
6797: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6798: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6799: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6800: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6801: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6802: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6803: vlv= nbcode[Tvaraff[k]][lv];
6804: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6805: }
6806: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6807: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6808: }
6809: fprintf(ficgp,"\n#\n");
6810: if(invalidvarcomb[k1]){
6811: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6812: continue;
6813: }
6814:
6815: /* k=2+nlstate*(2*cpt-2); */
6816: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6817: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6818: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6819: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),k1-1,k1-1,k,cpt);
1.238 brouard 6820: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6821: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6822: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6823: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6824: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6825: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6826:
1.238 brouard 6827: */
6828: for (i=1; i< nlstate ; i ++) {
6829: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+i,cpt,i+1);
6830: /* 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 6831:
1.238 brouard 6832: }
6833: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6834: }
6835: } /* end nres */
6836: } /* end kl 3eme */
1.126 brouard 6837:
1.223 brouard 6838: /* 4eme */
1.201 brouard 6839: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6840: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6841: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6842: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 6843: continue;
1.238 brouard 6844: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6845: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6846: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6847: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6848: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6849: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6850: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6851: vlv= nbcode[Tvaraff[k]][lv];
6852: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6853: }
6854: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6855: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6856: }
6857: fprintf(ficgp,"\n#\n");
6858: if(invalidvarcomb[k1]){
6859: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6860: continue;
1.223 brouard 6861: }
1.238 brouard 6862:
1.241 brouard 6863: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6864: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6865: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6866: k=3;
6867: for (i=1; i<= nlstate ; i ++){
6868: if(i==1){
6869: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6870: }else{
6871: fprintf(ficgp,", '' ");
6872: }
6873: l=(nlstate+ndeath)*(i-1)+1;
6874: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6875: for (j=2; j<= nlstate+ndeath ; j ++)
6876: fprintf(ficgp,"+$%d",k+l+j-1);
6877: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6878: } /* nlstate */
6879: fprintf(ficgp,"\nset out\n");
6880: } /* end cpt state*/
6881: } /* end nres */
6882: } /* end covariate k1 */
6883:
1.220 brouard 6884: /* 5eme */
1.201 brouard 6885: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6886: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6887: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6888: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 6889: continue;
1.238 brouard 6890: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6891: 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);
6892: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6893: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6894: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6895: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6896: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6897: vlv= nbcode[Tvaraff[k]][lv];
6898: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6899: }
6900: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6901: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6902: }
6903: fprintf(ficgp,"\n#\n");
6904: if(invalidvarcomb[k1]){
6905: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6906: continue;
6907: }
1.227 brouard 6908:
1.241 brouard 6909: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6910: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6911: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6912: k=3;
6913: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6914: if(j==1)
6915: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6916: else
6917: fprintf(ficgp,", '' ");
6918: l=(nlstate+ndeath)*(cpt-1) +j;
6919: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6920: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6921: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6922: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6923: } /* nlstate */
6924: fprintf(ficgp,", '' ");
6925: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6926: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6927: l=(nlstate+ndeath)*(cpt-1) +j;
6928: if(j < nlstate)
6929: fprintf(ficgp,"$%d +",k+l);
6930: else
6931: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6932: }
6933: fprintf(ficgp,"\nset out\n");
6934: } /* end cpt state*/
6935: } /* end covariate */
6936: } /* end nres */
1.227 brouard 6937:
1.220 brouard 6938: /* 6eme */
1.202 brouard 6939: /* CV preval stable (period) for each covariate */
1.237 brouard 6940: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6941: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6942: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6943: continue;
1.255 brouard 6944: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.227 brouard 6945:
1.211 brouard 6946: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6947: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6948: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6949: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6950: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6951: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6952: vlv= nbcode[Tvaraff[k]][lv];
6953: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6954: }
1.237 brouard 6955: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6956: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6957: }
1.211 brouard 6958: fprintf(ficgp,"\n#\n");
1.223 brouard 6959: if(invalidvarcomb[k1]){
1.227 brouard 6960: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6961: continue;
1.223 brouard 6962: }
1.227 brouard 6963:
1.241 brouard 6964: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6965: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6966: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6967: k=3; /* Offset */
1.255 brouard 6968: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 6969: if(i==1)
6970: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6971: else
6972: fprintf(ficgp,", '' ");
1.255 brouard 6973: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 6974: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6975: for (j=2; j<= nlstate ; j ++)
6976: fprintf(ficgp,"+$%d",k+l+j-1);
6977: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6978: } /* nlstate */
1.201 brouard 6979: fprintf(ficgp,"\nset out\n");
1.153 brouard 6980: } /* end cpt state*/
6981: } /* end covariate */
1.227 brouard 6982:
6983:
1.220 brouard 6984: /* 7eme */
1.218 brouard 6985: if(backcast == 1){
1.217 brouard 6986: /* CV back preval stable (period) for each covariate */
1.237 brouard 6987: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6988: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6989: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6990: continue;
1.255 brouard 6991: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life ending state */
6992: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 6993: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6994: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6995: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6996: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6997: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6998: vlv= nbcode[Tvaraff[k]][lv];
6999: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7000: }
1.237 brouard 7001: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7002: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7003: }
1.227 brouard 7004: fprintf(ficgp,"\n#\n");
7005: if(invalidvarcomb[k1]){
7006: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7007: continue;
7008: }
7009:
1.241 brouard 7010: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 7011: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7012: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7013: k=3; /* Offset */
1.255 brouard 7014: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7015: if(i==1)
7016: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7017: else
7018: fprintf(ficgp,", '' ");
7019: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7020: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7021: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7022: /* 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 7023: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7024: /* for (j=2; j<= nlstate ; j ++) */
7025: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7026: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7027: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7028: } /* nlstate */
7029: fprintf(ficgp,"\nset out\n");
1.218 brouard 7030: } /* end cpt state*/
7031: } /* end covariate */
7032: } /* End if backcast */
7033:
1.223 brouard 7034: /* 8eme */
1.218 brouard 7035: if(prevfcast==1){
7036: /* Projection from cross-sectional to stable (period) for each covariate */
7037:
1.237 brouard 7038: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7039: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7040: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7041: continue;
1.211 brouard 7042: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 7043: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7044: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7045: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7046: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7047: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7048: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7049: vlv= nbcode[Tvaraff[k]][lv];
7050: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7051: }
1.237 brouard 7052: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7053: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7054: }
1.227 brouard 7055: fprintf(ficgp,"\n#\n");
7056: if(invalidvarcomb[k1]){
7057: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7058: continue;
7059: }
7060:
7061: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7062: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 7063: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7064: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7065: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7066: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7067: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7068: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7069: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7070: if(i==1){
7071: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7072: }else{
7073: fprintf(ficgp,",\\\n '' ");
7074: }
7075: if(cptcoveff ==0){ /* No covariate */
7076: ioffset=2; /* Age is in 2 */
7077: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7078: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7079: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7080: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7081: fprintf(ficgp," u %d:(", ioffset);
7082: if(i==nlstate+1)
7083: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
7084: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7085: else
7086: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7087: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7088: }else{ /* more than 2 covariates */
7089: if(cptcoveff ==1){
7090: ioffset=4; /* Age is in 4 */
7091: }else{
7092: ioffset=6; /* Age is in 6 */
7093: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7094: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7095: }
7096: fprintf(ficgp," u %d:(",ioffset);
7097: kl=0;
7098: strcpy(gplotcondition,"(");
7099: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7100: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7101: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7102: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7103: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7104: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7105: kl++;
7106: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7107: kl++;
7108: if(k <cptcoveff && cptcoveff>1)
7109: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7110: }
7111: strcpy(gplotcondition+strlen(gplotcondition),")");
7112: /* 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 *\/ */
7113: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7114: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7115: /* '' 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*/
7116: if(i==nlstate+1){
7117: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7118: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7119: }else{
7120: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7121: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7122: }
7123: } /* end if covariate */
7124: } /* nlstate */
7125: fprintf(ficgp,"\nset out\n");
1.223 brouard 7126: } /* end cpt state*/
7127: } /* end covariate */
7128: } /* End if prevfcast */
1.227 brouard 7129:
7130:
1.238 brouard 7131: /* 9eme writing MLE parameters */
7132: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7133: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7134: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7135: for(k=1; k <=(nlstate+ndeath); k++){
7136: if (k != i) {
1.227 brouard 7137: fprintf(ficgp,"# current state %d\n",k);
7138: for(j=1; j <=ncovmodel; j++){
7139: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7140: jk++;
7141: }
7142: fprintf(ficgp,"\n");
1.126 brouard 7143: }
7144: }
1.223 brouard 7145: }
1.187 brouard 7146: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7147:
1.145 brouard 7148: /*goto avoid;*/
1.238 brouard 7149: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7150: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7151: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7152: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7153: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7154: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7155: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7156: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7157: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7158: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7159: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7160: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7161: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7162: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7163: fprintf(ficgp,"#\n");
1.223 brouard 7164: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7165: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7166: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7167: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7168: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7169: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7170: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7171: if(m != 1 && TKresult[nres]!= jk)
1.237 brouard 7172: continue;
7173: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7174: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7175: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7176: }
7177: fprintf(ficgp,"\n#\n");
1.241 brouard 7178: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7179: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7180: if (ng==1){
7181: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7182: fprintf(ficgp,"\nunset log y");
7183: }else if (ng==2){
7184: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7185: fprintf(ficgp,"\nset log y");
7186: }else if (ng==3){
7187: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7188: fprintf(ficgp,"\nset log y");
7189: }else
7190: fprintf(ficgp,"\nunset title ");
7191: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7192: i=1;
7193: for(k2=1; k2<=nlstate; k2++) {
7194: k3=i;
7195: for(k=1; k<=(nlstate+ndeath); k++) {
7196: if (k != k2){
7197: switch( ng) {
7198: case 1:
7199: if(nagesqr==0)
7200: fprintf(ficgp," p%d+p%d*x",i,i+1);
7201: else /* nagesqr =1 */
7202: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7203: break;
7204: case 2: /* ng=2 */
7205: if(nagesqr==0)
7206: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7207: else /* nagesqr =1 */
7208: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7209: break;
7210: case 3:
7211: if(nagesqr==0)
7212: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7213: else /* nagesqr =1 */
7214: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7215: break;
7216: }
7217: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7218: ijp=1; /* product no age */
7219: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7220: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7221: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7222: if(j==Tage[ij]) { /* Product by age */
7223: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7224: if(DummyV[j]==0){
1.237 brouard 7225: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7226: }else{ /* quantitative */
7227: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7228: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7229: }
7230: ij++;
7231: }
7232: }else if(j==Tprod[ijp]) { /* */
7233: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7234: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7235: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7236: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7237: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],nbcode[Tvard[ijp][2]][codtabm(jk,j)]); */
7238: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7239: }else{ /* Vn is dummy and Vm is quanti */
7240: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7241: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7242: }
7243: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7244: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7245: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7246: }else{ /* Both quanti */
7247: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7248: }
7249: }
1.238 brouard 7250: ijp++;
1.237 brouard 7251: }
7252: } else{ /* simple covariate */
7253: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7254: if(Dummy[j]==0){
7255: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7256: }else{ /* quantitative */
7257: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7258: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7259: }
1.237 brouard 7260: } /* end simple */
7261: } /* end j */
1.223 brouard 7262: }else{
7263: i=i-ncovmodel;
7264: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7265: fprintf(ficgp," (1.");
7266: }
1.227 brouard 7267:
1.223 brouard 7268: if(ng != 1){
7269: fprintf(ficgp,")/(1");
1.227 brouard 7270:
1.223 brouard 7271: for(k1=1; k1 <=nlstate; k1++){
7272: if(nagesqr==0)
7273: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7274: else /* nagesqr =1 */
7275: fprintf(ficgp,"+exp(p%d+p%d*x+p%d*x*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1,k3+(k1-1)*ncovmodel+1+nagesqr);
1.217 brouard 7276:
1.223 brouard 7277: ij=1;
7278: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7279: if((j-2)==Tage[ij]) { /* Bug valgrind */
7280: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7281: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7282: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7283: ij++;
7284: }
7285: }
7286: else
1.225 brouard 7287: fprintf(ficgp,"+p%d*%d",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 7288: }
7289: fprintf(ficgp,")");
7290: }
7291: fprintf(ficgp,")");
7292: if(ng ==2)
7293: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7294: else /* ng= 3 */
7295: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7296: }else{ /* end ng <> 1 */
7297: if( k !=k2) /* logit p11 is hard to draw */
7298: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7299: }
7300: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7301: fprintf(ficgp,",");
7302: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7303: fprintf(ficgp,",");
7304: i=i+ncovmodel;
7305: } /* end k */
7306: } /* end k2 */
7307: fprintf(ficgp,"\n set out\n");
7308: } /* end jk */
7309: } /* end ng */
7310: /* avoid: */
7311: fflush(ficgp);
1.126 brouard 7312: } /* end gnuplot */
7313:
7314:
7315: /*************** Moving average **************/
1.219 brouard 7316: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7317: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7318:
1.222 brouard 7319: int i, cpt, cptcod;
7320: int modcovmax =1;
7321: int mobilavrange, mob;
7322: int iage=0;
7323:
7324: double sum=0.;
7325: double age;
7326: double *sumnewp, *sumnewm;
7327: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7328:
7329:
1.225 brouard 7330: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7331: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7332:
7333: sumnewp = vector(1,ncovcombmax);
7334: sumnewm = vector(1,ncovcombmax);
7335: agemingood = vector(1,ncovcombmax);
7336: agemaxgood = vector(1,ncovcombmax);
7337:
7338: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7339: sumnewm[cptcod]=0.;
7340: sumnewp[cptcod]=0.;
7341: agemingood[cptcod]=0;
7342: agemaxgood[cptcod]=0;
7343: }
7344: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7345:
7346: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7347: if(mobilav==1) mobilavrange=5; /* default */
7348: else mobilavrange=mobilav;
7349: for (age=bage; age<=fage; age++)
7350: for (i=1; i<=nlstate;i++)
7351: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7352: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7353: /* We keep the original values on the extreme ages bage, fage and for
7354: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7355: we use a 5 terms etc. until the borders are no more concerned.
7356: */
7357: for (mob=3;mob <=mobilavrange;mob=mob+2){
7358: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7359: for (i=1; i<=nlstate;i++){
7360: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7361: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7362: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7363: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7364: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7365: }
7366: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7367: }
7368: }
7369: }/* end age */
7370: }/* end mob */
7371: }else
7372: return -1;
7373: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7374: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7375: if(invalidvarcomb[cptcod]){
7376: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7377: continue;
7378: }
1.219 brouard 7379:
1.222 brouard 7380: agemingood[cptcod]=fage-(mob-1)/2;
7381: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7382: sumnewm[cptcod]=0.;
7383: for (i=1; i<=nlstate;i++){
7384: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7385: }
7386: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7387: agemingood[cptcod]=age;
7388: }else{ /* bad */
7389: for (i=1; i<=nlstate;i++){
7390: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7391: } /* i */
7392: } /* end bad */
7393: }/* age */
7394: sum=0.;
7395: for (i=1; i<=nlstate;i++){
7396: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7397: }
7398: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7399: printf("For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one at any descending age!\n",cptcod);
7400: /* for (i=1; i<=nlstate;i++){ */
7401: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7402: /* } /\* i *\/ */
7403: } /* end bad */
7404: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7405: /* From youngest, finding the oldest wrong */
7406: agemaxgood[cptcod]=bage+(mob-1)/2;
7407: for (age=bage+(mob-1)/2; age<=fage; age++){
7408: sumnewm[cptcod]=0.;
7409: for (i=1; i<=nlstate;i++){
7410: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7411: }
7412: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7413: agemaxgood[cptcod]=age;
7414: }else{ /* bad */
7415: for (i=1; i<=nlstate;i++){
7416: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7417: } /* i */
7418: } /* end bad */
7419: }/* age */
7420: sum=0.;
7421: for (i=1; i<=nlstate;i++){
7422: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7423: }
7424: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7425: printf("For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one at any ascending age!\n",cptcod);
7426: /* for (i=1; i<=nlstate;i++){ */
7427: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7428: /* } /\* i *\/ */
7429: } /* end bad */
7430:
7431: for (age=bage; age<=fage; age++){
1.235 brouard 7432: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7433: sumnewp[cptcod]=0.;
7434: sumnewm[cptcod]=0.;
7435: for (i=1; i<=nlstate;i++){
7436: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7437: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7438: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7439: }
7440: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7441: }
7442: /* printf("\n"); */
7443: /* } */
7444: /* brutal averaging */
7445: for (i=1; i<=nlstate;i++){
7446: for (age=1; age<=bage; age++){
7447: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7448: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7449: }
7450: for (age=fage; age<=AGESUP; age++){
7451: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7452: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7453: }
7454: } /* end i status */
7455: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7456: for (age=1; age<=AGESUP; age++){
7457: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7458: mobaverage[(int)age][i][cptcod]=0.;
7459: }
7460: }
7461: }/* end cptcod */
7462: free_vector(sumnewm,1, ncovcombmax);
7463: free_vector(sumnewp,1, ncovcombmax);
7464: free_vector(agemaxgood,1, ncovcombmax);
7465: free_vector(agemingood,1, ncovcombmax);
7466: return 0;
7467: }/* End movingaverage */
1.218 brouard 7468:
1.126 brouard 7469:
7470: /************** Forecasting ******************/
1.235 brouard 7471: void prevforecast(char fileres[], double anproj1, double mproj1, double jproj1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anproj2, double p[], int cptcoveff){
1.126 brouard 7472: /* proj1, year, month, day of starting projection
7473: agemin, agemax range of age
7474: dateprev1 dateprev2 range of dates during which prevalence is computed
7475: anproj2 year of en of projection (same day and month as proj1).
7476: */
1.235 brouard 7477: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7478: double agec; /* generic age */
7479: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7480: double *popeffectif,*popcount;
7481: double ***p3mat;
1.218 brouard 7482: /* double ***mobaverage; */
1.126 brouard 7483: char fileresf[FILENAMELENGTH];
7484:
7485: agelim=AGESUP;
1.211 brouard 7486: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7487: in each health status at the date of interview (if between dateprev1 and dateprev2).
7488: We still use firstpass and lastpass as another selection.
7489: */
1.214 brouard 7490: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7491: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7492:
1.201 brouard 7493: strcpy(fileresf,"F_");
7494: strcat(fileresf,fileresu);
1.126 brouard 7495: if((ficresf=fopen(fileresf,"w"))==NULL) {
7496: printf("Problem with forecast resultfile: %s\n", fileresf);
7497: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7498: }
1.235 brouard 7499: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7500: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7501:
1.225 brouard 7502: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7503:
7504:
7505: stepsize=(int) (stepm+YEARM-1)/YEARM;
7506: if (stepm<=12) stepsize=1;
7507: if(estepm < stepm){
7508: printf ("Problem %d lower than %d\n",estepm, stepm);
7509: }
7510: else hstepm=estepm;
7511:
7512: hstepm=hstepm/stepm;
7513: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7514: fractional in yp1 */
7515: anprojmean=yp;
7516: yp2=modf((yp1*12),&yp);
7517: mprojmean=yp;
7518: yp1=modf((yp2*30.5),&yp);
7519: jprojmean=yp;
7520: if(jprojmean==0) jprojmean=1;
7521: if(mprojmean==0) jprojmean=1;
7522:
1.227 brouard 7523: i1=pow(2,cptcoveff);
1.126 brouard 7524: if (cptcovn < 1){i1=1;}
7525:
7526: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7527:
7528: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7529:
1.126 brouard 7530: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7531: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7532: for(k=1; k<=i1;k++){
1.253 brouard 7533: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7534: continue;
1.227 brouard 7535: if(invalidvarcomb[k]){
7536: printf("\nCombination (%d) projection ignored because no cases \n",k);
7537: continue;
7538: }
7539: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7540: for(j=1;j<=cptcoveff;j++) {
7541: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7542: }
1.235 brouard 7543: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7544: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7545: }
1.227 brouard 7546: fprintf(ficresf," yearproj age");
7547: for(j=1; j<=nlstate+ndeath;j++){
7548: for(i=1; i<=nlstate;i++)
7549: fprintf(ficresf," p%d%d",i,j);
7550: fprintf(ficresf," wp.%d",j);
7551: }
7552: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7553: fprintf(ficresf,"\n");
7554: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7555: for (agec=fage; agec>=(ageminpar-1); agec--){
7556: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7557: nhstepm = nhstepm/hstepm;
7558: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7559: oldm=oldms;savm=savms;
1.235 brouard 7560: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7561:
7562: for (h=0; h<=nhstepm; h++){
7563: if (h*hstepm/YEARM*stepm ==yearp) {
7564: fprintf(ficresf,"\n");
7565: for(j=1;j<=cptcoveff;j++)
7566: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7567: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7568: }
7569: for(j=1; j<=nlstate+ndeath;j++) {
7570: ppij=0.;
7571: for(i=1; i<=nlstate;i++) {
7572: if (mobilav==1)
7573: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7574: else {
7575: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7576: }
7577: if (h*hstepm/YEARM*stepm== yearp) {
7578: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7579: }
7580: } /* end i */
7581: if (h*hstepm/YEARM*stepm==yearp) {
7582: fprintf(ficresf," %.3f", ppij);
7583: }
7584: }/* end j */
7585: } /* end h */
7586: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7587: } /* end agec */
7588: } /* end yearp */
7589: } /* end k */
1.219 brouard 7590:
1.126 brouard 7591: fclose(ficresf);
1.215 brouard 7592: printf("End of Computing forecasting \n");
7593: fprintf(ficlog,"End of Computing forecasting\n");
7594:
1.126 brouard 7595: }
7596:
1.218 brouard 7597: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7598: /* void prevbackforecast(char fileres[], double anback1, double mback1, double jback1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anback2, double p[], int cptcoveff){ */
1.218 brouard 7599: /* /\* back1, year, month, day of starting backection */
7600: /* agemin, agemax range of age */
7601: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7602: /* anback2 year of en of backection (same day and month as back1). */
7603: /* *\/ */
7604: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7605: /* double agec; /\* generic age *\/ */
7606: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7607: /* double *popeffectif,*popcount; */
7608: /* double ***p3mat; */
7609: /* /\* double ***mobaverage; *\/ */
7610: /* char fileresfb[FILENAMELENGTH]; */
7611:
7612: /* agelim=AGESUP; */
7613: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7614: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7615: /* We still use firstpass and lastpass as another selection. */
7616: /* *\/ */
7617: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7618: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7619: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7620:
7621: /* strcpy(fileresfb,"FB_"); */
7622: /* strcat(fileresfb,fileresu); */
7623: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7624: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7625: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7626: /* } */
7627: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7628: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7629:
1.225 brouard 7630: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7631:
7632: /* /\* if (mobilav!=0) { *\/ */
7633: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7634: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7635: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7636: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7637: /* /\* } *\/ */
7638: /* /\* } *\/ */
7639:
7640: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7641: /* if (stepm<=12) stepsize=1; */
7642: /* if(estepm < stepm){ */
7643: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7644: /* } */
7645: /* else hstepm=estepm; */
7646:
7647: /* hstepm=hstepm/stepm; */
7648: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7649: /* fractional in yp1 *\/ */
7650: /* anprojmean=yp; */
7651: /* yp2=modf((yp1*12),&yp); */
7652: /* mprojmean=yp; */
7653: /* yp1=modf((yp2*30.5),&yp); */
7654: /* jprojmean=yp; */
7655: /* if(jprojmean==0) jprojmean=1; */
7656: /* if(mprojmean==0) jprojmean=1; */
7657:
1.225 brouard 7658: /* i1=cptcoveff; */
1.218 brouard 7659: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7660:
1.218 brouard 7661: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7662:
1.218 brouard 7663: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7664:
7665: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7666: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7667: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7668: /* k=k+1; */
7669: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7670: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7671: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7672: /* } */
7673: /* fprintf(ficresfb," yearbproj age"); */
7674: /* for(j=1; j<=nlstate+ndeath;j++){ */
7675: /* for(i=1; i<=nlstate;i++) */
7676: /* fprintf(ficresfb," p%d%d",i,j); */
7677: /* fprintf(ficresfb," p.%d",j); */
7678: /* } */
7679: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7680: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7681: /* fprintf(ficresfb,"\n"); */
7682: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7683: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7684: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7685: /* nhstepm = nhstepm/hstepm; */
7686: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7687: /* oldm=oldms;savm=savms; */
7688: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7689: /* for (h=0; h<=nhstepm; h++){ */
7690: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7691: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7692: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7693: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7694: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7695: /* } */
7696: /* for(j=1; j<=nlstate+ndeath;j++) { */
7697: /* ppij=0.; */
7698: /* for(i=1; i<=nlstate;i++) { */
7699: /* if (mobilav==1) */
7700: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7701: /* else { */
7702: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7703: /* } */
7704: /* if (h*hstepm/YEARM*stepm== yearp) { */
7705: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7706: /* } */
7707: /* } /\* end i *\/ */
7708: /* if (h*hstepm/YEARM*stepm==yearp) { */
7709: /* fprintf(ficresfb," %.3f", ppij); */
7710: /* } */
7711: /* }/\* end j *\/ */
7712: /* } /\* end h *\/ */
7713: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7714: /* } /\* end agec *\/ */
7715: /* } /\* end yearp *\/ */
7716: /* } /\* end cptcod *\/ */
7717: /* } /\* end cptcov *\/ */
7718:
7719: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7720:
7721: /* fclose(ficresfb); */
7722: /* printf("End of Computing Back forecasting \n"); */
7723: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7724:
1.218 brouard 7725: /* } */
1.217 brouard 7726:
1.126 brouard 7727: /************** Forecasting *****not tested NB*************/
1.227 brouard 7728: /* 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 7729:
1.227 brouard 7730: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7731: /* int *popage; */
7732: /* double calagedatem, agelim, kk1, kk2; */
7733: /* double *popeffectif,*popcount; */
7734: /* double ***p3mat,***tabpop,***tabpopprev; */
7735: /* /\* double ***mobaverage; *\/ */
7736: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7737:
1.227 brouard 7738: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7739: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7740: /* agelim=AGESUP; */
7741: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7742:
1.227 brouard 7743: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7744:
7745:
1.227 brouard 7746: /* strcpy(filerespop,"POP_"); */
7747: /* strcat(filerespop,fileresu); */
7748: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7749: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7750: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7751: /* } */
7752: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7753: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7754:
1.227 brouard 7755: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7756:
1.227 brouard 7757: /* /\* if (mobilav!=0) { *\/ */
7758: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7759: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7760: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7761: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7762: /* /\* } *\/ */
7763: /* /\* } *\/ */
1.126 brouard 7764:
1.227 brouard 7765: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7766: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7767:
1.227 brouard 7768: /* agelim=AGESUP; */
1.126 brouard 7769:
1.227 brouard 7770: /* hstepm=1; */
7771: /* hstepm=hstepm/stepm; */
1.218 brouard 7772:
1.227 brouard 7773: /* if (popforecast==1) { */
7774: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7775: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7776: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7777: /* } */
7778: /* popage=ivector(0,AGESUP); */
7779: /* popeffectif=vector(0,AGESUP); */
7780: /* popcount=vector(0,AGESUP); */
1.126 brouard 7781:
1.227 brouard 7782: /* i=1; */
7783: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7784:
1.227 brouard 7785: /* imx=i; */
7786: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7787: /* } */
1.218 brouard 7788:
1.227 brouard 7789: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7790: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7791: /* k=k+1; */
7792: /* fprintf(ficrespop,"\n#******"); */
7793: /* for(j=1;j<=cptcoveff;j++) { */
7794: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7795: /* } */
7796: /* fprintf(ficrespop,"******\n"); */
7797: /* fprintf(ficrespop,"# Age"); */
7798: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7799: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7800:
1.227 brouard 7801: /* for (cpt=0; cpt<=0;cpt++) { */
7802: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7803:
1.227 brouard 7804: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7805: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7806: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7807:
1.227 brouard 7808: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7809: /* oldm=oldms;savm=savms; */
7810: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7811:
1.227 brouard 7812: /* for (h=0; h<=nhstepm; h++){ */
7813: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7814: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7815: /* } */
7816: /* for(j=1; j<=nlstate+ndeath;j++) { */
7817: /* kk1=0.;kk2=0; */
7818: /* for(i=1; i<=nlstate;i++) { */
7819: /* if (mobilav==1) */
7820: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7821: /* else { */
7822: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7823: /* } */
7824: /* } */
7825: /* if (h==(int)(calagedatem+12*cpt)){ */
7826: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7827: /* /\*fprintf(ficrespop," %.3f", kk1); */
7828: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7829: /* } */
7830: /* } */
7831: /* for(i=1; i<=nlstate;i++){ */
7832: /* kk1=0.; */
7833: /* for(j=1; j<=nlstate;j++){ */
7834: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7835: /* } */
7836: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7837: /* } */
1.218 brouard 7838:
1.227 brouard 7839: /* if (h==(int)(calagedatem+12*cpt)) */
7840: /* for(j=1; j<=nlstate;j++) */
7841: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7842: /* } */
7843: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7844: /* } */
7845: /* } */
1.218 brouard 7846:
1.227 brouard 7847: /* /\******\/ */
1.218 brouard 7848:
1.227 brouard 7849: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7850: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7851: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7852: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7853: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7854:
1.227 brouard 7855: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7856: /* oldm=oldms;savm=savms; */
7857: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7858: /* for (h=0; h<=nhstepm; h++){ */
7859: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7860: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7861: /* } */
7862: /* for(j=1; j<=nlstate+ndeath;j++) { */
7863: /* kk1=0.;kk2=0; */
7864: /* for(i=1; i<=nlstate;i++) { */
7865: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7866: /* } */
7867: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7868: /* } */
7869: /* } */
7870: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7871: /* } */
7872: /* } */
7873: /* } */
7874: /* } */
1.218 brouard 7875:
1.227 brouard 7876: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7877:
1.227 brouard 7878: /* if (popforecast==1) { */
7879: /* free_ivector(popage,0,AGESUP); */
7880: /* free_vector(popeffectif,0,AGESUP); */
7881: /* free_vector(popcount,0,AGESUP); */
7882: /* } */
7883: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7884: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7885: /* fclose(ficrespop); */
7886: /* } /\* End of popforecast *\/ */
1.218 brouard 7887:
1.126 brouard 7888: int fileappend(FILE *fichier, char *optionfich)
7889: {
7890: if((fichier=fopen(optionfich,"a"))==NULL) {
7891: printf("Problem with file: %s\n", optionfich);
7892: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7893: return (0);
7894: }
7895: fflush(fichier);
7896: return (1);
7897: }
7898:
7899:
7900: /**************** function prwizard **********************/
7901: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7902: {
7903:
7904: /* Wizard to print covariance matrix template */
7905:
1.164 brouard 7906: char ca[32], cb[32];
7907: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7908: int numlinepar;
7909:
7910: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7911: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7912: for(i=1; i <=nlstate; i++){
7913: jj=0;
7914: for(j=1; j <=nlstate+ndeath; j++){
7915: if(j==i) continue;
7916: jj++;
7917: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7918: printf("%1d%1d",i,j);
7919: fprintf(ficparo,"%1d%1d",i,j);
7920: for(k=1; k<=ncovmodel;k++){
7921: /* printf(" %lf",param[i][j][k]); */
7922: /* fprintf(ficparo," %lf",param[i][j][k]); */
7923: printf(" 0.");
7924: fprintf(ficparo," 0.");
7925: }
7926: printf("\n");
7927: fprintf(ficparo,"\n");
7928: }
7929: }
7930: printf("# Scales (for hessian or gradient estimation)\n");
7931: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7932: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7933: for(i=1; i <=nlstate; i++){
7934: jj=0;
7935: for(j=1; j <=nlstate+ndeath; j++){
7936: if(j==i) continue;
7937: jj++;
7938: fprintf(ficparo,"%1d%1d",i,j);
7939: printf("%1d%1d",i,j);
7940: fflush(stdout);
7941: for(k=1; k<=ncovmodel;k++){
7942: /* printf(" %le",delti3[i][j][k]); */
7943: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7944: printf(" 0.");
7945: fprintf(ficparo," 0.");
7946: }
7947: numlinepar++;
7948: printf("\n");
7949: fprintf(ficparo,"\n");
7950: }
7951: }
7952: printf("# Covariance matrix\n");
7953: /* # 121 Var(a12)\n\ */
7954: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7955: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7956: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7957: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7958: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7959: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7960: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7961: fflush(stdout);
7962: fprintf(ficparo,"# Covariance matrix\n");
7963: /* # 121 Var(a12)\n\ */
7964: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7965: /* # ...\n\ */
7966: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7967:
7968: for(itimes=1;itimes<=2;itimes++){
7969: jj=0;
7970: for(i=1; i <=nlstate; i++){
7971: for(j=1; j <=nlstate+ndeath; j++){
7972: if(j==i) continue;
7973: for(k=1; k<=ncovmodel;k++){
7974: jj++;
7975: ca[0]= k+'a'-1;ca[1]='\0';
7976: if(itimes==1){
7977: printf("#%1d%1d%d",i,j,k);
7978: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7979: }else{
7980: printf("%1d%1d%d",i,j,k);
7981: fprintf(ficparo,"%1d%1d%d",i,j,k);
7982: /* printf(" %.5le",matcov[i][j]); */
7983: }
7984: ll=0;
7985: for(li=1;li <=nlstate; li++){
7986: for(lj=1;lj <=nlstate+ndeath; lj++){
7987: if(lj==li) continue;
7988: for(lk=1;lk<=ncovmodel;lk++){
7989: ll++;
7990: if(ll<=jj){
7991: cb[0]= lk +'a'-1;cb[1]='\0';
7992: if(ll<jj){
7993: if(itimes==1){
7994: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7995: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7996: }else{
7997: printf(" 0.");
7998: fprintf(ficparo," 0.");
7999: }
8000: }else{
8001: if(itimes==1){
8002: printf(" Var(%s%1d%1d)",ca,i,j);
8003: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8004: }else{
8005: printf(" 0.");
8006: fprintf(ficparo," 0.");
8007: }
8008: }
8009: }
8010: } /* end lk */
8011: } /* end lj */
8012: } /* end li */
8013: printf("\n");
8014: fprintf(ficparo,"\n");
8015: numlinepar++;
8016: } /* end k*/
8017: } /*end j */
8018: } /* end i */
8019: } /* end itimes */
8020:
8021: } /* end of prwizard */
8022: /******************* Gompertz Likelihood ******************************/
8023: double gompertz(double x[])
8024: {
8025: double A,B,L=0.0,sump=0.,num=0.;
8026: int i,n=0; /* n is the size of the sample */
8027:
1.220 brouard 8028: for (i=1;i<=imx ; i++) {
1.126 brouard 8029: sump=sump+weight[i];
8030: /* sump=sump+1;*/
8031: num=num+1;
8032: }
8033:
8034:
8035: /* for (i=0; i<=imx; i++)
8036: 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]);*/
8037:
8038: for (i=1;i<=imx ; i++)
8039: {
8040: if (cens[i] == 1 && wav[i]>1)
8041: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8042:
8043: if (cens[i] == 0 && wav[i]>1)
8044: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8045: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8046:
8047: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8048: if (wav[i] > 1 ) { /* ??? */
8049: L=L+A*weight[i];
8050: /* 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]);*/
8051: }
8052: }
8053:
8054: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8055:
8056: return -2*L*num/sump;
8057: }
8058:
1.136 brouard 8059: #ifdef GSL
8060: /******************* Gompertz_f Likelihood ******************************/
8061: double gompertz_f(const gsl_vector *v, void *params)
8062: {
8063: double A,B,LL=0.0,sump=0.,num=0.;
8064: double *x= (double *) v->data;
8065: int i,n=0; /* n is the size of the sample */
8066:
8067: for (i=0;i<=imx-1 ; i++) {
8068: sump=sump+weight[i];
8069: /* sump=sump+1;*/
8070: num=num+1;
8071: }
8072:
8073:
8074: /* for (i=0; i<=imx; i++)
8075: 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]);*/
8076: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8077: for (i=1;i<=imx ; i++)
8078: {
8079: if (cens[i] == 1 && wav[i]>1)
8080: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8081:
8082: if (cens[i] == 0 && wav[i]>1)
8083: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8084: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8085:
8086: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8087: if (wav[i] > 1 ) { /* ??? */
8088: LL=LL+A*weight[i];
8089: /* 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]);*/
8090: }
8091: }
8092:
8093: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8094: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8095:
8096: return -2*LL*num/sump;
8097: }
8098: #endif
8099:
1.126 brouard 8100: /******************* Printing html file ***********/
1.201 brouard 8101: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8102: int lastpass, int stepm, int weightopt, char model[],\
8103: int imx, double p[],double **matcov,double agemortsup){
8104: int i,k;
8105:
8106: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8107: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8108: for (i=1;i<=2;i++)
8109: 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 8110: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8111: fprintf(fichtm,"</ul>");
8112:
8113: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8114:
8115: 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>");
8116:
8117: for (k=agegomp;k<(agemortsup-2);k++)
8118: 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]);
8119:
8120:
8121: fflush(fichtm);
8122: }
8123:
8124: /******************* Gnuplot file **************/
1.201 brouard 8125: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8126:
8127: char dirfileres[132],optfileres[132];
1.164 brouard 8128:
1.126 brouard 8129: int ng;
8130:
8131:
8132: /*#ifdef windows */
8133: fprintf(ficgp,"cd \"%s\" \n",pathc);
8134: /*#endif */
8135:
8136:
8137: strcpy(dirfileres,optionfilefiname);
8138: strcpy(optfileres,"vpl");
1.199 brouard 8139: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8140: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8141: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8142: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8143: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8144:
8145: }
8146:
1.136 brouard 8147: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8148: {
1.126 brouard 8149:
1.136 brouard 8150: /*-------- data file ----------*/
8151: FILE *fic;
8152: char dummy[]=" ";
1.240 brouard 8153: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8154: int lstra;
1.136 brouard 8155: int linei, month, year,iout;
8156: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8157: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8158: char *stratrunc;
1.223 brouard 8159:
1.240 brouard 8160: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8161: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8162:
1.240 brouard 8163: for(v=1; v <=ncovcol;v++){
8164: DummyV[v]=0;
8165: FixedV[v]=0;
8166: }
8167: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8168: DummyV[v]=1;
8169: FixedV[v]=0;
8170: }
8171: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8172: DummyV[v]=0;
8173: FixedV[v]=1;
8174: }
8175: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8176: DummyV[v]=1;
8177: FixedV[v]=1;
8178: }
8179: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8180: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8181: 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]);
8182: }
1.126 brouard 8183:
1.136 brouard 8184: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8185: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8186: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8187: }
1.126 brouard 8188:
1.136 brouard 8189: i=1;
8190: linei=0;
8191: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8192: linei=linei+1;
8193: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8194: if(line[j] == '\t')
8195: line[j] = ' ';
8196: }
8197: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8198: ;
8199: };
8200: line[j+1]=0; /* Trims blanks at end of line */
8201: if(line[0]=='#'){
8202: fprintf(ficlog,"Comment line\n%s\n",line);
8203: printf("Comment line\n%s\n",line);
8204: continue;
8205: }
8206: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8207: strcpy(line, linetmp);
1.223 brouard 8208:
8209: /* Loops on waves */
8210: for (j=maxwav;j>=1;j--){
8211: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8212: cutv(stra, strb, line, ' ');
8213: if(strb[0]=='.') { /* Missing value */
8214: lval=-1;
8215: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8216: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8217: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8218: 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);
8219: 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);
8220: return 1;
8221: }
8222: }else{
8223: errno=0;
8224: /* what_kind_of_number(strb); */
8225: dval=strtod(strb,&endptr);
8226: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8227: /* if(strb != endptr && *endptr == '\0') */
8228: /* dval=dlval; */
8229: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8230: if( strb[0]=='\0' || (*endptr != '\0')){
8231: 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);
8232: 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);
8233: return 1;
8234: }
8235: cotqvar[j][iv][i]=dval;
8236: cotvar[j][ntv+iv][i]=dval;
8237: }
8238: strcpy(line,stra);
1.223 brouard 8239: }/* end loop ntqv */
1.225 brouard 8240:
1.223 brouard 8241: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8242: cutv(stra, strb, line, ' ');
8243: if(strb[0]=='.') { /* Missing value */
8244: lval=-1;
8245: }else{
8246: errno=0;
8247: lval=strtol(strb,&endptr,10);
8248: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8249: if( strb[0]=='\0' || (*endptr != '\0')){
8250: 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);
8251: 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);
8252: return 1;
8253: }
8254: }
8255: if(lval <-1 || lval >1){
8256: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8257: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8258: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8259: For example, for multinomial values like 1, 2 and 3,\n \
8260: build V1=0 V2=0 for the reference value (1),\n \
8261: V1=1 V2=0 for (2) \n \
1.223 brouard 8262: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8263: output of IMaCh is often meaningless.\n \
1.223 brouard 8264: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8265: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8266: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8267: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8268: For example, for multinomial values like 1, 2 and 3,\n \
8269: build V1=0 V2=0 for the reference value (1),\n \
8270: V1=1 V2=0 for (2) \n \
1.223 brouard 8271: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8272: output of IMaCh is often meaningless.\n \
1.223 brouard 8273: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8274: return 1;
8275: }
8276: cotvar[j][iv][i]=(double)(lval);
8277: strcpy(line,stra);
1.223 brouard 8278: }/* end loop ntv */
1.225 brouard 8279:
1.223 brouard 8280: /* Statuses at wave */
1.137 brouard 8281: cutv(stra, strb, line, ' ');
1.223 brouard 8282: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8283: lval=-1;
1.136 brouard 8284: }else{
1.238 brouard 8285: errno=0;
8286: lval=strtol(strb,&endptr,10);
8287: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8288: if( strb[0]=='\0' || (*endptr != '\0')){
8289: 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);
8290: 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);
8291: return 1;
8292: }
1.136 brouard 8293: }
1.225 brouard 8294:
1.136 brouard 8295: s[j][i]=lval;
1.225 brouard 8296:
1.223 brouard 8297: /* Date of Interview */
1.136 brouard 8298: strcpy(line,stra);
8299: cutv(stra, strb,line,' ');
1.169 brouard 8300: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8301: }
1.169 brouard 8302: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8303: month=99;
8304: year=9999;
1.136 brouard 8305: }else{
1.225 brouard 8306: 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);
8307: 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);
8308: return 1;
1.136 brouard 8309: }
8310: anint[j][i]= (double) year;
8311: mint[j][i]= (double)month;
8312: strcpy(line,stra);
1.223 brouard 8313: } /* End loop on waves */
1.225 brouard 8314:
1.223 brouard 8315: /* Date of death */
1.136 brouard 8316: cutv(stra, strb,line,' ');
1.169 brouard 8317: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8318: }
1.169 brouard 8319: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8320: month=99;
8321: year=9999;
8322: }else{
1.141 brouard 8323: 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 8324: 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);
8325: return 1;
1.136 brouard 8326: }
8327: andc[i]=(double) year;
8328: moisdc[i]=(double) month;
8329: strcpy(line,stra);
8330:
1.223 brouard 8331: /* Date of birth */
1.136 brouard 8332: cutv(stra, strb,line,' ');
1.169 brouard 8333: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8334: }
1.169 brouard 8335: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8336: month=99;
8337: year=9999;
8338: }else{
1.141 brouard 8339: 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);
8340: 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 8341: return 1;
1.136 brouard 8342: }
8343: if (year==9999) {
1.141 brouard 8344: 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);
8345: 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 8346: return 1;
8347:
1.136 brouard 8348: }
8349: annais[i]=(double)(year);
8350: moisnais[i]=(double)(month);
8351: strcpy(line,stra);
1.225 brouard 8352:
1.223 brouard 8353: /* Sample weight */
1.136 brouard 8354: cutv(stra, strb,line,' ');
8355: errno=0;
8356: dval=strtod(strb,&endptr);
8357: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8358: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8359: 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 8360: fflush(ficlog);
8361: return 1;
8362: }
8363: weight[i]=dval;
8364: strcpy(line,stra);
1.225 brouard 8365:
1.223 brouard 8366: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8367: cutv(stra, strb, line, ' ');
8368: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8369: lval=-1;
1.223 brouard 8370: }else{
1.225 brouard 8371: errno=0;
8372: /* what_kind_of_number(strb); */
8373: dval=strtod(strb,&endptr);
8374: /* if(strb != endptr && *endptr == '\0') */
8375: /* dval=dlval; */
8376: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8377: if( strb[0]=='\0' || (*endptr != '\0')){
8378: 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);
8379: 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);
8380: return 1;
8381: }
8382: coqvar[iv][i]=dval;
1.226 brouard 8383: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8384: }
8385: strcpy(line,stra);
8386: }/* end loop nqv */
1.136 brouard 8387:
1.223 brouard 8388: /* Covariate values */
1.136 brouard 8389: for (j=ncovcol;j>=1;j--){
8390: cutv(stra, strb,line,' ');
1.223 brouard 8391: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8392: lval=-1;
1.136 brouard 8393: }else{
1.225 brouard 8394: errno=0;
8395: lval=strtol(strb,&endptr,10);
8396: if( strb[0]=='\0' || (*endptr != '\0')){
8397: 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);
8398: 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);
8399: return 1;
8400: }
1.136 brouard 8401: }
8402: if(lval <-1 || lval >1){
1.225 brouard 8403: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8404: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8405: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8406: For example, for multinomial values like 1, 2 and 3,\n \
8407: build V1=0 V2=0 for the reference value (1),\n \
8408: V1=1 V2=0 for (2) \n \
1.136 brouard 8409: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8410: output of IMaCh is often meaningless.\n \
1.136 brouard 8411: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8412: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8413: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8414: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8415: For example, for multinomial values like 1, 2 and 3,\n \
8416: build V1=0 V2=0 for the reference value (1),\n \
8417: V1=1 V2=0 for (2) \n \
1.136 brouard 8418: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8419: output of IMaCh is often meaningless.\n \
1.136 brouard 8420: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8421: return 1;
1.136 brouard 8422: }
8423: covar[j][i]=(double)(lval);
8424: strcpy(line,stra);
8425: }
8426: lstra=strlen(stra);
1.225 brouard 8427:
1.136 brouard 8428: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8429: stratrunc = &(stra[lstra-9]);
8430: num[i]=atol(stratrunc);
8431: }
8432: else
8433: num[i]=atol(stra);
8434: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8435: 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;}*/
8436:
8437: i=i+1;
8438: } /* End loop reading data */
1.225 brouard 8439:
1.136 brouard 8440: *imax=i-1; /* Number of individuals */
8441: fclose(fic);
1.225 brouard 8442:
1.136 brouard 8443: return (0);
1.164 brouard 8444: /* endread: */
1.225 brouard 8445: printf("Exiting readdata: ");
8446: fclose(fic);
8447: return (1);
1.223 brouard 8448: }
1.126 brouard 8449:
1.234 brouard 8450: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8451: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8452: while (*p2 == ' ')
1.234 brouard 8453: p2++;
8454: /* while ((*p1++ = *p2++) !=0) */
8455: /* ; */
8456: /* do */
8457: /* while (*p2 == ' ') */
8458: /* p2++; */
8459: /* while (*p1++ == *p2++); */
8460: *stri=p2;
1.145 brouard 8461: }
8462:
1.235 brouard 8463: int decoderesult ( char resultline[], int nres)
1.230 brouard 8464: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8465: {
1.235 brouard 8466: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8467: char resultsav[MAXLINE];
1.234 brouard 8468: int resultmodel[MAXLINE];
8469: int modelresult[MAXLINE];
1.230 brouard 8470: char stra[80], strb[80], strc[80], strd[80],stre[80];
8471:
1.234 brouard 8472: removefirstspace(&resultline);
1.233 brouard 8473: printf("decoderesult:%s\n",resultline);
1.230 brouard 8474:
8475: if (strstr(resultline,"v") !=0){
8476: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8477: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8478: return 1;
8479: }
8480: trimbb(resultsav, resultline);
8481: if (strlen(resultsav) >1){
8482: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8483: }
1.253 brouard 8484: if(j == 0){ /* Resultline but no = */
8485: TKresult[nres]=0; /* Combination for the nresult and the model */
8486: return (0);
8487: }
8488:
1.234 brouard 8489: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8490: 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);
8491: 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);
8492: }
8493: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8494: if(nbocc(resultsav,'=') >1){
8495: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8496: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8497: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8498: }else
8499: cutl(strc,strd,resultsav,'=');
1.230 brouard 8500: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8501:
1.230 brouard 8502: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8503: Tvarsel[k]=atoi(strc);
8504: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8505: /* cptcovsel++; */
8506: if (nbocc(stra,'=') >0)
8507: strcpy(resultsav,stra); /* and analyzes it */
8508: }
1.235 brouard 8509: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8510: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8511: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8512: match=0;
1.236 brouard 8513: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8514: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8515: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8516: match=1;
8517: break;
8518: }
8519: }
8520: if(match == 0){
8521: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8522: }
8523: }
8524: }
1.235 brouard 8525: /* Checking for missing or useless values in comparison of current model needs */
8526: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8527: match=0;
1.235 brouard 8528: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8529: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8530: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8531: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8532: ++match;
8533: }
8534: }
8535: }
8536: if(match == 0){
8537: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8538: }else if(match > 1){
8539: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8540: }
8541: }
1.235 brouard 8542:
1.234 brouard 8543: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8544: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8545: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8546: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8547: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8548: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8549: /* 1 0 0 0 */
8550: /* 2 1 0 0 */
8551: /* 3 0 1 0 */
8552: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8553: /* 5 0 0 1 */
8554: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8555: /* 7 0 1 1 */
8556: /* 8 1 1 1 */
1.237 brouard 8557: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8558: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8559: /* V5*age V5 known which value for nres? */
8560: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8561: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8562: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8563: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8564: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8565: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8566: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8567: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8568: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8569: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8570: k4++;;
8571: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8572: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8573: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8574: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8575: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8576: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8577: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8578: k4q++;;
8579: }
8580: }
1.234 brouard 8581:
1.235 brouard 8582: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8583: return (0);
8584: }
1.235 brouard 8585:
1.230 brouard 8586: int decodemodel( char model[], int lastobs)
8587: /**< This routine decodes the model and returns:
1.224 brouard 8588: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8589: * - nagesqr = 1 if age*age in the model, otherwise 0.
8590: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8591: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8592: * - cptcovage number of covariates with age*products =2
8593: * - cptcovs number of simple covariates
8594: * - 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
8595: * which is a new column after the 9 (ncovcol) variables.
8596: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8597: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8598: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8599: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8600: */
1.136 brouard 8601: {
1.238 brouard 8602: int i, j, k, ks, v;
1.227 brouard 8603: int j1, k1, k2, k3, k4;
1.136 brouard 8604: char modelsav[80];
1.145 brouard 8605: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8606: char *strpt;
1.136 brouard 8607:
1.145 brouard 8608: /*removespace(model);*/
1.136 brouard 8609: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8610: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8611: if (strstr(model,"AGE") !=0){
1.192 brouard 8612: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8613: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8614: return 1;
8615: }
1.141 brouard 8616: if (strstr(model,"v") !=0){
8617: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8618: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8619: return 1;
8620: }
1.187 brouard 8621: strcpy(modelsav,model);
8622: if ((strpt=strstr(model,"age*age")) !=0){
8623: printf(" strpt=%s, model=%s\n",strpt, model);
8624: if(strpt != model){
1.234 brouard 8625: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8626: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8627: corresponding column of parameters.\n",model);
1.234 brouard 8628: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8629: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8630: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8631: return 1;
1.225 brouard 8632: }
1.187 brouard 8633: nagesqr=1;
8634: if (strstr(model,"+age*age") !=0)
1.234 brouard 8635: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8636: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8637: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8638: else
1.234 brouard 8639: substrchaine(modelsav, model, "age*age");
1.187 brouard 8640: }else
8641: nagesqr=0;
8642: if (strlen(modelsav) >1){
8643: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8644: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8645: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8646: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8647: * cst, age and age*age
8648: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8649: /* including age products which are counted in cptcovage.
8650: * but the covariates which are products must be treated
8651: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8652: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8653: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8654:
8655:
1.187 brouard 8656: /* Design
8657: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8658: * < ncovcol=8 >
8659: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8660: * k= 1 2 3 4 5 6 7 8
8661: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8662: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8663: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8664: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8665: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8666: * Tage[++cptcovage]=k
8667: * if products, new covar are created after ncovcol with k1
8668: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8669: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8670: * 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
8671: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8672: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8673: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8674: * < ncovcol=8 >
8675: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8676: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8677: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8678: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8679: * p Tprod[1]@2={ 6, 5}
8680: *p Tvard[1][1]@4= {7, 8, 5, 6}
8681: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8682: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8683: *How to reorganize?
8684: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8685: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8686: * {2, 1, 4, 8, 5, 6, 3, 7}
8687: * Struct []
8688: */
1.225 brouard 8689:
1.187 brouard 8690: /* This loop fills the array Tvar from the string 'model'.*/
8691: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8692: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8693: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8694: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8695: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8696: /* k=1 Tvar[1]=2 (from V2) */
8697: /* k=5 Tvar[5] */
8698: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8699: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8700: /* } */
1.198 brouard 8701: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8702: /*
8703: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8704: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8705: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8706: }
1.187 brouard 8707: cptcovage=0;
8708: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8709: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8710: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8711: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8712: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8713: /*scanf("%d",i);*/
8714: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8715: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8716: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8717: /* covar is not filled and then is empty */
8718: cptcovprod--;
8719: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8720: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8721: Typevar[k]=1; /* 1 for age product */
8722: cptcovage++; /* Sums the number of covariates which include age as a product */
8723: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8724: /*printf("stre=%s ", stre);*/
8725: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8726: cptcovprod--;
8727: cutl(stre,strb,strc,'V');
8728: Tvar[k]=atoi(stre);
8729: Typevar[k]=1; /* 1 for age product */
8730: cptcovage++;
8731: Tage[cptcovage]=k;
8732: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8733: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8734: cptcovn++;
8735: cptcovprodnoage++;k1++;
8736: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8737: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8738: because this model-covariate is a construction we invent a new column
8739: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8740: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8741: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8742: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8743: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8744: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8745: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8746: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8747: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8748: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8749: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8750: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8751: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8752: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8753: for (i=1; i<=lastobs;i++){
8754: /* Computes the new covariate which is a product of
8755: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8756: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8757: }
8758: } /* End age is not in the model */
8759: } /* End if model includes a product */
8760: else { /* no more sum */
8761: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8762: /* scanf("%d",i);*/
8763: cutl(strd,strc,strb,'V');
8764: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8765: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8766: Tvar[k]=atoi(strd);
8767: Typevar[k]=0; /* 0 for simple covariates */
8768: }
8769: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8770: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8771: scanf("%d",i);*/
1.187 brouard 8772: } /* end of loop + on total covariates */
8773: } /* end if strlen(modelsave == 0) age*age might exist */
8774: } /* end if strlen(model == 0) */
1.136 brouard 8775:
8776: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8777: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8778:
1.136 brouard 8779: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8780: printf("cptcovprod=%d ", cptcovprod);
8781: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8782: scanf("%d ",i);*/
8783:
8784:
1.230 brouard 8785: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8786: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8787: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8788: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8789: k = 1 2 3 4 5 6 7 8 9
8790: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8791: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8792: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8793: Dummy[k] 1 0 0 0 3 1 1 2 3
8794: Tmodelind[combination of covar]=k;
1.225 brouard 8795: */
8796: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8797: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8798: /* 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 8799: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8800: printf("Model=%s\n\
8801: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8802: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8803: 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);
8804: fprintf(ficlog,"Model=%s\n\
8805: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8806: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8807: 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 8808: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8809: 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 */
8810: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8811: Fixed[k]= 0;
8812: Dummy[k]= 0;
1.225 brouard 8813: ncoveff++;
1.232 brouard 8814: ncovf++;
1.234 brouard 8815: nsd++;
8816: modell[k].maintype= FTYPE;
8817: TvarsD[nsd]=Tvar[k];
8818: TvarsDind[nsd]=k;
8819: TvarF[ncovf]=Tvar[k];
8820: TvarFind[ncovf]=k;
8821: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8822: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8823: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8824: Fixed[k]= 0;
8825: Dummy[k]= 0;
8826: ncoveff++;
8827: ncovf++;
8828: modell[k].maintype= FTYPE;
8829: TvarF[ncovf]=Tvar[k];
8830: TvarFind[ncovf]=k;
1.230 brouard 8831: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8832: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8833: }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 8834: Fixed[k]= 0;
8835: Dummy[k]= 1;
1.230 brouard 8836: nqfveff++;
1.234 brouard 8837: modell[k].maintype= FTYPE;
8838: modell[k].subtype= FQ;
8839: nsq++;
8840: TvarsQ[nsq]=Tvar[k];
8841: TvarsQind[nsq]=k;
1.232 brouard 8842: ncovf++;
1.234 brouard 8843: TvarF[ncovf]=Tvar[k];
8844: TvarFind[ncovf]=k;
1.231 brouard 8845: 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 8846: 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 8847: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8848: Fixed[k]= 1;
8849: Dummy[k]= 0;
1.225 brouard 8850: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8851: modell[k].maintype= VTYPE;
8852: modell[k].subtype= VD;
8853: nsd++;
8854: TvarsD[nsd]=Tvar[k];
8855: TvarsDind[nsd]=k;
8856: ncovv++; /* Only simple time varying variables */
8857: TvarV[ncovv]=Tvar[k];
1.242 brouard 8858: 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 8859: 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 */
8860: 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 8861: 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);
8862: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8863: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8864: Fixed[k]= 1;
8865: Dummy[k]= 1;
8866: nqtveff++;
8867: modell[k].maintype= VTYPE;
8868: modell[k].subtype= VQ;
8869: ncovv++; /* Only simple time varying variables */
8870: nsq++;
8871: TvarsQ[nsq]=Tvar[k];
8872: TvarsQind[nsq]=k;
8873: TvarV[ncovv]=Tvar[k];
1.242 brouard 8874: 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 8875: 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 */
8876: 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 8877: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8878: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8879: 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 8880: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8881: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8882: ncova++;
8883: TvarA[ncova]=Tvar[k];
8884: TvarAind[ncova]=k;
1.231 brouard 8885: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8886: Fixed[k]= 2;
8887: Dummy[k]= 2;
8888: modell[k].maintype= ATYPE;
8889: modell[k].subtype= APFD;
8890: /* ncoveff++; */
1.227 brouard 8891: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8892: Fixed[k]= 2;
8893: Dummy[k]= 3;
8894: modell[k].maintype= ATYPE;
8895: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8896: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8897: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8898: Fixed[k]= 3;
8899: Dummy[k]= 2;
8900: modell[k].maintype= ATYPE;
8901: modell[k].subtype= APVD; /* Product age * varying dummy */
8902: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8903: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8904: Fixed[k]= 3;
8905: Dummy[k]= 3;
8906: modell[k].maintype= ATYPE;
8907: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8908: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8909: }
8910: }else if (Typevar[k] == 2) { /* product without age */
8911: k1=Tposprod[k];
8912: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8913: if(Tvard[k1][2] <=ncovcol){
8914: Fixed[k]= 1;
8915: Dummy[k]= 0;
8916: modell[k].maintype= FTYPE;
8917: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8918: ncovf++; /* Fixed variables without age */
8919: TvarF[ncovf]=Tvar[k];
8920: TvarFind[ncovf]=k;
8921: }else if(Tvard[k1][2] <=ncovcol+nqv){
8922: Fixed[k]= 0; /* or 2 ?*/
8923: Dummy[k]= 1;
8924: modell[k].maintype= FTYPE;
8925: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8926: ncovf++; /* Varying variables without age */
8927: TvarF[ncovf]=Tvar[k];
8928: TvarFind[ncovf]=k;
8929: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8930: Fixed[k]= 1;
8931: Dummy[k]= 0;
8932: modell[k].maintype= VTYPE;
8933: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8934: ncovv++; /* Varying variables without age */
8935: TvarV[ncovv]=Tvar[k];
8936: TvarVind[ncovv]=k;
8937: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8938: Fixed[k]= 1;
8939: Dummy[k]= 1;
8940: modell[k].maintype= VTYPE;
8941: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8942: ncovv++; /* Varying variables without age */
8943: TvarV[ncovv]=Tvar[k];
8944: TvarVind[ncovv]=k;
8945: }
1.227 brouard 8946: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8947: if(Tvard[k1][2] <=ncovcol){
8948: Fixed[k]= 0; /* or 2 ?*/
8949: Dummy[k]= 1;
8950: modell[k].maintype= FTYPE;
8951: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8952: ncovf++; /* Fixed variables without age */
8953: TvarF[ncovf]=Tvar[k];
8954: TvarFind[ncovf]=k;
8955: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8956: Fixed[k]= 1;
8957: Dummy[k]= 1;
8958: modell[k].maintype= VTYPE;
8959: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8960: ncovv++; /* Varying variables without age */
8961: TvarV[ncovv]=Tvar[k];
8962: TvarVind[ncovv]=k;
8963: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8964: Fixed[k]= 1;
8965: Dummy[k]= 1;
8966: modell[k].maintype= VTYPE;
8967: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8968: ncovv++; /* Varying variables without age */
8969: TvarV[ncovv]=Tvar[k];
8970: TvarVind[ncovv]=k;
8971: ncovv++; /* Varying variables without age */
8972: TvarV[ncovv]=Tvar[k];
8973: TvarVind[ncovv]=k;
8974: }
1.227 brouard 8975: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8976: if(Tvard[k1][2] <=ncovcol){
8977: Fixed[k]= 1;
8978: Dummy[k]= 1;
8979: modell[k].maintype= VTYPE;
8980: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8981: ncovv++; /* Varying variables without age */
8982: TvarV[ncovv]=Tvar[k];
8983: TvarVind[ncovv]=k;
8984: }else if(Tvard[k1][2] <=ncovcol+nqv){
8985: Fixed[k]= 1;
8986: Dummy[k]= 1;
8987: modell[k].maintype= VTYPE;
8988: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8989: ncovv++; /* Varying variables without age */
8990: TvarV[ncovv]=Tvar[k];
8991: TvarVind[ncovv]=k;
8992: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8993: Fixed[k]= 1;
8994: Dummy[k]= 0;
8995: modell[k].maintype= VTYPE;
8996: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8997: ncovv++; /* Varying variables without age */
8998: TvarV[ncovv]=Tvar[k];
8999: TvarVind[ncovv]=k;
9000: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9001: Fixed[k]= 1;
9002: Dummy[k]= 1;
9003: modell[k].maintype= VTYPE;
9004: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9005: ncovv++; /* Varying variables without age */
9006: TvarV[ncovv]=Tvar[k];
9007: TvarVind[ncovv]=k;
9008: }
1.227 brouard 9009: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9010: if(Tvard[k1][2] <=ncovcol){
9011: Fixed[k]= 1;
9012: Dummy[k]= 1;
9013: modell[k].maintype= VTYPE;
9014: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9015: ncovv++; /* Varying variables without age */
9016: TvarV[ncovv]=Tvar[k];
9017: TvarVind[ncovv]=k;
9018: }else if(Tvard[k1][2] <=ncovcol+nqv){
9019: Fixed[k]= 1;
9020: Dummy[k]= 1;
9021: modell[k].maintype= VTYPE;
9022: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9023: ncovv++; /* Varying variables without age */
9024: TvarV[ncovv]=Tvar[k];
9025: TvarVind[ncovv]=k;
9026: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9027: Fixed[k]= 1;
9028: Dummy[k]= 1;
9029: modell[k].maintype= VTYPE;
9030: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9031: ncovv++; /* Varying variables without age */
9032: TvarV[ncovv]=Tvar[k];
9033: TvarVind[ncovv]=k;
9034: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9035: Fixed[k]= 1;
9036: Dummy[k]= 1;
9037: modell[k].maintype= VTYPE;
9038: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9039: ncovv++; /* Varying variables without age */
9040: TvarV[ncovv]=Tvar[k];
9041: TvarVind[ncovv]=k;
9042: }
1.227 brouard 9043: }else{
1.240 brouard 9044: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9045: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9046: } /*end k1*/
1.225 brouard 9047: }else{
1.226 brouard 9048: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9049: 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 9050: }
1.227 brouard 9051: 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 9052: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9053: 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]);
9054: }
9055: /* Searching for doublons in the model */
9056: for(k1=1; k1<= cptcovt;k1++){
9057: for(k2=1; k2 <k1;k2++){
9058: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9059: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9060: if(Tvar[k1]==Tvar[k2]){
9061: 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]]);
9062: 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);
9063: return(1);
9064: }
9065: }else if (Typevar[k1] ==2){
9066: k3=Tposprod[k1];
9067: k4=Tposprod[k2];
9068: 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])) ){
9069: 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]]);
9070: 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);
9071: return(1);
9072: }
9073: }
1.227 brouard 9074: }
9075: }
1.225 brouard 9076: }
9077: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9078: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9079: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9080: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9081: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9082: /*endread:*/
1.225 brouard 9083: printf("Exiting decodemodel: ");
9084: return (1);
1.136 brouard 9085: }
9086:
1.169 brouard 9087: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9088: {/* Check ages at death */
1.136 brouard 9089: int i, m;
1.218 brouard 9090: int firstone=0;
9091:
1.136 brouard 9092: for (i=1; i<=imx; i++) {
9093: for(m=2; (m<= maxwav); m++) {
9094: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9095: anint[m][i]=9999;
1.216 brouard 9096: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9097: s[m][i]=-1;
1.136 brouard 9098: }
9099: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 9100: *nberr = *nberr + 1;
1.218 brouard 9101: if(firstone == 0){
9102: firstone=1;
9103: printf("Error! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown, you must set an arbitrary year of death or he/she is skipped and results can be biased (%d) because status is a death state %d at wave %d. Wave dropped.\nOther similar cases in log file\n",(int)moisdc[i],(int)andc[i],num[i],i, *nberr,s[m][i],m);
9104: }
9105: fprintf(ficlog,"Error! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown, you must set an arbitrary year of death or he/she is skipped and results can be biased (%d) because status is a death state %d at wave %d. Wave dropped.\n",(int)moisdc[i],(int)andc[i],num[i],i, *nberr,s[m][i],m);
1.136 brouard 9106: s[m][i]=-1;
9107: }
9108: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9109: (*nberr)++;
1.136 brouard 9110: printf("Error! Month of death of individual %ld on line %d was unknown %2d, you should set it otherwise the information on the death is skipped and results are biased.\n",num[i],i,(int)moisdc[i]);
9111: fprintf(ficlog,"Error! Month of death of individual %ld on line %d was unknown %f, you should set it otherwise the information on the death is skipped and results are biased.\n",num[i],i,moisdc[i]);
9112: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
9113: }
9114: }
9115: }
9116:
9117: for (i=1; i<=imx; i++) {
9118: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9119: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9120: 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 9121: if (s[m][i] >= nlstate+1) {
1.169 brouard 9122: if(agedc[i]>0){
9123: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9124: agev[m][i]=agedc[i];
1.214 brouard 9125: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9126: }else {
1.136 brouard 9127: if ((int)andc[i]!=9999){
9128: nbwarn++;
9129: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9130: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9131: agev[m][i]=-1;
9132: }
9133: }
1.169 brouard 9134: } /* agedc > 0 */
1.214 brouard 9135: } /* end if */
1.136 brouard 9136: else if(s[m][i] !=9){ /* Standard case, age in fractional
9137: years but with the precision of a month */
9138: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9139: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9140: agev[m][i]=1;
9141: else if(agev[m][i] < *agemin){
9142: *agemin=agev[m][i];
9143: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9144: }
9145: else if(agev[m][i] >*agemax){
9146: *agemax=agev[m][i];
1.156 brouard 9147: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9148: }
9149: /*agev[m][i]=anint[m][i]-annais[i];*/
9150: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9151: } /* en if 9*/
1.136 brouard 9152: else { /* =9 */
1.214 brouard 9153: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9154: agev[m][i]=1;
9155: s[m][i]=-1;
9156: }
9157: }
1.214 brouard 9158: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9159: agev[m][i]=1;
1.214 brouard 9160: else{
9161: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9162: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9163: agev[m][i]=0;
9164: }
9165: } /* End for lastpass */
9166: }
1.136 brouard 9167:
9168: for (i=1; i<=imx; i++) {
9169: for(m=firstpass; (m<=lastpass); m++){
9170: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9171: (*nberr)++;
1.136 brouard 9172: 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);
9173: 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);
9174: return 1;
9175: }
9176: }
9177: }
9178:
9179: /*for (i=1; i<=imx; i++){
9180: for (m=firstpass; (m<lastpass); m++){
9181: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9182: }
9183:
9184: }*/
9185:
9186:
1.139 brouard 9187: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9188: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9189:
9190: return (0);
1.164 brouard 9191: /* endread:*/
1.136 brouard 9192: printf("Exiting calandcheckages: ");
9193: return (1);
9194: }
9195:
1.172 brouard 9196: #if defined(_MSC_VER)
9197: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9198: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9199: //#include "stdafx.h"
9200: //#include <stdio.h>
9201: //#include <tchar.h>
9202: //#include <windows.h>
9203: //#include <iostream>
9204: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9205:
9206: LPFN_ISWOW64PROCESS fnIsWow64Process;
9207:
9208: BOOL IsWow64()
9209: {
9210: BOOL bIsWow64 = FALSE;
9211:
9212: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9213: // (HANDLE, PBOOL);
9214:
9215: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9216:
9217: HMODULE module = GetModuleHandle(_T("kernel32"));
9218: const char funcName[] = "IsWow64Process";
9219: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9220: GetProcAddress(module, funcName);
9221:
9222: if (NULL != fnIsWow64Process)
9223: {
9224: if (!fnIsWow64Process(GetCurrentProcess(),
9225: &bIsWow64))
9226: //throw std::exception("Unknown error");
9227: printf("Unknown error\n");
9228: }
9229: return bIsWow64 != FALSE;
9230: }
9231: #endif
1.177 brouard 9232:
1.191 brouard 9233: void syscompilerinfo(int logged)
1.167 brouard 9234: {
9235: /* #include "syscompilerinfo.h"*/
1.185 brouard 9236: /* command line Intel compiler 32bit windows, XP compatible:*/
9237: /* /GS /W3 /Gy
9238: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9239: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9240: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9241: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9242: */
9243: /* 64 bits */
1.185 brouard 9244: /*
9245: /GS /W3 /Gy
9246: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9247: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9248: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9249: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9250: /* Optimization are useless and O3 is slower than O2 */
9251: /*
9252: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9253: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9254: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9255: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9256: */
1.186 brouard 9257: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9258: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9259: /PDB:"visual studio
9260: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9261: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9262: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9263: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9264: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9265: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9266: uiAccess='false'"
9267: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9268: /NOLOGO /TLBID:1
9269: */
1.177 brouard 9270: #if defined __INTEL_COMPILER
1.178 brouard 9271: #if defined(__GNUC__)
9272: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9273: #endif
1.177 brouard 9274: #elif defined(__GNUC__)
1.179 brouard 9275: #ifndef __APPLE__
1.174 brouard 9276: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9277: #endif
1.177 brouard 9278: struct utsname sysInfo;
1.178 brouard 9279: int cross = CROSS;
9280: if (cross){
9281: printf("Cross-");
1.191 brouard 9282: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9283: }
1.174 brouard 9284: #endif
9285:
1.171 brouard 9286: #include <stdint.h>
1.178 brouard 9287:
1.191 brouard 9288: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9289: #if defined(__clang__)
1.191 brouard 9290: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9291: #endif
9292: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9293: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9294: #endif
9295: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9296: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9297: #endif
9298: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9299: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9300: #endif
9301: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9302: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9303: #endif
9304: #if defined(_MSC_VER)
1.191 brouard 9305: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9306: #endif
9307: #if defined(__PGI)
1.191 brouard 9308: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9309: #endif
9310: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9311: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9312: #endif
1.191 brouard 9313: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9314:
1.167 brouard 9315: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9316: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9317: // Windows (x64 and x86)
1.191 brouard 9318: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9319: #elif __unix__ // all unices, not all compilers
9320: // Unix
1.191 brouard 9321: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9322: #elif __linux__
9323: // linux
1.191 brouard 9324: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9325: #elif __APPLE__
1.174 brouard 9326: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9327: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9328: #endif
9329:
9330: /* __MINGW32__ */
9331: /* __CYGWIN__ */
9332: /* __MINGW64__ */
9333: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9334: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9335: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9336: /* _WIN64 // Defined for applications for Win64. */
9337: /* _M_X64 // Defined for compilations that target x64 processors. */
9338: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9339:
1.167 brouard 9340: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9341: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9342: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9343: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9344: #else
1.191 brouard 9345: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9346: #endif
9347:
1.169 brouard 9348: #if defined(__GNUC__)
9349: # if defined(__GNUC_PATCHLEVEL__)
9350: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9351: + __GNUC_MINOR__ * 100 \
9352: + __GNUC_PATCHLEVEL__)
9353: # else
9354: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9355: + __GNUC_MINOR__ * 100)
9356: # endif
1.174 brouard 9357: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9358: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9359:
9360: if (uname(&sysInfo) != -1) {
9361: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9362: 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 9363: }
9364: else
9365: perror("uname() error");
1.179 brouard 9366: //#ifndef __INTEL_COMPILER
9367: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9368: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9369: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9370: #endif
1.169 brouard 9371: #endif
1.172 brouard 9372:
9373: // void main()
9374: // {
1.169 brouard 9375: #if defined(_MSC_VER)
1.174 brouard 9376: if (IsWow64()){
1.191 brouard 9377: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9378: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9379: }
9380: else{
1.191 brouard 9381: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9382: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9383: }
1.172 brouard 9384: // printf("\nPress Enter to continue...");
9385: // getchar();
9386: // }
9387:
1.169 brouard 9388: #endif
9389:
1.167 brouard 9390:
1.219 brouard 9391: }
1.136 brouard 9392:
1.219 brouard 9393: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9394: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9395: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9396: /* double ftolpl = 1.e-10; */
1.180 brouard 9397: double age, agebase, agelim;
1.203 brouard 9398: double tot;
1.180 brouard 9399:
1.202 brouard 9400: strcpy(filerespl,"PL_");
9401: strcat(filerespl,fileresu);
9402: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9403: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9404: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9405: }
1.227 brouard 9406: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9407: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9408: pstamp(ficrespl);
1.203 brouard 9409: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9410: fprintf(ficrespl,"#Age ");
9411: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9412: fprintf(ficrespl,"\n");
1.180 brouard 9413:
1.219 brouard 9414: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9415:
1.219 brouard 9416: agebase=ageminpar;
9417: agelim=agemaxpar;
1.180 brouard 9418:
1.227 brouard 9419: /* i1=pow(2,ncoveff); */
1.234 brouard 9420: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9421: if (cptcovn < 1){i1=1;}
1.180 brouard 9422:
1.238 brouard 9423: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9424: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9425: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9426: continue;
1.235 brouard 9427:
1.238 brouard 9428: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9429: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9430: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9431: /* k=k+1; */
9432: /* to clean */
9433: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9434: fprintf(ficrespl,"#******");
9435: printf("#******");
9436: fprintf(ficlog,"#******");
9437: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9438: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9439: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9440: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9441: }
9442: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9443: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9444: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9445: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9446: }
9447: fprintf(ficrespl,"******\n");
9448: printf("******\n");
9449: fprintf(ficlog,"******\n");
9450: if(invalidvarcomb[k]){
9451: printf("\nCombination (%d) ignored because no case \n",k);
9452: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9453: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9454: continue;
9455: }
1.219 brouard 9456:
1.238 brouard 9457: fprintf(ficrespl,"#Age ");
9458: for(j=1;j<=cptcoveff;j++) {
9459: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9460: }
9461: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9462: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9463:
1.238 brouard 9464: for (age=agebase; age<=agelim; age++){
9465: /* for (age=agebase; age<=agebase; age++){ */
9466: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9467: fprintf(ficrespl,"%.0f ",age );
9468: for(j=1;j<=cptcoveff;j++)
9469: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9470: tot=0.;
9471: for(i=1; i<=nlstate;i++){
9472: tot += prlim[i][i];
9473: fprintf(ficrespl," %.5f", prlim[i][i]);
9474: }
9475: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9476: } /* Age */
9477: /* was end of cptcod */
9478: } /* cptcov */
9479: } /* nres */
1.219 brouard 9480: return 0;
1.180 brouard 9481: }
9482:
1.218 brouard 9483: 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){
9484: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9485:
9486: /* Computes the back prevalence limit for any combination of covariate values
9487: * at any age between ageminpar and agemaxpar
9488: */
1.235 brouard 9489: int i, j, k, i1, nres=0 ;
1.217 brouard 9490: /* double ftolpl = 1.e-10; */
9491: double age, agebase, agelim;
9492: double tot;
1.218 brouard 9493: /* double ***mobaverage; */
9494: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9495:
9496: strcpy(fileresplb,"PLB_");
9497: strcat(fileresplb,fileresu);
9498: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9499: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9500: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9501: }
9502: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9503: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9504: pstamp(ficresplb);
9505: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9506: fprintf(ficresplb,"#Age ");
9507: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9508: fprintf(ficresplb,"\n");
9509:
1.218 brouard 9510:
9511: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9512:
9513: agebase=ageminpar;
9514: agelim=agemaxpar;
9515:
9516:
1.227 brouard 9517: i1=pow(2,cptcoveff);
1.218 brouard 9518: if (cptcovn < 1){i1=1;}
1.227 brouard 9519:
1.238 brouard 9520: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9521: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9522: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9523: continue;
9524: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9525: fprintf(ficresplb,"#******");
9526: printf("#******");
9527: fprintf(ficlog,"#******");
9528: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9529: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9530: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9531: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9532: }
9533: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9534: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9535: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9536: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9537: }
9538: fprintf(ficresplb,"******\n");
9539: printf("******\n");
9540: fprintf(ficlog,"******\n");
9541: if(invalidvarcomb[k]){
9542: printf("\nCombination (%d) ignored because no cases \n",k);
9543: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9544: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9545: continue;
9546: }
1.218 brouard 9547:
1.238 brouard 9548: fprintf(ficresplb,"#Age ");
9549: for(j=1;j<=cptcoveff;j++) {
9550: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9551: }
9552: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9553: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9554:
9555:
1.238 brouard 9556: for (age=agebase; age<=agelim; age++){
9557: /* for (age=agebase; age<=agebase; age++){ */
9558: if(mobilavproj > 0){
9559: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9560: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9561: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9562: }else if (mobilavproj == 0){
9563: 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);
9564: 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);
9565: exit(1);
9566: }else{
9567: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9568: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9569: }
9570: fprintf(ficresplb,"%.0f ",age );
9571: for(j=1;j<=cptcoveff;j++)
9572: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9573: tot=0.;
9574: for(i=1; i<=nlstate;i++){
9575: tot += bprlim[i][i];
9576: fprintf(ficresplb," %.5f", bprlim[i][i]);
9577: }
9578: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9579: } /* Age */
9580: /* was end of cptcod */
1.255 brouard 9581: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 9582: } /* end of any combination */
9583: } /* end of nres */
1.218 brouard 9584: /* hBijx(p, bage, fage); */
9585: /* fclose(ficrespijb); */
9586:
9587: return 0;
1.217 brouard 9588: }
1.218 brouard 9589:
1.180 brouard 9590: int hPijx(double *p, int bage, int fage){
9591: /*------------- h Pij x at various ages ------------*/
9592:
9593: int stepsize;
9594: int agelim;
9595: int hstepm;
9596: int nhstepm;
1.235 brouard 9597: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9598:
9599: double agedeb;
9600: double ***p3mat;
9601:
1.201 brouard 9602: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9603: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9604: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9605: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9606: }
9607: printf("Computing pij: result on file '%s' \n", filerespij);
9608: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9609:
9610: stepsize=(int) (stepm+YEARM-1)/YEARM;
9611: /*if (stepm<=24) stepsize=2;*/
9612:
9613: agelim=AGESUP;
9614: hstepm=stepsize*YEARM; /* Every year of age */
9615: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9616:
1.180 brouard 9617: /* hstepm=1; aff par mois*/
9618: pstamp(ficrespij);
9619: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9620: i1= pow(2,cptcoveff);
1.218 brouard 9621: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9622: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9623: /* k=k+1; */
1.235 brouard 9624: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9625: for(k=1; k<=i1;k++){
1.253 brouard 9626: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9627: continue;
1.183 brouard 9628: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9629: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9630: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9631: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9632: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9633: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9634: }
1.183 brouard 9635: fprintf(ficrespij,"******\n");
9636:
9637: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9638: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9639: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9640:
9641: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9642:
1.183 brouard 9643: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9644: oldm=oldms;savm=savms;
1.235 brouard 9645: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9646: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9647: for(i=1; i<=nlstate;i++)
9648: for(j=1; j<=nlstate+ndeath;j++)
9649: fprintf(ficrespij," %1d-%1d",i,j);
9650: fprintf(ficrespij,"\n");
9651: for (h=0; h<=nhstepm; h++){
9652: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9653: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9654: for(i=1; i<=nlstate;i++)
9655: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9656: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9657: fprintf(ficrespij,"\n");
9658: }
1.183 brouard 9659: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9660: fprintf(ficrespij,"\n");
9661: }
1.180 brouard 9662: /*}*/
9663: }
1.218 brouard 9664: return 0;
1.180 brouard 9665: }
1.218 brouard 9666:
9667: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9668: /*------------- h Bij x at various ages ------------*/
9669:
9670: int stepsize;
1.218 brouard 9671: /* int agelim; */
9672: int ageminl;
1.217 brouard 9673: int hstepm;
9674: int nhstepm;
1.238 brouard 9675: int h, i, i1, j, k, nres;
1.218 brouard 9676:
1.217 brouard 9677: double agedeb;
9678: double ***p3mat;
1.218 brouard 9679:
9680: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9681: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9682: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9683: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9684: }
9685: printf("Computing pij back: result on file '%s' \n", filerespijb);
9686: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9687:
9688: stepsize=(int) (stepm+YEARM-1)/YEARM;
9689: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9690:
1.218 brouard 9691: /* agelim=AGESUP; */
9692: ageminl=30;
9693: hstepm=stepsize*YEARM; /* Every year of age */
9694: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9695:
9696: /* hstepm=1; aff par mois*/
9697: pstamp(ficrespijb);
1.255 brouard 9698: 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 9699: i1= pow(2,cptcoveff);
1.218 brouard 9700: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9701: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9702: /* k=k+1; */
1.238 brouard 9703: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9704: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9705: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9706: continue;
9707: fprintf(ficrespijb,"\n#****** ");
9708: for(j=1;j<=cptcoveff;j++)
9709: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9710: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9711: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9712: }
9713: fprintf(ficrespijb,"******\n");
9714: if(invalidvarcomb[k]){
9715: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9716: continue;
9717: }
9718:
9719: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9720: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9721: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9722: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9723: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9724:
9725: /* nhstepm=nhstepm*YEARM; aff par mois*/
9726:
9727: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9728: /* oldm=oldms;savm=savms; */
9729: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9730: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9731: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 9732: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 9733: for(i=1; i<=nlstate;i++)
9734: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9735: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9736: fprintf(ficrespijb,"\n");
1.238 brouard 9737: for (h=0; h<=nhstepm; h++){
9738: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9739: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9740: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9741: for(i=1; i<=nlstate;i++)
9742: for(j=1; j<=nlstate+ndeath;j++)
9743: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9744: fprintf(ficrespijb,"\n");
9745: }
9746: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9747: fprintf(ficrespijb,"\n");
9748: } /* end age deb */
9749: } /* end combination */
9750: } /* end nres */
1.218 brouard 9751: return 0;
9752: } /* hBijx */
1.217 brouard 9753:
1.180 brouard 9754:
1.136 brouard 9755: /***********************************************/
9756: /**************** Main Program *****************/
9757: /***********************************************/
9758:
9759: int main(int argc, char *argv[])
9760: {
9761: #ifdef GSL
9762: const gsl_multimin_fminimizer_type *T;
9763: size_t iteri = 0, it;
9764: int rval = GSL_CONTINUE;
9765: int status = GSL_SUCCESS;
9766: double ssval;
9767: #endif
9768: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9769: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9770: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9771: int jj, ll, li, lj, lk;
1.136 brouard 9772: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9773: int num_filled;
1.136 brouard 9774: int itimes;
9775: int NDIM=2;
9776: int vpopbased=0;
1.235 brouard 9777: int nres=0;
1.136 brouard 9778:
1.164 brouard 9779: char ca[32], cb[32];
1.136 brouard 9780: /* FILE *fichtm; *//* Html File */
9781: /* FILE *ficgp;*/ /*Gnuplot File */
9782: struct stat info;
1.191 brouard 9783: double agedeb=0.;
1.194 brouard 9784:
9785: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9786: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9787:
1.165 brouard 9788: double fret;
1.191 brouard 9789: double dum=0.; /* Dummy variable */
1.136 brouard 9790: double ***p3mat;
1.218 brouard 9791: /* double ***mobaverage; */
1.164 brouard 9792:
9793: char line[MAXLINE];
1.197 brouard 9794: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9795:
1.234 brouard 9796: char modeltemp[MAXLINE];
1.230 brouard 9797: char resultline[MAXLINE];
9798:
1.136 brouard 9799: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9800: char *tok, *val; /* pathtot */
1.136 brouard 9801: int firstobs=1, lastobs=10;
1.195 brouard 9802: int c, h , cpt, c2;
1.191 brouard 9803: int jl=0;
9804: int i1, j1, jk, stepsize=0;
1.194 brouard 9805: int count=0;
9806:
1.164 brouard 9807: int *tab;
1.136 brouard 9808: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9809: int backcast=0;
1.136 brouard 9810: int mobilav=0,popforecast=0;
1.191 brouard 9811: int hstepm=0, nhstepm=0;
1.136 brouard 9812: int agemortsup;
9813: float sumlpop=0.;
9814: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9815: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9816:
1.191 brouard 9817: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9818: double ftolpl=FTOL;
9819: double **prlim;
1.217 brouard 9820: double **bprlim;
1.136 brouard 9821: double ***param; /* Matrix of parameters */
1.251 brouard 9822: double ***paramstart; /* Matrix of starting parameter values */
9823: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9824: double **matcov; /* Matrix of covariance */
1.203 brouard 9825: double **hess; /* Hessian matrix */
1.136 brouard 9826: double ***delti3; /* Scale */
9827: double *delti; /* Scale */
9828: double ***eij, ***vareij;
9829: double **varpl; /* Variances of prevalence limits by age */
9830: double *epj, vepp;
1.164 brouard 9831:
1.136 brouard 9832: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9833: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9834:
1.136 brouard 9835: double **ximort;
1.145 brouard 9836: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9837: int *dcwave;
9838:
1.164 brouard 9839: char z[1]="c";
1.136 brouard 9840:
9841: /*char *strt;*/
9842: char strtend[80];
1.126 brouard 9843:
1.164 brouard 9844:
1.126 brouard 9845: /* setlocale (LC_ALL, ""); */
9846: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9847: /* textdomain (PACKAGE); */
9848: /* setlocale (LC_CTYPE, ""); */
9849: /* setlocale (LC_MESSAGES, ""); */
9850:
9851: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9852: rstart_time = time(NULL);
9853: /* (void) gettimeofday(&start_time,&tzp);*/
9854: start_time = *localtime(&rstart_time);
1.126 brouard 9855: curr_time=start_time;
1.157 brouard 9856: /*tml = *localtime(&start_time.tm_sec);*/
9857: /* strcpy(strstart,asctime(&tml)); */
9858: strcpy(strstart,asctime(&start_time));
1.126 brouard 9859:
9860: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9861: /* tp.tm_sec = tp.tm_sec +86400; */
9862: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9863: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9864: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9865: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9866: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9867: /* strt=asctime(&tmg); */
9868: /* printf("Time(after) =%s",strstart); */
9869: /* (void) time (&time_value);
9870: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9871: * tm = *localtime(&time_value);
9872: * strstart=asctime(&tm);
9873: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9874: */
9875:
9876: nberr=0; /* Number of errors and warnings */
9877: nbwarn=0;
1.184 brouard 9878: #ifdef WIN32
9879: _getcwd(pathcd, size);
9880: #else
1.126 brouard 9881: getcwd(pathcd, size);
1.184 brouard 9882: #endif
1.191 brouard 9883: syscompilerinfo(0);
1.196 brouard 9884: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9885: if(argc <=1){
9886: printf("\nEnter the parameter file name: ");
1.205 brouard 9887: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9888: printf("ERROR Empty parameter file name\n");
9889: goto end;
9890: }
1.126 brouard 9891: i=strlen(pathr);
9892: if(pathr[i-1]=='\n')
9893: pathr[i-1]='\0';
1.156 brouard 9894: i=strlen(pathr);
1.205 brouard 9895: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9896: pathr[i-1]='\0';
1.205 brouard 9897: }
9898: i=strlen(pathr);
9899: if( i==0 ){
9900: printf("ERROR Empty parameter file name\n");
9901: goto end;
9902: }
9903: for (tok = pathr; tok != NULL; ){
1.126 brouard 9904: printf("Pathr |%s|\n",pathr);
9905: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9906: printf("val= |%s| pathr=%s\n",val,pathr);
9907: strcpy (pathtot, val);
9908: if(pathr[0] == '\0') break; /* Dirty */
9909: }
9910: }
9911: else{
9912: strcpy(pathtot,argv[1]);
9913: }
9914: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9915: /*cygwin_split_path(pathtot,path,optionfile);
9916: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9917: /* cutv(path,optionfile,pathtot,'\\');*/
9918:
9919: /* Split argv[0], imach program to get pathimach */
9920: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9921: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9922: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9923: /* strcpy(pathimach,argv[0]); */
9924: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9925: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9926: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9927: #ifdef WIN32
9928: _chdir(path); /* Can be a relative path */
9929: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9930: #else
1.126 brouard 9931: chdir(path); /* Can be a relative path */
1.184 brouard 9932: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9933: #endif
9934: printf("Current directory %s!\n",pathcd);
1.126 brouard 9935: strcpy(command,"mkdir ");
9936: strcat(command,optionfilefiname);
9937: if((outcmd=system(command)) != 0){
1.169 brouard 9938: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9939: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9940: /* fclose(ficlog); */
9941: /* exit(1); */
9942: }
9943: /* if((imk=mkdir(optionfilefiname))<0){ */
9944: /* perror("mkdir"); */
9945: /* } */
9946:
9947: /*-------- arguments in the command line --------*/
9948:
1.186 brouard 9949: /* Main Log file */
1.126 brouard 9950: strcat(filelog, optionfilefiname);
9951: strcat(filelog,".log"); /* */
9952: if((ficlog=fopen(filelog,"w"))==NULL) {
9953: printf("Problem with logfile %s\n",filelog);
9954: goto end;
9955: }
9956: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9957: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9958: fprintf(ficlog,"\nEnter the parameter file name: \n");
9959: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9960: path=%s \n\
9961: optionfile=%s\n\
9962: optionfilext=%s\n\
1.156 brouard 9963: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9964:
1.197 brouard 9965: syscompilerinfo(1);
1.167 brouard 9966:
1.126 brouard 9967: printf("Local time (at start):%s",strstart);
9968: fprintf(ficlog,"Local time (at start): %s",strstart);
9969: fflush(ficlog);
9970: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9971: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9972:
9973: /* */
9974: strcpy(fileres,"r");
9975: strcat(fileres, optionfilefiname);
1.201 brouard 9976: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9977: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9978: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9979:
1.186 brouard 9980: /* Main ---------arguments file --------*/
1.126 brouard 9981:
9982: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9983: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9984: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9985: fflush(ficlog);
1.149 brouard 9986: /* goto end; */
9987: exit(70);
1.126 brouard 9988: }
9989:
9990:
9991:
9992: strcpy(filereso,"o");
1.201 brouard 9993: strcat(filereso,fileresu);
1.126 brouard 9994: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9995: printf("Problem with Output resultfile: %s\n", filereso);
9996: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9997: fflush(ficlog);
9998: goto end;
9999: }
10000:
10001: /* Reads comments: lines beginning with '#' */
10002: numlinepar=0;
1.197 brouard 10003:
10004: /* First parameter line */
10005: while(fgets(line, MAXLINE, ficpar)) {
10006: /* If line starts with a # it is a comment */
10007: if (line[0] == '#') {
10008: numlinepar++;
10009: fputs(line,stdout);
10010: fputs(line,ficparo);
10011: fputs(line,ficlog);
10012: continue;
10013: }else
10014: break;
10015: }
10016: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10017: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10018: if (num_filled != 5) {
10019: printf("Should be 5 parameters\n");
10020: }
1.126 brouard 10021: numlinepar++;
1.197 brouard 10022: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10023: }
10024: /* Second parameter line */
10025: while(fgets(line, MAXLINE, ficpar)) {
10026: /* If line starts with a # it is a comment */
10027: if (line[0] == '#') {
10028: numlinepar++;
10029: fputs(line,stdout);
10030: fputs(line,ficparo);
10031: fputs(line,ficlog);
10032: continue;
10033: }else
10034: break;
10035: }
1.223 brouard 10036: 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", \
10037: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10038: if (num_filled != 11) {
10039: 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 10040: printf("but line=%s\n",line);
1.197 brouard 10041: }
1.223 brouard 10042: 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 10043: }
1.203 brouard 10044: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10045: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10046: /* Third parameter line */
10047: while(fgets(line, MAXLINE, ficpar)) {
10048: /* If line starts with a # it is a comment */
10049: if (line[0] == '#') {
10050: numlinepar++;
10051: fputs(line,stdout);
10052: fputs(line,ficparo);
10053: fputs(line,ficlog);
10054: continue;
10055: }else
10056: break;
10057: }
1.201 brouard 10058: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
10059: if (num_filled == 0)
10060: model[0]='\0';
10061: else if (num_filled != 1){
1.197 brouard 10062: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10063: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10064: model[0]='\0';
10065: goto end;
10066: }
10067: else{
10068: if (model[0]=='+'){
10069: for(i=1; i<=strlen(model);i++)
10070: modeltemp[i-1]=model[i];
1.201 brouard 10071: strcpy(model,modeltemp);
1.197 brouard 10072: }
10073: }
1.199 brouard 10074: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10075: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10076: }
10077: /* 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); */
10078: /* numlinepar=numlinepar+3; /\* In general *\/ */
10079: /* 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 10080: 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);
10081: 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 10082: fflush(ficlog);
1.190 brouard 10083: /* if(model[0]=='#'|| model[0]== '\0'){ */
10084: if(model[0]=='#'){
1.187 brouard 10085: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10086: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10087: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10088: if(mle != -1){
10089: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10090: exit(1);
10091: }
10092: }
1.126 brouard 10093: while((c=getc(ficpar))=='#' && c!= EOF){
10094: ungetc(c,ficpar);
10095: fgets(line, MAXLINE, ficpar);
10096: numlinepar++;
1.195 brouard 10097: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10098: z[0]=line[1];
10099: }
10100: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10101: fputs(line, stdout);
10102: //puts(line);
1.126 brouard 10103: fputs(line,ficparo);
10104: fputs(line,ficlog);
10105: }
10106: ungetc(c,ficpar);
10107:
10108:
1.145 brouard 10109: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10110: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10111: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10112: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10113: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10114: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10115: v1+v2*age+v2*v3 makes cptcovn = 3
10116: */
10117: if (strlen(model)>1)
1.187 brouard 10118: 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 10119: else
1.187 brouard 10120: ncovmodel=2; /* Constant and age */
1.133 brouard 10121: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10122: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10123: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10124: 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);
10125: 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);
10126: fflush(stdout);
10127: fclose (ficlog);
10128: goto end;
10129: }
1.126 brouard 10130: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10131: delti=delti3[1][1];
10132: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10133: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10134: /* We could also provide initial parameters values giving by simple logistic regression
10135: * only one way, that is without matrix product. We will have nlstate maximizations */
10136: /* for(i=1;i<nlstate;i++){ */
10137: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10138: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10139: /* } */
1.126 brouard 10140: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10141: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10142: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10143: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10144: fclose (ficparo);
10145: fclose (ficlog);
10146: goto end;
10147: exit(0);
1.220 brouard 10148: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10149: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10150: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10151: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10152: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10153: matcov=matrix(1,npar,1,npar);
1.203 brouard 10154: hess=matrix(1,npar,1,npar);
1.220 brouard 10155: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10156: /* Read guessed parameters */
1.126 brouard 10157: /* Reads comments: lines beginning with '#' */
10158: while((c=getc(ficpar))=='#' && c!= EOF){
10159: ungetc(c,ficpar);
10160: fgets(line, MAXLINE, ficpar);
10161: numlinepar++;
1.141 brouard 10162: fputs(line,stdout);
1.126 brouard 10163: fputs(line,ficparo);
10164: fputs(line,ficlog);
10165: }
10166: ungetc(c,ficpar);
10167:
10168: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10169: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10170: for(i=1; i <=nlstate; i++){
1.234 brouard 10171: j=0;
1.126 brouard 10172: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10173: if(jj==i) continue;
10174: j++;
10175: fscanf(ficpar,"%1d%1d",&i1,&j1);
10176: if ((i1 != i) || (j1 != jj)){
10177: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10178: It might be a problem of design; if ncovcol and the model are correct\n \
10179: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10180: exit(1);
10181: }
10182: fprintf(ficparo,"%1d%1d",i1,j1);
10183: if(mle==1)
10184: printf("%1d%1d",i,jj);
10185: fprintf(ficlog,"%1d%1d",i,jj);
10186: for(k=1; k<=ncovmodel;k++){
10187: fscanf(ficpar," %lf",¶m[i][j][k]);
10188: if(mle==1){
10189: printf(" %lf",param[i][j][k]);
10190: fprintf(ficlog," %lf",param[i][j][k]);
10191: }
10192: else
10193: fprintf(ficlog," %lf",param[i][j][k]);
10194: fprintf(ficparo," %lf",param[i][j][k]);
10195: }
10196: fscanf(ficpar,"\n");
10197: numlinepar++;
10198: if(mle==1)
10199: printf("\n");
10200: fprintf(ficlog,"\n");
10201: fprintf(ficparo,"\n");
1.126 brouard 10202: }
10203: }
10204: fflush(ficlog);
1.234 brouard 10205:
1.251 brouard 10206: /* Reads parameters values */
1.126 brouard 10207: p=param[1][1];
1.251 brouard 10208: pstart=paramstart[1][1];
1.126 brouard 10209:
10210: /* Reads comments: lines beginning with '#' */
10211: while((c=getc(ficpar))=='#' && c!= EOF){
10212: ungetc(c,ficpar);
10213: fgets(line, MAXLINE, ficpar);
10214: numlinepar++;
1.141 brouard 10215: fputs(line,stdout);
1.126 brouard 10216: fputs(line,ficparo);
10217: fputs(line,ficlog);
10218: }
10219: ungetc(c,ficpar);
10220:
10221: for(i=1; i <=nlstate; i++){
10222: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10223: fscanf(ficpar,"%1d%1d",&i1,&j1);
10224: if ( (i1-i) * (j1-j) != 0){
10225: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10226: exit(1);
10227: }
10228: printf("%1d%1d",i,j);
10229: fprintf(ficparo,"%1d%1d",i1,j1);
10230: fprintf(ficlog,"%1d%1d",i1,j1);
10231: for(k=1; k<=ncovmodel;k++){
10232: fscanf(ficpar,"%le",&delti3[i][j][k]);
10233: printf(" %le",delti3[i][j][k]);
10234: fprintf(ficparo," %le",delti3[i][j][k]);
10235: fprintf(ficlog," %le",delti3[i][j][k]);
10236: }
10237: fscanf(ficpar,"\n");
10238: numlinepar++;
10239: printf("\n");
10240: fprintf(ficparo,"\n");
10241: fprintf(ficlog,"\n");
1.126 brouard 10242: }
10243: }
10244: fflush(ficlog);
1.234 brouard 10245:
1.145 brouard 10246: /* Reads covariance matrix */
1.126 brouard 10247: delti=delti3[1][1];
1.220 brouard 10248:
10249:
1.126 brouard 10250: /* 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 10251:
1.126 brouard 10252: /* Reads comments: lines beginning with '#' */
10253: while((c=getc(ficpar))=='#' && c!= EOF){
10254: ungetc(c,ficpar);
10255: fgets(line, MAXLINE, ficpar);
10256: numlinepar++;
1.141 brouard 10257: fputs(line,stdout);
1.126 brouard 10258: fputs(line,ficparo);
10259: fputs(line,ficlog);
10260: }
10261: ungetc(c,ficpar);
1.220 brouard 10262:
1.126 brouard 10263: matcov=matrix(1,npar,1,npar);
1.203 brouard 10264: hess=matrix(1,npar,1,npar);
1.131 brouard 10265: for(i=1; i <=npar; i++)
10266: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10267:
1.194 brouard 10268: /* Scans npar lines */
1.126 brouard 10269: for(i=1; i <=npar; i++){
1.226 brouard 10270: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10271: if(count != 3){
1.226 brouard 10272: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10273: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10274: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10275: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10276: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10277: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10278: exit(1);
1.220 brouard 10279: }else{
1.226 brouard 10280: if(mle==1)
10281: printf("%1d%1d%d",i1,j1,jk);
10282: }
10283: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10284: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10285: for(j=1; j <=i; j++){
1.226 brouard 10286: fscanf(ficpar," %le",&matcov[i][j]);
10287: if(mle==1){
10288: printf(" %.5le",matcov[i][j]);
10289: }
10290: fprintf(ficlog," %.5le",matcov[i][j]);
10291: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10292: }
10293: fscanf(ficpar,"\n");
10294: numlinepar++;
10295: if(mle==1)
1.220 brouard 10296: printf("\n");
1.126 brouard 10297: fprintf(ficlog,"\n");
10298: fprintf(ficparo,"\n");
10299: }
1.194 brouard 10300: /* End of read covariance matrix npar lines */
1.126 brouard 10301: for(i=1; i <=npar; i++)
10302: for(j=i+1;j<=npar;j++)
1.226 brouard 10303: matcov[i][j]=matcov[j][i];
1.126 brouard 10304:
10305: if(mle==1)
10306: printf("\n");
10307: fprintf(ficlog,"\n");
10308:
10309: fflush(ficlog);
10310:
10311: /*-------- Rewriting parameter file ----------*/
10312: strcpy(rfileres,"r"); /* "Rparameterfile */
10313: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10314: strcat(rfileres,"."); /* */
10315: strcat(rfileres,optionfilext); /* Other files have txt extension */
10316: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10317: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10318: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10319: }
10320: fprintf(ficres,"#%s\n",version);
10321: } /* End of mle != -3 */
1.218 brouard 10322:
1.186 brouard 10323: /* Main data
10324: */
1.126 brouard 10325: n= lastobs;
10326: num=lvector(1,n);
10327: moisnais=vector(1,n);
10328: annais=vector(1,n);
10329: moisdc=vector(1,n);
10330: andc=vector(1,n);
1.220 brouard 10331: weight=vector(1,n);
1.126 brouard 10332: agedc=vector(1,n);
10333: cod=ivector(1,n);
1.220 brouard 10334: for(i=1;i<=n;i++){
1.234 brouard 10335: num[i]=0;
10336: moisnais[i]=0;
10337: annais[i]=0;
10338: moisdc[i]=0;
10339: andc[i]=0;
10340: agedc[i]=0;
10341: cod[i]=0;
10342: weight[i]=1.0; /* Equal weights, 1 by default */
10343: }
1.126 brouard 10344: mint=matrix(1,maxwav,1,n);
10345: anint=matrix(1,maxwav,1,n);
1.131 brouard 10346: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10347: tab=ivector(1,NCOVMAX);
1.144 brouard 10348: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10349: 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 10350:
1.136 brouard 10351: /* Reads data from file datafile */
10352: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10353: goto end;
10354:
10355: /* Calculation of the number of parameters from char model */
1.234 brouard 10356: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10357: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10358: k=3 V4 Tvar[k=3]= 4 (from V4)
10359: k=2 V1 Tvar[k=2]= 1 (from V1)
10360: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10361: */
10362:
10363: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10364: TvarsDind=ivector(1,NCOVMAX); /* */
10365: TvarsD=ivector(1,NCOVMAX); /* */
10366: TvarsQind=ivector(1,NCOVMAX); /* */
10367: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10368: TvarF=ivector(1,NCOVMAX); /* */
10369: TvarFind=ivector(1,NCOVMAX); /* */
10370: TvarV=ivector(1,NCOVMAX); /* */
10371: TvarVind=ivector(1,NCOVMAX); /* */
10372: TvarA=ivector(1,NCOVMAX); /* */
10373: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10374: TvarFD=ivector(1,NCOVMAX); /* */
10375: TvarFDind=ivector(1,NCOVMAX); /* */
10376: TvarFQ=ivector(1,NCOVMAX); /* */
10377: TvarFQind=ivector(1,NCOVMAX); /* */
10378: TvarVD=ivector(1,NCOVMAX); /* */
10379: TvarVDind=ivector(1,NCOVMAX); /* */
10380: TvarVQ=ivector(1,NCOVMAX); /* */
10381: TvarVQind=ivector(1,NCOVMAX); /* */
10382:
1.230 brouard 10383: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10384: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10385: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10386: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10387: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10388: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10389: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10390: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10391: */
10392: /* For model-covariate k tells which data-covariate to use but
10393: because this model-covariate is a construction we invent a new column
10394: ncovcol + k1
10395: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10396: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10397: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10398: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10399: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10400: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10401: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10402: */
1.145 brouard 10403: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10404: 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 10405: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10406: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10407: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10408: 4 covariates (3 plus signs)
10409: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10410: */
1.230 brouard 10411: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10412: * individual dummy, fixed or varying:
10413: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10414: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10415: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10416: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10417: * Tmodelind[1]@9={9,0,3,2,}*/
10418: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10419: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10420: * individual quantitative, fixed or varying:
10421: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10422: * 3, 1, 0, 0, 0, 0, 0, 0},
10423: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10424: /* Main decodemodel */
10425:
1.187 brouard 10426:
1.223 brouard 10427: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10428: goto end;
10429:
1.137 brouard 10430: if((double)(lastobs-imx)/(double)imx > 1.10){
10431: nbwarn++;
10432: 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);
10433: 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);
10434: }
1.136 brouard 10435: /* if(mle==1){*/
1.137 brouard 10436: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10437: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10438: }
10439:
10440: /*-calculation of age at interview from date of interview and age at death -*/
10441: agev=matrix(1,maxwav,1,imx);
10442:
10443: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10444: goto end;
10445:
1.126 brouard 10446:
1.136 brouard 10447: agegomp=(int)agemin;
10448: free_vector(moisnais,1,n);
10449: free_vector(annais,1,n);
1.126 brouard 10450: /* free_matrix(mint,1,maxwav,1,n);
10451: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10452: /* free_vector(moisdc,1,n); */
10453: /* free_vector(andc,1,n); */
1.145 brouard 10454: /* */
10455:
1.126 brouard 10456: wav=ivector(1,imx);
1.214 brouard 10457: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10458: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10459: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10460: 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.*/
10461: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10462: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10463:
10464: /* Concatenates waves */
1.214 brouard 10465: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10466: Death is a valid wave (if date is known).
10467: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10468: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10469: and mw[mi+1][i]. dh depends on stepm.
10470: */
10471:
1.126 brouard 10472: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10473: /* Concatenates waves */
1.145 brouard 10474:
1.215 brouard 10475: free_vector(moisdc,1,n);
10476: free_vector(andc,1,n);
10477:
1.126 brouard 10478: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10479: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10480: ncodemax[1]=1;
1.145 brouard 10481: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10482: cptcoveff=0;
1.220 brouard 10483: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10484: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10485: }
10486:
10487: ncovcombmax=pow(2,cptcoveff);
10488: invalidvarcomb=ivector(1, ncovcombmax);
10489: for(i=1;i<ncovcombmax;i++)
10490: invalidvarcomb[i]=0;
10491:
1.211 brouard 10492: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10493: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10494: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10495:
1.200 brouard 10496: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10497: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10498: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10499: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10500: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10501: * (currently 0 or 1) in the data.
10502: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10503: * corresponding modality (h,j).
10504: */
10505:
1.145 brouard 10506: h=0;
10507: /*if (cptcovn > 0) */
1.126 brouard 10508: m=pow(2,cptcoveff);
10509:
1.144 brouard 10510: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10511: * For k=4 covariates, h goes from 1 to m=2**k
10512: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10513: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10514: * h\k 1 2 3 4
1.143 brouard 10515: *______________________________
10516: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10517: * 2 2 1 1 1
10518: * 3 i=2 1 2 1 1
10519: * 4 2 2 1 1
10520: * 5 i=3 1 i=2 1 2 1
10521: * 6 2 1 2 1
10522: * 7 i=4 1 2 2 1
10523: * 8 2 2 2 1
1.197 brouard 10524: * 9 i=5 1 i=3 1 i=2 1 2
10525: * 10 2 1 1 2
10526: * 11 i=6 1 2 1 2
10527: * 12 2 2 1 2
10528: * 13 i=7 1 i=4 1 2 2
10529: * 14 2 1 2 2
10530: * 15 i=8 1 2 2 2
10531: * 16 2 2 2 2
1.143 brouard 10532: */
1.212 brouard 10533: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10534: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10535: * and the value of each covariate?
10536: * V1=1, V2=1, V3=2, V4=1 ?
10537: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10538: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10539: * In order to get the real value in the data, we use nbcode
10540: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10541: * We are keeping this crazy system in order to be able (in the future?)
10542: * to have more than 2 values (0 or 1) for a covariate.
10543: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10544: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10545: * bbbbbbbb
10546: * 76543210
10547: * h-1 00000101 (6-1=5)
1.219 brouard 10548: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10549: * &
10550: * 1 00000001 (1)
1.219 brouard 10551: * 00000000 = 1 & ((h-1) >> (k-1))
10552: * +1= 00000001 =1
1.211 brouard 10553: *
10554: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10555: * h' 1101 =2^3+2^2+0x2^1+2^0
10556: * >>k' 11
10557: * & 00000001
10558: * = 00000001
10559: * +1 = 00000010=2 = codtabm(14,3)
10560: * Reverse h=6 and m=16?
10561: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10562: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10563: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10564: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10565: * V3=decodtabm(14,3,2**4)=2
10566: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10567: *(h-1) >> (j-1) 0011 =13 >> 2
10568: * &1 000000001
10569: * = 000000001
10570: * +1= 000000010 =2
10571: * 2211
10572: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10573: * V3=2
1.220 brouard 10574: * codtabm and decodtabm are identical
1.211 brouard 10575: */
10576:
1.145 brouard 10577:
10578: free_ivector(Ndum,-1,NCOVMAX);
10579:
10580:
1.126 brouard 10581:
1.186 brouard 10582: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10583: strcpy(optionfilegnuplot,optionfilefiname);
10584: if(mle==-3)
1.201 brouard 10585: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10586: strcat(optionfilegnuplot,".gp");
10587:
10588: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10589: printf("Problem with file %s",optionfilegnuplot);
10590: }
10591: else{
1.204 brouard 10592: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10593: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10594: //fprintf(ficgp,"set missing 'NaNq'\n");
10595: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10596: }
10597: /* fclose(ficgp);*/
1.186 brouard 10598:
10599:
10600: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10601:
10602: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10603: if(mle==-3)
1.201 brouard 10604: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10605: strcat(optionfilehtm,".htm");
10606: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10607: printf("Problem with %s \n",optionfilehtm);
10608: exit(0);
1.126 brouard 10609: }
10610:
10611: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10612: strcat(optionfilehtmcov,"-cov.htm");
10613: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10614: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10615: }
10616: else{
10617: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10618: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10619: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10620: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10621: }
10622:
1.213 brouard 10623: 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 10624: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10625: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10626: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10627: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10628: \n\
10629: <hr size=\"2\" color=\"#EC5E5E\">\
10630: <ul><li><h4>Parameter files</h4>\n\
10631: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10632: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10633: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10634: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10635: - Date and time at start: %s</ul>\n",\
10636: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10637: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10638: fileres,fileres,\
10639: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10640: fflush(fichtm);
10641:
10642: strcpy(pathr,path);
10643: strcat(pathr,optionfilefiname);
1.184 brouard 10644: #ifdef WIN32
10645: _chdir(optionfilefiname); /* Move to directory named optionfile */
10646: #else
1.126 brouard 10647: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10648: #endif
10649:
1.126 brouard 10650:
1.220 brouard 10651: /* Calculates basic frequencies. Computes observed prevalence at single age
10652: and for any valid combination of covariates
1.126 brouard 10653: and prints on file fileres'p'. */
1.251 brouard 10654: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10655: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10656:
10657: fprintf(fichtm,"\n");
10658: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10659: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10660: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10661: imx,agemin,agemax,jmin,jmax,jmean);
10662: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10663: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10664: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10665: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10666: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10667:
1.126 brouard 10668: /* For Powell, parameters are in a vector p[] starting at p[1]
10669: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10670: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10671:
10672: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10673: /* For mortality only */
1.126 brouard 10674: if (mle==-3){
1.136 brouard 10675: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10676: for(i=1;i<=NDIM;i++)
10677: for(j=1;j<=NDIM;j++)
10678: ximort[i][j]=0.;
1.186 brouard 10679: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10680: cens=ivector(1,n);
10681: ageexmed=vector(1,n);
10682: agecens=vector(1,n);
10683: dcwave=ivector(1,n);
1.223 brouard 10684:
1.126 brouard 10685: for (i=1; i<=imx; i++){
10686: dcwave[i]=-1;
10687: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10688: if (s[m][i]>nlstate) {
10689: dcwave[i]=m;
10690: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10691: break;
10692: }
1.126 brouard 10693: }
1.226 brouard 10694:
1.126 brouard 10695: for (i=1; i<=imx; i++) {
10696: if (wav[i]>0){
1.226 brouard 10697: ageexmed[i]=agev[mw[1][i]][i];
10698: j=wav[i];
10699: agecens[i]=1.;
10700:
10701: if (ageexmed[i]> 1 && wav[i] > 0){
10702: agecens[i]=agev[mw[j][i]][i];
10703: cens[i]= 1;
10704: }else if (ageexmed[i]< 1)
10705: cens[i]= -1;
10706: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10707: cens[i]=0 ;
1.126 brouard 10708: }
10709: else cens[i]=-1;
10710: }
10711:
10712: for (i=1;i<=NDIM;i++) {
10713: for (j=1;j<=NDIM;j++)
1.226 brouard 10714: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10715: }
10716:
1.145 brouard 10717: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10718: /*printf("%lf %lf", p[1], p[2]);*/
10719:
10720:
1.136 brouard 10721: #ifdef GSL
10722: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10723: #else
1.126 brouard 10724: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10725: #endif
1.201 brouard 10726: strcpy(filerespow,"POW-MORT_");
10727: strcat(filerespow,fileresu);
1.126 brouard 10728: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10729: printf("Problem with resultfile: %s\n", filerespow);
10730: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10731: }
1.136 brouard 10732: #ifdef GSL
10733: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10734: #else
1.126 brouard 10735: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10736: #endif
1.126 brouard 10737: /* for (i=1;i<=nlstate;i++)
10738: for(j=1;j<=nlstate+ndeath;j++)
10739: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10740: */
10741: fprintf(ficrespow,"\n");
1.136 brouard 10742: #ifdef GSL
10743: /* gsl starts here */
10744: T = gsl_multimin_fminimizer_nmsimplex;
10745: gsl_multimin_fminimizer *sfm = NULL;
10746: gsl_vector *ss, *x;
10747: gsl_multimin_function minex_func;
10748:
10749: /* Initial vertex size vector */
10750: ss = gsl_vector_alloc (NDIM);
10751:
10752: if (ss == NULL){
10753: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10754: }
10755: /* Set all step sizes to 1 */
10756: gsl_vector_set_all (ss, 0.001);
10757:
10758: /* Starting point */
1.126 brouard 10759:
1.136 brouard 10760: x = gsl_vector_alloc (NDIM);
10761:
10762: if (x == NULL){
10763: gsl_vector_free(ss);
10764: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10765: }
10766:
10767: /* Initialize method and iterate */
10768: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10769: /* gsl_vector_set(x, 0, 0.0268); */
10770: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10771: gsl_vector_set(x, 0, p[1]);
10772: gsl_vector_set(x, 1, p[2]);
10773:
10774: minex_func.f = &gompertz_f;
10775: minex_func.n = NDIM;
10776: minex_func.params = (void *)&p; /* ??? */
10777:
10778: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10779: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10780:
10781: printf("Iterations beginning .....\n\n");
10782: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10783:
10784: iteri=0;
10785: while (rval == GSL_CONTINUE){
10786: iteri++;
10787: status = gsl_multimin_fminimizer_iterate(sfm);
10788:
10789: if (status) printf("error: %s\n", gsl_strerror (status));
10790: fflush(0);
10791:
10792: if (status)
10793: break;
10794:
10795: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10796: ssval = gsl_multimin_fminimizer_size (sfm);
10797:
10798: if (rval == GSL_SUCCESS)
10799: printf ("converged to a local maximum at\n");
10800:
10801: printf("%5d ", iteri);
10802: for (it = 0; it < NDIM; it++){
10803: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10804: }
10805: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10806: }
10807:
10808: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10809:
10810: gsl_vector_free(x); /* initial values */
10811: gsl_vector_free(ss); /* inital step size */
10812: for (it=0; it<NDIM; it++){
10813: p[it+1]=gsl_vector_get(sfm->x,it);
10814: fprintf(ficrespow," %.12lf", p[it]);
10815: }
10816: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10817: #endif
10818: #ifdef POWELL
10819: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10820: #endif
1.126 brouard 10821: fclose(ficrespow);
10822:
1.203 brouard 10823: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10824:
10825: for(i=1; i <=NDIM; i++)
10826: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10827: matcov[i][j]=matcov[j][i];
1.126 brouard 10828:
10829: printf("\nCovariance matrix\n ");
1.203 brouard 10830: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10831: for(i=1; i <=NDIM; i++) {
10832: for(j=1;j<=NDIM;j++){
1.220 brouard 10833: printf("%f ",matcov[i][j]);
10834: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10835: }
1.203 brouard 10836: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10837: }
10838:
10839: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10840: for (i=1;i<=NDIM;i++) {
1.126 brouard 10841: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10842: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10843: }
1.126 brouard 10844: lsurv=vector(1,AGESUP);
10845: lpop=vector(1,AGESUP);
10846: tpop=vector(1,AGESUP);
10847: lsurv[agegomp]=100000;
10848:
10849: for (k=agegomp;k<=AGESUP;k++) {
10850: agemortsup=k;
10851: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10852: }
10853:
10854: for (k=agegomp;k<agemortsup;k++)
10855: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10856:
10857: for (k=agegomp;k<agemortsup;k++){
10858: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10859: sumlpop=sumlpop+lpop[k];
10860: }
10861:
10862: tpop[agegomp]=sumlpop;
10863: for (k=agegomp;k<(agemortsup-3);k++){
10864: /* tpop[k+1]=2;*/
10865: tpop[k+1]=tpop[k]-lpop[k];
10866: }
10867:
10868:
10869: printf("\nAge lx qx dx Lx Tx e(x)\n");
10870: for (k=agegomp;k<(agemortsup-2);k++)
10871: 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]);
10872:
10873:
10874: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10875: ageminpar=50;
10876: agemaxpar=100;
1.194 brouard 10877: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10878: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10879: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10880: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10881: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10882: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10883: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10884: }else{
10885: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10886: 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 10887: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10888: }
1.201 brouard 10889: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10890: stepm, weightopt,\
10891: model,imx,p,matcov,agemortsup);
10892:
10893: free_vector(lsurv,1,AGESUP);
10894: free_vector(lpop,1,AGESUP);
10895: free_vector(tpop,1,AGESUP);
1.220 brouard 10896: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10897: free_ivector(cens,1,n);
10898: free_vector(agecens,1,n);
10899: free_ivector(dcwave,1,n);
1.220 brouard 10900: #ifdef GSL
1.136 brouard 10901: #endif
1.186 brouard 10902: } /* Endof if mle==-3 mortality only */
1.205 brouard 10903: /* Standard */
10904: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10905: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10906: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10907: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10908: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10909: for (k=1; k<=npar;k++)
10910: printf(" %d %8.5f",k,p[k]);
10911: printf("\n");
1.205 brouard 10912: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10913: /* mlikeli uses func not funcone */
1.247 brouard 10914: /* for(i=1;i<nlstate;i++){ */
10915: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10916: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10917: /* } */
1.205 brouard 10918: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10919: }
10920: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10921: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10922: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10923: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10924: }
10925: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10926: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10927: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10928: for (k=1; k<=npar;k++)
10929: printf(" %d %8.5f",k,p[k]);
10930: printf("\n");
10931:
10932: /*--------- results files --------------*/
1.224 brouard 10933: 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 10934:
10935:
10936: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10937: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10938: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10939: for(i=1,jk=1; i <=nlstate; i++){
10940: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10941: if (k != i) {
10942: printf("%d%d ",i,k);
10943: fprintf(ficlog,"%d%d ",i,k);
10944: fprintf(ficres,"%1d%1d ",i,k);
10945: for(j=1; j <=ncovmodel; j++){
10946: printf("%12.7f ",p[jk]);
10947: fprintf(ficlog,"%12.7f ",p[jk]);
10948: fprintf(ficres,"%12.7f ",p[jk]);
10949: jk++;
10950: }
10951: printf("\n");
10952: fprintf(ficlog,"\n");
10953: fprintf(ficres,"\n");
10954: }
1.126 brouard 10955: }
10956: }
1.203 brouard 10957: if(mle != 0){
10958: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10959: ftolhess=ftol; /* Usually correct */
1.203 brouard 10960: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10961: 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");
10962: 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");
10963: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10964: for(k=1; k <=(nlstate+ndeath); k++){
10965: if (k != i) {
10966: printf("%d%d ",i,k);
10967: fprintf(ficlog,"%d%d ",i,k);
10968: for(j=1; j <=ncovmodel; j++){
10969: 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]));
10970: 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]));
10971: jk++;
10972: }
10973: printf("\n");
10974: fprintf(ficlog,"\n");
10975: }
10976: }
1.193 brouard 10977: }
1.203 brouard 10978: } /* end of hesscov and Wald tests */
1.225 brouard 10979:
1.203 brouard 10980: /* */
1.126 brouard 10981: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10982: printf("# Scales (for hessian or gradient estimation)\n");
10983: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10984: for(i=1,jk=1; i <=nlstate; i++){
10985: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10986: if (j!=i) {
10987: fprintf(ficres,"%1d%1d",i,j);
10988: printf("%1d%1d",i,j);
10989: fprintf(ficlog,"%1d%1d",i,j);
10990: for(k=1; k<=ncovmodel;k++){
10991: printf(" %.5e",delti[jk]);
10992: fprintf(ficlog," %.5e",delti[jk]);
10993: fprintf(ficres," %.5e",delti[jk]);
10994: jk++;
10995: }
10996: printf("\n");
10997: fprintf(ficlog,"\n");
10998: fprintf(ficres,"\n");
10999: }
1.126 brouard 11000: }
11001: }
11002:
11003: 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 11004: if(mle >= 1) /* To big for the screen */
1.126 brouard 11005: 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");
11006: 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");
11007: /* # 121 Var(a12)\n\ */
11008: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11009: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11010: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11011: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11012: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11013: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11014: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11015:
11016:
11017: /* Just to have a covariance matrix which will be more understandable
11018: even is we still don't want to manage dictionary of variables
11019: */
11020: for(itimes=1;itimes<=2;itimes++){
11021: jj=0;
11022: for(i=1; i <=nlstate; i++){
1.225 brouard 11023: for(j=1; j <=nlstate+ndeath; j++){
11024: if(j==i) continue;
11025: for(k=1; k<=ncovmodel;k++){
11026: jj++;
11027: ca[0]= k+'a'-1;ca[1]='\0';
11028: if(itimes==1){
11029: if(mle>=1)
11030: printf("#%1d%1d%d",i,j,k);
11031: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11032: fprintf(ficres,"#%1d%1d%d",i,j,k);
11033: }else{
11034: if(mle>=1)
11035: printf("%1d%1d%d",i,j,k);
11036: fprintf(ficlog,"%1d%1d%d",i,j,k);
11037: fprintf(ficres,"%1d%1d%d",i,j,k);
11038: }
11039: ll=0;
11040: for(li=1;li <=nlstate; li++){
11041: for(lj=1;lj <=nlstate+ndeath; lj++){
11042: if(lj==li) continue;
11043: for(lk=1;lk<=ncovmodel;lk++){
11044: ll++;
11045: if(ll<=jj){
11046: cb[0]= lk +'a'-1;cb[1]='\0';
11047: if(ll<jj){
11048: if(itimes==1){
11049: if(mle>=1)
11050: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11051: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11052: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11053: }else{
11054: if(mle>=1)
11055: printf(" %.5e",matcov[jj][ll]);
11056: fprintf(ficlog," %.5e",matcov[jj][ll]);
11057: fprintf(ficres," %.5e",matcov[jj][ll]);
11058: }
11059: }else{
11060: if(itimes==1){
11061: if(mle>=1)
11062: printf(" Var(%s%1d%1d)",ca,i,j);
11063: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11064: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11065: }else{
11066: if(mle>=1)
11067: printf(" %.7e",matcov[jj][ll]);
11068: fprintf(ficlog," %.7e",matcov[jj][ll]);
11069: fprintf(ficres," %.7e",matcov[jj][ll]);
11070: }
11071: }
11072: }
11073: } /* end lk */
11074: } /* end lj */
11075: } /* end li */
11076: if(mle>=1)
11077: printf("\n");
11078: fprintf(ficlog,"\n");
11079: fprintf(ficres,"\n");
11080: numlinepar++;
11081: } /* end k*/
11082: } /*end j */
1.126 brouard 11083: } /* end i */
11084: } /* end itimes */
11085:
11086: fflush(ficlog);
11087: fflush(ficres);
1.225 brouard 11088: while(fgets(line, MAXLINE, ficpar)) {
11089: /* If line starts with a # it is a comment */
11090: if (line[0] == '#') {
11091: numlinepar++;
11092: fputs(line,stdout);
11093: fputs(line,ficparo);
11094: fputs(line,ficlog);
11095: continue;
11096: }else
11097: break;
11098: }
11099:
1.209 brouard 11100: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11101: /* ungetc(c,ficpar); */
11102: /* fgets(line, MAXLINE, ficpar); */
11103: /* fputs(line,stdout); */
11104: /* fputs(line,ficparo); */
11105: /* } */
11106: /* ungetc(c,ficpar); */
1.126 brouard 11107:
11108: estepm=0;
1.209 brouard 11109: 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 11110:
11111: if (num_filled != 6) {
11112: 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);
11113: 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);
11114: goto end;
11115: }
11116: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11117: }
11118: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11119: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11120:
1.209 brouard 11121: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11122: if (estepm==0 || estepm < stepm) estepm=stepm;
11123: if (fage <= 2) {
11124: bage = ageminpar;
11125: fage = agemaxpar;
11126: }
11127:
11128: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11129: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11130: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11131:
1.186 brouard 11132: /* Other stuffs, more or less useful */
1.254 brouard 11133: while(fgets(line, MAXLINE, ficpar)) {
11134: /* If line starts with a # it is a comment */
11135: if (line[0] == '#') {
11136: numlinepar++;
11137: fputs(line,stdout);
11138: fputs(line,ficparo);
11139: fputs(line,ficlog);
11140: continue;
11141: }else
11142: break;
11143: }
11144:
11145: 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){
11146:
11147: if (num_filled != 7) {
11148: 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);
11149: 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);
11150: goto end;
11151: }
11152: /* fscanf(ficpar,"begin-prev-date=%lf/%lf/%lf end-prev-date=%lf/%lf/%lf mov_average=%d\n",&jprev1, &mprev1,&anprev1,&jprev2, &mprev2,&anprev2,&mobilav); */
11153: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11154: 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);
11155: 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);
11156: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11157: 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 11158: }
1.254 brouard 11159:
11160: while(fgets(line, MAXLINE, ficpar)) {
11161: /* If line starts with a # it is a comment */
11162: if (line[0] == '#') {
11163: numlinepar++;
11164: fputs(line,stdout);
11165: fputs(line,ficparo);
11166: fputs(line,ficlog);
11167: continue;
11168: }else
11169: break;
1.126 brouard 11170: }
11171:
11172:
11173: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11174: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11175:
1.254 brouard 11176: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11177: if (num_filled != 1) {
11178: 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);
11179: 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);
11180: goto end;
11181: }
11182: printf("pop_based=%d\n",popbased);
11183: fprintf(ficlog,"pop_based=%d\n",popbased);
11184: fprintf(ficparo,"pop_based=%d\n",popbased);
11185: fprintf(ficres,"pop_based=%d\n",popbased);
11186: }
11187:
11188: while(fgets(line, MAXLINE, ficpar)) {
11189: /* If line starts with a # it is a comment */
11190: if (line[0] == '#') {
11191: numlinepar++;
11192: fputs(line,stdout);
11193: fputs(line,ficparo);
11194: fputs(line,ficlog);
11195: continue;
11196: }else
11197: break;
1.126 brouard 11198: }
1.254 brouard 11199: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11200: /* ungetc(c,ficpar); */
11201: /* fgets(line, MAXLINE, ficpar); */
11202: /* fputs(line,stdout); */
11203: /* fputs(line,ficres); */
11204: /* fputs(line,ficparo); */
11205: /* } */
11206: /* ungetc(c,ficpar); */
11207:
11208: /* fscanf(ficpar,"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); */
11209: 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){
11210: if (num_filled != 8) {
11211: 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);
11212: 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);
11213: goto end;
11214: }
11215: 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);
11216: 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);
11217: 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);
11218: 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);
1.126 brouard 11219: /* day and month of proj2 are not used but only year anproj2.*/
1.217 brouard 11220: }
1.254 brouard 11221: while(fgets(line, MAXLINE, ficpar)) {
11222: /* If line starts with a # it is a comment */
11223: if (line[0] == '#') {
11224: numlinepar++;
11225: fputs(line,stdout);
11226: fputs(line,ficparo);
11227: fputs(line,ficlog);
11228: continue;
11229: }else
11230: break;
11231: }
11232: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11233: /* ungetc(c,ficpar); */
11234: /* fgets(line, MAXLINE, ficpar); */
11235: /* fputs(line,stdout); */
11236: /* fputs(line,ficparo); */
11237: /* fputs(line,ficres); */
11238: /* } */
11239: /* ungetc(c,ficpar); */
1.217 brouard 11240:
11241: 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);
1.254 brouard 11242: 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){
11243: if (num_filled != 8) {
11244: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11245: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11246: goto end;
11247: }
11248: 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);
11249: 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);
11250: 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);
11251: 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);
1.217 brouard 11252: /* day and month of proj2 are not used but only year anproj2.*/
1.254 brouard 11253: }
1.230 brouard 11254: /* Results */
1.235 brouard 11255: nresult=0;
1.230 brouard 11256: while(fgets(line, MAXLINE, ficpar)) {
11257: /* If line starts with a # it is a comment */
11258: if (line[0] == '#') {
11259: numlinepar++;
11260: fputs(line,stdout);
11261: fputs(line,ficparo);
11262: fputs(line,ficlog);
1.238 brouard 11263: fputs(line,ficres);
1.230 brouard 11264: continue;
11265: }else
11266: break;
11267: }
1.240 brouard 11268: if (!feof(ficpar))
1.230 brouard 11269: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 11270: if (num_filled == 0){
1.230 brouard 11271: resultline[0]='\0';
1.253 brouard 11272: printf("Warning %d: no result line should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
1.240 brouard 11273: break;
11274: } else if (num_filled != 1){
1.253 brouard 11275: printf("ERROR %d: result line should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
1.230 brouard 11276: }
1.235 brouard 11277: nresult++; /* Sum of resultlines */
11278: printf("Result %d: result=%s\n",nresult, resultline);
11279: if(nresult > MAXRESULTLINES){
11280: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11281: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11282: goto end;
11283: }
11284: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 11285: fprintf(ficparo,"result: %s\n",resultline);
11286: fprintf(ficres,"result: %s\n",resultline);
11287: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11288: while(fgets(line, MAXLINE, ficpar)) {
11289: /* If line starts with a # it is a comment */
11290: if (line[0] == '#') {
11291: numlinepar++;
11292: fputs(line,stdout);
11293: fputs(line,ficparo);
1.238 brouard 11294: fputs(line,ficres);
1.230 brouard 11295: fputs(line,ficlog);
11296: continue;
11297: }else
11298: break;
11299: }
11300: if (feof(ficpar))
11301: break;
11302: else{ /* Processess output results for this combination of covariate values */
11303: }
1.240 brouard 11304: } /* end while */
1.230 brouard 11305:
11306:
1.126 brouard 11307:
1.230 brouard 11308: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11309: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11310:
11311: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11312: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11313: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11314: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11315: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11316: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11317: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11318: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11319: }else{
1.218 brouard 11320: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11321: }
11322: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 11323: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
11324: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11325:
1.225 brouard 11326: /*------------ free_vector -------------*/
11327: /* chdir(path); */
1.220 brouard 11328:
1.215 brouard 11329: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11330: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11331: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11332: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11333: free_lvector(num,1,n);
11334: free_vector(agedc,1,n);
11335: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11336: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11337: fclose(ficparo);
11338: fclose(ficres);
1.220 brouard 11339:
11340:
1.186 brouard 11341: /* Other results (useful)*/
1.220 brouard 11342:
11343:
1.126 brouard 11344: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11345: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11346: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11347: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11348: fclose(ficrespl);
11349:
11350: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11351: /*#include "hpijx.h"*/
11352: hPijx(p, bage, fage);
1.145 brouard 11353: fclose(ficrespij);
1.227 brouard 11354:
1.220 brouard 11355: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11356: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11357: k=1;
1.126 brouard 11358: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11359:
1.219 brouard 11360: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11361: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11362: for(i=1;i<=AGESUP;i++)
1.219 brouard 11363: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11364: for(k=1;k<=ncovcombmax;k++)
11365: probs[i][j][k]=0.;
1.219 brouard 11366: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11367: if (mobilav!=0 ||mobilavproj !=0 ) {
11368: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11369: for(i=1;i<=AGESUP;i++)
11370: for(j=1;j<=nlstate;j++)
11371: for(k=1;k<=ncovcombmax;k++)
11372: mobaverages[i][j][k]=0.;
1.219 brouard 11373: mobaverage=mobaverages;
11374: if (mobilav!=0) {
1.235 brouard 11375: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11376: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11377: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11378: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11379: }
1.219 brouard 11380: }
11381: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11382: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11383: else if (mobilavproj !=0) {
1.235 brouard 11384: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11385: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11386: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11387: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11388: }
1.219 brouard 11389: }
11390: }/* end if moving average */
1.227 brouard 11391:
1.126 brouard 11392: /*---------- Forecasting ------------------*/
11393: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11394: if(prevfcast==1){
11395: /* if(stepm ==1){*/
1.225 brouard 11396: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11397: }
1.217 brouard 11398: if(backcast==1){
1.219 brouard 11399: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11400: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11401: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11402:
11403: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11404:
11405: bprlim=matrix(1,nlstate,1,nlstate);
11406: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11407: fclose(ficresplb);
11408:
1.222 brouard 11409: hBijx(p, bage, fage, mobaverage);
11410: fclose(ficrespijb);
1.219 brouard 11411: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11412:
11413: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11414: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11415: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11416: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11417: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11418: }
1.217 brouard 11419:
1.186 brouard 11420:
11421: /* ------ Other prevalence ratios------------ */
1.126 brouard 11422:
1.215 brouard 11423: free_ivector(wav,1,imx);
11424: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11425: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11426: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11427:
11428:
1.127 brouard 11429: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11430:
1.201 brouard 11431: strcpy(filerese,"E_");
11432: strcat(filerese,fileresu);
1.126 brouard 11433: if((ficreseij=fopen(filerese,"w"))==NULL) {
11434: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11435: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11436: }
1.208 brouard 11437: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11438: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11439:
11440: pstamp(ficreseij);
1.219 brouard 11441:
1.235 brouard 11442: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11443: if (cptcovn < 1){i1=1;}
11444:
11445: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11446: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11447: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11448: continue;
1.219 brouard 11449: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11450: printf("\n#****** ");
1.225 brouard 11451: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11452: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11453: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11454: }
11455: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11456: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11457: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11458: }
11459: fprintf(ficreseij,"******\n");
1.235 brouard 11460: printf("******\n");
1.219 brouard 11461:
11462: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11463: oldm=oldms;savm=savms;
1.235 brouard 11464: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11465:
1.219 brouard 11466: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11467: }
11468: fclose(ficreseij);
1.208 brouard 11469: printf("done evsij\n");fflush(stdout);
11470: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11471:
1.227 brouard 11472: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11473:
11474:
1.201 brouard 11475: strcpy(filerest,"T_");
11476: strcat(filerest,fileresu);
1.127 brouard 11477: if((ficrest=fopen(filerest,"w"))==NULL) {
11478: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11479: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11480: }
1.208 brouard 11481: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11482: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11483:
1.126 brouard 11484:
1.201 brouard 11485: strcpy(fileresstde,"STDE_");
11486: strcat(fileresstde,fileresu);
1.126 brouard 11487: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11488: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11489: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11490: }
1.227 brouard 11491: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11492: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11493:
1.201 brouard 11494: strcpy(filerescve,"CVE_");
11495: strcat(filerescve,fileresu);
1.126 brouard 11496: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11497: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11498: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11499: }
1.227 brouard 11500: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11501: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11502:
1.201 brouard 11503: strcpy(fileresv,"V_");
11504: strcat(fileresv,fileresu);
1.126 brouard 11505: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11506: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11507: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11508: }
1.227 brouard 11509: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11510: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11511:
1.145 brouard 11512: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11513: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11514:
1.235 brouard 11515: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11516: if (cptcovn < 1){i1=1;}
11517:
11518: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11519: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11520: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11521: continue;
1.242 brouard 11522: printf("\n#****** Result for:");
11523: fprintf(ficrest,"\n#****** Result for:");
11524: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11525: for(j=1;j<=cptcoveff;j++){
11526: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11527: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11528: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11529: }
1.235 brouard 11530: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11531: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11532: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11533: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11534: }
1.208 brouard 11535: fprintf(ficrest,"******\n");
1.227 brouard 11536: fprintf(ficlog,"******\n");
11537: printf("******\n");
1.208 brouard 11538:
11539: fprintf(ficresstdeij,"\n#****** ");
11540: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11541: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11542: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11543: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11544: }
1.235 brouard 11545: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11546: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11547: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11548: }
1.208 brouard 11549: fprintf(ficresstdeij,"******\n");
11550: fprintf(ficrescveij,"******\n");
11551:
11552: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11553: /* pstamp(ficresvij); */
1.225 brouard 11554: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11555: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11556: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11557: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11558: }
1.208 brouard 11559: fprintf(ficresvij,"******\n");
11560:
11561: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11562: oldm=oldms;savm=savms;
1.235 brouard 11563: printf(" cvevsij ");
11564: fprintf(ficlog, " cvevsij ");
11565: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11566: printf(" end cvevsij \n ");
11567: fprintf(ficlog, " end cvevsij \n ");
11568:
11569: /*
11570: */
11571: /* goto endfree; */
11572:
11573: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11574: pstamp(ficrest);
11575:
11576:
11577: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11578: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11579: cptcod= 0; /* To be deleted */
11580: printf("varevsij vpopbased=%d \n",vpopbased);
11581: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11582: 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 11583: 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 ");
11584: if(vpopbased==1)
11585: 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);
11586: else
11587: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11588: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11589: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11590: fprintf(ficrest,"\n");
11591: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11592: epj=vector(1,nlstate+1);
11593: printf("Computing age specific period (stable) prevalences in each health state \n");
11594: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11595: for(age=bage; age <=fage ;age++){
1.235 brouard 11596: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11597: if (vpopbased==1) {
11598: if(mobilav ==0){
11599: for(i=1; i<=nlstate;i++)
11600: prlim[i][i]=probs[(int)age][i][k];
11601: }else{ /* mobilav */
11602: for(i=1; i<=nlstate;i++)
11603: prlim[i][i]=mobaverage[(int)age][i][k];
11604: }
11605: }
1.219 brouard 11606:
1.227 brouard 11607: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11608: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11609: /* printf(" age %4.0f ",age); */
11610: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11611: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11612: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11613: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11614: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11615: }
11616: epj[nlstate+1] +=epj[j];
11617: }
11618: /* printf(" age %4.0f \n",age); */
1.219 brouard 11619:
1.227 brouard 11620: for(i=1, vepp=0.;i <=nlstate;i++)
11621: for(j=1;j <=nlstate;j++)
11622: vepp += vareij[i][j][(int)age];
11623: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11624: for(j=1;j <=nlstate;j++){
11625: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11626: }
11627: fprintf(ficrest,"\n");
11628: }
1.208 brouard 11629: } /* End vpopbased */
11630: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11631: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11632: free_vector(epj,1,nlstate+1);
1.235 brouard 11633: printf("done selection\n");fflush(stdout);
11634: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11635:
1.145 brouard 11636: /*}*/
1.235 brouard 11637: } /* End k selection */
1.227 brouard 11638:
11639: printf("done State-specific expectancies\n");fflush(stdout);
11640: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11641:
1.126 brouard 11642: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11643:
1.201 brouard 11644: strcpy(fileresvpl,"VPL_");
11645: strcat(fileresvpl,fileresu);
1.126 brouard 11646: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11647: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11648: exit(0);
11649: }
1.208 brouard 11650: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11651: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11652:
1.145 brouard 11653: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11654: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11655:
1.235 brouard 11656: i1=pow(2,cptcoveff);
11657: if (cptcovn < 1){i1=1;}
11658:
11659: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11660: for(k=1; k<=i1;k++){
1.253 brouard 11661: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11662: continue;
1.227 brouard 11663: fprintf(ficresvpl,"\n#****** ");
11664: printf("\n#****** ");
11665: fprintf(ficlog,"\n#****** ");
11666: for(j=1;j<=cptcoveff;j++) {
11667: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11668: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11669: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11670: }
1.235 brouard 11671: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11672: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11673: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11674: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11675: }
1.227 brouard 11676: fprintf(ficresvpl,"******\n");
11677: printf("******\n");
11678: fprintf(ficlog,"******\n");
11679:
11680: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11681: oldm=oldms;savm=savms;
1.235 brouard 11682: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11683: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11684: /*}*/
1.126 brouard 11685: }
1.227 brouard 11686:
1.126 brouard 11687: fclose(ficresvpl);
1.208 brouard 11688: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11689: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11690:
11691: free_vector(weight,1,n);
11692: free_imatrix(Tvard,1,NCOVMAX,1,2);
11693: free_imatrix(s,1,maxwav+1,1,n);
11694: free_matrix(anint,1,maxwav,1,n);
11695: free_matrix(mint,1,maxwav,1,n);
11696: free_ivector(cod,1,n);
11697: free_ivector(tab,1,NCOVMAX);
11698: fclose(ficresstdeij);
11699: fclose(ficrescveij);
11700: fclose(ficresvij);
11701: fclose(ficrest);
11702: fclose(ficpar);
11703:
11704:
1.126 brouard 11705: /*---------- End : free ----------------*/
1.219 brouard 11706: if (mobilav!=0 ||mobilavproj !=0)
11707: free_ma3x(mobaverages,1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
1.218 brouard 11708: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11709: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11710: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11711: } /* mle==-3 arrives here for freeing */
1.227 brouard 11712: /* endfree:*/
11713: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11714: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11715: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11716: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11717: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11718: free_matrix(coqvar,1,maxwav,1,n);
11719: free_matrix(covar,0,NCOVMAX,1,n);
11720: free_matrix(matcov,1,npar,1,npar);
11721: free_matrix(hess,1,npar,1,npar);
11722: /*free_vector(delti,1,npar);*/
11723: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11724: free_matrix(agev,1,maxwav,1,imx);
11725: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11726:
11727: free_ivector(ncodemax,1,NCOVMAX);
11728: free_ivector(ncodemaxwundef,1,NCOVMAX);
11729: free_ivector(Dummy,-1,NCOVMAX);
11730: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11731: free_ivector(DummyV,1,NCOVMAX);
11732: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11733: free_ivector(Typevar,-1,NCOVMAX);
11734: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11735: free_ivector(TvarsQ,1,NCOVMAX);
11736: free_ivector(TvarsQind,1,NCOVMAX);
11737: free_ivector(TvarsD,1,NCOVMAX);
11738: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11739: free_ivector(TvarFD,1,NCOVMAX);
11740: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11741: free_ivector(TvarF,1,NCOVMAX);
11742: free_ivector(TvarFind,1,NCOVMAX);
11743: free_ivector(TvarV,1,NCOVMAX);
11744: free_ivector(TvarVind,1,NCOVMAX);
11745: free_ivector(TvarA,1,NCOVMAX);
11746: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11747: free_ivector(TvarFQ,1,NCOVMAX);
11748: free_ivector(TvarFQind,1,NCOVMAX);
11749: free_ivector(TvarVD,1,NCOVMAX);
11750: free_ivector(TvarVDind,1,NCOVMAX);
11751: free_ivector(TvarVQ,1,NCOVMAX);
11752: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11753: free_ivector(Tvarsel,1,NCOVMAX);
11754: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11755: free_ivector(Tposprod,1,NCOVMAX);
11756: free_ivector(Tprod,1,NCOVMAX);
11757: free_ivector(Tvaraff,1,NCOVMAX);
11758: free_ivector(invalidvarcomb,1,ncovcombmax);
11759: free_ivector(Tage,1,NCOVMAX);
11760: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11761: free_ivector(TmodelInvind,1,NCOVMAX);
11762: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11763:
11764: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11765: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11766: fflush(fichtm);
11767: fflush(ficgp);
11768:
1.227 brouard 11769:
1.126 brouard 11770: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11771: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11772: 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 11773: }else{
11774: printf("End of Imach\n");
11775: fprintf(ficlog,"End of Imach\n");
11776: }
11777: printf("See log file on %s\n",filelog);
11778: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11779: /*(void) gettimeofday(&end_time,&tzp);*/
11780: rend_time = time(NULL);
11781: end_time = *localtime(&rend_time);
11782: /* tml = *localtime(&end_time.tm_sec); */
11783: strcpy(strtend,asctime(&end_time));
1.126 brouard 11784: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11785: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11786: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11787:
1.157 brouard 11788: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11789: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11790: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11791: /* printf("Total time was %d uSec.\n", total_usecs);*/
11792: /* if(fileappend(fichtm,optionfilehtm)){ */
11793: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11794: fclose(fichtm);
11795: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11796: fclose(fichtmcov);
11797: fclose(ficgp);
11798: fclose(ficlog);
11799: /*------ End -----------*/
1.227 brouard 11800:
11801:
11802: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11803: #ifdef WIN32
1.227 brouard 11804: if (_chdir(pathcd) != 0)
11805: printf("Can't move to directory %s!\n",path);
11806: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11807: #else
1.227 brouard 11808: if(chdir(pathcd) != 0)
11809: printf("Can't move to directory %s!\n", path);
11810: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11811: #endif
1.126 brouard 11812: printf("Current directory %s!\n",pathcd);
11813: /*strcat(plotcmd,CHARSEPARATOR);*/
11814: sprintf(plotcmd,"gnuplot");
1.157 brouard 11815: #ifdef _WIN32
1.126 brouard 11816: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11817: #endif
11818: if(!stat(plotcmd,&info)){
1.158 brouard 11819: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11820: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11821: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11822: }else
11823: strcpy(pplotcmd,plotcmd);
1.157 brouard 11824: #ifdef __unix
1.126 brouard 11825: strcpy(plotcmd,GNUPLOTPROGRAM);
11826: if(!stat(plotcmd,&info)){
1.158 brouard 11827: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11828: }else
11829: strcpy(pplotcmd,plotcmd);
11830: #endif
11831: }else
11832: strcpy(pplotcmd,plotcmd);
11833:
11834: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11835: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11836:
1.126 brouard 11837: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11838: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11839: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11840: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11841: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11842: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11843: }
1.158 brouard 11844: printf(" Successful, please wait...");
1.126 brouard 11845: while (z[0] != 'q') {
11846: /* chdir(path); */
1.154 brouard 11847: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11848: scanf("%s",z);
11849: /* if (z[0] == 'c') system("./imach"); */
11850: if (z[0] == 'e') {
1.158 brouard 11851: #ifdef __APPLE__
1.152 brouard 11852: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11853: #elif __linux
11854: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11855: #else
1.152 brouard 11856: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11857: #endif
11858: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11859: system(pplotcmd);
1.126 brouard 11860: }
11861: else if (z[0] == 'g') system(plotcmd);
11862: else if (z[0] == 'q') exit(0);
11863: }
1.227 brouard 11864: end:
1.126 brouard 11865: while (z[0] != 'q') {
1.195 brouard 11866: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11867: scanf("%s",z);
11868: }
11869: }
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