Annotation of imach/src/imach.c, revision 1.242
1.242 ! brouard 1: /* $Id: imach.c,v 1.241 2016/08/29 17:17:25 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.242 ! brouard 4: Revision 1.241 2016/08/29 17:17:25 brouard
! 5: Summary: gnuplot problem in Back projection to fix
! 6:
1.241 brouard 7: Revision 1.240 2016/08/29 07:53:18 brouard
8: Summary: Better
9:
1.240 brouard 10: Revision 1.239 2016/08/26 15:51:03 brouard
11: Summary: Improvement in Powell output in order to copy and paste
12:
13: Author:
14:
1.239 brouard 15: Revision 1.238 2016/08/26 14:23:35 brouard
16: Summary: Starting tests of 0.99
17:
1.238 brouard 18: Revision 1.237 2016/08/26 09:20:19 brouard
19: Summary: to valgrind
20:
1.237 brouard 21: Revision 1.236 2016/08/25 10:50:18 brouard
22: *** empty log message ***
23:
1.236 brouard 24: Revision 1.235 2016/08/25 06:59:23 brouard
25: *** empty log message ***
26:
1.235 brouard 27: Revision 1.234 2016/08/23 16:51:20 brouard
28: *** empty log message ***
29:
1.234 brouard 30: Revision 1.233 2016/08/23 07:40:50 brouard
31: Summary: not working
32:
1.233 brouard 33: Revision 1.232 2016/08/22 14:20:21 brouard
34: Summary: not working
35:
1.232 brouard 36: Revision 1.231 2016/08/22 07:17:15 brouard
37: Summary: not working
38:
1.231 brouard 39: Revision 1.230 2016/08/22 06:55:53 brouard
40: Summary: Not working
41:
1.230 brouard 42: Revision 1.229 2016/07/23 09:45:53 brouard
43: Summary: Completing for func too
44:
1.229 brouard 45: Revision 1.228 2016/07/22 17:45:30 brouard
46: Summary: Fixing some arrays, still debugging
47:
1.227 brouard 48: Revision 1.226 2016/07/12 18:42:34 brouard
49: Summary: temp
50:
1.226 brouard 51: Revision 1.225 2016/07/12 08:40:03 brouard
52: Summary: saving but not running
53:
1.225 brouard 54: Revision 1.224 2016/07/01 13:16:01 brouard
55: Summary: Fixes
56:
1.224 brouard 57: Revision 1.223 2016/02/19 09:23:35 brouard
58: Summary: temporary
59:
1.223 brouard 60: Revision 1.222 2016/02/17 08:14:50 brouard
61: Summary: Probably last 0.98 stable version 0.98r6
62:
1.222 brouard 63: Revision 1.221 2016/02/15 23:35:36 brouard
64: Summary: minor bug
65:
1.220 brouard 66: Revision 1.219 2016/02/15 00:48:12 brouard
67: *** empty log message ***
68:
1.219 brouard 69: Revision 1.218 2016/02/12 11:29:23 brouard
70: Summary: 0.99 Back projections
71:
1.218 brouard 72: Revision 1.217 2015/12/23 17:18:31 brouard
73: Summary: Experimental backcast
74:
1.217 brouard 75: Revision 1.216 2015/12/18 17:32:11 brouard
76: Summary: 0.98r4 Warning and status=-2
77:
78: Version 0.98r4 is now:
79: - displaying an error when status is -1, date of interview unknown and date of death known;
80: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
81: Older changes concerning s=-2, dating from 2005 have been supersed.
82:
1.216 brouard 83: Revision 1.215 2015/12/16 08:52:24 brouard
84: Summary: 0.98r4 working
85:
1.215 brouard 86: Revision 1.214 2015/12/16 06:57:54 brouard
87: Summary: temporary not working
88:
1.214 brouard 89: Revision 1.213 2015/12/11 18:22:17 brouard
90: Summary: 0.98r4
91:
1.213 brouard 92: Revision 1.212 2015/11/21 12:47:24 brouard
93: Summary: minor typo
94:
1.212 brouard 95: Revision 1.211 2015/11/21 12:41:11 brouard
96: Summary: 0.98r3 with some graph of projected cross-sectional
97:
98: Author: Nicolas Brouard
99:
1.211 brouard 100: Revision 1.210 2015/11/18 17:41:20 brouard
101: Summary: Start working on projected prevalences
102:
1.210 brouard 103: Revision 1.209 2015/11/17 22:12:03 brouard
104: Summary: Adding ftolpl parameter
105: Author: N Brouard
106:
107: We had difficulties to get smoothed confidence intervals. It was due
108: to the period prevalence which wasn't computed accurately. The inner
109: parameter ftolpl is now an outer parameter of the .imach parameter
110: file after estepm. If ftolpl is small 1.e-4 and estepm too,
111: computation are long.
112:
1.209 brouard 113: Revision 1.208 2015/11/17 14:31:57 brouard
114: Summary: temporary
115:
1.208 brouard 116: Revision 1.207 2015/10/27 17:36:57 brouard
117: *** empty log message ***
118:
1.207 brouard 119: Revision 1.206 2015/10/24 07:14:11 brouard
120: *** empty log message ***
121:
1.206 brouard 122: Revision 1.205 2015/10/23 15:50:53 brouard
123: Summary: 0.98r3 some clarification for graphs on likelihood contributions
124:
1.205 brouard 125: Revision 1.204 2015/10/01 16:20:26 brouard
126: Summary: Some new graphs of contribution to likelihood
127:
1.204 brouard 128: Revision 1.203 2015/09/30 17:45:14 brouard
129: Summary: looking at better estimation of the hessian
130:
131: Also a better criteria for convergence to the period prevalence And
132: therefore adding the number of years needed to converge. (The
133: prevalence in any alive state shold sum to one
134:
1.203 brouard 135: Revision 1.202 2015/09/22 19:45:16 brouard
136: Summary: Adding some overall graph on contribution to likelihood. Might change
137:
1.202 brouard 138: Revision 1.201 2015/09/15 17:34:58 brouard
139: Summary: 0.98r0
140:
141: - Some new graphs like suvival functions
142: - Some bugs fixed like model=1+age+V2.
143:
1.201 brouard 144: Revision 1.200 2015/09/09 16:53:55 brouard
145: Summary: Big bug thanks to Flavia
146:
147: Even model=1+age+V2. did not work anymore
148:
1.200 brouard 149: Revision 1.199 2015/09/07 14:09:23 brouard
150: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
151:
1.199 brouard 152: Revision 1.198 2015/09/03 07:14:39 brouard
153: Summary: 0.98q5 Flavia
154:
1.198 brouard 155: Revision 1.197 2015/09/01 18:24:39 brouard
156: *** empty log message ***
157:
1.197 brouard 158: Revision 1.196 2015/08/18 23:17:52 brouard
159: Summary: 0.98q5
160:
1.196 brouard 161: Revision 1.195 2015/08/18 16:28:39 brouard
162: Summary: Adding a hack for testing purpose
163:
164: After reading the title, ftol and model lines, if the comment line has
165: a q, starting with #q, the answer at the end of the run is quit. It
166: permits to run test files in batch with ctest. The former workaround was
167: $ echo q | imach foo.imach
168:
1.195 brouard 169: Revision 1.194 2015/08/18 13:32:00 brouard
170: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
171:
1.194 brouard 172: Revision 1.193 2015/08/04 07:17:42 brouard
173: Summary: 0.98q4
174:
1.193 brouard 175: Revision 1.192 2015/07/16 16:49:02 brouard
176: Summary: Fixing some outputs
177:
1.192 brouard 178: Revision 1.191 2015/07/14 10:00:33 brouard
179: Summary: Some fixes
180:
1.191 brouard 181: Revision 1.190 2015/05/05 08:51:13 brouard
182: Summary: Adding digits in output parameters (7 digits instead of 6)
183:
184: Fix 1+age+.
185:
1.190 brouard 186: Revision 1.189 2015/04/30 14:45:16 brouard
187: Summary: 0.98q2
188:
1.189 brouard 189: Revision 1.188 2015/04/30 08:27:53 brouard
190: *** empty log message ***
191:
1.188 brouard 192: Revision 1.187 2015/04/29 09:11:15 brouard
193: *** empty log message ***
194:
1.187 brouard 195: Revision 1.186 2015/04/23 12:01:52 brouard
196: Summary: V1*age is working now, version 0.98q1
197:
198: Some codes had been disabled in order to simplify and Vn*age was
199: working in the optimization phase, ie, giving correct MLE parameters,
200: but, as usual, outputs were not correct and program core dumped.
201:
1.186 brouard 202: Revision 1.185 2015/03/11 13:26:42 brouard
203: Summary: Inclusion of compile and links command line for Intel Compiler
204:
1.185 brouard 205: Revision 1.184 2015/03/11 11:52:39 brouard
206: Summary: Back from Windows 8. Intel Compiler
207:
1.184 brouard 208: Revision 1.183 2015/03/10 20:34:32 brouard
209: Summary: 0.98q0, trying with directest, mnbrak fixed
210:
211: We use directest instead of original Powell test; probably no
212: incidence on the results, but better justifications;
213: We fixed Numerical Recipes mnbrak routine which was wrong and gave
214: wrong results.
215:
1.183 brouard 216: Revision 1.182 2015/02/12 08:19:57 brouard
217: Summary: Trying to keep directest which seems simpler and more general
218: Author: Nicolas Brouard
219:
1.182 brouard 220: Revision 1.181 2015/02/11 23:22:24 brouard
221: Summary: Comments on Powell added
222:
223: Author:
224:
1.181 brouard 225: Revision 1.180 2015/02/11 17:33:45 brouard
226: Summary: Finishing move from main to function (hpijx and prevalence_limit)
227:
1.180 brouard 228: Revision 1.179 2015/01/04 09:57:06 brouard
229: Summary: back to OS/X
230:
1.179 brouard 231: Revision 1.178 2015/01/04 09:35:48 brouard
232: *** empty log message ***
233:
1.178 brouard 234: Revision 1.177 2015/01/03 18:40:56 brouard
235: Summary: Still testing ilc32 on OSX
236:
1.177 brouard 237: Revision 1.176 2015/01/03 16:45:04 brouard
238: *** empty log message ***
239:
1.176 brouard 240: Revision 1.175 2015/01/03 16:33:42 brouard
241: *** empty log message ***
242:
1.175 brouard 243: Revision 1.174 2015/01/03 16:15:49 brouard
244: Summary: Still in cross-compilation
245:
1.174 brouard 246: Revision 1.173 2015/01/03 12:06:26 brouard
247: Summary: trying to detect cross-compilation
248:
1.173 brouard 249: Revision 1.172 2014/12/27 12:07:47 brouard
250: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
251:
1.172 brouard 252: Revision 1.171 2014/12/23 13:26:59 brouard
253: Summary: Back from Visual C
254:
255: Still problem with utsname.h on Windows
256:
1.171 brouard 257: Revision 1.170 2014/12/23 11:17:12 brouard
258: Summary: Cleaning some \%% back to %%
259:
260: The escape was mandatory for a specific compiler (which one?), but too many warnings.
261:
1.170 brouard 262: Revision 1.169 2014/12/22 23:08:31 brouard
263: Summary: 0.98p
264:
265: Outputs some informations on compiler used, OS etc. Testing on different platforms.
266:
1.169 brouard 267: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 268: Summary: update
1.169 brouard 269:
1.168 brouard 270: Revision 1.167 2014/12/22 13:50:56 brouard
271: Summary: Testing uname and compiler version and if compiled 32 or 64
272:
273: Testing on Linux 64
274:
1.167 brouard 275: Revision 1.166 2014/12/22 11:40:47 brouard
276: *** empty log message ***
277:
1.166 brouard 278: Revision 1.165 2014/12/16 11:20:36 brouard
279: Summary: After compiling on Visual C
280:
281: * imach.c (Module): Merging 1.61 to 1.162
282:
1.165 brouard 283: Revision 1.164 2014/12/16 10:52:11 brouard
284: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
285:
286: * imach.c (Module): Merging 1.61 to 1.162
287:
1.164 brouard 288: Revision 1.163 2014/12/16 10:30:11 brouard
289: * imach.c (Module): Merging 1.61 to 1.162
290:
1.163 brouard 291: Revision 1.162 2014/09/25 11:43:39 brouard
292: Summary: temporary backup 0.99!
293:
1.162 brouard 294: Revision 1.1 2014/09/16 11:06:58 brouard
295: Summary: With some code (wrong) for nlopt
296:
297: Author:
298:
299: Revision 1.161 2014/09/15 20:41:41 brouard
300: Summary: Problem with macro SQR on Intel compiler
301:
1.161 brouard 302: Revision 1.160 2014/09/02 09:24:05 brouard
303: *** empty log message ***
304:
1.160 brouard 305: Revision 1.159 2014/09/01 10:34:10 brouard
306: Summary: WIN32
307: Author: Brouard
308:
1.159 brouard 309: Revision 1.158 2014/08/27 17:11:51 brouard
310: *** empty log message ***
311:
1.158 brouard 312: Revision 1.157 2014/08/27 16:26:55 brouard
313: Summary: Preparing windows Visual studio version
314: Author: Brouard
315:
316: In order to compile on Visual studio, time.h is now correct and time_t
317: and tm struct should be used. difftime should be used but sometimes I
318: just make the differences in raw time format (time(&now).
319: Trying to suppress #ifdef LINUX
320: Add xdg-open for __linux in order to open default browser.
321:
1.157 brouard 322: Revision 1.156 2014/08/25 20:10:10 brouard
323: *** empty log message ***
324:
1.156 brouard 325: Revision 1.155 2014/08/25 18:32:34 brouard
326: Summary: New compile, minor changes
327: Author: Brouard
328:
1.155 brouard 329: Revision 1.154 2014/06/20 17:32:08 brouard
330: Summary: Outputs now all graphs of convergence to period prevalence
331:
1.154 brouard 332: Revision 1.153 2014/06/20 16:45:46 brouard
333: Summary: If 3 live state, convergence to period prevalence on same graph
334: Author: Brouard
335:
1.153 brouard 336: Revision 1.152 2014/06/18 17:54:09 brouard
337: Summary: open browser, use gnuplot on same dir than imach if not found in the path
338:
1.152 brouard 339: Revision 1.151 2014/06/18 16:43:30 brouard
340: *** empty log message ***
341:
1.151 brouard 342: Revision 1.150 2014/06/18 16:42:35 brouard
343: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
344: Author: brouard
345:
1.150 brouard 346: Revision 1.149 2014/06/18 15:51:14 brouard
347: Summary: Some fixes in parameter files errors
348: Author: Nicolas Brouard
349:
1.149 brouard 350: Revision 1.148 2014/06/17 17:38:48 brouard
351: Summary: Nothing new
352: Author: Brouard
353:
354: Just a new packaging for OS/X version 0.98nS
355:
1.148 brouard 356: Revision 1.147 2014/06/16 10:33:11 brouard
357: *** empty log message ***
358:
1.147 brouard 359: Revision 1.146 2014/06/16 10:20:28 brouard
360: Summary: Merge
361: Author: Brouard
362:
363: Merge, before building revised version.
364:
1.146 brouard 365: Revision 1.145 2014/06/10 21:23:15 brouard
366: Summary: Debugging with valgrind
367: Author: Nicolas Brouard
368:
369: Lot of changes in order to output the results with some covariates
370: After the Edimburgh REVES conference 2014, it seems mandatory to
371: improve the code.
372: No more memory valgrind error but a lot has to be done in order to
373: continue the work of splitting the code into subroutines.
374: Also, decodemodel has been improved. Tricode is still not
375: optimal. nbcode should be improved. Documentation has been added in
376: the source code.
377:
1.144 brouard 378: Revision 1.143 2014/01/26 09:45:38 brouard
379: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
380:
381: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
382: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
383:
1.143 brouard 384: Revision 1.142 2014/01/26 03:57:36 brouard
385: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
386:
387: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
388:
1.142 brouard 389: Revision 1.141 2014/01/26 02:42:01 brouard
390: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
391:
1.141 brouard 392: Revision 1.140 2011/09/02 10:37:54 brouard
393: Summary: times.h is ok with mingw32 now.
394:
1.140 brouard 395: Revision 1.139 2010/06/14 07:50:17 brouard
396: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
397: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
398:
1.139 brouard 399: Revision 1.138 2010/04/30 18:19:40 brouard
400: *** empty log message ***
401:
1.138 brouard 402: Revision 1.137 2010/04/29 18:11:38 brouard
403: (Module): Checking covariates for more complex models
404: than V1+V2. A lot of change to be done. Unstable.
405:
1.137 brouard 406: Revision 1.136 2010/04/26 20:30:53 brouard
407: (Module): merging some libgsl code. Fixing computation
408: of likelione (using inter/intrapolation if mle = 0) in order to
409: get same likelihood as if mle=1.
410: Some cleaning of code and comments added.
411:
1.136 brouard 412: Revision 1.135 2009/10/29 15:33:14 brouard
413: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
414:
1.135 brouard 415: Revision 1.134 2009/10/29 13:18:53 brouard
416: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
417:
1.134 brouard 418: Revision 1.133 2009/07/06 10:21:25 brouard
419: just nforces
420:
1.133 brouard 421: Revision 1.132 2009/07/06 08:22:05 brouard
422: Many tings
423:
1.132 brouard 424: Revision 1.131 2009/06/20 16:22:47 brouard
425: Some dimensions resccaled
426:
1.131 brouard 427: Revision 1.130 2009/05/26 06:44:34 brouard
428: (Module): Max Covariate is now set to 20 instead of 8. A
429: lot of cleaning with variables initialized to 0. Trying to make
430: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
431:
1.130 brouard 432: Revision 1.129 2007/08/31 13:49:27 lievre
433: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
434:
1.129 lievre 435: Revision 1.128 2006/06/30 13:02:05 brouard
436: (Module): Clarifications on computing e.j
437:
1.128 brouard 438: Revision 1.127 2006/04/28 18:11:50 brouard
439: (Module): Yes the sum of survivors was wrong since
440: imach-114 because nhstepm was no more computed in the age
441: loop. Now we define nhstepma in the age loop.
442: (Module): In order to speed up (in case of numerous covariates) we
443: compute health expectancies (without variances) in a first step
444: and then all the health expectancies with variances or standard
445: deviation (needs data from the Hessian matrices) which slows the
446: computation.
447: In the future we should be able to stop the program is only health
448: expectancies and graph are needed without standard deviations.
449:
1.127 brouard 450: Revision 1.126 2006/04/28 17:23:28 brouard
451: (Module): Yes the sum of survivors was wrong since
452: imach-114 because nhstepm was no more computed in the age
453: loop. Now we define nhstepma in the age loop.
454: Version 0.98h
455:
1.126 brouard 456: Revision 1.125 2006/04/04 15:20:31 lievre
457: Errors in calculation of health expectancies. Age was not initialized.
458: Forecasting file added.
459:
460: Revision 1.124 2006/03/22 17:13:53 lievre
461: Parameters are printed with %lf instead of %f (more numbers after the comma).
462: The log-likelihood is printed in the log file
463:
464: Revision 1.123 2006/03/20 10:52:43 brouard
465: * imach.c (Module): <title> changed, corresponds to .htm file
466: name. <head> headers where missing.
467:
468: * imach.c (Module): Weights can have a decimal point as for
469: English (a comma might work with a correct LC_NUMERIC environment,
470: otherwise the weight is truncated).
471: Modification of warning when the covariates values are not 0 or
472: 1.
473: Version 0.98g
474:
475: Revision 1.122 2006/03/20 09:45:41 brouard
476: (Module): Weights can have a decimal point as for
477: English (a comma might work with a correct LC_NUMERIC environment,
478: otherwise the weight is truncated).
479: Modification of warning when the covariates values are not 0 or
480: 1.
481: Version 0.98g
482:
483: Revision 1.121 2006/03/16 17:45:01 lievre
484: * imach.c (Module): Comments concerning covariates added
485:
486: * imach.c (Module): refinements in the computation of lli if
487: status=-2 in order to have more reliable computation if stepm is
488: not 1 month. Version 0.98f
489:
490: Revision 1.120 2006/03/16 15:10:38 lievre
491: (Module): refinements in the computation of lli if
492: status=-2 in order to have more reliable computation if stepm is
493: not 1 month. Version 0.98f
494:
495: Revision 1.119 2006/03/15 17:42:26 brouard
496: (Module): Bug if status = -2, the loglikelihood was
497: computed as likelihood omitting the logarithm. Version O.98e
498:
499: Revision 1.118 2006/03/14 18:20:07 brouard
500: (Module): varevsij Comments added explaining the second
501: table of variances if popbased=1 .
502: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
503: (Module): Function pstamp added
504: (Module): Version 0.98d
505:
506: Revision 1.117 2006/03/14 17:16:22 brouard
507: (Module): varevsij Comments added explaining the second
508: table of variances if popbased=1 .
509: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
510: (Module): Function pstamp added
511: (Module): Version 0.98d
512:
513: Revision 1.116 2006/03/06 10:29:27 brouard
514: (Module): Variance-covariance wrong links and
515: varian-covariance of ej. is needed (Saito).
516:
517: Revision 1.115 2006/02/27 12:17:45 brouard
518: (Module): One freematrix added in mlikeli! 0.98c
519:
520: Revision 1.114 2006/02/26 12:57:58 brouard
521: (Module): Some improvements in processing parameter
522: filename with strsep.
523:
524: Revision 1.113 2006/02/24 14:20:24 brouard
525: (Module): Memory leaks checks with valgrind and:
526: datafile was not closed, some imatrix were not freed and on matrix
527: allocation too.
528:
529: Revision 1.112 2006/01/30 09:55:26 brouard
530: (Module): Back to gnuplot.exe instead of wgnuplot.exe
531:
532: Revision 1.111 2006/01/25 20:38:18 brouard
533: (Module): Lots of cleaning and bugs added (Gompertz)
534: (Module): Comments can be added in data file. Missing date values
535: can be a simple dot '.'.
536:
537: Revision 1.110 2006/01/25 00:51:50 brouard
538: (Module): Lots of cleaning and bugs added (Gompertz)
539:
540: Revision 1.109 2006/01/24 19:37:15 brouard
541: (Module): Comments (lines starting with a #) are allowed in data.
542:
543: Revision 1.108 2006/01/19 18:05:42 lievre
544: Gnuplot problem appeared...
545: To be fixed
546:
547: Revision 1.107 2006/01/19 16:20:37 brouard
548: Test existence of gnuplot in imach path
549:
550: Revision 1.106 2006/01/19 13:24:36 brouard
551: Some cleaning and links added in html output
552:
553: Revision 1.105 2006/01/05 20:23:19 lievre
554: *** empty log message ***
555:
556: Revision 1.104 2005/09/30 16:11:43 lievre
557: (Module): sump fixed, loop imx fixed, and simplifications.
558: (Module): If the status is missing at the last wave but we know
559: that the person is alive, then we can code his/her status as -2
560: (instead of missing=-1 in earlier versions) and his/her
561: contributions to the likelihood is 1 - Prob of dying from last
562: health status (= 1-p13= p11+p12 in the easiest case of somebody in
563: the healthy state at last known wave). Version is 0.98
564:
565: Revision 1.103 2005/09/30 15:54:49 lievre
566: (Module): sump fixed, loop imx fixed, and simplifications.
567:
568: Revision 1.102 2004/09/15 17:31:30 brouard
569: Add the possibility to read data file including tab characters.
570:
571: Revision 1.101 2004/09/15 10:38:38 brouard
572: Fix on curr_time
573:
574: Revision 1.100 2004/07/12 18:29:06 brouard
575: Add version for Mac OS X. Just define UNIX in Makefile
576:
577: Revision 1.99 2004/06/05 08:57:40 brouard
578: *** empty log message ***
579:
580: Revision 1.98 2004/05/16 15:05:56 brouard
581: New version 0.97 . First attempt to estimate force of mortality
582: directly from the data i.e. without the need of knowing the health
583: state at each age, but using a Gompertz model: log u =a + b*age .
584: This is the basic analysis of mortality and should be done before any
585: other analysis, in order to test if the mortality estimated from the
586: cross-longitudinal survey is different from the mortality estimated
587: from other sources like vital statistic data.
588:
589: The same imach parameter file can be used but the option for mle should be -3.
590:
1.133 brouard 591: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 592: former routines in order to include the new code within the former code.
593:
594: The output is very simple: only an estimate of the intercept and of
595: the slope with 95% confident intervals.
596:
597: Current limitations:
598: A) Even if you enter covariates, i.e. with the
599: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
600: B) There is no computation of Life Expectancy nor Life Table.
601:
602: Revision 1.97 2004/02/20 13:25:42 lievre
603: Version 0.96d. Population forecasting command line is (temporarily)
604: suppressed.
605:
606: Revision 1.96 2003/07/15 15:38:55 brouard
607: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
608: rewritten within the same printf. Workaround: many printfs.
609:
610: Revision 1.95 2003/07/08 07:54:34 brouard
611: * imach.c (Repository):
612: (Repository): Using imachwizard code to output a more meaningful covariance
613: matrix (cov(a12,c31) instead of numbers.
614:
615: Revision 1.94 2003/06/27 13:00:02 brouard
616: Just cleaning
617:
618: Revision 1.93 2003/06/25 16:33:55 brouard
619: (Module): On windows (cygwin) function asctime_r doesn't
620: exist so I changed back to asctime which exists.
621: (Module): Version 0.96b
622:
623: Revision 1.92 2003/06/25 16:30:45 brouard
624: (Module): On windows (cygwin) function asctime_r doesn't
625: exist so I changed back to asctime which exists.
626:
627: Revision 1.91 2003/06/25 15:30:29 brouard
628: * imach.c (Repository): Duplicated warning errors corrected.
629: (Repository): Elapsed time after each iteration is now output. It
630: helps to forecast when convergence will be reached. Elapsed time
631: is stamped in powell. We created a new html file for the graphs
632: concerning matrix of covariance. It has extension -cov.htm.
633:
634: Revision 1.90 2003/06/24 12:34:15 brouard
635: (Module): Some bugs corrected for windows. Also, when
636: mle=-1 a template is output in file "or"mypar.txt with the design
637: of the covariance matrix to be input.
638:
639: Revision 1.89 2003/06/24 12:30:52 brouard
640: (Module): Some bugs corrected for windows. Also, when
641: mle=-1 a template is output in file "or"mypar.txt with the design
642: of the covariance matrix to be input.
643:
644: Revision 1.88 2003/06/23 17:54:56 brouard
645: * 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.
646:
647: Revision 1.87 2003/06/18 12:26:01 brouard
648: Version 0.96
649:
650: Revision 1.86 2003/06/17 20:04:08 brouard
651: (Module): Change position of html and gnuplot routines and added
652: routine fileappend.
653:
654: Revision 1.85 2003/06/17 13:12:43 brouard
655: * imach.c (Repository): Check when date of death was earlier that
656: current date of interview. It may happen when the death was just
657: prior to the death. In this case, dh was negative and likelihood
658: was wrong (infinity). We still send an "Error" but patch by
659: assuming that the date of death was just one stepm after the
660: interview.
661: (Repository): Because some people have very long ID (first column)
662: we changed int to long in num[] and we added a new lvector for
663: memory allocation. But we also truncated to 8 characters (left
664: truncation)
665: (Repository): No more line truncation errors.
666:
667: Revision 1.84 2003/06/13 21:44:43 brouard
668: * imach.c (Repository): Replace "freqsummary" at a correct
669: place. It differs from routine "prevalence" which may be called
670: many times. Probs is memory consuming and must be used with
671: parcimony.
672: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
673:
674: Revision 1.83 2003/06/10 13:39:11 lievre
675: *** empty log message ***
676:
677: Revision 1.82 2003/06/05 15:57:20 brouard
678: Add log in imach.c and fullversion number is now printed.
679:
680: */
681: /*
682: Interpolated Markov Chain
683:
684: Short summary of the programme:
685:
1.227 brouard 686: This program computes Healthy Life Expectancies or State-specific
687: (if states aren't health statuses) Expectancies from
688: cross-longitudinal data. Cross-longitudinal data consist in:
689:
690: -1- a first survey ("cross") where individuals from different ages
691: are interviewed on their health status or degree of disability (in
692: the case of a health survey which is our main interest)
693:
694: -2- at least a second wave of interviews ("longitudinal") which
695: measure each change (if any) in individual health status. Health
696: expectancies are computed from the time spent in each health state
697: according to a model. More health states you consider, more time is
698: necessary to reach the Maximum Likelihood of the parameters involved
699: in the model. The simplest model is the multinomial logistic model
700: where pij is the probability to be observed in state j at the second
701: wave conditional to be observed in state i at the first
702: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
703: etc , where 'age' is age and 'sex' is a covariate. If you want to
704: have a more complex model than "constant and age", you should modify
705: the program where the markup *Covariates have to be included here
706: again* invites you to do it. More covariates you add, slower the
1.126 brouard 707: convergence.
708:
709: The advantage of this computer programme, compared to a simple
710: multinomial logistic model, is clear when the delay between waves is not
711: identical for each individual. Also, if a individual missed an
712: intermediate interview, the information is lost, but taken into
713: account using an interpolation or extrapolation.
714:
715: hPijx is the probability to be observed in state i at age x+h
716: conditional to the observed state i at age x. The delay 'h' can be
717: split into an exact number (nh*stepm) of unobserved intermediate
718: states. This elementary transition (by month, quarter,
719: semester or year) is modelled as a multinomial logistic. The hPx
720: matrix is simply the matrix product of nh*stepm elementary matrices
721: and the contribution of each individual to the likelihood is simply
722: hPijx.
723:
724: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 725: of the life expectancies. It also computes the period (stable) prevalence.
726:
727: Back prevalence and projections:
1.227 brouard 728:
729: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
730: double agemaxpar, double ftolpl, int *ncvyearp, double
731: dateprev1,double dateprev2, int firstpass, int lastpass, int
732: mobilavproj)
733:
734: Computes the back prevalence limit for any combination of
735: covariate values k at any age between ageminpar and agemaxpar and
736: returns it in **bprlim. In the loops,
737:
738: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
739: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
740:
741: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 742: Computes for any combination of covariates k and any age between bage and fage
743: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
744: oldm=oldms;savm=savms;
1.227 brouard 745:
746: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 747: Computes the transition matrix starting at age 'age' over
748: 'nhstepm*hstepm*stepm' months (i.e. until
749: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 750: nhstepm*hstepm matrices.
751:
752: Returns p3mat[i][j][h] after calling
753: p3mat[i][j][h]=matprod2(newm,
754: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
755: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
756: oldm);
1.226 brouard 757:
758: Important routines
759:
760: - func (or funcone), computes logit (pij) distinguishing
761: o fixed variables (single or product dummies or quantitative);
762: o varying variables by:
763: (1) wave (single, product dummies, quantitative),
764: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
765: % fixed dummy (treated) or quantitative (not done because time-consuming);
766: % varying dummy (not done) or quantitative (not done);
767: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
768: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
769: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
770: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
771: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 772:
1.226 brouard 773:
774:
1.133 brouard 775: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
776: Institut national d'études démographiques, Paris.
1.126 brouard 777: This software have been partly granted by Euro-REVES, a concerted action
778: from the European Union.
779: It is copyrighted identically to a GNU software product, ie programme and
780: software can be distributed freely for non commercial use. Latest version
781: can be accessed at http://euroreves.ined.fr/imach .
782:
783: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
784: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
785:
786: **********************************************************************/
787: /*
788: main
789: read parameterfile
790: read datafile
791: concatwav
792: freqsummary
793: if (mle >= 1)
794: mlikeli
795: print results files
796: if mle==1
797: computes hessian
798: read end of parameter file: agemin, agemax, bage, fage, estepm
799: begin-prev-date,...
800: open gnuplot file
801: open html file
1.145 brouard 802: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
803: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
804: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
805: freexexit2 possible for memory heap.
806:
807: h Pij x | pij_nom ficrestpij
808: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
809: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
810: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
811:
812: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
813: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
814: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
815: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
816: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
817:
1.126 brouard 818: forecasting if prevfcast==1 prevforecast call prevalence()
819: health expectancies
820: Variance-covariance of DFLE
821: prevalence()
822: movingaverage()
823: varevsij()
824: if popbased==1 varevsij(,popbased)
825: total life expectancies
826: Variance of period (stable) prevalence
827: end
828: */
829:
1.187 brouard 830: /* #define DEBUG */
831: /* #define DEBUGBRENT */
1.203 brouard 832: /* #define DEBUGLINMIN */
833: /* #define DEBUGHESS */
834: #define DEBUGHESSIJ
1.224 brouard 835: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 836: #define POWELL /* Instead of NLOPT */
1.224 brouard 837: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 838: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
839: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 840:
841: #include <math.h>
842: #include <stdio.h>
843: #include <stdlib.h>
844: #include <string.h>
1.226 brouard 845: #include <ctype.h>
1.159 brouard 846:
847: #ifdef _WIN32
848: #include <io.h>
1.172 brouard 849: #include <windows.h>
850: #include <tchar.h>
1.159 brouard 851: #else
1.126 brouard 852: #include <unistd.h>
1.159 brouard 853: #endif
1.126 brouard 854:
855: #include <limits.h>
856: #include <sys/types.h>
1.171 brouard 857:
858: #if defined(__GNUC__)
859: #include <sys/utsname.h> /* Doesn't work on Windows */
860: #endif
861:
1.126 brouard 862: #include <sys/stat.h>
863: #include <errno.h>
1.159 brouard 864: /* extern int errno; */
1.126 brouard 865:
1.157 brouard 866: /* #ifdef LINUX */
867: /* #include <time.h> */
868: /* #include "timeval.h" */
869: /* #else */
870: /* #include <sys/time.h> */
871: /* #endif */
872:
1.126 brouard 873: #include <time.h>
874:
1.136 brouard 875: #ifdef GSL
876: #include <gsl/gsl_errno.h>
877: #include <gsl/gsl_multimin.h>
878: #endif
879:
1.167 brouard 880:
1.162 brouard 881: #ifdef NLOPT
882: #include <nlopt.h>
883: typedef struct {
884: double (* function)(double [] );
885: } myfunc_data ;
886: #endif
887:
1.126 brouard 888: /* #include <libintl.h> */
889: /* #define _(String) gettext (String) */
890:
1.141 brouard 891: #define MAXLINE 1024 /* Was 256. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 892:
893: #define GNUPLOTPROGRAM "gnuplot"
894: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
895: #define FILENAMELENGTH 132
896:
897: #define GLOCK_ERROR_NOPATH -1 /* empty path */
898: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
899:
1.144 brouard 900: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
901: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 902:
903: #define NINTERVMAX 8
1.144 brouard 904: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
905: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
906: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 907: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 908: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
909: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 910: #define MAXN 20000
1.144 brouard 911: #define YEARM 12. /**< Number of months per year */
1.218 brouard 912: /* #define AGESUP 130 */
913: #define AGESUP 150
914: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 915: #define AGEBASE 40
1.194 brouard 916: #define AGEOVERFLOW 1.e20
1.164 brouard 917: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 918: #ifdef _WIN32
919: #define DIRSEPARATOR '\\'
920: #define CHARSEPARATOR "\\"
921: #define ODIRSEPARATOR '/'
922: #else
1.126 brouard 923: #define DIRSEPARATOR '/'
924: #define CHARSEPARATOR "/"
925: #define ODIRSEPARATOR '\\'
926: #endif
927:
1.242 ! brouard 928: /* $Id: imach.c,v 1.241 2016/08/29 17:17:25 brouard Exp $ */
1.126 brouard 929: /* $State: Exp $ */
1.196 brouard 930: #include "version.h"
931: char version[]=__IMACH_VERSION__;
1.224 brouard 932: 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";
1.242 ! brouard 933: char fullversion[]="$Revision: 1.241 $ $Date: 2016/08/29 17:17:25 $";
1.126 brouard 934: char strstart[80];
935: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 936: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 937: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 938: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
939: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
940: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 941: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
942: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 943: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
944: int cptcovprodnoage=0; /**< Number of covariate products without age */
945: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 946: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
947: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 948: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 949: int nsd=0; /**< Total number of single dummy variables (output) */
950: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 951: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 952: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 953: int ntveff=0; /**< ntveff number of effective time varying variables */
954: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 955: int cptcov=0; /* Working variable */
1.218 brouard 956: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 957: int npar=NPARMAX;
958: int nlstate=2; /* Number of live states */
959: int ndeath=1; /* Number of dead states */
1.130 brouard 960: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 961: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 962: int popbased=0;
963:
964: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 965: int maxwav=0; /* Maxim number of waves */
966: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
967: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
968: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 969: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 970: int mle=1, weightopt=0;
1.126 brouard 971: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
972: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
973: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
974: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 975: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 976: int selected(int kvar); /* Is covariate kvar selected for printing results */
977:
1.130 brouard 978: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 979: double **matprod2(); /* test */
1.126 brouard 980: double **oldm, **newm, **savm; /* Working pointers to matrices */
981: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 982: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
983:
1.136 brouard 984: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 985: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 986: FILE *ficlog, *ficrespow;
1.130 brouard 987: int globpr=0; /* Global variable for printing or not */
1.126 brouard 988: double fretone; /* Only one call to likelihood */
1.130 brouard 989: long ipmx=0; /* Number of contributions */
1.126 brouard 990: double sw; /* Sum of weights */
991: char filerespow[FILENAMELENGTH];
992: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
993: FILE *ficresilk;
994: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
995: FILE *ficresprobmorprev;
996: FILE *fichtm, *fichtmcov; /* Html File */
997: FILE *ficreseij;
998: char filerese[FILENAMELENGTH];
999: FILE *ficresstdeij;
1000: char fileresstde[FILENAMELENGTH];
1001: FILE *ficrescveij;
1002: char filerescve[FILENAMELENGTH];
1003: FILE *ficresvij;
1004: char fileresv[FILENAMELENGTH];
1005: FILE *ficresvpl;
1006: char fileresvpl[FILENAMELENGTH];
1007: char title[MAXLINE];
1.234 brouard 1008: char model[MAXLINE]; /**< The model line */
1.217 brouard 1009: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1010: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1011: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1012: char command[FILENAMELENGTH];
1013: int outcmd=0;
1014:
1.217 brouard 1015: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1016: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1017: char filelog[FILENAMELENGTH]; /* Log file */
1018: char filerest[FILENAMELENGTH];
1019: char fileregp[FILENAMELENGTH];
1020: char popfile[FILENAMELENGTH];
1021:
1022: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1023:
1.157 brouard 1024: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1025: /* struct timezone tzp; */
1026: /* extern int gettimeofday(); */
1027: struct tm tml, *gmtime(), *localtime();
1028:
1029: extern time_t time();
1030:
1031: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1032: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1033: struct tm tm;
1034:
1.126 brouard 1035: char strcurr[80], strfor[80];
1036:
1037: char *endptr;
1038: long lval;
1039: double dval;
1040:
1041: #define NR_END 1
1042: #define FREE_ARG char*
1043: #define FTOL 1.0e-10
1044:
1045: #define NRANSI
1.240 brouard 1046: #define ITMAX 200
1047: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1048:
1049: #define TOL 2.0e-4
1050:
1051: #define CGOLD 0.3819660
1052: #define ZEPS 1.0e-10
1053: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1054:
1055: #define GOLD 1.618034
1056: #define GLIMIT 100.0
1057: #define TINY 1.0e-20
1058:
1059: static double maxarg1,maxarg2;
1060: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1061: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1062:
1063: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1064: #define rint(a) floor(a+0.5)
1.166 brouard 1065: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1066: #define mytinydouble 1.0e-16
1.166 brouard 1067: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1068: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1069: /* static double dsqrarg; */
1070: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1071: static double sqrarg;
1072: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1073: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1074: int agegomp= AGEGOMP;
1075:
1076: int imx;
1077: int stepm=1;
1078: /* Stepm, step in month: minimum step interpolation*/
1079:
1080: int estepm;
1081: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1082:
1083: int m,nb;
1084: long *num;
1.197 brouard 1085: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1086: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1087: covariate for which somebody answered excluding
1088: undefined. Usually 2: 0 and 1. */
1089: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1090: covariate for which somebody answered including
1091: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1092: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1093: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1094: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1095: double *ageexmed,*agecens;
1096: double dateintmean=0;
1097:
1098: double *weight;
1099: int **s; /* Status */
1.141 brouard 1100: double *agedc;
1.145 brouard 1101: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1102: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1103: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1104: double **coqvar; /* Fixed quantitative covariate iqv */
1105: double ***cotvar; /* Time varying covariate itv */
1106: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1107: double idx;
1108: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1109: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1110: /*k 1 2 3 4 5 6 7 8 9 */
1111: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1112: /* Tndvar[k] 1 2 3 4 5 */
1113: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1114: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1115: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1116: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1117: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1118: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1119: /* Tprod[i]=k 4 7 */
1120: /* Tage[i]=k 5 8 */
1121: /* */
1122: /* Type */
1123: /* V 1 2 3 4 5 */
1124: /* F F V V V */
1125: /* D Q D D Q */
1126: /* */
1127: int *TvarsD;
1128: int *TvarsDind;
1129: int *TvarsQ;
1130: int *TvarsQind;
1131:
1.235 brouard 1132: #define MAXRESULTLINES 10
1133: int nresult=0;
1134: int TKresult[MAXRESULTLINES];
1.237 brouard 1135: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1136: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1137: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1138: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1139: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1140: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1141:
1.234 brouard 1142: /* 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 1143: 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 */
1144: 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 */
1145: 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 */
1146: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1147: 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 */
1148: 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 1149: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1150: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1151: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1152: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1153: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1154: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1155: 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 */
1156: 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 */
1157:
1.230 brouard 1158: int *Tvarsel; /**< Selected covariates for output */
1159: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1160: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1161: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1162: 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 1163: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1164: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1165: int *Tage;
1.227 brouard 1166: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1167: 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 1168: 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*/
1169: 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 1170: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1171: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1172: int **Tvard;
1173: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1174: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1175: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1176: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1177: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1178: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1179: double *lsurv, *lpop, *tpop;
1180:
1.231 brouard 1181: #define FD 1; /* Fixed dummy covariate */
1182: #define FQ 2; /* Fixed quantitative covariate */
1183: #define FP 3; /* Fixed product covariate */
1184: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1185: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1186: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1187: #define VD 10; /* Varying dummy covariate */
1188: #define VQ 11; /* Varying quantitative covariate */
1189: #define VP 12; /* Varying product covariate */
1190: #define VPDD 13; /* Varying product dummy*dummy covariate */
1191: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1192: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1193: #define APFD 16; /* Age product * fixed dummy covariate */
1194: #define APFQ 17; /* Age product * fixed quantitative covariate */
1195: #define APVD 18; /* Age product * varying dummy covariate */
1196: #define APVQ 19; /* Age product * varying quantitative covariate */
1197:
1198: #define FTYPE 1; /* Fixed covariate */
1199: #define VTYPE 2; /* Varying covariate (loop in wave) */
1200: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1201:
1202: struct kmodel{
1203: int maintype; /* main type */
1204: int subtype; /* subtype */
1205: };
1206: struct kmodel modell[NCOVMAX];
1207:
1.143 brouard 1208: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1209: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1210:
1211: /**************** split *************************/
1212: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1213: {
1214: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1215: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1216: */
1217: char *ss; /* pointer */
1.186 brouard 1218: int l1=0, l2=0; /* length counters */
1.126 brouard 1219:
1220: l1 = strlen(path ); /* length of path */
1221: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1222: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1223: if ( ss == NULL ) { /* no directory, so determine current directory */
1224: strcpy( name, path ); /* we got the fullname name because no directory */
1225: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1226: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1227: /* get current working directory */
1228: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1229: #ifdef WIN32
1230: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1231: #else
1232: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1233: #endif
1.126 brouard 1234: return( GLOCK_ERROR_GETCWD );
1235: }
1236: /* got dirc from getcwd*/
1237: printf(" DIRC = %s \n",dirc);
1.205 brouard 1238: } else { /* strip directory from path */
1.126 brouard 1239: ss++; /* after this, the filename */
1240: l2 = strlen( ss ); /* length of filename */
1241: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1242: strcpy( name, ss ); /* save file name */
1243: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1244: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1245: printf(" DIRC2 = %s \n",dirc);
1246: }
1247: /* We add a separator at the end of dirc if not exists */
1248: l1 = strlen( dirc ); /* length of directory */
1249: if( dirc[l1-1] != DIRSEPARATOR ){
1250: dirc[l1] = DIRSEPARATOR;
1251: dirc[l1+1] = 0;
1252: printf(" DIRC3 = %s \n",dirc);
1253: }
1254: ss = strrchr( name, '.' ); /* find last / */
1255: if (ss >0){
1256: ss++;
1257: strcpy(ext,ss); /* save extension */
1258: l1= strlen( name);
1259: l2= strlen(ss)+1;
1260: strncpy( finame, name, l1-l2);
1261: finame[l1-l2]= 0;
1262: }
1263:
1264: return( 0 ); /* we're done */
1265: }
1266:
1267:
1268: /******************************************/
1269:
1270: void replace_back_to_slash(char *s, char*t)
1271: {
1272: int i;
1273: int lg=0;
1274: i=0;
1275: lg=strlen(t);
1276: for(i=0; i<= lg; i++) {
1277: (s[i] = t[i]);
1278: if (t[i]== '\\') s[i]='/';
1279: }
1280: }
1281:
1.132 brouard 1282: char *trimbb(char *out, char *in)
1.137 brouard 1283: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1284: char *s;
1285: s=out;
1286: while (*in != '\0'){
1.137 brouard 1287: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1288: in++;
1289: }
1290: *out++ = *in++;
1291: }
1292: *out='\0';
1293: return s;
1294: }
1295:
1.187 brouard 1296: /* char *substrchaine(char *out, char *in, char *chain) */
1297: /* { */
1298: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1299: /* char *s, *t; */
1300: /* t=in;s=out; */
1301: /* while ((*in != *chain) && (*in != '\0')){ */
1302: /* *out++ = *in++; */
1303: /* } */
1304:
1305: /* /\* *in matches *chain *\/ */
1306: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1307: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1308: /* } */
1309: /* in--; chain--; */
1310: /* while ( (*in != '\0')){ */
1311: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1312: /* *out++ = *in++; */
1313: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1314: /* } */
1315: /* *out='\0'; */
1316: /* out=s; */
1317: /* return out; */
1318: /* } */
1319: char *substrchaine(char *out, char *in, char *chain)
1320: {
1321: /* Substract chain 'chain' from 'in', return and output 'out' */
1322: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1323:
1324: char *strloc;
1325:
1326: strcpy (out, in);
1327: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1328: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1329: if(strloc != NULL){
1330: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1331: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1332: /* strcpy (strloc, strloc +strlen(chain));*/
1333: }
1334: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1335: return out;
1336: }
1337:
1338:
1.145 brouard 1339: char *cutl(char *blocc, char *alocc, char *in, char occ)
1340: {
1.187 brouard 1341: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1342: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1343: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1344: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1345: */
1.160 brouard 1346: char *s, *t;
1.145 brouard 1347: t=in;s=in;
1348: while ((*in != occ) && (*in != '\0')){
1349: *alocc++ = *in++;
1350: }
1351: if( *in == occ){
1352: *(alocc)='\0';
1353: s=++in;
1354: }
1355:
1356: if (s == t) {/* occ not found */
1357: *(alocc-(in-s))='\0';
1358: in=s;
1359: }
1360: while ( *in != '\0'){
1361: *blocc++ = *in++;
1362: }
1363:
1364: *blocc='\0';
1365: return t;
1366: }
1.137 brouard 1367: char *cutv(char *blocc, char *alocc, char *in, char occ)
1368: {
1.187 brouard 1369: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1370: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1371: gives blocc="abcdef2ghi" and alocc="j".
1372: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1373: */
1374: char *s, *t;
1375: t=in;s=in;
1376: while (*in != '\0'){
1377: while( *in == occ){
1378: *blocc++ = *in++;
1379: s=in;
1380: }
1381: *blocc++ = *in++;
1382: }
1383: if (s == t) /* occ not found */
1384: *(blocc-(in-s))='\0';
1385: else
1386: *(blocc-(in-s)-1)='\0';
1387: in=s;
1388: while ( *in != '\0'){
1389: *alocc++ = *in++;
1390: }
1391:
1392: *alocc='\0';
1393: return s;
1394: }
1395:
1.126 brouard 1396: int nbocc(char *s, char occ)
1397: {
1398: int i,j=0;
1399: int lg=20;
1400: i=0;
1401: lg=strlen(s);
1402: for(i=0; i<= lg; i++) {
1.234 brouard 1403: if (s[i] == occ ) j++;
1.126 brouard 1404: }
1405: return j;
1406: }
1407:
1.137 brouard 1408: /* void cutv(char *u,char *v, char*t, char occ) */
1409: /* { */
1410: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1411: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1412: /* gives u="abcdef2ghi" and v="j" *\/ */
1413: /* int i,lg,j,p=0; */
1414: /* i=0; */
1415: /* lg=strlen(t); */
1416: /* for(j=0; j<=lg-1; j++) { */
1417: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1418: /* } */
1.126 brouard 1419:
1.137 brouard 1420: /* for(j=0; j<p; j++) { */
1421: /* (u[j] = t[j]); */
1422: /* } */
1423: /* u[p]='\0'; */
1.126 brouard 1424:
1.137 brouard 1425: /* for(j=0; j<= lg; j++) { */
1426: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1427: /* } */
1428: /* } */
1.126 brouard 1429:
1.160 brouard 1430: #ifdef _WIN32
1431: char * strsep(char **pp, const char *delim)
1432: {
1433: char *p, *q;
1434:
1435: if ((p = *pp) == NULL)
1436: return 0;
1437: if ((q = strpbrk (p, delim)) != NULL)
1438: {
1439: *pp = q + 1;
1440: *q = '\0';
1441: }
1442: else
1443: *pp = 0;
1444: return p;
1445: }
1446: #endif
1447:
1.126 brouard 1448: /********************** nrerror ********************/
1449:
1450: void nrerror(char error_text[])
1451: {
1452: fprintf(stderr,"ERREUR ...\n");
1453: fprintf(stderr,"%s\n",error_text);
1454: exit(EXIT_FAILURE);
1455: }
1456: /*********************** vector *******************/
1457: double *vector(int nl, int nh)
1458: {
1459: double *v;
1460: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1461: if (!v) nrerror("allocation failure in vector");
1462: return v-nl+NR_END;
1463: }
1464:
1465: /************************ free vector ******************/
1466: void free_vector(double*v, int nl, int nh)
1467: {
1468: free((FREE_ARG)(v+nl-NR_END));
1469: }
1470:
1471: /************************ivector *******************************/
1472: int *ivector(long nl,long nh)
1473: {
1474: int *v;
1475: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1476: if (!v) nrerror("allocation failure in ivector");
1477: return v-nl+NR_END;
1478: }
1479:
1480: /******************free ivector **************************/
1481: void free_ivector(int *v, long nl, long nh)
1482: {
1483: free((FREE_ARG)(v+nl-NR_END));
1484: }
1485:
1486: /************************lvector *******************************/
1487: long *lvector(long nl,long nh)
1488: {
1489: long *v;
1490: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1491: if (!v) nrerror("allocation failure in ivector");
1492: return v-nl+NR_END;
1493: }
1494:
1495: /******************free lvector **************************/
1496: void free_lvector(long *v, long nl, long nh)
1497: {
1498: free((FREE_ARG)(v+nl-NR_END));
1499: }
1500:
1501: /******************* imatrix *******************************/
1502: int **imatrix(long nrl, long nrh, long ncl, long nch)
1503: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1504: {
1505: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1506: int **m;
1507:
1508: /* allocate pointers to rows */
1509: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1510: if (!m) nrerror("allocation failure 1 in matrix()");
1511: m += NR_END;
1512: m -= nrl;
1513:
1514:
1515: /* allocate rows and set pointers to them */
1516: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1517: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1518: m[nrl] += NR_END;
1519: m[nrl] -= ncl;
1520:
1521: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1522:
1523: /* return pointer to array of pointers to rows */
1524: return m;
1525: }
1526:
1527: /****************** free_imatrix *************************/
1528: void free_imatrix(m,nrl,nrh,ncl,nch)
1529: int **m;
1530: long nch,ncl,nrh,nrl;
1531: /* free an int matrix allocated by imatrix() */
1532: {
1533: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1534: free((FREE_ARG) (m+nrl-NR_END));
1535: }
1536:
1537: /******************* matrix *******************************/
1538: double **matrix(long nrl, long nrh, long ncl, long nch)
1539: {
1540: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1541: double **m;
1542:
1543: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1544: if (!m) nrerror("allocation failure 1 in matrix()");
1545: m += NR_END;
1546: m -= nrl;
1547:
1548: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1549: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1550: m[nrl] += NR_END;
1551: m[nrl] -= ncl;
1552:
1553: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1554: return m;
1.145 brouard 1555: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1556: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1557: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1558: */
1559: }
1560:
1561: /*************************free matrix ************************/
1562: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1563: {
1564: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1565: free((FREE_ARG)(m+nrl-NR_END));
1566: }
1567:
1568: /******************* ma3x *******************************/
1569: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1570: {
1571: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1572: double ***m;
1573:
1574: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1575: if (!m) nrerror("allocation failure 1 in matrix()");
1576: m += NR_END;
1577: m -= nrl;
1578:
1579: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1580: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1581: m[nrl] += NR_END;
1582: m[nrl] -= ncl;
1583:
1584: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1585:
1586: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1587: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1588: m[nrl][ncl] += NR_END;
1589: m[nrl][ncl] -= nll;
1590: for (j=ncl+1; j<=nch; j++)
1591: m[nrl][j]=m[nrl][j-1]+nlay;
1592:
1593: for (i=nrl+1; i<=nrh; i++) {
1594: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1595: for (j=ncl+1; j<=nch; j++)
1596: m[i][j]=m[i][j-1]+nlay;
1597: }
1598: return m;
1599: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1600: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1601: */
1602: }
1603:
1604: /*************************free ma3x ************************/
1605: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1606: {
1607: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1608: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1609: free((FREE_ARG)(m+nrl-NR_END));
1610: }
1611:
1612: /*************** function subdirf ***********/
1613: char *subdirf(char fileres[])
1614: {
1615: /* Caution optionfilefiname is hidden */
1616: strcpy(tmpout,optionfilefiname);
1617: strcat(tmpout,"/"); /* Add to the right */
1618: strcat(tmpout,fileres);
1619: return tmpout;
1620: }
1621:
1622: /*************** function subdirf2 ***********/
1623: char *subdirf2(char fileres[], char *preop)
1624: {
1625:
1626: /* Caution optionfilefiname is hidden */
1627: strcpy(tmpout,optionfilefiname);
1628: strcat(tmpout,"/");
1629: strcat(tmpout,preop);
1630: strcat(tmpout,fileres);
1631: return tmpout;
1632: }
1633:
1634: /*************** function subdirf3 ***********/
1635: char *subdirf3(char fileres[], char *preop, char *preop2)
1636: {
1637:
1638: /* Caution optionfilefiname is hidden */
1639: strcpy(tmpout,optionfilefiname);
1640: strcat(tmpout,"/");
1641: strcat(tmpout,preop);
1642: strcat(tmpout,preop2);
1643: strcat(tmpout,fileres);
1644: return tmpout;
1645: }
1.213 brouard 1646:
1647: /*************** function subdirfext ***********/
1648: char *subdirfext(char fileres[], char *preop, char *postop)
1649: {
1650:
1651: strcpy(tmpout,preop);
1652: strcat(tmpout,fileres);
1653: strcat(tmpout,postop);
1654: return tmpout;
1655: }
1.126 brouard 1656:
1.213 brouard 1657: /*************** function subdirfext3 ***********/
1658: char *subdirfext3(char fileres[], char *preop, char *postop)
1659: {
1660:
1661: /* Caution optionfilefiname is hidden */
1662: strcpy(tmpout,optionfilefiname);
1663: strcat(tmpout,"/");
1664: strcat(tmpout,preop);
1665: strcat(tmpout,fileres);
1666: strcat(tmpout,postop);
1667: return tmpout;
1668: }
1669:
1.162 brouard 1670: char *asc_diff_time(long time_sec, char ascdiff[])
1671: {
1672: long sec_left, days, hours, minutes;
1673: days = (time_sec) / (60*60*24);
1674: sec_left = (time_sec) % (60*60*24);
1675: hours = (sec_left) / (60*60) ;
1676: sec_left = (sec_left) %(60*60);
1677: minutes = (sec_left) /60;
1678: sec_left = (sec_left) % (60);
1679: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1680: return ascdiff;
1681: }
1682:
1.126 brouard 1683: /***************** f1dim *************************/
1684: extern int ncom;
1685: extern double *pcom,*xicom;
1686: extern double (*nrfunc)(double []);
1687:
1688: double f1dim(double x)
1689: {
1690: int j;
1691: double f;
1692: double *xt;
1693:
1694: xt=vector(1,ncom);
1695: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1696: f=(*nrfunc)(xt);
1697: free_vector(xt,1,ncom);
1698: return f;
1699: }
1700:
1701: /*****************brent *************************/
1702: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1703: {
1704: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1705: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1706: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1707: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1708: * returned function value.
1709: */
1.126 brouard 1710: int iter;
1711: double a,b,d,etemp;
1.159 brouard 1712: double fu=0,fv,fw,fx;
1.164 brouard 1713: double ftemp=0.;
1.126 brouard 1714: double p,q,r,tol1,tol2,u,v,w,x,xm;
1715: double e=0.0;
1716:
1717: a=(ax < cx ? ax : cx);
1718: b=(ax > cx ? ax : cx);
1719: x=w=v=bx;
1720: fw=fv=fx=(*f)(x);
1721: for (iter=1;iter<=ITMAX;iter++) {
1722: xm=0.5*(a+b);
1723: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1724: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1725: printf(".");fflush(stdout);
1726: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1727: #ifdef DEBUGBRENT
1.126 brouard 1728: 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);
1729: 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);
1730: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1731: #endif
1732: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1733: *xmin=x;
1734: return fx;
1735: }
1736: ftemp=fu;
1737: if (fabs(e) > tol1) {
1738: r=(x-w)*(fx-fv);
1739: q=(x-v)*(fx-fw);
1740: p=(x-v)*q-(x-w)*r;
1741: q=2.0*(q-r);
1742: if (q > 0.0) p = -p;
1743: q=fabs(q);
1744: etemp=e;
1745: e=d;
1746: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1747: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1748: else {
1.224 brouard 1749: d=p/q;
1750: u=x+d;
1751: if (u-a < tol2 || b-u < tol2)
1752: d=SIGN(tol1,xm-x);
1.126 brouard 1753: }
1754: } else {
1755: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1756: }
1757: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1758: fu=(*f)(u);
1759: if (fu <= fx) {
1760: if (u >= x) a=x; else b=x;
1761: SHFT(v,w,x,u)
1.183 brouard 1762: SHFT(fv,fw,fx,fu)
1763: } else {
1764: if (u < x) a=u; else b=u;
1765: if (fu <= fw || w == x) {
1.224 brouard 1766: v=w;
1767: w=u;
1768: fv=fw;
1769: fw=fu;
1.183 brouard 1770: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1771: v=u;
1772: fv=fu;
1.183 brouard 1773: }
1774: }
1.126 brouard 1775: }
1776: nrerror("Too many iterations in brent");
1777: *xmin=x;
1778: return fx;
1779: }
1780:
1781: /****************** mnbrak ***********************/
1782:
1783: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1784: double (*func)(double))
1.183 brouard 1785: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1786: the downhill direction (defined by the function as evaluated at the initial points) and returns
1787: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1788: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1789: */
1.126 brouard 1790: double ulim,u,r,q, dum;
1791: double fu;
1.187 brouard 1792:
1793: double scale=10.;
1794: int iterscale=0;
1795:
1796: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1797: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1798:
1799:
1800: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1801: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1802: /* *bx = *ax - (*ax - *bx)/scale; */
1803: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1804: /* } */
1805:
1.126 brouard 1806: if (*fb > *fa) {
1807: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1808: SHFT(dum,*fb,*fa,dum)
1809: }
1.126 brouard 1810: *cx=(*bx)+GOLD*(*bx-*ax);
1811: *fc=(*func)(*cx);
1.183 brouard 1812: #ifdef DEBUG
1.224 brouard 1813: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1814: 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 1815: #endif
1.224 brouard 1816: 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 1817: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1818: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1819: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1820: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1821: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1822: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1823: fu=(*func)(u);
1.163 brouard 1824: #ifdef DEBUG
1825: /* f(x)=A(x-u)**2+f(u) */
1826: double A, fparabu;
1827: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1828: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1829: 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);
1830: 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 1831: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1832: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1833: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1834: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1835: #endif
1.184 brouard 1836: #ifdef MNBRAKORIGINAL
1.183 brouard 1837: #else
1.191 brouard 1838: /* if (fu > *fc) { */
1839: /* #ifdef DEBUG */
1840: /* printf("mnbrak4 fu > fc \n"); */
1841: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1842: /* #endif */
1843: /* /\* 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 *\\/ *\/ */
1844: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1845: /* dum=u; /\* Shifting c and u *\/ */
1846: /* u = *cx; */
1847: /* *cx = dum; */
1848: /* dum = fu; */
1849: /* fu = *fc; */
1850: /* *fc =dum; */
1851: /* } else { /\* end *\/ */
1852: /* #ifdef DEBUG */
1853: /* printf("mnbrak3 fu < fc \n"); */
1854: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1855: /* #endif */
1856: /* dum=u; /\* Shifting c and u *\/ */
1857: /* u = *cx; */
1858: /* *cx = dum; */
1859: /* dum = fu; */
1860: /* fu = *fc; */
1861: /* *fc =dum; */
1862: /* } */
1.224 brouard 1863: #ifdef DEBUGMNBRAK
1864: double A, fparabu;
1865: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1866: fparabu= *fa - A*(*ax-u)*(*ax-u);
1867: 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);
1868: 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 1869: #endif
1.191 brouard 1870: dum=u; /* Shifting c and u */
1871: u = *cx;
1872: *cx = dum;
1873: dum = fu;
1874: fu = *fc;
1875: *fc =dum;
1.183 brouard 1876: #endif
1.162 brouard 1877: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1878: #ifdef DEBUG
1.224 brouard 1879: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1880: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1881: #endif
1.126 brouard 1882: fu=(*func)(u);
1883: if (fu < *fc) {
1.183 brouard 1884: #ifdef DEBUG
1.224 brouard 1885: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1886: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1887: #endif
1888: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1889: SHFT(*fb,*fc,fu,(*func)(u))
1890: #ifdef DEBUG
1891: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1892: #endif
1893: }
1.162 brouard 1894: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1895: #ifdef DEBUG
1.224 brouard 1896: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1897: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1898: #endif
1.126 brouard 1899: u=ulim;
1900: fu=(*func)(u);
1.183 brouard 1901: } else { /* u could be left to b (if r > q parabola has a maximum) */
1902: #ifdef DEBUG
1.224 brouard 1903: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1904: 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 1905: #endif
1.126 brouard 1906: u=(*cx)+GOLD*(*cx-*bx);
1907: fu=(*func)(u);
1.224 brouard 1908: #ifdef DEBUG
1909: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1910: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1911: #endif
1.183 brouard 1912: } /* end tests */
1.126 brouard 1913: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1914: SHFT(*fa,*fb,*fc,fu)
1915: #ifdef DEBUG
1.224 brouard 1916: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1917: 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 1918: #endif
1919: } /* 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 1920: }
1921:
1922: /*************** linmin ************************/
1.162 brouard 1923: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1924: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1925: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1926: the value of func at the returned location p . This is actually all accomplished by calling the
1927: routines mnbrak and brent .*/
1.126 brouard 1928: int ncom;
1929: double *pcom,*xicom;
1930: double (*nrfunc)(double []);
1931:
1.224 brouard 1932: #ifdef LINMINORIGINAL
1.126 brouard 1933: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1934: #else
1935: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1936: #endif
1.126 brouard 1937: {
1938: double brent(double ax, double bx, double cx,
1939: double (*f)(double), double tol, double *xmin);
1940: double f1dim(double x);
1941: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1942: double *fc, double (*func)(double));
1943: int j;
1944: double xx,xmin,bx,ax;
1945: double fx,fb,fa;
1.187 brouard 1946:
1.203 brouard 1947: #ifdef LINMINORIGINAL
1948: #else
1949: double scale=10., axs, xxs; /* Scale added for infinity */
1950: #endif
1951:
1.126 brouard 1952: ncom=n;
1953: pcom=vector(1,n);
1954: xicom=vector(1,n);
1955: nrfunc=func;
1956: for (j=1;j<=n;j++) {
1957: pcom[j]=p[j];
1.202 brouard 1958: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1959: }
1.187 brouard 1960:
1.203 brouard 1961: #ifdef LINMINORIGINAL
1962: xx=1.;
1963: #else
1964: axs=0.0;
1965: xxs=1.;
1966: do{
1967: xx= xxs;
1968: #endif
1.187 brouard 1969: ax=0.;
1970: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
1971: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
1972: /* 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)) */
1973: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
1974: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
1975: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
1976: /* 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 1977: #ifdef LINMINORIGINAL
1978: #else
1979: if (fx != fx){
1.224 brouard 1980: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
1981: printf("|");
1982: fprintf(ficlog,"|");
1.203 brouard 1983: #ifdef DEBUGLINMIN
1.224 brouard 1984: 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 1985: #endif
1986: }
1.224 brouard 1987: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 1988: #endif
1989:
1.191 brouard 1990: #ifdef DEBUGLINMIN
1991: 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 1992: 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 1993: #endif
1.224 brouard 1994: #ifdef LINMINORIGINAL
1995: #else
1996: if(fb == fx){ /* Flat function in the direction */
1997: xmin=xx;
1998: *flat=1;
1999: }else{
2000: *flat=0;
2001: #endif
2002: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2003: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2004: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2005: /* fmin = f(p[j] + xmin * xi[j]) */
2006: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2007: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2008: #ifdef DEBUG
1.224 brouard 2009: 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);
2010: 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);
2011: #endif
2012: #ifdef LINMINORIGINAL
2013: #else
2014: }
1.126 brouard 2015: #endif
1.191 brouard 2016: #ifdef DEBUGLINMIN
2017: printf("linmin end ");
1.202 brouard 2018: fprintf(ficlog,"linmin end ");
1.191 brouard 2019: #endif
1.126 brouard 2020: for (j=1;j<=n;j++) {
1.203 brouard 2021: #ifdef LINMINORIGINAL
2022: xi[j] *= xmin;
2023: #else
2024: #ifdef DEBUGLINMIN
2025: if(xxs <1.0)
2026: printf(" before xi[%d]=%12.8f", j,xi[j]);
2027: #endif
2028: 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) */
2029: #ifdef DEBUGLINMIN
2030: if(xxs <1.0)
2031: 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 );
2032: #endif
2033: #endif
1.187 brouard 2034: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2035: }
1.191 brouard 2036: #ifdef DEBUGLINMIN
1.203 brouard 2037: printf("\n");
1.191 brouard 2038: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2039: 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 2040: for (j=1;j<=n;j++) {
1.202 brouard 2041: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2042: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2043: if(j % ncovmodel == 0){
1.191 brouard 2044: printf("\n");
1.202 brouard 2045: fprintf(ficlog,"\n");
2046: }
1.191 brouard 2047: }
1.203 brouard 2048: #else
1.191 brouard 2049: #endif
1.126 brouard 2050: free_vector(xicom,1,n);
2051: free_vector(pcom,1,n);
2052: }
2053:
2054:
2055: /*************** powell ************************/
1.162 brouard 2056: /*
2057: Minimization of a function func of n variables. Input consists of an initial starting point
2058: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2059: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2060: such that failure to decrease by more than this amount on one iteration signals doneness. On
2061: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2062: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2063: */
1.224 brouard 2064: #ifdef LINMINORIGINAL
2065: #else
2066: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2067: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2068: #endif
1.126 brouard 2069: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2070: double (*func)(double []))
2071: {
1.224 brouard 2072: #ifdef LINMINORIGINAL
2073: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2074: double (*func)(double []));
1.224 brouard 2075: #else
1.241 brouard 2076: void linmin(double p[], double xi[], int n, double *fret,
2077: double (*func)(double []),int *flat);
1.224 brouard 2078: #endif
1.239 brouard 2079: int i,ibig,j,jk,k;
1.126 brouard 2080: double del,t,*pt,*ptt,*xit;
1.181 brouard 2081: double directest;
1.126 brouard 2082: double fp,fptt;
2083: double *xits;
2084: int niterf, itmp;
1.224 brouard 2085: #ifdef LINMINORIGINAL
2086: #else
2087:
2088: flatdir=ivector(1,n);
2089: for (j=1;j<=n;j++) flatdir[j]=0;
2090: #endif
1.126 brouard 2091:
2092: pt=vector(1,n);
2093: ptt=vector(1,n);
2094: xit=vector(1,n);
2095: xits=vector(1,n);
2096: *fret=(*func)(p);
2097: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2098: rcurr_time = time(NULL);
1.126 brouard 2099: for (*iter=1;;++(*iter)) {
1.187 brouard 2100: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2101: ibig=0;
2102: del=0.0;
1.157 brouard 2103: rlast_time=rcurr_time;
2104: /* (void) gettimeofday(&curr_time,&tzp); */
2105: rcurr_time = time(NULL);
2106: curr_time = *localtime(&rcurr_time);
2107: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2108: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2109: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2110: for (i=1;i<=n;i++) {
1.126 brouard 2111: fprintf(ficrespow," %.12lf", p[i]);
2112: }
1.239 brouard 2113: fprintf(ficrespow,"\n");fflush(ficrespow);
2114: printf("\n#model= 1 + age ");
2115: fprintf(ficlog,"\n#model= 1 + age ");
2116: if(nagesqr==1){
1.241 brouard 2117: printf(" + age*age ");
2118: fprintf(ficlog," + age*age ");
1.239 brouard 2119: }
2120: for(j=1;j <=ncovmodel-2;j++){
2121: if(Typevar[j]==0) {
2122: printf(" + V%d ",Tvar[j]);
2123: fprintf(ficlog," + V%d ",Tvar[j]);
2124: }else if(Typevar[j]==1) {
2125: printf(" + V%d*age ",Tvar[j]);
2126: fprintf(ficlog," + V%d*age ",Tvar[j]);
2127: }else if(Typevar[j]==2) {
2128: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2129: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2130: }
2131: }
1.126 brouard 2132: printf("\n");
1.239 brouard 2133: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2134: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2135: fprintf(ficlog,"\n");
1.239 brouard 2136: for(i=1,jk=1; i <=nlstate; i++){
2137: for(k=1; k <=(nlstate+ndeath); k++){
2138: if (k != i) {
2139: printf("%d%d ",i,k);
2140: fprintf(ficlog,"%d%d ",i,k);
2141: for(j=1; j <=ncovmodel; j++){
2142: printf("%12.7f ",p[jk]);
2143: fprintf(ficlog,"%12.7f ",p[jk]);
2144: jk++;
2145: }
2146: printf("\n");
2147: fprintf(ficlog,"\n");
2148: }
2149: }
2150: }
1.241 brouard 2151: if(*iter <=3 && *iter >1){
1.157 brouard 2152: tml = *localtime(&rcurr_time);
2153: strcpy(strcurr,asctime(&tml));
2154: rforecast_time=rcurr_time;
1.126 brouard 2155: itmp = strlen(strcurr);
2156: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2157: strcurr[itmp-1]='\0';
1.162 brouard 2158: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2159: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2160: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2161: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2162: forecast_time = *localtime(&rforecast_time);
2163: strcpy(strfor,asctime(&forecast_time));
2164: itmp = strlen(strfor);
2165: if(strfor[itmp-1]=='\n')
2166: strfor[itmp-1]='\0';
2167: 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);
2168: 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 2169: }
2170: }
1.187 brouard 2171: for (i=1;i<=n;i++) { /* For each direction i */
2172: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2173: fptt=(*fret);
2174: #ifdef DEBUG
1.203 brouard 2175: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2176: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2177: #endif
1.203 brouard 2178: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2179: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2180: #ifdef LINMINORIGINAL
1.188 brouard 2181: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2182: #else
2183: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2184: flatdir[i]=flat; /* Function is vanishing in that direction i */
2185: #endif
2186: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2187: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2188: /* because that direction will be replaced unless the gain del is small */
2189: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2190: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2191: /* with the new direction. */
2192: del=fabs(fptt-(*fret));
2193: ibig=i;
1.126 brouard 2194: }
2195: #ifdef DEBUG
2196: printf("%d %.12e",i,(*fret));
2197: fprintf(ficlog,"%d %.12e",i,(*fret));
2198: for (j=1;j<=n;j++) {
1.224 brouard 2199: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2200: printf(" x(%d)=%.12e",j,xit[j]);
2201: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2202: }
2203: for(j=1;j<=n;j++) {
1.225 brouard 2204: printf(" p(%d)=%.12e",j,p[j]);
2205: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2206: }
2207: printf("\n");
2208: fprintf(ficlog,"\n");
2209: #endif
1.187 brouard 2210: } /* end loop on each direction i */
2211: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2212: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2213: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2214: for(j=1;j<=n;j++) {
1.225 brouard 2215: if(flatdir[j] >0){
2216: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2217: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2218: }
2219: /* printf("\n"); */
2220: /* fprintf(ficlog,"\n"); */
2221: }
1.182 brouard 2222: if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /* Did we reach enough precision? */
1.188 brouard 2223: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2224: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2225: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2226: /* decreased of more than 3.84 */
2227: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2228: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2229: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2230:
1.188 brouard 2231: /* Starting the program with initial values given by a former maximization will simply change */
2232: /* the scales of the directions and the directions, because the are reset to canonical directions */
2233: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2234: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2235: #ifdef DEBUG
2236: int k[2],l;
2237: k[0]=1;
2238: k[1]=-1;
2239: printf("Max: %.12e",(*func)(p));
2240: fprintf(ficlog,"Max: %.12e",(*func)(p));
2241: for (j=1;j<=n;j++) {
2242: printf(" %.12e",p[j]);
2243: fprintf(ficlog," %.12e",p[j]);
2244: }
2245: printf("\n");
2246: fprintf(ficlog,"\n");
2247: for(l=0;l<=1;l++) {
2248: for (j=1;j<=n;j++) {
2249: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2250: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2251: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2252: }
2253: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2254: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2255: }
2256: #endif
2257:
1.224 brouard 2258: #ifdef LINMINORIGINAL
2259: #else
2260: free_ivector(flatdir,1,n);
2261: #endif
1.126 brouard 2262: free_vector(xit,1,n);
2263: free_vector(xits,1,n);
2264: free_vector(ptt,1,n);
2265: free_vector(pt,1,n);
2266: return;
1.192 brouard 2267: } /* enough precision */
1.240 brouard 2268: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2269: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2270: ptt[j]=2.0*p[j]-pt[j];
2271: xit[j]=p[j]-pt[j];
2272: pt[j]=p[j];
2273: }
1.181 brouard 2274: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2275: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2276: if (*iter <=4) {
1.225 brouard 2277: #else
2278: #endif
1.224 brouard 2279: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2280: #else
1.161 brouard 2281: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2282: #endif
1.162 brouard 2283: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2284: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2285: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2286: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2287: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2288: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2289: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2290: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2291: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2292: /* Even if f3 <f1, directest can be negative and t >0 */
2293: /* mu² and del² are equal when f3=f1 */
2294: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2295: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2296: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2297: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2298: #ifdef NRCORIGINAL
2299: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2300: #else
2301: 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 2302: t= t- del*SQR(fp-fptt);
1.183 brouard 2303: #endif
1.202 brouard 2304: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2305: #ifdef DEBUG
1.181 brouard 2306: 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);
2307: 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 2308: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2309: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2310: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2311: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2312: 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);
2313: 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);
2314: #endif
1.183 brouard 2315: #ifdef POWELLORIGINAL
2316: if (t < 0.0) { /* Then we use it for new direction */
2317: #else
1.182 brouard 2318: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2319: 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 2320: 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 2321: 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 2322: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2323: }
1.181 brouard 2324: if (directest < 0.0) { /* Then we use it for new direction */
2325: #endif
1.191 brouard 2326: #ifdef DEBUGLINMIN
1.234 brouard 2327: printf("Before linmin in direction P%d-P0\n",n);
2328: for (j=1;j<=n;j++) {
2329: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2330: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2331: if(j % ncovmodel == 0){
2332: printf("\n");
2333: fprintf(ficlog,"\n");
2334: }
2335: }
1.224 brouard 2336: #endif
2337: #ifdef LINMINORIGINAL
1.234 brouard 2338: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2339: #else
1.234 brouard 2340: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2341: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2342: #endif
1.234 brouard 2343:
1.191 brouard 2344: #ifdef DEBUGLINMIN
1.234 brouard 2345: for (j=1;j<=n;j++) {
2346: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2347: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2348: if(j % ncovmodel == 0){
2349: printf("\n");
2350: fprintf(ficlog,"\n");
2351: }
2352: }
1.224 brouard 2353: #endif
1.234 brouard 2354: for (j=1;j<=n;j++) {
2355: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2356: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2357: }
1.224 brouard 2358: #ifdef LINMINORIGINAL
2359: #else
1.234 brouard 2360: for (j=1, flatd=0;j<=n;j++) {
2361: if(flatdir[j]>0)
2362: flatd++;
2363: }
2364: if(flatd >0){
2365: printf("%d flat directions\n",flatd);
2366: fprintf(ficlog,"%d flat directions\n",flatd);
2367: for (j=1;j<=n;j++) {
2368: if(flatdir[j]>0){
2369: printf("%d ",j);
2370: fprintf(ficlog,"%d ",j);
2371: }
2372: }
2373: printf("\n");
2374: fprintf(ficlog,"\n");
2375: }
1.191 brouard 2376: #endif
1.234 brouard 2377: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2378: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2379:
1.126 brouard 2380: #ifdef DEBUG
1.234 brouard 2381: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2382: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2383: for(j=1;j<=n;j++){
2384: printf(" %lf",xit[j]);
2385: fprintf(ficlog," %lf",xit[j]);
2386: }
2387: printf("\n");
2388: fprintf(ficlog,"\n");
1.126 brouard 2389: #endif
1.192 brouard 2390: } /* end of t or directest negative */
1.224 brouard 2391: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2392: #else
1.234 brouard 2393: } /* end if (fptt < fp) */
1.192 brouard 2394: #endif
1.225 brouard 2395: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2396: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2397: #else
1.224 brouard 2398: #endif
1.234 brouard 2399: } /* loop iteration */
1.126 brouard 2400: }
1.234 brouard 2401:
1.126 brouard 2402: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2403:
1.235 brouard 2404: 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 2405: {
1.235 brouard 2406: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2407: (and selected quantitative values in nres)
2408: by left multiplying the unit
1.234 brouard 2409: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2410: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2411: /* Wx is row vector: population in state 1, population in state 2, population dead */
2412: /* or prevalence in state 1, prevalence in state 2, 0 */
2413: /* newm is the matrix after multiplications, its rows are identical at a factor */
2414: /* Initial matrix pimij */
2415: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2416: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2417: /* 0, 0 , 1} */
2418: /*
2419: * and after some iteration: */
2420: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2421: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2422: /* 0, 0 , 1} */
2423: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2424: /* {0.51571254859325999, 0.4842874514067399, */
2425: /* 0.51326036147820708, 0.48673963852179264} */
2426: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2427:
1.126 brouard 2428: int i, ii,j,k;
1.209 brouard 2429: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2430: /* double **matprod2(); */ /* test */
1.218 brouard 2431: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2432: double **newm;
1.209 brouard 2433: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2434: int ncvloop=0;
1.169 brouard 2435:
1.209 brouard 2436: min=vector(1,nlstate);
2437: max=vector(1,nlstate);
2438: meandiff=vector(1,nlstate);
2439:
1.218 brouard 2440: /* Starting with matrix unity */
1.126 brouard 2441: for (ii=1;ii<=nlstate+ndeath;ii++)
2442: for (j=1;j<=nlstate+ndeath;j++){
2443: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2444: }
1.169 brouard 2445:
2446: cov[1]=1.;
2447:
2448: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2449: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2450: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2451: ncvloop++;
1.126 brouard 2452: newm=savm;
2453: /* Covariates have to be included here again */
1.138 brouard 2454: cov[2]=agefin;
1.187 brouard 2455: if(nagesqr==1)
2456: cov[3]= agefin*agefin;;
1.234 brouard 2457: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2458: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2459: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2460: /* 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 2461: }
2462: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2463: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2464: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2465: /* 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 2466: }
1.237 brouard 2467: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2468: if(Dummy[Tvar[Tage[k]]]){
2469: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2470: } else{
1.235 brouard 2471: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2472: }
1.235 brouard 2473: /* 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 2474: }
1.237 brouard 2475: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2476: /* 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 2477: if(Dummy[Tvard[k][1]==0]){
2478: if(Dummy[Tvard[k][2]==0]){
2479: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2480: }else{
2481: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2482: }
2483: }else{
2484: if(Dummy[Tvard[k][2]==0]){
2485: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2486: }else{
2487: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2488: }
2489: }
1.234 brouard 2490: }
1.138 brouard 2491: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2492: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2493: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2494: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2495: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2496: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2497: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2498:
1.126 brouard 2499: savm=oldm;
2500: oldm=newm;
1.209 brouard 2501:
2502: for(j=1; j<=nlstate; j++){
2503: max[j]=0.;
2504: min[j]=1.;
2505: }
2506: for(i=1;i<=nlstate;i++){
2507: sumnew=0;
2508: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2509: for(j=1; j<=nlstate; j++){
2510: prlim[i][j]= newm[i][j]/(1-sumnew);
2511: max[j]=FMAX(max[j],prlim[i][j]);
2512: min[j]=FMIN(min[j],prlim[i][j]);
2513: }
2514: }
2515:
1.126 brouard 2516: maxmax=0.;
1.209 brouard 2517: for(j=1; j<=nlstate; j++){
2518: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2519: maxmax=FMAX(maxmax,meandiff[j]);
2520: /* 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 2521: } /* j loop */
1.203 brouard 2522: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2523: /* 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 2524: if(maxmax < ftolpl){
1.209 brouard 2525: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2526: free_vector(min,1,nlstate);
2527: free_vector(max,1,nlstate);
2528: free_vector(meandiff,1,nlstate);
1.126 brouard 2529: return prlim;
2530: }
1.169 brouard 2531: } /* age loop */
1.208 brouard 2532: /* After some age loop it doesn't converge */
1.209 brouard 2533: 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 2534: 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 2535: /* 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); */
2536: free_vector(min,1,nlstate);
2537: free_vector(max,1,nlstate);
2538: free_vector(meandiff,1,nlstate);
1.208 brouard 2539:
1.169 brouard 2540: return prlim; /* should not reach here */
1.126 brouard 2541: }
2542:
1.217 brouard 2543:
2544: /**** Back Prevalence limit (stable or period prevalence) ****************/
2545:
1.218 brouard 2546: /* 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) */
2547: /* 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 2548: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2549: {
1.218 brouard 2550: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2551: matrix by transitions matrix until convergence is reached with precision ftolpl */
2552: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2553: /* Wx is row vector: population in state 1, population in state 2, population dead */
2554: /* or prevalence in state 1, prevalence in state 2, 0 */
2555: /* newm is the matrix after multiplications, its rows are identical at a factor */
2556: /* Initial matrix pimij */
2557: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2558: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2559: /* 0, 0 , 1} */
2560: /*
2561: * and after some iteration: */
2562: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2563: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2564: /* 0, 0 , 1} */
2565: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2566: /* {0.51571254859325999, 0.4842874514067399, */
2567: /* 0.51326036147820708, 0.48673963852179264} */
2568: /* If we start from prlim again, prlim tends to a constant matrix */
2569:
2570: int i, ii,j,k;
2571: double *min, *max, *meandiff, maxmax,sumnew=0.;
2572: /* double **matprod2(); */ /* test */
2573: double **out, cov[NCOVMAX+1], **bmij();
2574: double **newm;
1.218 brouard 2575: double **dnewm, **doldm, **dsavm; /* for use */
2576: double **oldm, **savm; /* for use */
2577:
1.217 brouard 2578: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2579: int ncvloop=0;
2580:
2581: min=vector(1,nlstate);
2582: max=vector(1,nlstate);
2583: meandiff=vector(1,nlstate);
2584:
1.218 brouard 2585: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2586: oldm=oldms; savm=savms;
2587:
2588: /* Starting with matrix unity */
2589: for (ii=1;ii<=nlstate+ndeath;ii++)
2590: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2591: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2592: }
2593:
2594: cov[1]=1.;
2595:
2596: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2597: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2598: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2599: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2600: ncvloop++;
1.218 brouard 2601: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2602: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2603: /* Covariates have to be included here again */
2604: cov[2]=agefin;
2605: if(nagesqr==1)
2606: cov[3]= agefin*agefin;;
1.242 ! brouard 2607: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
! 2608: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
! 2609: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
! 2610: /* 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)); */
! 2611: }
! 2612: /* for (k=1; k<=cptcovn;k++) { */
! 2613: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
! 2614: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
! 2615: /* /\* 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])]); *\/ */
! 2616: /* } */
! 2617: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
! 2618: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
! 2619: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
! 2620: /* 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]); */
! 2621: }
! 2622: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
! 2623: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
! 2624: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
! 2625: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
! 2626: for (k=1; k<=cptcovage;k++){ /* For product with age */
! 2627: if(Dummy[Tvar[Tage[k]]]){
! 2628: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
! 2629: } else{
! 2630: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
! 2631: }
! 2632: /* 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]); */
! 2633: }
! 2634: for (k=1; k<=cptcovprod;k++){ /* For product without age */
! 2635: /* 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]); */
! 2636: if(Dummy[Tvard[k][1]==0]){
! 2637: if(Dummy[Tvard[k][2]==0]){
! 2638: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
! 2639: }else{
! 2640: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
! 2641: }
! 2642: }else{
! 2643: if(Dummy[Tvard[k][2]==0]){
! 2644: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
! 2645: }else{
! 2646: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
! 2647: }
! 2648: }
1.217 brouard 2649: }
2650:
2651: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2652: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2653: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2654: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2655: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2656: /* ij should be linked to the correct index of cov */
2657: /* age and covariate values ij are in 'cov', but we need to pass
2658: * ij for the observed prevalence at age and status and covariate
2659: * number: prevacurrent[(int)agefin][ii][ij]
2660: */
2661: /* 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 *\/ */
2662: /* 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 *\/ */
2663: 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 2664: savm=oldm;
2665: oldm=newm;
2666: for(j=1; j<=nlstate; j++){
2667: max[j]=0.;
2668: min[j]=1.;
2669: }
2670: for(j=1; j<=nlstate; j++){
2671: for(i=1;i<=nlstate;i++){
1.234 brouard 2672: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2673: bprlim[i][j]= newm[i][j];
2674: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2675: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2676: }
2677: }
1.218 brouard 2678:
1.217 brouard 2679: maxmax=0.;
2680: for(i=1; i<=nlstate; i++){
2681: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2682: maxmax=FMAX(maxmax,meandiff[i]);
2683: /* 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); */
2684: } /* j loop */
2685: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2686: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2687: if(maxmax < ftolpl){
1.220 brouard 2688: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2689: free_vector(min,1,nlstate);
2690: free_vector(max,1,nlstate);
2691: free_vector(meandiff,1,nlstate);
2692: return bprlim;
2693: }
2694: } /* age loop */
2695: /* After some age loop it doesn't converge */
2696: 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'. \n\
2697: 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);
2698: /* 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); */
2699: free_vector(min,1,nlstate);
2700: free_vector(max,1,nlstate);
2701: free_vector(meandiff,1,nlstate);
2702:
2703: return bprlim; /* should not reach here */
2704: }
2705:
1.126 brouard 2706: /*************** transition probabilities ***************/
2707:
2708: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2709: {
1.138 brouard 2710: /* According to parameters values stored in x and the covariate's values stored in cov,
2711: computes the probability to be observed in state j being in state i by appying the
2712: model to the ncovmodel covariates (including constant and age).
2713: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2714: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2715: ncth covariate in the global vector x is given by the formula:
2716: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2717: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2718: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2719: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2720: Outputs ps[i][j] the probability to be observed in j being in j according to
2721: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2722: */
2723: double s1, lnpijopii;
1.126 brouard 2724: /*double t34;*/
1.164 brouard 2725: int i,j, nc, ii, jj;
1.126 brouard 2726:
1.223 brouard 2727: for(i=1; i<= nlstate; i++){
2728: for(j=1; j<i;j++){
2729: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2730: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2731: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2732: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2733: }
2734: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2735: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2736: }
2737: for(j=i+1; j<=nlstate+ndeath;j++){
2738: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2739: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2740: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2741: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2742: }
2743: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2744: }
2745: }
1.218 brouard 2746:
1.223 brouard 2747: for(i=1; i<= nlstate; i++){
2748: s1=0;
2749: for(j=1; j<i; j++){
2750: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2751: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2752: }
2753: for(j=i+1; j<=nlstate+ndeath; j++){
2754: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2755: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2756: }
2757: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2758: ps[i][i]=1./(s1+1.);
2759: /* Computing other pijs */
2760: for(j=1; j<i; j++)
2761: ps[i][j]= exp(ps[i][j])*ps[i][i];
2762: for(j=i+1; j<=nlstate+ndeath; j++)
2763: ps[i][j]= exp(ps[i][j])*ps[i][i];
2764: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2765: } /* end i */
1.218 brouard 2766:
1.223 brouard 2767: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2768: for(jj=1; jj<= nlstate+ndeath; jj++){
2769: ps[ii][jj]=0;
2770: ps[ii][ii]=1;
2771: }
2772: }
1.218 brouard 2773:
2774:
1.223 brouard 2775: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2776: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2777: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2778: /* } */
2779: /* printf("\n "); */
2780: /* } */
2781: /* printf("\n ");printf("%lf ",cov[2]);*/
2782: /*
2783: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2784: goto end;*/
1.223 brouard 2785: return ps;
1.126 brouard 2786: }
2787:
1.218 brouard 2788: /*************** backward transition probabilities ***************/
2789:
2790: /* 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 ) */
2791: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2792: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2793: {
1.222 brouard 2794: /* Computes the backward probability at age agefin and covariate ij
2795: * and returns in **ps as well as **bmij.
2796: */
1.218 brouard 2797: int i, ii, j,k;
1.222 brouard 2798:
2799: double **out, **pmij();
2800: double sumnew=0.;
1.218 brouard 2801: double agefin;
1.222 brouard 2802:
2803: double **dnewm, **dsavm, **doldm;
2804: double **bbmij;
2805:
1.218 brouard 2806: doldm=ddoldms; /* global pointers */
1.222 brouard 2807: dnewm=ddnewms;
2808: dsavm=ddsavms;
2809:
2810: agefin=cov[2];
2811: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2812: the observed prevalence (with this covariate ij) */
2813: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2814: /* We do have the matrix Px in savm and we need pij */
2815: for (j=1;j<=nlstate+ndeath;j++){
2816: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2817: for (ii=1;ii<=nlstate;ii++){
2818: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2819: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2820: for (ii=1;ii<=nlstate+ndeath;ii++){
2821: if(sumnew >= 1.e-10){
2822: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2823: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2824: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2825: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2826: /* }else */
2827: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2828: }else{
1.242 ! brouard 2829: ;
! 2830: /* 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 2831: }
2832: } /*End ii */
2833: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2834: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2835: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2836: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2837: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2838: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2839: /* left Product of this matrix by diag matrix of prevalences (savm) */
2840: for (j=1;j<=nlstate+ndeath;j++){
2841: for (ii=1;ii<=nlstate+ndeath;ii++){
2842: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2843: }
2844: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2845: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2846: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2847: /* end bmij */
2848: return ps;
1.218 brouard 2849: }
1.217 brouard 2850: /*************** transition probabilities ***************/
2851:
1.218 brouard 2852: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2853: {
2854: /* According to parameters values stored in x and the covariate's values stored in cov,
2855: computes the probability to be observed in state j being in state i by appying the
2856: model to the ncovmodel covariates (including constant and age).
2857: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2858: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2859: ncth covariate in the global vector x is given by the formula:
2860: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2861: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2862: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2863: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2864: Outputs ps[i][j] the probability to be observed in j being in j according to
2865: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2866: */
2867: double s1, lnpijopii;
2868: /*double t34;*/
2869: int i,j, nc, ii, jj;
2870:
1.234 brouard 2871: for(i=1; i<= nlstate; i++){
2872: for(j=1; j<i;j++){
2873: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2874: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2875: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2876: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2877: }
2878: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2879: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2880: }
2881: for(j=i+1; j<=nlstate+ndeath;j++){
2882: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2883: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2884: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2885: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2886: }
2887: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2888: }
2889: }
2890:
2891: for(i=1; i<= nlstate; i++){
2892: s1=0;
2893: for(j=1; j<i; j++){
2894: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2895: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2896: }
2897: for(j=i+1; j<=nlstate+ndeath; j++){
2898: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2899: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2900: }
2901: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2902: ps[i][i]=1./(s1+1.);
2903: /* Computing other pijs */
2904: for(j=1; j<i; j++)
2905: ps[i][j]= exp(ps[i][j])*ps[i][i];
2906: for(j=i+1; j<=nlstate+ndeath; j++)
2907: ps[i][j]= exp(ps[i][j])*ps[i][i];
2908: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2909: } /* end i */
2910:
2911: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2912: for(jj=1; jj<= nlstate+ndeath; jj++){
2913: ps[ii][jj]=0;
2914: ps[ii][ii]=1;
2915: }
2916: }
2917: /* Added for backcast */ /* Transposed matrix too */
2918: for(jj=1; jj<= nlstate+ndeath; jj++){
2919: s1=0.;
2920: for(ii=1; ii<= nlstate+ndeath; ii++){
2921: s1+=ps[ii][jj];
2922: }
2923: for(ii=1; ii<= nlstate; ii++){
2924: ps[ii][jj]=ps[ii][jj]/s1;
2925: }
2926: }
2927: /* Transposition */
2928: for(jj=1; jj<= nlstate+ndeath; jj++){
2929: for(ii=jj; ii<= nlstate+ndeath; ii++){
2930: s1=ps[ii][jj];
2931: ps[ii][jj]=ps[jj][ii];
2932: ps[jj][ii]=s1;
2933: }
2934: }
2935: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2936: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2937: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2938: /* } */
2939: /* printf("\n "); */
2940: /* } */
2941: /* printf("\n ");printf("%lf ",cov[2]);*/
2942: /*
2943: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2944: goto end;*/
2945: return ps;
1.217 brouard 2946: }
2947:
2948:
1.126 brouard 2949: /**************** Product of 2 matrices ******************/
2950:
1.145 brouard 2951: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2952: {
2953: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2954: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2955: /* in, b, out are matrice of pointers which should have been initialized
2956: before: only the contents of out is modified. The function returns
2957: a pointer to pointers identical to out */
1.145 brouard 2958: int i, j, k;
1.126 brouard 2959: for(i=nrl; i<= nrh; i++)
1.145 brouard 2960: for(k=ncolol; k<=ncoloh; k++){
2961: out[i][k]=0.;
2962: for(j=ncl; j<=nch; j++)
2963: out[i][k] +=in[i][j]*b[j][k];
2964: }
1.126 brouard 2965: return out;
2966: }
2967:
2968:
2969: /************* Higher Matrix Product ***************/
2970:
1.235 brouard 2971: 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 2972: {
1.218 brouard 2973: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 2974: 'nhstepm*hstepm*stepm' months (i.e. until
2975: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
2976: nhstepm*hstepm matrices.
2977: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
2978: (typically every 2 years instead of every month which is too big
2979: for the memory).
2980: Model is determined by parameters x and covariates have to be
2981: included manually here.
2982:
2983: */
2984:
2985: int i, j, d, h, k;
1.131 brouard 2986: double **out, cov[NCOVMAX+1];
1.126 brouard 2987: double **newm;
1.187 brouard 2988: double agexact;
1.214 brouard 2989: double agebegin, ageend;
1.126 brouard 2990:
2991: /* Hstepm could be zero and should return the unit matrix */
2992: for (i=1;i<=nlstate+ndeath;i++)
2993: for (j=1;j<=nlstate+ndeath;j++){
2994: oldm[i][j]=(i==j ? 1.0 : 0.0);
2995: po[i][j][0]=(i==j ? 1.0 : 0.0);
2996: }
2997: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2998: for(h=1; h <=nhstepm; h++){
2999: for(d=1; d <=hstepm; d++){
3000: newm=savm;
3001: /* Covariates have to be included here again */
3002: cov[1]=1.;
1.214 brouard 3003: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3004: cov[2]=agexact;
3005: if(nagesqr==1)
1.227 brouard 3006: cov[3]= agexact*agexact;
1.235 brouard 3007: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3008: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3009: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3010: /* 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)); */
3011: }
3012: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3013: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3014: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3015: /* 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]); */
3016: }
3017: for (k=1; k<=cptcovage;k++){
3018: if(Dummy[Tvar[Tage[k]]]){
3019: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3020: } else{
3021: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3022: }
3023: /* 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]); */
3024: }
3025: for (k=1; k<=cptcovprod;k++){ /* */
3026: /* 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]); */
3027: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3028: }
3029: /* for (k=1; k<=cptcovn;k++) */
3030: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3031: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3032: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3033: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3034: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3035:
3036:
1.126 brouard 3037: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3038: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3039: /* right multiplication of oldm by the current matrix */
1.126 brouard 3040: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3041: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3042: /* if((int)age == 70){ */
3043: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3044: /* for(i=1; i<=nlstate+ndeath; i++) { */
3045: /* printf("%d pmmij ",i); */
3046: /* for(j=1;j<=nlstate+ndeath;j++) { */
3047: /* printf("%f ",pmmij[i][j]); */
3048: /* } */
3049: /* printf(" oldm "); */
3050: /* for(j=1;j<=nlstate+ndeath;j++) { */
3051: /* printf("%f ",oldm[i][j]); */
3052: /* } */
3053: /* printf("\n"); */
3054: /* } */
3055: /* } */
1.126 brouard 3056: savm=oldm;
3057: oldm=newm;
3058: }
3059: for(i=1; i<=nlstate+ndeath; i++)
3060: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3061: po[i][j][h]=newm[i][j];
3062: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3063: }
1.128 brouard 3064: /*printf("h=%d ",h);*/
1.126 brouard 3065: } /* end h */
1.218 brouard 3066: /* printf("\n H=%d \n",h); */
1.126 brouard 3067: return po;
3068: }
3069:
1.217 brouard 3070: /************* Higher Back Matrix Product ***************/
1.218 brouard 3071: /* 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 3072: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3073: {
1.218 brouard 3074: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3075: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3076: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3077: nhstepm*hstepm matrices.
3078: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3079: (typically every 2 years instead of every month which is too big
1.217 brouard 3080: for the memory).
1.218 brouard 3081: Model is determined by parameters x and covariates have to be
3082: included manually here.
1.217 brouard 3083:
1.222 brouard 3084: */
1.217 brouard 3085:
3086: int i, j, d, h, k;
3087: double **out, cov[NCOVMAX+1];
3088: double **newm;
3089: double agexact;
3090: double agebegin, ageend;
1.222 brouard 3091: double **oldm, **savm;
1.217 brouard 3092:
1.222 brouard 3093: oldm=oldms;savm=savms;
1.217 brouard 3094: /* Hstepm could be zero and should return the unit matrix */
3095: for (i=1;i<=nlstate+ndeath;i++)
3096: for (j=1;j<=nlstate+ndeath;j++){
3097: oldm[i][j]=(i==j ? 1.0 : 0.0);
3098: po[i][j][0]=(i==j ? 1.0 : 0.0);
3099: }
3100: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3101: for(h=1; h <=nhstepm; h++){
3102: for(d=1; d <=hstepm; d++){
3103: newm=savm;
3104: /* Covariates have to be included here again */
3105: cov[1]=1.;
3106: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3107: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3108: cov[2]=agexact;
3109: if(nagesqr==1)
1.222 brouard 3110: cov[3]= agexact*agexact;
1.218 brouard 3111: for (k=1; k<=cptcovn;k++)
1.222 brouard 3112: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3113: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3114: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3115: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3116: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3117: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3118: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3119: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3120: /* 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 3121:
3122:
1.217 brouard 3123: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3124: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3125: /* Careful transposed matrix */
1.222 brouard 3126: /* age is in cov[2] */
1.218 brouard 3127: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3128: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3129: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3130: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3131: /* if((int)age == 70){ */
3132: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3133: /* for(i=1; i<=nlstate+ndeath; i++) { */
3134: /* printf("%d pmmij ",i); */
3135: /* for(j=1;j<=nlstate+ndeath;j++) { */
3136: /* printf("%f ",pmmij[i][j]); */
3137: /* } */
3138: /* printf(" oldm "); */
3139: /* for(j=1;j<=nlstate+ndeath;j++) { */
3140: /* printf("%f ",oldm[i][j]); */
3141: /* } */
3142: /* printf("\n"); */
3143: /* } */
3144: /* } */
3145: savm=oldm;
3146: oldm=newm;
3147: }
3148: for(i=1; i<=nlstate+ndeath; i++)
3149: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3150: po[i][j][h]=newm[i][j];
3151: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3152: }
3153: /*printf("h=%d ",h);*/
3154: } /* end h */
1.222 brouard 3155: /* printf("\n H=%d \n",h); */
1.217 brouard 3156: return po;
3157: }
3158:
3159:
1.162 brouard 3160: #ifdef NLOPT
3161: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3162: double fret;
3163: double *xt;
3164: int j;
3165: myfunc_data *d2 = (myfunc_data *) pd;
3166: /* xt = (p1-1); */
3167: xt=vector(1,n);
3168: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3169:
3170: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3171: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3172: printf("Function = %.12lf ",fret);
3173: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3174: printf("\n");
3175: free_vector(xt,1,n);
3176: return fret;
3177: }
3178: #endif
1.126 brouard 3179:
3180: /*************** log-likelihood *************/
3181: double func( double *x)
3182: {
1.226 brouard 3183: int i, ii, j, k, mi, d, kk;
3184: int ioffset=0;
3185: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3186: double **out;
3187: double lli; /* Individual log likelihood */
3188: int s1, s2;
1.228 brouard 3189: 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 3190: double bbh, survp;
3191: long ipmx;
3192: double agexact;
3193: /*extern weight */
3194: /* We are differentiating ll according to initial status */
3195: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3196: /*for(i=1;i<imx;i++)
3197: printf(" %d\n",s[4][i]);
3198: */
1.162 brouard 3199:
1.226 brouard 3200: ++countcallfunc;
1.162 brouard 3201:
1.226 brouard 3202: cov[1]=1.;
1.126 brouard 3203:
1.226 brouard 3204: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3205: ioffset=0;
1.226 brouard 3206: if(mle==1){
3207: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3208: /* Computes the values of the ncovmodel covariates of the model
3209: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3210: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3211: to be observed in j being in i according to the model.
3212: */
3213: ioffset=2+nagesqr+cptcovage;
1.233 brouard 3214: /* Fixed */
1.234 brouard 3215: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3216: 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)*/
3217: }
1.226 brouard 3218: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3219: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3220: has been calculated etc */
3221: /* For an individual i, wav[i] gives the number of effective waves */
3222: /* We compute the contribution to Likelihood of each effective transition
3223: mw[mi][i] is real wave of the mi th effectve wave */
3224: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3225: s2=s[mw[mi+1][i]][i];
3226: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3227: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3228: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3229: */
3230: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3231: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 ! brouard 3232: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
! 3233: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3234: }
3235: for (ii=1;ii<=nlstate+ndeath;ii++)
3236: for (j=1;j<=nlstate+ndeath;j++){
3237: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3238: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3239: }
3240: for(d=0; d<dh[mi][i]; d++){
3241: newm=savm;
3242: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3243: cov[2]=agexact;
3244: if(nagesqr==1)
3245: cov[3]= agexact*agexact; /* Should be changed here */
3246: for (kk=1; kk<=cptcovage;kk++) {
1.242 ! brouard 3247: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3248: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 ! brouard 3249: else
! 3250: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3251: }
3252: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3253: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3254: savm=oldm;
3255: oldm=newm;
3256: } /* end mult */
3257:
3258: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3259: /* But now since version 0.9 we anticipate for bias at large stepm.
3260: * If stepm is larger than one month (smallest stepm) and if the exact delay
3261: * (in months) between two waves is not a multiple of stepm, we rounded to
3262: * the nearest (and in case of equal distance, to the lowest) interval but now
3263: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3264: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3265: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3266: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3267: * -stepm/2 to stepm/2 .
3268: * For stepm=1 the results are the same as for previous versions of Imach.
3269: * For stepm > 1 the results are less biased than in previous versions.
3270: */
1.234 brouard 3271: s1=s[mw[mi][i]][i];
3272: s2=s[mw[mi+1][i]][i];
3273: bbh=(double)bh[mi][i]/(double)stepm;
3274: /* bias bh is positive if real duration
3275: * is higher than the multiple of stepm and negative otherwise.
3276: */
3277: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3278: if( s2 > nlstate){
3279: /* i.e. if s2 is a death state and if the date of death is known
3280: then the contribution to the likelihood is the probability to
3281: die between last step unit time and current step unit time,
3282: which is also equal to probability to die before dh
3283: minus probability to die before dh-stepm .
3284: In version up to 0.92 likelihood was computed
3285: as if date of death was unknown. Death was treated as any other
3286: health state: the date of the interview describes the actual state
3287: and not the date of a change in health state. The former idea was
3288: to consider that at each interview the state was recorded
3289: (healthy, disable or death) and IMaCh was corrected; but when we
3290: introduced the exact date of death then we should have modified
3291: the contribution of an exact death to the likelihood. This new
3292: contribution is smaller and very dependent of the step unit
3293: stepm. It is no more the probability to die between last interview
3294: and month of death but the probability to survive from last
3295: interview up to one month before death multiplied by the
3296: probability to die within a month. Thanks to Chris
3297: Jackson for correcting this bug. Former versions increased
3298: mortality artificially. The bad side is that we add another loop
3299: which slows down the processing. The difference can be up to 10%
3300: lower mortality.
3301: */
3302: /* If, at the beginning of the maximization mostly, the
3303: cumulative probability or probability to be dead is
3304: constant (ie = 1) over time d, the difference is equal to
3305: 0. out[s1][3] = savm[s1][3]: probability, being at state
3306: s1 at precedent wave, to be dead a month before current
3307: wave is equal to probability, being at state s1 at
3308: precedent wave, to be dead at mont of the current
3309: wave. Then the observed probability (that this person died)
3310: is null according to current estimated parameter. In fact,
3311: it should be very low but not zero otherwise the log go to
3312: infinity.
3313: */
1.183 brouard 3314: /* #ifdef INFINITYORIGINAL */
3315: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3316: /* #else */
3317: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3318: /* lli=log(mytinydouble); */
3319: /* else */
3320: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3321: /* #endif */
1.226 brouard 3322: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3323:
1.226 brouard 3324: } else if ( s2==-1 ) { /* alive */
3325: for (j=1,survp=0. ; j<=nlstate; j++)
3326: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3327: /*survp += out[s1][j]; */
3328: lli= log(survp);
3329: }
3330: else if (s2==-4) {
3331: for (j=3,survp=0. ; j<=nlstate; j++)
3332: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3333: lli= log(survp);
3334: }
3335: else if (s2==-5) {
3336: for (j=1,survp=0. ; j<=2; j++)
3337: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3338: lli= log(survp);
3339: }
3340: else{
3341: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3342: /* 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 */
3343: }
3344: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3345: /*if(lli ==000.0)*/
3346: /*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); */
3347: ipmx +=1;
3348: sw += weight[i];
3349: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3350: /* if (lli < log(mytinydouble)){ */
3351: /* 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); */
3352: /* 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]); */
3353: /* } */
3354: } /* end of wave */
3355: } /* end of individual */
3356: } else if(mle==2){
3357: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3358: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3359: for(mi=1; mi<= wav[i]-1; mi++){
3360: for (ii=1;ii<=nlstate+ndeath;ii++)
3361: for (j=1;j<=nlstate+ndeath;j++){
3362: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3363: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3364: }
3365: for(d=0; d<=dh[mi][i]; d++){
3366: newm=savm;
3367: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3368: cov[2]=agexact;
3369: if(nagesqr==1)
3370: cov[3]= agexact*agexact;
3371: for (kk=1; kk<=cptcovage;kk++) {
3372: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3373: }
3374: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3375: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3376: savm=oldm;
3377: oldm=newm;
3378: } /* end mult */
3379:
3380: s1=s[mw[mi][i]][i];
3381: s2=s[mw[mi+1][i]][i];
3382: bbh=(double)bh[mi][i]/(double)stepm;
3383: 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 */
3384: ipmx +=1;
3385: sw += weight[i];
3386: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3387: } /* end of wave */
3388: } /* end of individual */
3389: } else if(mle==3){ /* exponential inter-extrapolation */
3390: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3391: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3392: for(mi=1; mi<= wav[i]-1; mi++){
3393: for (ii=1;ii<=nlstate+ndeath;ii++)
3394: for (j=1;j<=nlstate+ndeath;j++){
3395: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3396: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3397: }
3398: for(d=0; d<dh[mi][i]; d++){
3399: newm=savm;
3400: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3401: cov[2]=agexact;
3402: if(nagesqr==1)
3403: cov[3]= agexact*agexact;
3404: for (kk=1; kk<=cptcovage;kk++) {
3405: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3406: }
3407: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3408: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3409: savm=oldm;
3410: oldm=newm;
3411: } /* end mult */
3412:
3413: s1=s[mw[mi][i]][i];
3414: s2=s[mw[mi+1][i]][i];
3415: bbh=(double)bh[mi][i]/(double)stepm;
3416: 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 */
3417: ipmx +=1;
3418: sw += weight[i];
3419: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3420: } /* end of wave */
3421: } /* end of individual */
3422: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3423: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3424: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3425: for(mi=1; mi<= wav[i]-1; mi++){
3426: for (ii=1;ii<=nlstate+ndeath;ii++)
3427: for (j=1;j<=nlstate+ndeath;j++){
3428: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3429: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3430: }
3431: for(d=0; d<dh[mi][i]; d++){
3432: newm=savm;
3433: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3434: cov[2]=agexact;
3435: if(nagesqr==1)
3436: cov[3]= agexact*agexact;
3437: for (kk=1; kk<=cptcovage;kk++) {
3438: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3439: }
1.126 brouard 3440:
1.226 brouard 3441: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3442: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3443: savm=oldm;
3444: oldm=newm;
3445: } /* end mult */
3446:
3447: s1=s[mw[mi][i]][i];
3448: s2=s[mw[mi+1][i]][i];
3449: if( s2 > nlstate){
3450: lli=log(out[s1][s2] - savm[s1][s2]);
3451: } else if ( s2==-1 ) { /* alive */
3452: for (j=1,survp=0. ; j<=nlstate; j++)
3453: survp += out[s1][j];
3454: lli= log(survp);
3455: }else{
3456: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3457: }
3458: ipmx +=1;
3459: sw += weight[i];
3460: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3461: /* 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 3462: } /* end of wave */
3463: } /* end of individual */
3464: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3465: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3466: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3467: for(mi=1; mi<= wav[i]-1; mi++){
3468: for (ii=1;ii<=nlstate+ndeath;ii++)
3469: for (j=1;j<=nlstate+ndeath;j++){
3470: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3471: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3472: }
3473: for(d=0; d<dh[mi][i]; d++){
3474: newm=savm;
3475: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3476: cov[2]=agexact;
3477: if(nagesqr==1)
3478: cov[3]= agexact*agexact;
3479: for (kk=1; kk<=cptcovage;kk++) {
3480: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3481: }
1.126 brouard 3482:
1.226 brouard 3483: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3484: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3485: savm=oldm;
3486: oldm=newm;
3487: } /* end mult */
3488:
3489: s1=s[mw[mi][i]][i];
3490: s2=s[mw[mi+1][i]][i];
3491: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3492: ipmx +=1;
3493: sw += weight[i];
3494: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3495: /*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]);*/
3496: } /* end of wave */
3497: } /* end of individual */
3498: } /* End of if */
3499: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3500: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3501: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3502: return -l;
1.126 brouard 3503: }
3504:
3505: /*************** log-likelihood *************/
3506: double funcone( double *x)
3507: {
1.228 brouard 3508: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3509: int i, ii, j, k, mi, d, kk;
1.228 brouard 3510: int ioffset=0;
1.131 brouard 3511: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3512: double **out;
3513: double lli; /* Individual log likelihood */
3514: double llt;
3515: int s1, s2;
1.228 brouard 3516: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3517:
1.126 brouard 3518: double bbh, survp;
1.187 brouard 3519: double agexact;
1.214 brouard 3520: double agebegin, ageend;
1.126 brouard 3521: /*extern weight */
3522: /* We are differentiating ll according to initial status */
3523: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3524: /*for(i=1;i<imx;i++)
3525: printf(" %d\n",s[4][i]);
3526: */
3527: cov[1]=1.;
3528:
3529: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3530: ioffset=0;
3531: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.225 brouard 3532: ioffset=2+nagesqr+cptcovage;
1.232 brouard 3533: /* Fixed */
1.224 brouard 3534: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3535: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3536: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3537: 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)*/
3538: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3539: /* cov[2+6]=covar[Tvar[6]][i]; */
3540: /* cov[2+6]=covar[2][i]; V2 */
3541: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3542: /* cov[2+7]=covar[Tvar[7]][i]; */
3543: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3544: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3545: /* cov[2+9]=covar[Tvar[9]][i]; */
3546: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3547: }
1.232 brouard 3548: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3549: /* 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?)*\/ */
3550: /* } */
1.231 brouard 3551: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3552: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3553: /* } */
1.225 brouard 3554:
1.233 brouard 3555:
3556: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3557: /* Wave varying (but not age varying) */
3558: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 ! brouard 3559: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
! 3560: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
! 3561: }
1.232 brouard 3562: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 ! brouard 3563: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
! 3564: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
! 3565: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
! 3566: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
! 3567: /* 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 3568: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 ! brouard 3569: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
! 3570: /* /\* 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]); *\/ */
! 3571: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3572: /* } */
1.126 brouard 3573: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 ! brouard 3574: for (j=1;j<=nlstate+ndeath;j++){
! 3575: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
! 3576: savm[ii][j]=(ii==j ? 1.0 : 0.0);
! 3577: }
1.214 brouard 3578:
3579: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3580: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3581: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.242 ! brouard 3582: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
! 3583: and mw[mi+1][i]. dh depends on stepm.*/
! 3584: newm=savm;
! 3585: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
! 3586: cov[2]=agexact;
! 3587: if(nagesqr==1)
! 3588: cov[3]= agexact*agexact;
! 3589: for (kk=1; kk<=cptcovage;kk++) {
! 3590: if(!FixedV[Tvar[Tage[kk]]])
! 3591: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
! 3592: else
! 3593: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
! 3594: }
! 3595: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
! 3596: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
! 3597: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
! 3598: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
! 3599: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
! 3600: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
! 3601: savm=oldm;
! 3602: oldm=newm;
1.126 brouard 3603: } /* end mult */
3604:
3605: s1=s[mw[mi][i]][i];
3606: s2=s[mw[mi+1][i]][i];
1.217 brouard 3607: /* if(s2==-1){ */
3608: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3609: /* /\* exit(1); *\/ */
3610: /* } */
1.126 brouard 3611: bbh=(double)bh[mi][i]/(double)stepm;
3612: /* bias is positive if real duration
3613: * is higher than the multiple of stepm and negative otherwise.
3614: */
3615: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 ! brouard 3616: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3617: } else if ( s2==-1 ) { /* alive */
1.242 ! brouard 3618: for (j=1,survp=0. ; j<=nlstate; j++)
! 3619: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
! 3620: lli= log(survp);
1.126 brouard 3621: }else if (mle==1){
1.242 ! brouard 3622: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3623: } else if(mle==2){
1.242 ! brouard 3624: 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 3625: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 ! brouard 3626: 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 3627: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 ! brouard 3628: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3629: } else{ /* mle=0 back to 1 */
1.242 ! brouard 3630: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
! 3631: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3632: } /* End of if */
3633: ipmx +=1;
3634: sw += weight[i];
3635: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3636: /*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 3637: if(globpr){
1.242 ! brouard 3638: fprintf(ficresilk,"%9ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3639: %11.6f %11.6f %11.6f ", \
1.242 ! brouard 3640: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
! 3641: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
! 3642: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
! 3643: llt +=ll[k]*gipmx/gsw;
! 3644: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
! 3645: }
! 3646: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3647: }
1.232 brouard 3648: } /* end of wave */
3649: } /* end of individual */
3650: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3651: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3652: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3653: if(globpr==0){ /* First time we count the contributions and weights */
3654: gipmx=ipmx;
3655: gsw=sw;
3656: }
3657: return -l;
1.126 brouard 3658: }
3659:
3660:
3661: /*************** function likelione ***********/
3662: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3663: {
3664: /* This routine should help understanding what is done with
3665: the selection of individuals/waves and
3666: to check the exact contribution to the likelihood.
3667: Plotting could be done.
3668: */
3669: int k;
3670:
3671: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3672: strcpy(fileresilk,"ILK_");
1.202 brouard 3673: strcat(fileresilk,fileresu);
1.126 brouard 3674: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3675: printf("Problem with resultfile: %s\n", fileresilk);
3676: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3677: }
1.214 brouard 3678: 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");
3679: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3680: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3681: for(k=1; k<=nlstate; k++)
3682: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3683: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3684: }
3685:
3686: *fretone=(*funcone)(p);
3687: if(*globpri !=0){
3688: fclose(ficresilk);
1.205 brouard 3689: if (mle ==0)
3690: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3691: else if(mle >=1)
3692: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3693: 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 3694:
1.208 brouard 3695:
3696: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3697: 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 3698: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3699: }
1.207 brouard 3700: 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 3701: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3702: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3703: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3704: fflush(fichtm);
1.205 brouard 3705: }
1.126 brouard 3706: return;
3707: }
3708:
3709:
3710: /*********** Maximum Likelihood Estimation ***************/
3711:
3712: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3713: {
1.165 brouard 3714: int i,j, iter=0;
1.126 brouard 3715: double **xi;
3716: double fret;
3717: double fretone; /* Only one call to likelihood */
3718: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3719:
3720: #ifdef NLOPT
3721: int creturn;
3722: nlopt_opt opt;
3723: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3724: double *lb;
3725: double minf; /* the minimum objective value, upon return */
3726: double * p1; /* Shifted parameters from 0 instead of 1 */
3727: myfunc_data dinst, *d = &dinst;
3728: #endif
3729:
3730:
1.126 brouard 3731: xi=matrix(1,npar,1,npar);
3732: for (i=1;i<=npar;i++)
3733: for (j=1;j<=npar;j++)
3734: xi[i][j]=(i==j ? 1.0 : 0.0);
3735: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3736: strcpy(filerespow,"POW_");
1.126 brouard 3737: strcat(filerespow,fileres);
3738: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3739: printf("Problem with resultfile: %s\n", filerespow);
3740: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3741: }
3742: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3743: for (i=1;i<=nlstate;i++)
3744: for(j=1;j<=nlstate+ndeath;j++)
3745: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3746: fprintf(ficrespow,"\n");
1.162 brouard 3747: #ifdef POWELL
1.126 brouard 3748: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3749: #endif
1.126 brouard 3750:
1.162 brouard 3751: #ifdef NLOPT
3752: #ifdef NEWUOA
3753: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3754: #else
3755: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3756: #endif
3757: lb=vector(0,npar-1);
3758: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3759: nlopt_set_lower_bounds(opt, lb);
3760: nlopt_set_initial_step1(opt, 0.1);
3761:
3762: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3763: d->function = func;
3764: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3765: nlopt_set_min_objective(opt, myfunc, d);
3766: nlopt_set_xtol_rel(opt, ftol);
3767: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3768: printf("nlopt failed! %d\n",creturn);
3769: }
3770: else {
3771: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3772: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3773: iter=1; /* not equal */
3774: }
3775: nlopt_destroy(opt);
3776: #endif
1.126 brouard 3777: free_matrix(xi,1,npar,1,npar);
3778: fclose(ficrespow);
1.203 brouard 3779: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3780: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3781: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3782:
3783: }
3784:
3785: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3786: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3787: {
3788: double **a,**y,*x,pd;
1.203 brouard 3789: /* double **hess; */
1.164 brouard 3790: int i, j;
1.126 brouard 3791: int *indx;
3792:
3793: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3794: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3795: void lubksb(double **a, int npar, int *indx, double b[]) ;
3796: void ludcmp(double **a, int npar, int *indx, double *d) ;
3797: double gompertz(double p[]);
1.203 brouard 3798: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3799:
3800: printf("\nCalculation of the hessian matrix. Wait...\n");
3801: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3802: for (i=1;i<=npar;i++){
1.203 brouard 3803: printf("%d-",i);fflush(stdout);
3804: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3805:
3806: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3807:
3808: /* printf(" %f ",p[i]);
3809: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3810: }
3811:
3812: for (i=1;i<=npar;i++) {
3813: for (j=1;j<=npar;j++) {
3814: if (j>i) {
1.203 brouard 3815: printf(".%d-%d",i,j);fflush(stdout);
3816: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3817: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3818:
3819: hess[j][i]=hess[i][j];
3820: /*printf(" %lf ",hess[i][j]);*/
3821: }
3822: }
3823: }
3824: printf("\n");
3825: fprintf(ficlog,"\n");
3826:
3827: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3828: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3829:
3830: a=matrix(1,npar,1,npar);
3831: y=matrix(1,npar,1,npar);
3832: x=vector(1,npar);
3833: indx=ivector(1,npar);
3834: for (i=1;i<=npar;i++)
3835: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3836: ludcmp(a,npar,indx,&pd);
3837:
3838: for (j=1;j<=npar;j++) {
3839: for (i=1;i<=npar;i++) x[i]=0;
3840: x[j]=1;
3841: lubksb(a,npar,indx,x);
3842: for (i=1;i<=npar;i++){
3843: matcov[i][j]=x[i];
3844: }
3845: }
3846:
3847: printf("\n#Hessian matrix#\n");
3848: fprintf(ficlog,"\n#Hessian matrix#\n");
3849: for (i=1;i<=npar;i++) {
3850: for (j=1;j<=npar;j++) {
1.203 brouard 3851: printf("%.6e ",hess[i][j]);
3852: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3853: }
3854: printf("\n");
3855: fprintf(ficlog,"\n");
3856: }
3857:
1.203 brouard 3858: /* printf("\n#Covariance matrix#\n"); */
3859: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3860: /* for (i=1;i<=npar;i++) { */
3861: /* for (j=1;j<=npar;j++) { */
3862: /* printf("%.6e ",matcov[i][j]); */
3863: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3864: /* } */
3865: /* printf("\n"); */
3866: /* fprintf(ficlog,"\n"); */
3867: /* } */
3868:
1.126 brouard 3869: /* Recompute Inverse */
1.203 brouard 3870: /* for (i=1;i<=npar;i++) */
3871: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3872: /* ludcmp(a,npar,indx,&pd); */
3873:
3874: /* printf("\n#Hessian matrix recomputed#\n"); */
3875:
3876: /* for (j=1;j<=npar;j++) { */
3877: /* for (i=1;i<=npar;i++) x[i]=0; */
3878: /* x[j]=1; */
3879: /* lubksb(a,npar,indx,x); */
3880: /* for (i=1;i<=npar;i++){ */
3881: /* y[i][j]=x[i]; */
3882: /* printf("%.3e ",y[i][j]); */
3883: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3884: /* } */
3885: /* printf("\n"); */
3886: /* fprintf(ficlog,"\n"); */
3887: /* } */
3888:
3889: /* Verifying the inverse matrix */
3890: #ifdef DEBUGHESS
3891: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3892:
1.203 brouard 3893: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3894: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3895:
3896: for (j=1;j<=npar;j++) {
3897: for (i=1;i<=npar;i++){
1.203 brouard 3898: printf("%.2f ",y[i][j]);
3899: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3900: }
3901: printf("\n");
3902: fprintf(ficlog,"\n");
3903: }
1.203 brouard 3904: #endif
1.126 brouard 3905:
3906: free_matrix(a,1,npar,1,npar);
3907: free_matrix(y,1,npar,1,npar);
3908: free_vector(x,1,npar);
3909: free_ivector(indx,1,npar);
1.203 brouard 3910: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3911:
3912:
3913: }
3914:
3915: /*************** hessian matrix ****************/
3916: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3917: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3918: int i;
3919: int l=1, lmax=20;
1.203 brouard 3920: double k1,k2, res, fx;
1.132 brouard 3921: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3922: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3923: int k=0,kmax=10;
3924: double l1;
3925:
3926: fx=func(x);
3927: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3928: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3929: l1=pow(10,l);
3930: delts=delt;
3931: for(k=1 ; k <kmax; k=k+1){
3932: delt = delta*(l1*k);
3933: p2[theta]=x[theta] +delt;
1.145 brouard 3934: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3935: p2[theta]=x[theta]-delt;
3936: k2=func(p2)-fx;
3937: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3938: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3939:
1.203 brouard 3940: #ifdef DEBUGHESSII
1.126 brouard 3941: 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);
3942: 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);
3943: #endif
3944: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3945: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3946: k=kmax;
3947: }
3948: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3949: k=kmax; l=lmax*10;
1.126 brouard 3950: }
3951: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3952: delts=delt;
3953: }
1.203 brouard 3954: } /* End loop k */
1.126 brouard 3955: }
3956: delti[theta]=delts;
3957: return res;
3958:
3959: }
3960:
1.203 brouard 3961: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 3962: {
3963: int i;
1.164 brouard 3964: int l=1, lmax=20;
1.126 brouard 3965: double k1,k2,k3,k4,res,fx;
1.132 brouard 3966: double p2[MAXPARM+1];
1.203 brouard 3967: int k, kmax=1;
3968: double v1, v2, cv12, lc1, lc2;
1.208 brouard 3969:
3970: int firstime=0;
1.203 brouard 3971:
1.126 brouard 3972: fx=func(x);
1.203 brouard 3973: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 3974: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 3975: p2[thetai]=x[thetai]+delti[thetai]*k;
3976: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 3977: k1=func(p2)-fx;
3978:
1.203 brouard 3979: p2[thetai]=x[thetai]+delti[thetai]*k;
3980: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 3981: k2=func(p2)-fx;
3982:
1.203 brouard 3983: p2[thetai]=x[thetai]-delti[thetai]*k;
3984: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 3985: k3=func(p2)-fx;
3986:
1.203 brouard 3987: p2[thetai]=x[thetai]-delti[thetai]*k;
3988: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 3989: k4=func(p2)-fx;
1.203 brouard 3990: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
3991: if(k1*k2*k3*k4 <0.){
1.208 brouard 3992: firstime=1;
1.203 brouard 3993: kmax=kmax+10;
1.208 brouard 3994: }
3995: if(kmax >=10 || firstime ==1){
1.218 brouard 3996: printf("Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you may increase ftol=%.2e\n",thetai,thetaj, ftol);
3997: fprintf(ficlog,"Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you may increase ftol=%.2e\n",thetai,thetaj, ftol);
1.203 brouard 3998: 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);
3999: 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);
4000: }
4001: #ifdef DEBUGHESSIJ
4002: v1=hess[thetai][thetai];
4003: v2=hess[thetaj][thetaj];
4004: cv12=res;
4005: /* Computing eigen value of Hessian matrix */
4006: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4007: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4008: if ((lc2 <0) || (lc1 <0) ){
4009: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4010: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4011: 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);
4012: 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);
4013: }
1.126 brouard 4014: #endif
4015: }
4016: return res;
4017: }
4018:
1.203 brouard 4019: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4020: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4021: /* { */
4022: /* int i; */
4023: /* int l=1, lmax=20; */
4024: /* double k1,k2,k3,k4,res,fx; */
4025: /* double p2[MAXPARM+1]; */
4026: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4027: /* int k=0,kmax=10; */
4028: /* double l1; */
4029:
4030: /* fx=func(x); */
4031: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4032: /* l1=pow(10,l); */
4033: /* delts=delt; */
4034: /* for(k=1 ; k <kmax; k=k+1){ */
4035: /* delt = delti*(l1*k); */
4036: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4037: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4038: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4039: /* k1=func(p2)-fx; */
4040:
4041: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4042: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4043: /* k2=func(p2)-fx; */
4044:
4045: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4046: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4047: /* k3=func(p2)-fx; */
4048:
4049: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4050: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4051: /* k4=func(p2)-fx; */
4052: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4053: /* #ifdef DEBUGHESSIJ */
4054: /* 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); */
4055: /* 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); */
4056: /* #endif */
4057: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4058: /* k=kmax; */
4059: /* } */
4060: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4061: /* k=kmax; l=lmax*10; */
4062: /* } */
4063: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4064: /* delts=delt; */
4065: /* } */
4066: /* } /\* End loop k *\/ */
4067: /* } */
4068: /* delti[theta]=delts; */
4069: /* return res; */
4070: /* } */
4071:
4072:
1.126 brouard 4073: /************** Inverse of matrix **************/
4074: void ludcmp(double **a, int n, int *indx, double *d)
4075: {
4076: int i,imax,j,k;
4077: double big,dum,sum,temp;
4078: double *vv;
4079:
4080: vv=vector(1,n);
4081: *d=1.0;
4082: for (i=1;i<=n;i++) {
4083: big=0.0;
4084: for (j=1;j<=n;j++)
4085: if ((temp=fabs(a[i][j])) > big) big=temp;
4086: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4087: vv[i]=1.0/big;
4088: }
4089: for (j=1;j<=n;j++) {
4090: for (i=1;i<j;i++) {
4091: sum=a[i][j];
4092: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4093: a[i][j]=sum;
4094: }
4095: big=0.0;
4096: for (i=j;i<=n;i++) {
4097: sum=a[i][j];
4098: for (k=1;k<j;k++)
4099: sum -= a[i][k]*a[k][j];
4100: a[i][j]=sum;
4101: if ( (dum=vv[i]*fabs(sum)) >= big) {
4102: big=dum;
4103: imax=i;
4104: }
4105: }
4106: if (j != imax) {
4107: for (k=1;k<=n;k++) {
4108: dum=a[imax][k];
4109: a[imax][k]=a[j][k];
4110: a[j][k]=dum;
4111: }
4112: *d = -(*d);
4113: vv[imax]=vv[j];
4114: }
4115: indx[j]=imax;
4116: if (a[j][j] == 0.0) a[j][j]=TINY;
4117: if (j != n) {
4118: dum=1.0/(a[j][j]);
4119: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4120: }
4121: }
4122: free_vector(vv,1,n); /* Doesn't work */
4123: ;
4124: }
4125:
4126: void lubksb(double **a, int n, int *indx, double b[])
4127: {
4128: int i,ii=0,ip,j;
4129: double sum;
4130:
4131: for (i=1;i<=n;i++) {
4132: ip=indx[i];
4133: sum=b[ip];
4134: b[ip]=b[i];
4135: if (ii)
4136: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4137: else if (sum) ii=i;
4138: b[i]=sum;
4139: }
4140: for (i=n;i>=1;i--) {
4141: sum=b[i];
4142: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4143: b[i]=sum/a[i][i];
4144: }
4145: }
4146:
4147: void pstamp(FILE *fichier)
4148: {
1.196 brouard 4149: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4150: }
4151:
4152: /************ Frequencies ********************/
1.226 brouard 4153: void freqsummary(char fileres[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
4154: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4155: int firstpass, int lastpass, int stepm, int weightopt, char model[])
4156: { /* Some frequencies */
4157:
1.227 brouard 4158: int i, m, jk, j1, bool, z1,j, k, iv;
1.226 brouard 4159: int iind=0, iage=0;
4160: int mi; /* Effective wave */
4161: int first;
4162: double ***freq; /* Frequencies */
4163: double *meanq;
4164: double **meanqt;
4165: double *pp, **prop, *posprop, *pospropt;
4166: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4167: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4168: double agebegin, ageend;
4169:
4170: pp=vector(1,nlstate);
4171: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4172: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4173: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4174: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4175: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4176: meanqt=matrix(1,lastpass,1,nqtveff);
4177: strcpy(fileresp,"P_");
4178: strcat(fileresp,fileresu);
4179: /*strcat(fileresphtm,fileresu);*/
4180: if((ficresp=fopen(fileresp,"w"))==NULL) {
4181: printf("Problem with prevalence resultfile: %s\n", fileresp);
4182: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4183: exit(0);
4184: }
1.240 brouard 4185:
1.226 brouard 4186: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4187: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4188: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4189: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4190: fflush(ficlog);
4191: exit(70);
4192: }
4193: else{
4194: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4195: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4196: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4197: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4198: }
1.237 brouard 4199: 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 4200:
1.226 brouard 4201: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4202: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4203: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4204: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4205: fflush(ficlog);
4206: exit(70);
1.240 brouard 4207: } else{
1.226 brouard 4208: 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 4209: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4210: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4211: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4212: }
1.240 brouard 4213: 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);
4214:
1.226 brouard 4215: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4216: j1=0;
1.126 brouard 4217:
1.227 brouard 4218: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4219: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4220: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4221:
1.226 brouard 4222: first=1;
1.240 brouard 4223:
1.226 brouard 4224: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4225: reference=low_education V1=0,V2=0
4226: med_educ V1=1 V2=0,
4227: high_educ V1=0 V2=1
4228: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4229: */
1.240 brouard 4230:
1.227 brouard 4231: 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 */
1.226 brouard 4232: posproptt=0.;
4233: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4234: scanf("%d", i);*/
4235: for (i=-5; i<=nlstate+ndeath; i++)
4236: for (jk=-5; jk<=nlstate+ndeath; jk++)
1.240 brouard 4237: for(m=iagemin; m <= iagemax+3; m++)
4238: freq[i][jk][m]=0;
4239:
1.226 brouard 4240: for (i=1; i<=nlstate; i++) {
4241: for(m=iagemin; m <= iagemax+3; m++)
1.240 brouard 4242: prop[i][m]=0;
1.226 brouard 4243: posprop[i]=0;
4244: pospropt[i]=0;
4245: }
1.227 brouard 4246: /* for (z1=1; z1<= nqfveff; z1++) { */
4247: /* meanq[z1]+=0.; */
4248: /* for(m=1;m<=lastpass;m++){ */
4249: /* meanqt[m][z1]=0.; */
4250: /* } */
4251: /* } */
1.240 brouard 4252:
1.226 brouard 4253: dateintsum=0;
4254: k2cpt=0;
1.227 brouard 4255: /* For that combination of covariate j1, we count and print the frequencies in one pass */
1.226 brouard 4256: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4257: bool=1;
1.227 brouard 4258: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.234 brouard 4259: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
1.227 brouard 4260: /* for (z1=1; z1<= nqfveff; z1++) { */
4261: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4262: /* } */
1.234 brouard 4263: for (z1=1; z1<=cptcoveff; z1++) {
4264: /* if(Tvaraff[z1] ==-20){ */
4265: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4266: /* }else if(Tvaraff[z1] ==-10){ */
4267: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4268: /* }else */
4269: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){
4270: /* Tests if this individual iind responded to j1 (V4=1 V3=0) */
4271: bool=0;
4272: /* 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",
4273: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4274: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4275: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4276: } /* Onlyf fixed */
4277: } /* end z1 */
4278: } /* cptcovn > 0 */
1.227 brouard 4279: } /* end any */
4280: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
1.234 brouard 4281: /* for(m=firstpass; m<=lastpass; m++){ */
4282: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4283: m=mw[mi][iind];
4284: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4285: for (z1=1; z1<=cptcoveff; z1++) {
4286: if( Fixed[Tmodelind[z1]]==1){
4287: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4288: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4289: bool=0;
4290: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4291: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4292: bool=0;
4293: }
4294: }
4295: }
4296: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4297: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4298: if(bool==1){
4299: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4300: and mw[mi+1][iind]. dh depends on stepm. */
4301: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4302: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4303: if(m >=firstpass && m <=lastpass){
4304: k2=anint[m][iind]+(mint[m][iind]/12.);
4305: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4306: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4307: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4308: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4309: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4310: if (m<lastpass) {
4311: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4312: /* 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]); */
4313: if(s[m][iind]==-1)
4314: 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.));
4315: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4316: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4317: 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 */
4318: }
4319: } /* end if between passes */
4320: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99)) {
4321: dateintsum=dateintsum+k2;
4322: k2cpt++;
4323: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
4324: }
4325: } /* end bool 2 */
4326: } /* end m */
1.226 brouard 4327: } /* end bool */
4328: } /* end iind = 1 to imx */
4329: /* prop[s][age] is feeded for any initial and valid live state as well as
4330: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
1.240 brouard 4331:
4332:
1.226 brouard 4333: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4334: pstamp(ficresp);
1.240 brouard 4335: if (cptcoveff>0){
1.226 brouard 4336: fprintf(ficresp, "\n#********** Variable ");
4337: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4338: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
1.240 brouard 4339: fprintf(ficlog, "\n#********** Variable ");
1.227 brouard 4340: for (z1=1; z1<=cptcoveff; z1++){
1.240 brouard 4341: if(DummyV[z1]){
4342: fprintf(ficresp, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4343: fprintf(ficresphtm, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4344: fprintf(ficresphtmfr, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4345: fprintf(ficlog, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4346: }else{
4347: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4348: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4349: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4350: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4351: }
1.226 brouard 4352: }
4353: fprintf(ficresp, "**********\n#");
4354: fprintf(ficresphtm, "**********</h3>\n");
4355: fprintf(ficresphtmfr, "**********</h3>\n");
4356: fprintf(ficlog, "**********\n");
4357: }
4358: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4359: for(i=1; i<=nlstate;i++) {
1.240 brouard 4360: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
1.226 brouard 4361: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4362: }
4363: fprintf(ficresp, "\n");
4364: fprintf(ficresphtm, "\n");
1.240 brouard 4365:
1.226 brouard 4366: /* Header of frequency table by age */
4367: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4368: fprintf(ficresphtmfr,"<th>Age</th> ");
4369: for(jk=-1; jk <=nlstate+ndeath; jk++){
4370: for(m=-1; m <=nlstate+ndeath; m++){
1.234 brouard 4371: if(jk!=0 && m!=0)
4372: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.226 brouard 4373: }
4374: }
4375: fprintf(ficresphtmfr, "\n");
1.240 brouard 4376:
1.226 brouard 4377: /* For each age */
4378: for(iage=iagemin; iage <= iagemax+3; iage++){
4379: fprintf(ficresphtm,"<tr>");
4380: if(iage==iagemax+1){
1.240 brouard 4381: fprintf(ficlog,"1");
4382: fprintf(ficresphtmfr,"<tr><th>0</th> ");
1.226 brouard 4383: }else if(iage==iagemax+2){
1.240 brouard 4384: fprintf(ficlog,"0");
4385: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
1.226 brouard 4386: }else if(iage==iagemax+3){
1.240 brouard 4387: fprintf(ficlog,"Total");
4388: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
1.226 brouard 4389: }else{
1.240 brouard 4390: if(first==1){
4391: first=0;
4392: printf("See log file for details...\n");
4393: }
4394: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4395: fprintf(ficlog,"Age %d", iage);
1.226 brouard 4396: }
4397: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4398: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4399: pp[jk] += freq[jk][m][iage];
1.226 brouard 4400: }
4401: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4402: for(m=-1, pos=0; m <=0 ; m++)
4403: pos += freq[jk][m][iage];
4404: if(pp[jk]>=1.e-10){
4405: if(first==1){
4406: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4407: }
4408: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4409: }else{
4410: if(first==1)
4411: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4412: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4413: }
1.226 brouard 4414: }
1.240 brouard 4415:
1.226 brouard 4416: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4417: /* posprop[jk]=0; */
4418: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4419: pp[jk] += freq[jk][m][iage];
1.226 brouard 4420: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
1.240 brouard 4421:
1.226 brouard 4422: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
1.240 brouard 4423: pos += pp[jk]; /* pos is the total number of transitions until this age */
4424: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4425: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4426: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
4427: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
1.226 brouard 4428: }
4429: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4430: if(pos>=1.e-5){
4431: if(first==1)
4432: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4433: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4434: }else{
4435: if(first==1)
4436: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4437: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4438: }
4439: if( iage <= iagemax){
4440: if(pos>=1.e-5){
4441: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4442: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4443: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4444: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4445: }
4446: else{
4447: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4448: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4449: }
4450: }
4451: pospropt[jk] +=posprop[jk];
1.226 brouard 4452: } /* end loop jk */
4453: /* pospropt=0.; */
4454: for(jk=-1; jk <=nlstate+ndeath; jk++){
1.240 brouard 4455: for(m=-1; m <=nlstate+ndeath; m++){
4456: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4457: if(first==1){
4458: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4459: }
4460: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4461: }
4462: if(jk!=0 && m!=0)
4463: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
4464: }
1.226 brouard 4465: } /* end loop jk */
4466: posproptt=0.;
4467: for(jk=1; jk <=nlstate; jk++){
1.240 brouard 4468: posproptt += pospropt[jk];
1.226 brouard 4469: }
4470: fprintf(ficresphtmfr,"</tr>\n ");
4471: if(iage <= iagemax){
1.240 brouard 4472: fprintf(ficresp,"\n");
4473: fprintf(ficresphtm,"</tr>\n");
1.226 brouard 4474: }
4475: if(first==1)
1.240 brouard 4476: printf("Others in log...\n");
1.226 brouard 4477: fprintf(ficlog,"\n");
4478: } /* end loop age iage */
4479: fprintf(ficresphtm,"<tr><th>Tot</th>");
4480: for(jk=1; jk <=nlstate ; jk++){
4481: if(posproptt < 1.e-5){
1.240 brouard 4482: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
1.226 brouard 4483: }else{
1.240 brouard 4484: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.226 brouard 4485: }
4486: }
4487: fprintf(ficresphtm,"</tr>\n");
4488: fprintf(ficresphtm,"</table>\n");
4489: fprintf(ficresphtmfr,"</table>\n");
4490: if(posproptt < 1.e-5){
4491: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4492: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4493: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4494: invalidvarcomb[j1]=1;
4495: }else{
4496: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4497: invalidvarcomb[j1]=0;
4498: }
4499: fprintf(ficresphtmfr,"</table>\n");
4500: } /* end selected combination of covariate j1 */
4501: dateintmean=dateintsum/k2cpt;
1.240 brouard 4502:
1.226 brouard 4503: fclose(ficresp);
4504: fclose(ficresphtm);
4505: fclose(ficresphtmfr);
4506: free_vector(meanq,1,nqfveff);
4507: free_matrix(meanqt,1,lastpass,1,nqtveff);
4508: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4509: free_vector(pospropt,1,nlstate);
4510: free_vector(posprop,1,nlstate);
4511: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4512: free_vector(pp,1,nlstate);
4513: /* End of freqsummary */
4514: }
1.126 brouard 4515:
4516: /************ Prevalence ********************/
1.227 brouard 4517: 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)
4518: {
4519: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4520: in each health status at the date of interview (if between dateprev1 and dateprev2).
4521: We still use firstpass and lastpass as another selection.
4522: */
1.126 brouard 4523:
1.227 brouard 4524: int i, m, jk, j1, bool, z1,j, iv;
4525: int mi; /* Effective wave */
4526: int iage;
4527: double agebegin, ageend;
4528:
4529: double **prop;
4530: double posprop;
4531: double y2; /* in fractional years */
4532: int iagemin, iagemax;
4533: int first; /** to stop verbosity which is redirected to log file */
4534:
4535: iagemin= (int) agemin;
4536: iagemax= (int) agemax;
4537: /*pp=vector(1,nlstate);*/
4538: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4539: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4540: j1=0;
1.222 brouard 4541:
1.227 brouard 4542: /*j=cptcoveff;*/
4543: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4544:
1.227 brouard 4545: first=1;
4546: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4547: for (i=1; i<=nlstate; i++)
4548: for(iage=iagemin-AGEMARGE; iage <= iagemax+3+AGEMARGE; iage++)
4549: prop[i][iage]=0.0;
4550: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4551: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4552: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4553:
4554: for (i=1; i<=imx; i++) { /* Each individual */
4555: bool=1;
4556: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4557: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4558: m=mw[mi][i];
4559: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4560: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4561: for (z1=1; z1<=cptcoveff; z1++){
4562: if( Fixed[Tmodelind[z1]]==1){
4563: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4564: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4565: bool=0;
4566: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4567: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4568: bool=0;
4569: }
4570: }
4571: if(bool==1){ /* Otherwise we skip that wave/person */
4572: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4573: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4574: if(m >=firstpass && m <=lastpass){
4575: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4576: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4577: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4578: if(agev[m][i]==1) agev[m][i]=iagemax+2;
4579: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+3+AGEMARGE){
4580: 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);
4581: exit(1);
4582: }
4583: if (s[m][i]>0 && s[m][i]<=nlstate) {
4584: /*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]]);*/
4585: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4586: prop[s[m][i]][iagemax+3] += weight[i];
4587: } /* end valid statuses */
4588: } /* end selection of dates */
4589: } /* end selection of waves */
4590: } /* end bool */
4591: } /* end wave */
4592: } /* end individual */
4593: for(i=iagemin; i <= iagemax+3; i++){
4594: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4595: posprop += prop[jk][i];
4596: }
4597:
4598: for(jk=1; jk <=nlstate ; jk++){
4599: if( i <= iagemax){
4600: if(posprop>=1.e-5){
4601: probs[i][jk][j1]= prop[jk][i]/posprop;
4602: } else{
4603: if(first==1){
4604: first=0;
4605: 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]);
4606: }
4607: }
4608: }
4609: }/* end jk */
4610: }/* end i */
1.222 brouard 4611: /*} *//* end i1 */
1.227 brouard 4612: } /* end j1 */
1.222 brouard 4613:
1.227 brouard 4614: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4615: /*free_vector(pp,1,nlstate);*/
4616: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4617: } /* End of prevalence */
1.126 brouard 4618:
4619: /************* Waves Concatenation ***************/
4620:
4621: 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)
4622: {
4623: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4624: Death is a valid wave (if date is known).
4625: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4626: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4627: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4628: */
1.126 brouard 4629:
1.224 brouard 4630: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4631: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4632: double sum=0., jmean=0.;*/
1.224 brouard 4633: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4634: int j, k=0,jk, ju, jl;
4635: double sum=0.;
4636: first=0;
1.214 brouard 4637: firstwo=0;
1.217 brouard 4638: firsthree=0;
1.218 brouard 4639: firstfour=0;
1.164 brouard 4640: jmin=100000;
1.126 brouard 4641: jmax=-1;
4642: jmean=0.;
1.224 brouard 4643:
4644: /* Treating live states */
1.214 brouard 4645: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4646: mi=0; /* First valid wave */
1.227 brouard 4647: mli=0; /* Last valid wave */
1.126 brouard 4648: m=firstpass;
1.214 brouard 4649: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4650: 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 */
4651: mli=m-1;/* mw[++mi][i]=m-1; */
4652: }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 */
4653: mw[++mi][i]=m;
4654: mli=m;
1.224 brouard 4655: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4656: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4657: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4658: }
1.227 brouard 4659: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4660: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4661: break;
1.224 brouard 4662: #else
1.227 brouard 4663: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4664: if(firsthree == 0){
4665: 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);
4666: firsthree=1;
4667: }
4668: 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);
4669: mw[++mi][i]=m;
4670: mli=m;
4671: }
4672: if(s[m][i]==-2){ /* Vital status is really unknown */
4673: nbwarn++;
4674: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4675: 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);
4676: 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);
4677: }
4678: break;
4679: }
4680: break;
1.224 brouard 4681: #endif
1.227 brouard 4682: }/* End m >= lastpass */
1.126 brouard 4683: }/* end while */
1.224 brouard 4684:
1.227 brouard 4685: /* 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 4686: /* After last pass */
1.224 brouard 4687: /* Treating death states */
1.214 brouard 4688: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4689: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4690: /* } */
1.126 brouard 4691: mi++; /* Death is another wave */
4692: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4693: /* Only death is a correct wave */
1.126 brouard 4694: mw[mi][i]=m;
1.224 brouard 4695: }
4696: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4697: 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 4698: /* m++; */
4699: /* mi++; */
4700: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4701: /* mw[mi][i]=m; */
1.218 brouard 4702: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4703: 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 */
4704: nbwarn++;
4705: if(firstfiv==0){
4706: 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 );
4707: firstfiv=1;
4708: }else{
4709: 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 );
4710: }
4711: }else{ /* Death occured afer last wave potential bias */
4712: nberr++;
4713: if(firstwo==0){
4714: 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 );
4715: firstwo=1;
4716: }
4717: 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 );
4718: }
1.218 brouard 4719: }else{ /* end date of interview is known */
1.227 brouard 4720: /* death is known but not confirmed by death status at any wave */
4721: if(firstfour==0){
4722: 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 );
4723: firstfour=1;
4724: }
4725: 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 4726: }
1.224 brouard 4727: } /* end if date of death is known */
4728: #endif
4729: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4730: /* wav[i]=mw[mi][i]; */
1.126 brouard 4731: if(mi==0){
4732: nbwarn++;
4733: if(first==0){
1.227 brouard 4734: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4735: first=1;
1.126 brouard 4736: }
4737: if(first==1){
1.227 brouard 4738: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4739: }
4740: } /* end mi==0 */
4741: } /* End individuals */
1.214 brouard 4742: /* wav and mw are no more changed */
1.223 brouard 4743:
1.214 brouard 4744:
1.126 brouard 4745: for(i=1; i<=imx; i++){
4746: for(mi=1; mi<wav[i];mi++){
4747: if (stepm <=0)
1.227 brouard 4748: dh[mi][i]=1;
1.126 brouard 4749: else{
1.227 brouard 4750: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4751: if (agedc[i] < 2*AGESUP) {
4752: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4753: if(j==0) j=1; /* Survives at least one month after exam */
4754: else if(j<0){
4755: nberr++;
4756: 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]);
4757: j=1; /* Temporary Dangerous patch */
4758: 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);
4759: 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]);
4760: 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);
4761: }
4762: k=k+1;
4763: if (j >= jmax){
4764: jmax=j;
4765: ijmax=i;
4766: }
4767: if (j <= jmin){
4768: jmin=j;
4769: ijmin=i;
4770: }
4771: sum=sum+j;
4772: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
4773: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
4774: }
4775: }
4776: else{
4777: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 4778: /* 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 4779:
1.227 brouard 4780: k=k+1;
4781: if (j >= jmax) {
4782: jmax=j;
4783: ijmax=i;
4784: }
4785: else if (j <= jmin){
4786: jmin=j;
4787: ijmin=i;
4788: }
4789: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
4790: /*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]);*/
4791: if(j<0){
4792: nberr++;
4793: 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]);
4794: 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]);
4795: }
4796: sum=sum+j;
4797: }
4798: jk= j/stepm;
4799: jl= j -jk*stepm;
4800: ju= j -(jk+1)*stepm;
4801: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
4802: if(jl==0){
4803: dh[mi][i]=jk;
4804: bh[mi][i]=0;
4805: }else{ /* We want a negative bias in order to only have interpolation ie
4806: * to avoid the price of an extra matrix product in likelihood */
4807: dh[mi][i]=jk+1;
4808: bh[mi][i]=ju;
4809: }
4810: }else{
4811: if(jl <= -ju){
4812: dh[mi][i]=jk;
4813: bh[mi][i]=jl; /* bias is positive if real duration
4814: * is higher than the multiple of stepm and negative otherwise.
4815: */
4816: }
4817: else{
4818: dh[mi][i]=jk+1;
4819: bh[mi][i]=ju;
4820: }
4821: if(dh[mi][i]==0){
4822: dh[mi][i]=1; /* At least one step */
4823: bh[mi][i]=ju; /* At least one step */
4824: /* 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);*/
4825: }
4826: } /* end if mle */
1.126 brouard 4827: }
4828: } /* end wave */
4829: }
4830: jmean=sum/k;
4831: 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 4832: 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 4833: }
1.126 brouard 4834:
4835: /*********** Tricode ****************************/
1.220 brouard 4836: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 ! brouard 4837: {
! 4838: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
! 4839: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
! 4840: * Boring subroutine which should only output nbcode[Tvar[j]][k]
! 4841: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
! 4842: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
! 4843: */
1.130 brouard 4844:
1.242 ! brouard 4845: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
! 4846: int modmaxcovj=0; /* Modality max of covariates j */
! 4847: int cptcode=0; /* Modality max of covariates j */
! 4848: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 4849:
4850:
1.242 ! brouard 4851: /* cptcoveff=0; */
! 4852: /* *cptcov=0; */
1.126 brouard 4853:
1.242 ! brouard 4854: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 4855:
1.242 ! brouard 4856: /* Loop on covariates without age and products and no quantitative variable */
! 4857: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
! 4858: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
! 4859: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
! 4860: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
! 4861: switch(Fixed[k]) {
! 4862: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
! 4863: 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*/
! 4864: ij=(int)(covar[Tvar[k]][i]);
! 4865: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
! 4866: * If product of Vn*Vm, still boolean *:
! 4867: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
! 4868: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
! 4869: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
! 4870: modality of the nth covariate of individual i. */
! 4871: if (ij > modmaxcovj)
! 4872: modmaxcovj=ij;
! 4873: else if (ij < modmincovj)
! 4874: modmincovj=ij;
! 4875: if ((ij < -1) && (ij > NCOVMAX)){
! 4876: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
! 4877: exit(1);
! 4878: }else
! 4879: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
! 4880: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
! 4881: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
! 4882: /* getting the maximum value of the modality of the covariate
! 4883: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
! 4884: female ies 1, then modmaxcovj=1.
! 4885: */
! 4886: } /* end for loop on individuals i */
! 4887: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
! 4888: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
! 4889: cptcode=modmaxcovj;
! 4890: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
! 4891: /*for (i=0; i<=cptcode; i++) {*/
! 4892: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
! 4893: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
! 4894: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
! 4895: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
! 4896: if( j != -1){
! 4897: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
! 4898: covariate for which somebody answered excluding
! 4899: undefined. Usually 2: 0 and 1. */
! 4900: }
! 4901: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
! 4902: covariate for which somebody answered including
! 4903: undefined. Usually 3: -1, 0 and 1. */
! 4904: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
! 4905: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
! 4906: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 4907:
1.242 ! brouard 4908: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
! 4909: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
! 4910: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
! 4911: /* modmincovj=3; modmaxcovj = 7; */
! 4912: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
! 4913: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
! 4914: /* defining two dummy variables: variables V1_1 and V1_2.*/
! 4915: /* nbcode[Tvar[j]][ij]=k; */
! 4916: /* nbcode[Tvar[j]][1]=0; */
! 4917: /* nbcode[Tvar[j]][2]=1; */
! 4918: /* nbcode[Tvar[j]][3]=2; */
! 4919: /* To be continued (not working yet). */
! 4920: ij=0; /* ij is similar to i but can jump over null modalities */
! 4921: 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*/
! 4922: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
! 4923: break;
! 4924: }
! 4925: ij++;
! 4926: 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*/
! 4927: cptcode = ij; /* New max modality for covar j */
! 4928: } /* end of loop on modality i=-1 to 1 or more */
! 4929: break;
! 4930: case 1: /* Testing on varying covariate, could be simple and
! 4931: * should look at waves or product of fixed *
! 4932: * varying. No time to test -1, assuming 0 and 1 only */
! 4933: ij=0;
! 4934: for(i=0; i<=1;i++){
! 4935: nbcode[Tvar[k]][++ij]=i;
! 4936: }
! 4937: break;
! 4938: default:
! 4939: break;
! 4940: } /* end switch */
! 4941: } /* end dummy test */
! 4942:
! 4943: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
! 4944: /* /\*recode from 0 *\/ */
! 4945: /* k is a modality. If we have model=V1+V1*sex */
! 4946: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
! 4947: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
! 4948: /* } */
! 4949: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
! 4950: /* if (ij > ncodemax[j]) { */
! 4951: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
! 4952: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
! 4953: /* break; */
! 4954: /* } */
! 4955: /* } /\* end of loop on modality k *\/ */
! 4956: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
! 4957:
! 4958: for (k=-1; k< maxncov; k++) Ndum[k]=0;
! 4959: /* Look at fixed dummy (single or product) covariates to check empty modalities */
! 4960: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
! 4961: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
! 4962: 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 */
! 4963: 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 */
! 4964: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
! 4965: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
! 4966:
! 4967: ij=0;
! 4968: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
! 4969: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
! 4970: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
! 4971: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
! 4972: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
! 4973: /* If product not in single variable we don't print results */
! 4974: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
! 4975: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
! 4976: 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*/
! 4977: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
! 4978: 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 */
! 4979: if(Fixed[k]!=0)
! 4980: anyvaryingduminmodel=1;
! 4981: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
! 4982: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
! 4983: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
! 4984: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
! 4985: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
! 4986: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
! 4987: }
! 4988: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
! 4989: /* ij--; */
! 4990: /* cptcoveff=ij; /\*Number of total covariates*\/ */
! 4991: *cptcov=ij; /*Number of total real effective covariates: effective
! 4992: * because they can be excluded from the model and real
! 4993: * if in the model but excluded because missing values, but how to get k from ij?*/
! 4994: for(j=ij+1; j<= cptcovt; j++){
! 4995: Tvaraff[j]=0;
! 4996: Tmodelind[j]=0;
! 4997: }
! 4998: for(j=ntveff+1; j<= cptcovt; j++){
! 4999: TmodelInvind[j]=0;
! 5000: }
! 5001: /* To be sorted */
! 5002: ;
! 5003: }
1.126 brouard 5004:
1.145 brouard 5005:
1.126 brouard 5006: /*********** Health Expectancies ****************/
5007:
1.235 brouard 5008: 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 5009:
5010: {
5011: /* Health expectancies, no variances */
1.164 brouard 5012: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5013: int nhstepma, nstepma; /* Decreasing with age */
5014: double age, agelim, hf;
5015: double ***p3mat;
5016: double eip;
5017:
1.238 brouard 5018: /* pstamp(ficreseij); */
1.126 brouard 5019: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5020: fprintf(ficreseij,"# Age");
5021: for(i=1; i<=nlstate;i++){
5022: for(j=1; j<=nlstate;j++){
5023: fprintf(ficreseij," e%1d%1d ",i,j);
5024: }
5025: fprintf(ficreseij," e%1d. ",i);
5026: }
5027: fprintf(ficreseij,"\n");
5028:
5029:
5030: if(estepm < stepm){
5031: printf ("Problem %d lower than %d\n",estepm, stepm);
5032: }
5033: else hstepm=estepm;
5034: /* We compute the life expectancy from trapezoids spaced every estepm months
5035: * This is mainly to measure the difference between two models: for example
5036: * if stepm=24 months pijx are given only every 2 years and by summing them
5037: * we are calculating an estimate of the Life Expectancy assuming a linear
5038: * progression in between and thus overestimating or underestimating according
5039: * to the curvature of the survival function. If, for the same date, we
5040: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5041: * to compare the new estimate of Life expectancy with the same linear
5042: * hypothesis. A more precise result, taking into account a more precise
5043: * curvature will be obtained if estepm is as small as stepm. */
5044:
5045: /* For example we decided to compute the life expectancy with the smallest unit */
5046: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5047: nhstepm is the number of hstepm from age to agelim
5048: nstepm is the number of stepm from age to agelin.
5049: Look at hpijx to understand the reason of that which relies in memory size
5050: and note for a fixed period like estepm months */
5051: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5052: survival function given by stepm (the optimization length). Unfortunately it
5053: means that if the survival funtion is printed only each two years of age and if
5054: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5055: results. So we changed our mind and took the option of the best precision.
5056: */
5057: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5058:
5059: agelim=AGESUP;
5060: /* If stepm=6 months */
5061: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5062: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5063:
5064: /* nhstepm age range expressed in number of stepm */
5065: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5066: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5067: /* if (stepm >= YEARM) hstepm=1;*/
5068: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5069: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5070:
5071: for (age=bage; age<=fage; age ++){
5072: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5073: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5074: /* if (stepm >= YEARM) hstepm=1;*/
5075: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5076:
5077: /* If stepm=6 months */
5078: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5079: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5080:
1.235 brouard 5081: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5082:
5083: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5084:
5085: printf("%d|",(int)age);fflush(stdout);
5086: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5087:
5088: /* Computing expectancies */
5089: for(i=1; i<=nlstate;i++)
5090: for(j=1; j<=nlstate;j++)
5091: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5092: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5093:
5094: /* 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]);*/
5095:
5096: }
5097:
5098: fprintf(ficreseij,"%3.0f",age );
5099: for(i=1; i<=nlstate;i++){
5100: eip=0;
5101: for(j=1; j<=nlstate;j++){
5102: eip +=eij[i][j][(int)age];
5103: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5104: }
5105: fprintf(ficreseij,"%9.4f", eip );
5106: }
5107: fprintf(ficreseij,"\n");
5108:
5109: }
5110: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5111: printf("\n");
5112: fprintf(ficlog,"\n");
5113:
5114: }
5115:
1.235 brouard 5116: 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 5117:
5118: {
5119: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5120: to initial status i, ei. .
1.126 brouard 5121: */
5122: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5123: int nhstepma, nstepma; /* Decreasing with age */
5124: double age, agelim, hf;
5125: double ***p3matp, ***p3matm, ***varhe;
5126: double **dnewm,**doldm;
5127: double *xp, *xm;
5128: double **gp, **gm;
5129: double ***gradg, ***trgradg;
5130: int theta;
5131:
5132: double eip, vip;
5133:
5134: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5135: xp=vector(1,npar);
5136: xm=vector(1,npar);
5137: dnewm=matrix(1,nlstate*nlstate,1,npar);
5138: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5139:
5140: pstamp(ficresstdeij);
5141: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5142: fprintf(ficresstdeij,"# Age");
5143: for(i=1; i<=nlstate;i++){
5144: for(j=1; j<=nlstate;j++)
5145: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5146: fprintf(ficresstdeij," e%1d. ",i);
5147: }
5148: fprintf(ficresstdeij,"\n");
5149:
5150: pstamp(ficrescveij);
5151: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5152: fprintf(ficrescveij,"# Age");
5153: for(i=1; i<=nlstate;i++)
5154: for(j=1; j<=nlstate;j++){
5155: cptj= (j-1)*nlstate+i;
5156: for(i2=1; i2<=nlstate;i2++)
5157: for(j2=1; j2<=nlstate;j2++){
5158: cptj2= (j2-1)*nlstate+i2;
5159: if(cptj2 <= cptj)
5160: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5161: }
5162: }
5163: fprintf(ficrescveij,"\n");
5164:
5165: if(estepm < stepm){
5166: printf ("Problem %d lower than %d\n",estepm, stepm);
5167: }
5168: else hstepm=estepm;
5169: /* We compute the life expectancy from trapezoids spaced every estepm months
5170: * This is mainly to measure the difference between two models: for example
5171: * if stepm=24 months pijx are given only every 2 years and by summing them
5172: * we are calculating an estimate of the Life Expectancy assuming a linear
5173: * progression in between and thus overestimating or underestimating according
5174: * to the curvature of the survival function. If, for the same date, we
5175: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5176: * to compare the new estimate of Life expectancy with the same linear
5177: * hypothesis. A more precise result, taking into account a more precise
5178: * curvature will be obtained if estepm is as small as stepm. */
5179:
5180: /* For example we decided to compute the life expectancy with the smallest unit */
5181: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5182: nhstepm is the number of hstepm from age to agelim
5183: nstepm is the number of stepm from age to agelin.
5184: Look at hpijx to understand the reason of that which relies in memory size
5185: and note for a fixed period like estepm months */
5186: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5187: survival function given by stepm (the optimization length). Unfortunately it
5188: means that if the survival funtion is printed only each two years of age and if
5189: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5190: results. So we changed our mind and took the option of the best precision.
5191: */
5192: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5193:
5194: /* If stepm=6 months */
5195: /* nhstepm age range expressed in number of stepm */
5196: agelim=AGESUP;
5197: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5198: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5199: /* if (stepm >= YEARM) hstepm=1;*/
5200: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5201:
5202: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5203: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5204: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5205: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5206: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5207: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5208:
5209: for (age=bage; age<=fage; age ++){
5210: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5211: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5212: /* if (stepm >= YEARM) hstepm=1;*/
5213: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5214:
1.126 brouard 5215: /* If stepm=6 months */
5216: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5217: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5218:
5219: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5220:
1.126 brouard 5221: /* Computing Variances of health expectancies */
5222: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5223: decrease memory allocation */
5224: for(theta=1; theta <=npar; theta++){
5225: for(i=1; i<=npar; i++){
1.222 brouard 5226: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5227: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5228: }
1.235 brouard 5229: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5230: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5231:
1.126 brouard 5232: for(j=1; j<= nlstate; j++){
1.222 brouard 5233: for(i=1; i<=nlstate; i++){
5234: for(h=0; h<=nhstepm-1; h++){
5235: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5236: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5237: }
5238: }
1.126 brouard 5239: }
1.218 brouard 5240:
1.126 brouard 5241: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5242: for(h=0; h<=nhstepm-1; h++){
5243: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5244: }
1.126 brouard 5245: }/* End theta */
5246:
5247:
5248: for(h=0; h<=nhstepm-1; h++)
5249: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5250: for(theta=1; theta <=npar; theta++)
5251: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5252:
1.218 brouard 5253:
1.222 brouard 5254: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5255: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5256: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5257:
1.222 brouard 5258: printf("%d|",(int)age);fflush(stdout);
5259: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5260: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5261: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5262: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5263: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5264: for(ij=1;ij<=nlstate*nlstate;ij++)
5265: for(ji=1;ji<=nlstate*nlstate;ji++)
5266: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5267: }
5268: }
1.218 brouard 5269:
1.126 brouard 5270: /* Computing expectancies */
1.235 brouard 5271: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5272: for(i=1; i<=nlstate;i++)
5273: for(j=1; j<=nlstate;j++)
1.222 brouard 5274: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5275: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5276:
1.222 brouard 5277: /* 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 5278:
1.222 brouard 5279: }
1.218 brouard 5280:
1.126 brouard 5281: fprintf(ficresstdeij,"%3.0f",age );
5282: for(i=1; i<=nlstate;i++){
5283: eip=0.;
5284: vip=0.;
5285: for(j=1; j<=nlstate;j++){
1.222 brouard 5286: eip += eij[i][j][(int)age];
5287: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5288: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5289: 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 5290: }
5291: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5292: }
5293: fprintf(ficresstdeij,"\n");
1.218 brouard 5294:
1.126 brouard 5295: fprintf(ficrescveij,"%3.0f",age );
5296: for(i=1; i<=nlstate;i++)
5297: for(j=1; j<=nlstate;j++){
1.222 brouard 5298: cptj= (j-1)*nlstate+i;
5299: for(i2=1; i2<=nlstate;i2++)
5300: for(j2=1; j2<=nlstate;j2++){
5301: cptj2= (j2-1)*nlstate+i2;
5302: if(cptj2 <= cptj)
5303: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5304: }
1.126 brouard 5305: }
5306: fprintf(ficrescveij,"\n");
1.218 brouard 5307:
1.126 brouard 5308: }
5309: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5310: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5311: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5312: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5313: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5314: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5315: printf("\n");
5316: fprintf(ficlog,"\n");
1.218 brouard 5317:
1.126 brouard 5318: free_vector(xm,1,npar);
5319: free_vector(xp,1,npar);
5320: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5321: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5322: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5323: }
1.218 brouard 5324:
1.126 brouard 5325: /************ Variance ******************/
1.235 brouard 5326: 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 5327: {
5328: /* Variance of health expectancies */
5329: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5330: /* double **newm;*/
5331: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5332:
5333: /* int movingaverage(); */
5334: double **dnewm,**doldm;
5335: double **dnewmp,**doldmp;
5336: int i, j, nhstepm, hstepm, h, nstepm ;
5337: int k;
5338: double *xp;
5339: double **gp, **gm; /* for var eij */
5340: double ***gradg, ***trgradg; /*for var eij */
5341: double **gradgp, **trgradgp; /* for var p point j */
5342: double *gpp, *gmp; /* for var p point j */
5343: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5344: double ***p3mat;
5345: double age,agelim, hf;
5346: /* double ***mobaverage; */
5347: int theta;
5348: char digit[4];
5349: char digitp[25];
5350:
5351: char fileresprobmorprev[FILENAMELENGTH];
5352:
5353: if(popbased==1){
5354: if(mobilav!=0)
5355: strcpy(digitp,"-POPULBASED-MOBILAV_");
5356: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5357: }
5358: else
5359: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5360:
1.218 brouard 5361: /* if (mobilav!=0) { */
5362: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5363: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5364: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5365: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5366: /* } */
5367: /* } */
5368:
5369: strcpy(fileresprobmorprev,"PRMORPREV-");
5370: sprintf(digit,"%-d",ij);
5371: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5372: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5373: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5374: strcat(fileresprobmorprev,fileresu);
5375: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5376: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5377: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5378: }
5379: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5380: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5381: pstamp(ficresprobmorprev);
5382: 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 5383: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5384: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5385: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5386: }
5387: for(j=1;j<=cptcoveff;j++)
5388: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5389: fprintf(ficresprobmorprev,"\n");
5390:
1.218 brouard 5391: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5392: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5393: fprintf(ficresprobmorprev," p.%-d SE",j);
5394: for(i=1; i<=nlstate;i++)
5395: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5396: }
5397: fprintf(ficresprobmorprev,"\n");
5398:
5399: fprintf(ficgp,"\n# Routine varevsij");
5400: fprintf(ficgp,"\nunset title \n");
5401: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5402: 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");
5403: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5404: /* } */
5405: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5406: pstamp(ficresvij);
5407: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5408: if(popbased==1)
5409: 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);
5410: else
5411: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5412: fprintf(ficresvij,"# Age");
5413: for(i=1; i<=nlstate;i++)
5414: for(j=1; j<=nlstate;j++)
5415: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5416: fprintf(ficresvij,"\n");
5417:
5418: xp=vector(1,npar);
5419: dnewm=matrix(1,nlstate,1,npar);
5420: doldm=matrix(1,nlstate,1,nlstate);
5421: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5422: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5423:
5424: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5425: gpp=vector(nlstate+1,nlstate+ndeath);
5426: gmp=vector(nlstate+1,nlstate+ndeath);
5427: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5428:
1.218 brouard 5429: if(estepm < stepm){
5430: printf ("Problem %d lower than %d\n",estepm, stepm);
5431: }
5432: else hstepm=estepm;
5433: /* For example we decided to compute the life expectancy with the smallest unit */
5434: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5435: nhstepm is the number of hstepm from age to agelim
5436: nstepm is the number of stepm from age to agelim.
5437: Look at function hpijx to understand why because of memory size limitations,
5438: we decided (b) to get a life expectancy respecting the most precise curvature of the
5439: survival function given by stepm (the optimization length). Unfortunately it
5440: means that if the survival funtion is printed every two years of age and if
5441: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5442: results. So we changed our mind and took the option of the best precision.
5443: */
5444: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5445: agelim = AGESUP;
5446: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5447: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5448: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5449: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5450: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5451: gp=matrix(0,nhstepm,1,nlstate);
5452: gm=matrix(0,nhstepm,1,nlstate);
5453:
5454:
5455: for(theta=1; theta <=npar; theta++){
5456: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5457: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5458: }
5459:
1.242 ! brouard 5460: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5461:
5462: if (popbased==1) {
5463: if(mobilav ==0){
5464: for(i=1; i<=nlstate;i++)
5465: prlim[i][i]=probs[(int)age][i][ij];
5466: }else{ /* mobilav */
5467: for(i=1; i<=nlstate;i++)
5468: prlim[i][i]=mobaverage[(int)age][i][ij];
5469: }
5470: }
5471:
1.235 brouard 5472: 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 5473: for(j=1; j<= nlstate; j++){
5474: for(h=0; h<=nhstepm; h++){
5475: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5476: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5477: }
5478: }
5479: /* Next for computing probability of death (h=1 means
5480: computed over hstepm matrices product = hstepm*stepm months)
5481: as a weighted average of prlim.
5482: */
5483: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5484: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5485: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5486: }
5487: /* end probability of death */
5488:
5489: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5490: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5491:
1.242 ! brouard 5492: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5493:
5494: if (popbased==1) {
5495: if(mobilav ==0){
5496: for(i=1; i<=nlstate;i++)
5497: prlim[i][i]=probs[(int)age][i][ij];
5498: }else{ /* mobilav */
5499: for(i=1; i<=nlstate;i++)
5500: prlim[i][i]=mobaverage[(int)age][i][ij];
5501: }
5502: }
5503:
1.235 brouard 5504: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5505:
5506: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5507: for(h=0; h<=nhstepm; h++){
5508: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5509: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5510: }
5511: }
5512: /* This for computing probability of death (h=1 means
5513: computed over hstepm matrices product = hstepm*stepm months)
5514: as a weighted average of prlim.
5515: */
5516: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5517: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5518: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5519: }
5520: /* end probability of death */
5521:
5522: for(j=1; j<= nlstate; j++) /* vareij */
5523: for(h=0; h<=nhstepm; h++){
5524: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5525: }
5526:
5527: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5528: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5529: }
5530:
5531: } /* End theta */
5532:
5533: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5534:
5535: for(h=0; h<=nhstepm; h++) /* veij */
5536: for(j=1; j<=nlstate;j++)
5537: for(theta=1; theta <=npar; theta++)
5538: trgradg[h][j][theta]=gradg[h][theta][j];
5539:
5540: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5541: for(theta=1; theta <=npar; theta++)
5542: trgradgp[j][theta]=gradgp[theta][j];
5543:
5544:
5545: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5546: for(i=1;i<=nlstate;i++)
5547: for(j=1;j<=nlstate;j++)
5548: vareij[i][j][(int)age] =0.;
5549:
5550: for(h=0;h<=nhstepm;h++){
5551: for(k=0;k<=nhstepm;k++){
5552: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5553: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5554: for(i=1;i<=nlstate;i++)
5555: for(j=1;j<=nlstate;j++)
5556: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5557: }
5558: }
5559:
5560: /* pptj */
5561: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5562: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5563: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5564: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5565: varppt[j][i]=doldmp[j][i];
5566: /* end ppptj */
5567: /* x centered again */
5568:
1.242 ! brouard 5569: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5570:
5571: if (popbased==1) {
5572: if(mobilav ==0){
5573: for(i=1; i<=nlstate;i++)
5574: prlim[i][i]=probs[(int)age][i][ij];
5575: }else{ /* mobilav */
5576: for(i=1; i<=nlstate;i++)
5577: prlim[i][i]=mobaverage[(int)age][i][ij];
5578: }
5579: }
5580:
5581: /* This for computing probability of death (h=1 means
5582: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5583: as a weighted average of prlim.
5584: */
1.235 brouard 5585: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5586: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5587: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5588: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5589: }
5590: /* end probability of death */
5591:
5592: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5593: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5594: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5595: for(i=1; i<=nlstate;i++){
5596: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5597: }
5598: }
5599: fprintf(ficresprobmorprev,"\n");
5600:
5601: fprintf(ficresvij,"%.0f ",age );
5602: for(i=1; i<=nlstate;i++)
5603: for(j=1; j<=nlstate;j++){
5604: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5605: }
5606: fprintf(ficresvij,"\n");
5607: free_matrix(gp,0,nhstepm,1,nlstate);
5608: free_matrix(gm,0,nhstepm,1,nlstate);
5609: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5610: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5611: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5612: } /* End age */
5613: free_vector(gpp,nlstate+1,nlstate+ndeath);
5614: free_vector(gmp,nlstate+1,nlstate+ndeath);
5615: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5616: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5617: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5618: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5619: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5620: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5621: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5622: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5623: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5624: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5625: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5626: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5627: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5628: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5629: 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);
5630: /* 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 5631: */
1.218 brouard 5632: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5633: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5634:
1.218 brouard 5635: free_vector(xp,1,npar);
5636: free_matrix(doldm,1,nlstate,1,nlstate);
5637: free_matrix(dnewm,1,nlstate,1,npar);
5638: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5639: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5640: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5641: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5642: fclose(ficresprobmorprev);
5643: fflush(ficgp);
5644: fflush(fichtm);
5645: } /* end varevsij */
1.126 brouard 5646:
5647: /************ Variance of prevlim ******************/
1.235 brouard 5648: 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 5649: {
1.205 brouard 5650: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5651: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5652:
1.126 brouard 5653: double **dnewm,**doldm;
5654: int i, j, nhstepm, hstepm;
5655: double *xp;
5656: double *gp, *gm;
5657: double **gradg, **trgradg;
1.208 brouard 5658: double **mgm, **mgp;
1.126 brouard 5659: double age,agelim;
5660: int theta;
5661:
5662: pstamp(ficresvpl);
5663: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5664: fprintf(ficresvpl,"# Age ");
5665: if(nresult >=1)
5666: fprintf(ficresvpl," Result# ");
1.126 brouard 5667: for(i=1; i<=nlstate;i++)
5668: fprintf(ficresvpl," %1d-%1d",i,i);
5669: fprintf(ficresvpl,"\n");
5670:
5671: xp=vector(1,npar);
5672: dnewm=matrix(1,nlstate,1,npar);
5673: doldm=matrix(1,nlstate,1,nlstate);
5674:
5675: hstepm=1*YEARM; /* Every year of age */
5676: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5677: agelim = AGESUP;
5678: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5679: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5680: if (stepm >= YEARM) hstepm=1;
5681: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5682: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5683: mgp=matrix(1,npar,1,nlstate);
5684: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5685: gp=vector(1,nlstate);
5686: gm=vector(1,nlstate);
5687:
5688: for(theta=1; theta <=npar; theta++){
5689: for(i=1; i<=npar; i++){ /* Computes gradient */
5690: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5691: }
1.209 brouard 5692: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5693: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5694: else
1.235 brouard 5695: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5696: for(i=1;i<=nlstate;i++){
1.126 brouard 5697: gp[i] = prlim[i][i];
1.208 brouard 5698: mgp[theta][i] = prlim[i][i];
5699: }
1.126 brouard 5700: for(i=1; i<=npar; i++) /* Computes gradient */
5701: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5702: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5703: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5704: else
1.235 brouard 5705: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5706: for(i=1;i<=nlstate;i++){
1.126 brouard 5707: gm[i] = prlim[i][i];
1.208 brouard 5708: mgm[theta][i] = prlim[i][i];
5709: }
1.126 brouard 5710: for(i=1;i<=nlstate;i++)
5711: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5712: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5713: } /* End theta */
5714:
5715: trgradg =matrix(1,nlstate,1,npar);
5716:
5717: for(j=1; j<=nlstate;j++)
5718: for(theta=1; theta <=npar; theta++)
5719: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5720: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5721: /* printf("\nmgm mgp %d ",(int)age); */
5722: /* for(j=1; j<=nlstate;j++){ */
5723: /* printf(" %d ",j); */
5724: /* for(theta=1; theta <=npar; theta++) */
5725: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5726: /* printf("\n "); */
5727: /* } */
5728: /* } */
5729: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5730: /* printf("\n gradg %d ",(int)age); */
5731: /* for(j=1; j<=nlstate;j++){ */
5732: /* printf("%d ",j); */
5733: /* for(theta=1; theta <=npar; theta++) */
5734: /* printf("%d %lf ",theta,gradg[theta][j]); */
5735: /* printf("\n "); */
5736: /* } */
5737: /* } */
1.126 brouard 5738:
5739: for(i=1;i<=nlstate;i++)
5740: varpl[i][(int)age] =0.;
1.209 brouard 5741: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5742: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5743: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5744: }else{
1.126 brouard 5745: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5746: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5747: }
1.126 brouard 5748: for(i=1;i<=nlstate;i++)
5749: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5750:
5751: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 5752: if(nresult >=1)
5753: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 5754: for(i=1; i<=nlstate;i++)
5755: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
5756: fprintf(ficresvpl,"\n");
5757: free_vector(gp,1,nlstate);
5758: free_vector(gm,1,nlstate);
1.208 brouard 5759: free_matrix(mgm,1,npar,1,nlstate);
5760: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 5761: free_matrix(gradg,1,npar,1,nlstate);
5762: free_matrix(trgradg,1,nlstate,1,npar);
5763: } /* End age */
5764:
5765: free_vector(xp,1,npar);
5766: free_matrix(doldm,1,nlstate,1,npar);
5767: free_matrix(dnewm,1,nlstate,1,nlstate);
5768:
5769: }
5770:
5771: /************ Variance of one-step probabilities ******************/
5772: 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 5773: {
5774: int i, j=0, k1, l1, tj;
5775: int k2, l2, j1, z1;
5776: int k=0, l;
5777: int first=1, first1, first2;
5778: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
5779: double **dnewm,**doldm;
5780: double *xp;
5781: double *gp, *gm;
5782: double **gradg, **trgradg;
5783: double **mu;
5784: double age, cov[NCOVMAX+1];
5785: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
5786: int theta;
5787: char fileresprob[FILENAMELENGTH];
5788: char fileresprobcov[FILENAMELENGTH];
5789: char fileresprobcor[FILENAMELENGTH];
5790: double ***varpij;
5791:
5792: strcpy(fileresprob,"PROB_");
5793: strcat(fileresprob,fileres);
5794: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
5795: printf("Problem with resultfile: %s\n", fileresprob);
5796: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
5797: }
5798: strcpy(fileresprobcov,"PROBCOV_");
5799: strcat(fileresprobcov,fileresu);
5800: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
5801: printf("Problem with resultfile: %s\n", fileresprobcov);
5802: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
5803: }
5804: strcpy(fileresprobcor,"PROBCOR_");
5805: strcat(fileresprobcor,fileresu);
5806: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
5807: printf("Problem with resultfile: %s\n", fileresprobcor);
5808: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
5809: }
5810: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5811: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5812: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5813: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5814: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5815: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5816: pstamp(ficresprob);
5817: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
5818: fprintf(ficresprob,"# Age");
5819: pstamp(ficresprobcov);
5820: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
5821: fprintf(ficresprobcov,"# Age");
5822: pstamp(ficresprobcor);
5823: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
5824: fprintf(ficresprobcor,"# Age");
1.126 brouard 5825:
5826:
1.222 brouard 5827: for(i=1; i<=nlstate;i++)
5828: for(j=1; j<=(nlstate+ndeath);j++){
5829: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
5830: fprintf(ficresprobcov," p%1d-%1d ",i,j);
5831: fprintf(ficresprobcor," p%1d-%1d ",i,j);
5832: }
5833: /* fprintf(ficresprob,"\n");
5834: fprintf(ficresprobcov,"\n");
5835: fprintf(ficresprobcor,"\n");
5836: */
5837: xp=vector(1,npar);
5838: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5839: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
5840: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
5841: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
5842: first=1;
5843: fprintf(ficgp,"\n# Routine varprob");
5844: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
5845: fprintf(fichtm,"\n");
5846:
5847: 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);
5848: 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);
5849: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 5850: and drawn. It helps understanding how is the covariance between two incidences.\
5851: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 5852: 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 5853: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
5854: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
5855: standard deviations wide on each axis. <br>\
5856: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
5857: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
5858: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
5859:
1.222 brouard 5860: cov[1]=1;
5861: /* tj=cptcoveff; */
1.225 brouard 5862: tj = (int) pow(2,cptcoveff);
1.222 brouard 5863: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
5864: j1=0;
1.224 brouard 5865: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 5866: if (cptcovn>0) {
5867: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 5868: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5869: fprintf(ficresprob, "**********\n#\n");
5870: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 5871: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5872: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 5873:
1.222 brouard 5874: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 5875: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5876: fprintf(ficgp, "**********\n#\n");
1.220 brouard 5877:
5878:
1.222 brouard 5879: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 5880: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5881: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 5882:
1.222 brouard 5883: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 5884: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5885: fprintf(ficresprobcor, "**********\n#");
5886: if(invalidvarcomb[j1]){
5887: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
5888: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
5889: continue;
5890: }
5891: }
5892: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
5893: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5894: gp=vector(1,(nlstate)*(nlstate+ndeath));
5895: gm=vector(1,(nlstate)*(nlstate+ndeath));
5896: for (age=bage; age<=fage; age ++){
5897: cov[2]=age;
5898: if(nagesqr==1)
5899: cov[3]= age*age;
5900: for (k=1; k<=cptcovn;k++) {
5901: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
5902: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
5903: * 1 1 1 1 1
5904: * 2 2 1 1 1
5905: * 3 1 2 1 1
5906: */
5907: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
5908: }
5909: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
5910: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
5911: for (k=1; k<=cptcovprod;k++)
5912: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 5913:
5914:
1.222 brouard 5915: for(theta=1; theta <=npar; theta++){
5916: for(i=1; i<=npar; i++)
5917: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 5918:
1.222 brouard 5919: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 5920:
1.222 brouard 5921: k=0;
5922: for(i=1; i<= (nlstate); i++){
5923: for(j=1; j<=(nlstate+ndeath);j++){
5924: k=k+1;
5925: gp[k]=pmmij[i][j];
5926: }
5927: }
1.220 brouard 5928:
1.222 brouard 5929: for(i=1; i<=npar; i++)
5930: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 5931:
1.222 brouard 5932: pmij(pmmij,cov,ncovmodel,xp,nlstate);
5933: k=0;
5934: for(i=1; i<=(nlstate); i++){
5935: for(j=1; j<=(nlstate+ndeath);j++){
5936: k=k+1;
5937: gm[k]=pmmij[i][j];
5938: }
5939: }
1.220 brouard 5940:
1.222 brouard 5941: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
5942: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
5943: }
1.126 brouard 5944:
1.222 brouard 5945: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
5946: for(theta=1; theta <=npar; theta++)
5947: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 5948:
1.222 brouard 5949: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
5950: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 5951:
1.222 brouard 5952: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 5953:
1.222 brouard 5954: k=0;
5955: for(i=1; i<=(nlstate); i++){
5956: for(j=1; j<=(nlstate+ndeath);j++){
5957: k=k+1;
5958: mu[k][(int) age]=pmmij[i][j];
5959: }
5960: }
5961: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
5962: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
5963: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 5964:
1.222 brouard 5965: /*printf("\n%d ",(int)age);
5966: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5967: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5968: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5969: }*/
1.220 brouard 5970:
1.222 brouard 5971: fprintf(ficresprob,"\n%d ",(int)age);
5972: fprintf(ficresprobcov,"\n%d ",(int)age);
5973: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 5974:
1.222 brouard 5975: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
5976: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
5977: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5978: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
5979: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
5980: }
5981: i=0;
5982: for (k=1; k<=(nlstate);k++){
5983: for (l=1; l<=(nlstate+ndeath);l++){
5984: i++;
5985: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
5986: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
5987: for (j=1; j<=i;j++){
5988: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
5989: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
5990: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
5991: }
5992: }
5993: }/* end of loop for state */
5994: } /* end of loop for age */
5995: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
5996: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
5997: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
5998: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
5999:
6000: /* Confidence intervalle of pij */
6001: /*
6002: fprintf(ficgp,"\nunset parametric;unset label");
6003: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6004: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6005: 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);
6006: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6007: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6008: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6009: */
6010:
6011: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6012: first1=1;first2=2;
6013: for (k2=1; k2<=(nlstate);k2++){
6014: for (l2=1; l2<=(nlstate+ndeath);l2++){
6015: if(l2==k2) continue;
6016: j=(k2-1)*(nlstate+ndeath)+l2;
6017: for (k1=1; k1<=(nlstate);k1++){
6018: for (l1=1; l1<=(nlstate+ndeath);l1++){
6019: if(l1==k1) continue;
6020: i=(k1-1)*(nlstate+ndeath)+l1;
6021: if(i<=j) continue;
6022: for (age=bage; age<=fage; age ++){
6023: if ((int)age %5==0){
6024: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6025: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6026: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6027: mu1=mu[i][(int) age]/stepm*YEARM ;
6028: mu2=mu[j][(int) age]/stepm*YEARM;
6029: c12=cv12/sqrt(v1*v2);
6030: /* Computing eigen value of matrix of covariance */
6031: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6032: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6033: if ((lc2 <0) || (lc1 <0) ){
6034: if(first2==1){
6035: first1=0;
6036: 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);
6037: }
6038: 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);
6039: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6040: /* lc2=fabs(lc2); */
6041: }
1.220 brouard 6042:
1.222 brouard 6043: /* Eigen vectors */
6044: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6045: /*v21=sqrt(1.-v11*v11); *//* error */
6046: v21=(lc1-v1)/cv12*v11;
6047: v12=-v21;
6048: v22=v11;
6049: tnalp=v21/v11;
6050: if(first1==1){
6051: first1=0;
6052: 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);
6053: }
6054: 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);
6055: /*printf(fignu*/
6056: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6057: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6058: if(first==1){
6059: first=0;
6060: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6061: fprintf(ficgp,"\nset parametric;unset label");
6062: 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);
6063: fprintf(ficgp,"\nset ter svg size 640, 480");
6064: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6065: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6066: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6067: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6068: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6069: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6070: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6071: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6072: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6073: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6074: 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", \
6075: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6076: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6077: }else{
6078: first=0;
6079: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6080: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6081: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6082: 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", \
6083: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6084: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6085: }/* if first */
6086: } /* age mod 5 */
6087: } /* end loop age */
6088: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6089: first=1;
6090: } /*l12 */
6091: } /* k12 */
6092: } /*l1 */
6093: }/* k1 */
6094: } /* loop on combination of covariates j1 */
6095: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6096: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6097: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6098: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6099: free_vector(xp,1,npar);
6100: fclose(ficresprob);
6101: fclose(ficresprobcov);
6102: fclose(ficresprobcor);
6103: fflush(ficgp);
6104: fflush(fichtmcov);
6105: }
1.126 brouard 6106:
6107:
6108: /******************* Printing html file ***********/
1.201 brouard 6109: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6110: int lastpass, int stepm, int weightopt, char model[],\
6111: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6112: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6113: double jprev1, double mprev1,double anprev1, double dateprev1, \
6114: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6115: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6116:
6117: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6118: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6119: </ul>");
1.237 brouard 6120: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6121: </ul>", model);
1.214 brouard 6122: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6123: 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",
6124: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6125: 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 6126: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6127: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6128: fprintf(fichtm,"\
6129: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6130: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6131: fprintf(fichtm,"\
1.217 brouard 6132: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6133: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6134: fprintf(fichtm,"\
1.126 brouard 6135: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6136: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6137: fprintf(fichtm,"\
1.217 brouard 6138: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6139: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6140: fprintf(fichtm,"\
1.211 brouard 6141: - (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 6142: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6143: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6144: if(prevfcast==1){
6145: fprintf(fichtm,"\
6146: - Prevalence projections by age and states: \
1.201 brouard 6147: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6148: }
1.126 brouard 6149:
1.222 brouard 6150: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6151:
1.225 brouard 6152: m=pow(2,cptcoveff);
1.222 brouard 6153: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6154:
1.222 brouard 6155: jj1=0;
1.237 brouard 6156:
6157: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6158: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.237 brouard 6159: if(TKresult[nres]!= k1)
6160: continue;
1.220 brouard 6161:
1.222 brouard 6162: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6163: jj1++;
6164: if (cptcovn > 0) {
6165: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6166: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6167: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6168: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6169: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6170: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6171: }
1.237 brouard 6172: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6173: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6174: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6175: }
6176:
1.230 brouard 6177: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6178: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6179: if(invalidvarcomb[k1]){
6180: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6181: printf("\nCombination (%d) ignored because no cases \n",k1);
6182: continue;
6183: }
6184: }
6185: /* aij, bij */
1.241 brouard 6186: 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> \
6187: <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 6188: /* Pij */
1.241 brouard 6189: 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> \
6190: <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 6191: /* Quasi-incidences */
6192: 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 6193: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6194: 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 6195: 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> \
6196: <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 6197: /* Survival functions (period) in state j */
6198: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6199: 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> \
6200: <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 6201: }
6202: /* State specific survival functions (period) */
6203: for(cpt=1; cpt<=nlstate;cpt++){
6204: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6205: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6206: <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 6207: }
6208: /* Period (stable) prevalence in each health state */
6209: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6210: fprintf(fichtm,"<br>\n- Convergence 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> \
6211: <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 6212: }
6213: if(backcast==1){
6214: /* Period (stable) back prevalence in each health state */
6215: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6216: fprintf(fichtm,"<br>\n- Convergence to period (stable) back 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> \
6217: <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 6218: }
1.217 brouard 6219: }
1.222 brouard 6220: if(prevfcast==1){
6221: /* Projection of prevalence up to period (stable) prevalence in each health state */
6222: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6223: 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> \
6224: <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 6225: }
6226: }
1.220 brouard 6227:
1.222 brouard 6228: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6229: 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> \
6230: <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 6231: }
6232: /* } /\* end i1 *\/ */
6233: }/* End k1 */
6234: fprintf(fichtm,"</ul>");
1.126 brouard 6235:
1.222 brouard 6236: fprintf(fichtm,"\
1.126 brouard 6237: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6238: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6239: - 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 6240: But because parameters are usually highly correlated (a higher incidence of disability \
6241: and a higher incidence of recovery can give very close observed transition) it might \
6242: be very useful to look not only at linear confidence intervals estimated from the \
6243: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6244: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6245: covariance matrix of the one-step probabilities. \
6246: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6247:
1.222 brouard 6248: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6249: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6250: fprintf(fichtm,"\
1.126 brouard 6251: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6252: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6253:
1.222 brouard 6254: fprintf(fichtm,"\
1.126 brouard 6255: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6256: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6257: fprintf(fichtm,"\
1.126 brouard 6258: - 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): \
6259: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6260: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6261: fprintf(fichtm,"\
1.126 brouard 6262: - (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): \
6263: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6264: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6265: fprintf(fichtm,"\
1.128 brouard 6266: - 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 6267: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6268: fprintf(fichtm,"\
1.128 brouard 6269: - 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 6270: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6271: fprintf(fichtm,"\
1.126 brouard 6272: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6273: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6274:
6275: /* if(popforecast==1) fprintf(fichtm,"\n */
6276: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6277: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6278: /* <br>",fileres,fileres,fileres,fileres); */
6279: /* else */
6280: /* 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 6281: fflush(fichtm);
6282: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6283:
1.225 brouard 6284: m=pow(2,cptcoveff);
1.222 brouard 6285: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6286:
1.222 brouard 6287: jj1=0;
1.237 brouard 6288:
1.241 brouard 6289: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6290: for(k1=1; k1<=m;k1++){
1.237 brouard 6291: if(TKresult[nres]!= k1)
6292: continue;
1.222 brouard 6293: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6294: jj1++;
1.126 brouard 6295: if (cptcovn > 0) {
6296: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6297: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6298: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6299: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6300: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6301: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6302: }
6303:
1.126 brouard 6304: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6305:
1.222 brouard 6306: if(invalidvarcomb[k1]){
6307: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6308: continue;
6309: }
1.126 brouard 6310: }
6311: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6312: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6313: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6314: <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 6315: }
6316: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6317: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6318: true period expectancies (those weighted with period prevalences are also\
6319: drawn in addition to the population based expectancies computed using\
1.241 brouard 6320: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6321: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6322: /* } /\* end i1 *\/ */
6323: }/* End k1 */
1.241 brouard 6324: }/* End nres */
1.222 brouard 6325: fprintf(fichtm,"</ul>");
6326: fflush(fichtm);
1.126 brouard 6327: }
6328:
6329: /******************* Gnuplot file **************/
1.223 brouard 6330: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6331:
6332: char dirfileres[132],optfileres[132];
1.223 brouard 6333: char gplotcondition[132];
1.237 brouard 6334: 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 6335: int lv=0, vlv=0, kl=0;
1.130 brouard 6336: int ng=0;
1.201 brouard 6337: int vpopbased;
1.223 brouard 6338: int ioffset; /* variable offset for columns */
1.235 brouard 6339: int nres=0; /* Index of resultline */
1.219 brouard 6340:
1.126 brouard 6341: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6342: /* printf("Problem with file %s",optionfilegnuplot); */
6343: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6344: /* } */
6345:
6346: /*#ifdef windows */
6347: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6348: /*#endif */
1.225 brouard 6349: m=pow(2,cptcoveff);
1.126 brouard 6350:
1.202 brouard 6351: /* Contribution to likelihood */
6352: /* Plot the probability implied in the likelihood */
1.223 brouard 6353: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6354: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6355: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6356: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6357: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6358: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6359: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6360: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6361: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6362: 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));
6363: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6364: 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));
6365: for (i=1; i<= nlstate ; i ++) {
6366: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6367: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6368: 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);
6369: for (j=2; j<= nlstate+ndeath ; j ++) {
6370: 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);
6371: }
6372: fprintf(ficgp,";\nset out; unset ylabel;\n");
6373: }
6374: /* 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 */
6375: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6376: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6377: fprintf(ficgp,"\nset out;unset log\n");
6378: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6379:
1.126 brouard 6380: strcpy(dirfileres,optionfilefiname);
6381: strcpy(optfileres,"vpl");
1.223 brouard 6382: /* 1eme*/
1.238 brouard 6383: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6384: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6385: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6386: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
6387: if(TKresult[nres]!= k1)
6388: continue;
6389: /* We are interested in selected combination by the resultline */
6390: printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6391: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6392: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6393: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6394: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6395: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6396: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6397: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6398: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
6399: printf(" V%d=%d ",Tvaraff[k],vlv);
6400: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6401: }
6402: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6403: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6404: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6405: }
6406: printf("\n#\n");
6407: fprintf(ficgp,"\n#\n");
6408: if(invalidvarcomb[k1]){
6409: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6410: continue;
6411: }
1.235 brouard 6412:
1.241 brouard 6413: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6414: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6415: 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 6416:
1.238 brouard 6417: for (i=1; i<= nlstate ; i ++) {
6418: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6419: else fprintf(ficgp," %%*lf (%%*lf)");
6420: }
1.242 ! brouard 6421: 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 6422: for (i=1; i<= nlstate ; i ++) {
6423: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6424: else fprintf(ficgp," %%*lf (%%*lf)");
6425: }
1.242 ! brouard 6426: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6427: for (i=1; i<= nlstate ; i ++) {
6428: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6429: else fprintf(ficgp," %%*lf (%%*lf)");
6430: }
6431: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6432: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6433: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 ! brouard 6434: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6435: if(cptcoveff ==0){
6436: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line ", 2+(cpt-1), cpt );
6437: }else{
6438: kl=0;
6439: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6440: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6441: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6442: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6443: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6444: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6445: kl++;
1.238 brouard 6446: /* 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 *\/ */
6447: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6448: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6449: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6450: if(k==cptcoveff){
6451: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 ! brouard 6452: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6453: }else{
6454: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6455: kl++;
6456: }
6457: } /* end covariate */
6458: } /* end if no covariate */
6459: } /* end if backcast */
6460: fprintf(ficgp,"\nset out \n");
6461: } /* nres */
1.201 brouard 6462: } /* k1 */
6463: } /* cpt */
1.235 brouard 6464:
6465:
1.126 brouard 6466: /*2 eme*/
1.238 brouard 6467: for (k1=1; k1<= m ; k1 ++){
6468: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6469: if(TKresult[nres]!= k1)
6470: continue;
6471: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6472: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6473: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6474: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6475: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6476: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6477: vlv= nbcode[Tvaraff[k]][lv];
6478: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6479: }
1.237 brouard 6480: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6481: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6482: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6483: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6484: }
1.211 brouard 6485: fprintf(ficgp,"\n#\n");
1.223 brouard 6486: if(invalidvarcomb[k1]){
6487: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6488: continue;
6489: }
1.219 brouard 6490:
1.241 brouard 6491: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6492: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6493: if(vpopbased==0)
6494: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6495: else
6496: fprintf(ficgp,"\nreplot ");
6497: for (i=1; i<= nlstate+1 ; i ++) {
6498: k=2*i;
6499: 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);
6500: for (j=1; j<= nlstate+1 ; j ++) {
6501: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6502: else fprintf(ficgp," %%*lf (%%*lf)");
6503: }
6504: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6505: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6506: 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);
6507: for (j=1; j<= nlstate+1 ; j ++) {
6508: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6509: else fprintf(ficgp," %%*lf (%%*lf)");
6510: }
6511: fprintf(ficgp,"\" t\"\" w l lt 0,");
6512: 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);
6513: for (j=1; j<= nlstate+1 ; j ++) {
6514: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6515: else fprintf(ficgp," %%*lf (%%*lf)");
6516: }
6517: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6518: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6519: } /* state */
6520: } /* vpopbased */
6521: fprintf(ficgp,"\nset out;set out \"%s_%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1); /* Buggy gnuplot */
6522: } /* end nres */
6523: } /* k1 end 2 eme*/
6524:
6525:
6526: /*3eme*/
6527: for (k1=1; k1<= m ; k1 ++){
6528: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.240 brouard 6529: if(TKresult[nres]!= k1)
1.238 brouard 6530: continue;
6531:
6532: for (cpt=1; cpt<= nlstate ; cpt ++) {
6533: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6534: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6535: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6536: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6537: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6538: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6539: vlv= nbcode[Tvaraff[k]][lv];
6540: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6541: }
6542: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6543: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6544: }
6545: fprintf(ficgp,"\n#\n");
6546: if(invalidvarcomb[k1]){
6547: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6548: continue;
6549: }
6550:
6551: /* k=2+nlstate*(2*cpt-2); */
6552: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6553: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6554: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6555: 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 6556: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6557: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6558: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6559: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6560: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6561: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6562:
1.238 brouard 6563: */
6564: for (i=1; i< nlstate ; i ++) {
6565: 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);
6566: /* 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 6567:
1.238 brouard 6568: }
6569: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6570: }
6571: } /* end nres */
6572: } /* end kl 3eme */
1.126 brouard 6573:
1.223 brouard 6574: /* 4eme */
1.201 brouard 6575: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6576: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6577: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6578: if(TKresult[nres]!= k1)
1.223 brouard 6579: continue;
1.238 brouard 6580: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6581: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6582: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6583: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6584: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6585: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6586: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6587: vlv= nbcode[Tvaraff[k]][lv];
6588: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6589: }
6590: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6591: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6592: }
6593: fprintf(ficgp,"\n#\n");
6594: if(invalidvarcomb[k1]){
6595: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6596: continue;
1.223 brouard 6597: }
1.238 brouard 6598:
1.241 brouard 6599: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6600: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6601: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6602: k=3;
6603: for (i=1; i<= nlstate ; i ++){
6604: if(i==1){
6605: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6606: }else{
6607: fprintf(ficgp,", '' ");
6608: }
6609: l=(nlstate+ndeath)*(i-1)+1;
6610: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6611: for (j=2; j<= nlstate+ndeath ; j ++)
6612: fprintf(ficgp,"+$%d",k+l+j-1);
6613: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6614: } /* nlstate */
6615: fprintf(ficgp,"\nset out\n");
6616: } /* end cpt state*/
6617: } /* end nres */
6618: } /* end covariate k1 */
6619:
1.220 brouard 6620: /* 5eme */
1.201 brouard 6621: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6622: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6623: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6624: if(TKresult[nres]!= k1)
1.227 brouard 6625: continue;
1.238 brouard 6626: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6627: 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);
6628: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6629: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6630: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6631: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6632: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6633: vlv= nbcode[Tvaraff[k]][lv];
6634: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6635: }
6636: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6637: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6638: }
6639: fprintf(ficgp,"\n#\n");
6640: if(invalidvarcomb[k1]){
6641: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6642: continue;
6643: }
1.227 brouard 6644:
1.241 brouard 6645: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6646: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6647: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6648: k=3;
6649: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6650: if(j==1)
6651: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6652: else
6653: fprintf(ficgp,", '' ");
6654: l=(nlstate+ndeath)*(cpt-1) +j;
6655: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6656: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6657: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6658: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6659: } /* nlstate */
6660: fprintf(ficgp,", '' ");
6661: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6662: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6663: l=(nlstate+ndeath)*(cpt-1) +j;
6664: if(j < nlstate)
6665: fprintf(ficgp,"$%d +",k+l);
6666: else
6667: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6668: }
6669: fprintf(ficgp,"\nset out\n");
6670: } /* end cpt state*/
6671: } /* end covariate */
6672: } /* end nres */
1.227 brouard 6673:
1.220 brouard 6674: /* 6eme */
1.202 brouard 6675: /* CV preval stable (period) for each covariate */
1.237 brouard 6676: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6677: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6678: if(TKresult[nres]!= k1)
6679: continue;
1.153 brouard 6680: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6681:
1.211 brouard 6682: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6683: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6684: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6685: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6686: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6687: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6688: vlv= nbcode[Tvaraff[k]][lv];
6689: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6690: }
1.237 brouard 6691: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6692: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6693: }
1.211 brouard 6694: fprintf(ficgp,"\n#\n");
1.223 brouard 6695: if(invalidvarcomb[k1]){
1.227 brouard 6696: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6697: continue;
1.223 brouard 6698: }
1.227 brouard 6699:
1.241 brouard 6700: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6701: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6702: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6703: k=3; /* Offset */
1.153 brouard 6704: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6705: if(i==1)
6706: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6707: else
6708: fprintf(ficgp,", '' ");
6709: l=(nlstate+ndeath)*(i-1)+1;
6710: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6711: for (j=2; j<= nlstate ; j ++)
6712: fprintf(ficgp,"+$%d",k+l+j-1);
6713: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6714: } /* nlstate */
1.201 brouard 6715: fprintf(ficgp,"\nset out\n");
1.153 brouard 6716: } /* end cpt state*/
6717: } /* end covariate */
1.227 brouard 6718:
6719:
1.220 brouard 6720: /* 7eme */
1.218 brouard 6721: if(backcast == 1){
1.217 brouard 6722: /* CV back preval stable (period) for each covariate */
1.237 brouard 6723: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6724: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6725: if(TKresult[nres]!= k1)
6726: continue;
1.218 brouard 6727: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6728: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6729: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6730: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6731: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6732: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6733: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6734: vlv= nbcode[Tvaraff[k]][lv];
6735: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6736: }
1.237 brouard 6737: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6738: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6739: }
1.227 brouard 6740: fprintf(ficgp,"\n#\n");
6741: if(invalidvarcomb[k1]){
6742: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6743: continue;
6744: }
6745:
1.241 brouard 6746: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6747: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6748: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6749: k=3; /* Offset */
6750: for (i=1; i<= nlstate ; i ++){
6751: if(i==1)
6752: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6753: else
6754: fprintf(ficgp,", '' ");
6755: /* l=(nlstate+ndeath)*(i-1)+1; */
6756: l=(nlstate+ndeath)*(cpt-1)+1;
6757: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
6758: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
6759: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
6760: /* for (j=2; j<= nlstate ; j ++) */
6761: /* fprintf(ficgp,"+$%d",k+l+j-1); */
6762: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
6763: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
6764: } /* nlstate */
6765: fprintf(ficgp,"\nset out\n");
1.218 brouard 6766: } /* end cpt state*/
6767: } /* end covariate */
6768: } /* End if backcast */
6769:
1.223 brouard 6770: /* 8eme */
1.218 brouard 6771: if(prevfcast==1){
6772: /* Projection from cross-sectional to stable (period) for each covariate */
6773:
1.237 brouard 6774: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6775: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6776: if(TKresult[nres]!= k1)
6777: continue;
1.211 brouard 6778: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6779: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
6780: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
6781: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6782: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6783: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6784: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6785: vlv= nbcode[Tvaraff[k]][lv];
6786: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6787: }
1.237 brouard 6788: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6789: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6790: }
1.227 brouard 6791: fprintf(ficgp,"\n#\n");
6792: if(invalidvarcomb[k1]){
6793: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6794: continue;
6795: }
6796:
6797: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 6798: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 6799: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 6800: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6801: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
6802: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6803: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6804: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6805: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6806: if(i==1){
6807: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
6808: }else{
6809: fprintf(ficgp,",\\\n '' ");
6810: }
6811: if(cptcoveff ==0){ /* No covariate */
6812: ioffset=2; /* Age is in 2 */
6813: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6814: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6815: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6816: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6817: fprintf(ficgp," u %d:(", ioffset);
6818: if(i==nlstate+1)
6819: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
6820: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6821: else
6822: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
6823: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6824: }else{ /* more than 2 covariates */
6825: if(cptcoveff ==1){
6826: ioffset=4; /* Age is in 4 */
6827: }else{
6828: ioffset=6; /* Age is in 6 */
6829: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6830: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6831: }
6832: fprintf(ficgp," u %d:(",ioffset);
6833: kl=0;
6834: strcpy(gplotcondition,"(");
6835: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
6836: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
6837: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6838: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6839: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6840: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
6841: kl++;
6842: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
6843: kl++;
6844: if(k <cptcoveff && cptcoveff>1)
6845: sprintf(gplotcondition+strlen(gplotcondition)," && ");
6846: }
6847: strcpy(gplotcondition+strlen(gplotcondition),")");
6848: /* 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 *\/ */
6849: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6850: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6851: /* '' 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*/
6852: if(i==nlstate+1){
6853: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
6854: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6855: }else{
6856: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
6857: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6858: }
6859: } /* end if covariate */
6860: } /* nlstate */
6861: fprintf(ficgp,"\nset out\n");
1.223 brouard 6862: } /* end cpt state*/
6863: } /* end covariate */
6864: } /* End if prevfcast */
1.227 brouard 6865:
6866:
1.238 brouard 6867: /* 9eme writing MLE parameters */
6868: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 6869: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 6870: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 6871: for(k=1; k <=(nlstate+ndeath); k++){
6872: if (k != i) {
1.227 brouard 6873: fprintf(ficgp,"# current state %d\n",k);
6874: for(j=1; j <=ncovmodel; j++){
6875: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
6876: jk++;
6877: }
6878: fprintf(ficgp,"\n");
1.126 brouard 6879: }
6880: }
1.223 brouard 6881: }
1.187 brouard 6882: fprintf(ficgp,"##############\n#\n");
1.227 brouard 6883:
1.145 brouard 6884: /*goto avoid;*/
1.238 brouard 6885: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
6886: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 6887: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
6888: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
6889: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
6890: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
6891: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6892: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6893: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6894: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6895: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
6896: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6897: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
6898: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
6899: fprintf(ficgp,"#\n");
1.223 brouard 6900: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 6901: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 6902: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 6903: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 6904: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
6905: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
6906: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6907: if(TKresult[nres]!= jk)
6908: continue;
6909: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
6910: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6911: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6912: }
6913: fprintf(ficgp,"\n#\n");
1.241 brouard 6914: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 6915: fprintf(ficgp,"\nset ter svg size 640, 480 ");
6916: if (ng==1){
6917: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
6918: fprintf(ficgp,"\nunset log y");
6919: }else if (ng==2){
6920: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
6921: fprintf(ficgp,"\nset log y");
6922: }else if (ng==3){
6923: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
6924: fprintf(ficgp,"\nset log y");
6925: }else
6926: fprintf(ficgp,"\nunset title ");
6927: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
6928: i=1;
6929: for(k2=1; k2<=nlstate; k2++) {
6930: k3=i;
6931: for(k=1; k<=(nlstate+ndeath); k++) {
6932: if (k != k2){
6933: switch( ng) {
6934: case 1:
6935: if(nagesqr==0)
6936: fprintf(ficgp," p%d+p%d*x",i,i+1);
6937: else /* nagesqr =1 */
6938: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6939: break;
6940: case 2: /* ng=2 */
6941: if(nagesqr==0)
6942: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
6943: else /* nagesqr =1 */
6944: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6945: break;
6946: case 3:
6947: if(nagesqr==0)
6948: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
6949: else /* nagesqr =1 */
6950: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
6951: break;
6952: }
6953: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 6954: ijp=1; /* product no age */
6955: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
6956: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 6957: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 6958: if(j==Tage[ij]) { /* Product by age */
6959: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 6960: if(DummyV[j]==0){
1.237 brouard 6961: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
6962: }else{ /* quantitative */
6963: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
6964: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
6965: }
6966: ij++;
6967: }
6968: }else if(j==Tprod[ijp]) { /* */
6969: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
6970: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 6971: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
6972: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 6973: /* 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)]); */
6974: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
6975: }else{ /* Vn is dummy and Vm is quanti */
6976: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
6977: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
6978: }
6979: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 6980: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 6981: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
6982: }else{ /* Both quanti */
6983: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
6984: }
6985: }
1.238 brouard 6986: ijp++;
1.237 brouard 6987: }
6988: } else{ /* simple covariate */
6989: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
6990: if(Dummy[j]==0){
6991: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
6992: }else{ /* quantitative */
6993: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 6994: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
6995: }
1.237 brouard 6996: } /* end simple */
6997: } /* end j */
1.223 brouard 6998: }else{
6999: i=i-ncovmodel;
7000: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7001: fprintf(ficgp," (1.");
7002: }
1.227 brouard 7003:
1.223 brouard 7004: if(ng != 1){
7005: fprintf(ficgp,")/(1");
1.227 brouard 7006:
1.223 brouard 7007: for(k1=1; k1 <=nlstate; k1++){
7008: if(nagesqr==0)
7009: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7010: else /* nagesqr =1 */
7011: 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 7012:
1.223 brouard 7013: ij=1;
7014: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7015: if((j-2)==Tage[ij]) { /* Bug valgrind */
7016: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7017: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7018: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7019: ij++;
7020: }
7021: }
7022: else
1.225 brouard 7023: 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 7024: }
7025: fprintf(ficgp,")");
7026: }
7027: fprintf(ficgp,")");
7028: if(ng ==2)
7029: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7030: else /* ng= 3 */
7031: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7032: }else{ /* end ng <> 1 */
7033: if( k !=k2) /* logit p11 is hard to draw */
7034: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7035: }
7036: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7037: fprintf(ficgp,",");
7038: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7039: fprintf(ficgp,",");
7040: i=i+ncovmodel;
7041: } /* end k */
7042: } /* end k2 */
7043: fprintf(ficgp,"\n set out\n");
7044: } /* end jk */
7045: } /* end ng */
7046: /* avoid: */
7047: fflush(ficgp);
1.126 brouard 7048: } /* end gnuplot */
7049:
7050:
7051: /*************** Moving average **************/
1.219 brouard 7052: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7053: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7054:
1.222 brouard 7055: int i, cpt, cptcod;
7056: int modcovmax =1;
7057: int mobilavrange, mob;
7058: int iage=0;
7059:
7060: double sum=0.;
7061: double age;
7062: double *sumnewp, *sumnewm;
7063: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7064:
7065:
1.225 brouard 7066: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7067: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7068:
7069: sumnewp = vector(1,ncovcombmax);
7070: sumnewm = vector(1,ncovcombmax);
7071: agemingood = vector(1,ncovcombmax);
7072: agemaxgood = vector(1,ncovcombmax);
7073:
7074: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7075: sumnewm[cptcod]=0.;
7076: sumnewp[cptcod]=0.;
7077: agemingood[cptcod]=0;
7078: agemaxgood[cptcod]=0;
7079: }
7080: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7081:
7082: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7083: if(mobilav==1) mobilavrange=5; /* default */
7084: else mobilavrange=mobilav;
7085: for (age=bage; age<=fage; age++)
7086: for (i=1; i<=nlstate;i++)
7087: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7088: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7089: /* We keep the original values on the extreme ages bage, fage and for
7090: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7091: we use a 5 terms etc. until the borders are no more concerned.
7092: */
7093: for (mob=3;mob <=mobilavrange;mob=mob+2){
7094: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7095: for (i=1; i<=nlstate;i++){
7096: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7097: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7098: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7099: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7100: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7101: }
7102: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7103: }
7104: }
7105: }/* end age */
7106: }/* end mob */
7107: }else
7108: return -1;
7109: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7110: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7111: if(invalidvarcomb[cptcod]){
7112: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7113: continue;
7114: }
1.219 brouard 7115:
1.222 brouard 7116: agemingood[cptcod]=fage-(mob-1)/2;
7117: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7118: sumnewm[cptcod]=0.;
7119: for (i=1; i<=nlstate;i++){
7120: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7121: }
7122: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7123: agemingood[cptcod]=age;
7124: }else{ /* bad */
7125: for (i=1; i<=nlstate;i++){
7126: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7127: } /* i */
7128: } /* end bad */
7129: }/* age */
7130: sum=0.;
7131: for (i=1; i<=nlstate;i++){
7132: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7133: }
7134: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7135: 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);
7136: /* for (i=1; i<=nlstate;i++){ */
7137: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7138: /* } /\* i *\/ */
7139: } /* end bad */
7140: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7141: /* From youngest, finding the oldest wrong */
7142: agemaxgood[cptcod]=bage+(mob-1)/2;
7143: for (age=bage+(mob-1)/2; age<=fage; age++){
7144: sumnewm[cptcod]=0.;
7145: for (i=1; i<=nlstate;i++){
7146: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7147: }
7148: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7149: agemaxgood[cptcod]=age;
7150: }else{ /* bad */
7151: for (i=1; i<=nlstate;i++){
7152: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7153: } /* i */
7154: } /* end bad */
7155: }/* age */
7156: sum=0.;
7157: for (i=1; i<=nlstate;i++){
7158: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7159: }
7160: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7161: 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);
7162: /* for (i=1; i<=nlstate;i++){ */
7163: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7164: /* } /\* i *\/ */
7165: } /* end bad */
7166:
7167: for (age=bage; age<=fage; age++){
1.235 brouard 7168: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7169: sumnewp[cptcod]=0.;
7170: sumnewm[cptcod]=0.;
7171: for (i=1; i<=nlstate;i++){
7172: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7173: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7174: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7175: }
7176: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7177: }
7178: /* printf("\n"); */
7179: /* } */
7180: /* brutal averaging */
7181: for (i=1; i<=nlstate;i++){
7182: for (age=1; age<=bage; age++){
7183: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7184: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7185: }
7186: for (age=fage; age<=AGESUP; age++){
7187: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7188: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7189: }
7190: } /* end i status */
7191: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7192: for (age=1; age<=AGESUP; age++){
7193: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7194: mobaverage[(int)age][i][cptcod]=0.;
7195: }
7196: }
7197: }/* end cptcod */
7198: free_vector(sumnewm,1, ncovcombmax);
7199: free_vector(sumnewp,1, ncovcombmax);
7200: free_vector(agemaxgood,1, ncovcombmax);
7201: free_vector(agemingood,1, ncovcombmax);
7202: return 0;
7203: }/* End movingaverage */
1.218 brouard 7204:
1.126 brouard 7205:
7206: /************** Forecasting ******************/
1.235 brouard 7207: 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 7208: /* proj1, year, month, day of starting projection
7209: agemin, agemax range of age
7210: dateprev1 dateprev2 range of dates during which prevalence is computed
7211: anproj2 year of en of projection (same day and month as proj1).
7212: */
1.235 brouard 7213: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7214: double agec; /* generic age */
7215: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7216: double *popeffectif,*popcount;
7217: double ***p3mat;
1.218 brouard 7218: /* double ***mobaverage; */
1.126 brouard 7219: char fileresf[FILENAMELENGTH];
7220:
7221: agelim=AGESUP;
1.211 brouard 7222: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7223: in each health status at the date of interview (if between dateprev1 and dateprev2).
7224: We still use firstpass and lastpass as another selection.
7225: */
1.214 brouard 7226: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7227: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7228:
1.201 brouard 7229: strcpy(fileresf,"F_");
7230: strcat(fileresf,fileresu);
1.126 brouard 7231: if((ficresf=fopen(fileresf,"w"))==NULL) {
7232: printf("Problem with forecast resultfile: %s\n", fileresf);
7233: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7234: }
1.235 brouard 7235: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7236: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7237:
1.225 brouard 7238: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7239:
7240:
7241: stepsize=(int) (stepm+YEARM-1)/YEARM;
7242: if (stepm<=12) stepsize=1;
7243: if(estepm < stepm){
7244: printf ("Problem %d lower than %d\n",estepm, stepm);
7245: }
7246: else hstepm=estepm;
7247:
7248: hstepm=hstepm/stepm;
7249: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7250: fractional in yp1 */
7251: anprojmean=yp;
7252: yp2=modf((yp1*12),&yp);
7253: mprojmean=yp;
7254: yp1=modf((yp2*30.5),&yp);
7255: jprojmean=yp;
7256: if(jprojmean==0) jprojmean=1;
7257: if(mprojmean==0) jprojmean=1;
7258:
1.227 brouard 7259: i1=pow(2,cptcoveff);
1.126 brouard 7260: if (cptcovn < 1){i1=1;}
7261:
7262: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7263:
7264: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7265:
1.126 brouard 7266: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7267: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7268: for(k=1; k<=i1;k++){
7269: if(TKresult[nres]!= k)
7270: continue;
1.227 brouard 7271: if(invalidvarcomb[k]){
7272: printf("\nCombination (%d) projection ignored because no cases \n",k);
7273: continue;
7274: }
7275: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7276: for(j=1;j<=cptcoveff;j++) {
7277: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7278: }
1.235 brouard 7279: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7280: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7281: }
1.227 brouard 7282: fprintf(ficresf," yearproj age");
7283: for(j=1; j<=nlstate+ndeath;j++){
7284: for(i=1; i<=nlstate;i++)
7285: fprintf(ficresf," p%d%d",i,j);
7286: fprintf(ficresf," wp.%d",j);
7287: }
7288: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7289: fprintf(ficresf,"\n");
7290: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7291: for (agec=fage; agec>=(ageminpar-1); agec--){
7292: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7293: nhstepm = nhstepm/hstepm;
7294: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7295: oldm=oldms;savm=savms;
1.235 brouard 7296: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7297:
7298: for (h=0; h<=nhstepm; h++){
7299: if (h*hstepm/YEARM*stepm ==yearp) {
7300: fprintf(ficresf,"\n");
7301: for(j=1;j<=cptcoveff;j++)
7302: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7303: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7304: }
7305: for(j=1; j<=nlstate+ndeath;j++) {
7306: ppij=0.;
7307: for(i=1; i<=nlstate;i++) {
7308: if (mobilav==1)
7309: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7310: else {
7311: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7312: }
7313: if (h*hstepm/YEARM*stepm== yearp) {
7314: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7315: }
7316: } /* end i */
7317: if (h*hstepm/YEARM*stepm==yearp) {
7318: fprintf(ficresf," %.3f", ppij);
7319: }
7320: }/* end j */
7321: } /* end h */
7322: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7323: } /* end agec */
7324: } /* end yearp */
7325: } /* end k */
1.219 brouard 7326:
1.126 brouard 7327: fclose(ficresf);
1.215 brouard 7328: printf("End of Computing forecasting \n");
7329: fprintf(ficlog,"End of Computing forecasting\n");
7330:
1.126 brouard 7331: }
7332:
1.218 brouard 7333: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7334: /* 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 7335: /* /\* back1, year, month, day of starting backection */
7336: /* agemin, agemax range of age */
7337: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7338: /* anback2 year of en of backection (same day and month as back1). */
7339: /* *\/ */
7340: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7341: /* double agec; /\* generic age *\/ */
7342: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7343: /* double *popeffectif,*popcount; */
7344: /* double ***p3mat; */
7345: /* /\* double ***mobaverage; *\/ */
7346: /* char fileresfb[FILENAMELENGTH]; */
7347:
7348: /* agelim=AGESUP; */
7349: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7350: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7351: /* We still use firstpass and lastpass as another selection. */
7352: /* *\/ */
7353: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7354: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7355: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7356:
7357: /* strcpy(fileresfb,"FB_"); */
7358: /* strcat(fileresfb,fileresu); */
7359: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7360: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7361: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7362: /* } */
7363: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7364: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7365:
1.225 brouard 7366: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7367:
7368: /* /\* if (mobilav!=0) { *\/ */
7369: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7370: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7371: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7372: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7373: /* /\* } *\/ */
7374: /* /\* } *\/ */
7375:
7376: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7377: /* if (stepm<=12) stepsize=1; */
7378: /* if(estepm < stepm){ */
7379: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7380: /* } */
7381: /* else hstepm=estepm; */
7382:
7383: /* hstepm=hstepm/stepm; */
7384: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7385: /* fractional in yp1 *\/ */
7386: /* anprojmean=yp; */
7387: /* yp2=modf((yp1*12),&yp); */
7388: /* mprojmean=yp; */
7389: /* yp1=modf((yp2*30.5),&yp); */
7390: /* jprojmean=yp; */
7391: /* if(jprojmean==0) jprojmean=1; */
7392: /* if(mprojmean==0) jprojmean=1; */
7393:
1.225 brouard 7394: /* i1=cptcoveff; */
1.218 brouard 7395: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7396:
1.218 brouard 7397: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7398:
1.218 brouard 7399: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7400:
7401: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7402: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7403: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7404: /* k=k+1; */
7405: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7406: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7407: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7408: /* } */
7409: /* fprintf(ficresfb," yearbproj age"); */
7410: /* for(j=1; j<=nlstate+ndeath;j++){ */
7411: /* for(i=1; i<=nlstate;i++) */
7412: /* fprintf(ficresfb," p%d%d",i,j); */
7413: /* fprintf(ficresfb," p.%d",j); */
7414: /* } */
7415: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7416: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7417: /* fprintf(ficresfb,"\n"); */
7418: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7419: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7420: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7421: /* nhstepm = nhstepm/hstepm; */
7422: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7423: /* oldm=oldms;savm=savms; */
7424: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7425: /* for (h=0; h<=nhstepm; h++){ */
7426: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7427: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7428: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7429: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7430: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7431: /* } */
7432: /* for(j=1; j<=nlstate+ndeath;j++) { */
7433: /* ppij=0.; */
7434: /* for(i=1; i<=nlstate;i++) { */
7435: /* if (mobilav==1) */
7436: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7437: /* else { */
7438: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7439: /* } */
7440: /* if (h*hstepm/YEARM*stepm== yearp) { */
7441: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7442: /* } */
7443: /* } /\* end i *\/ */
7444: /* if (h*hstepm/YEARM*stepm==yearp) { */
7445: /* fprintf(ficresfb," %.3f", ppij); */
7446: /* } */
7447: /* }/\* end j *\/ */
7448: /* } /\* end h *\/ */
7449: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7450: /* } /\* end agec *\/ */
7451: /* } /\* end yearp *\/ */
7452: /* } /\* end cptcod *\/ */
7453: /* } /\* end cptcov *\/ */
7454:
7455: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7456:
7457: /* fclose(ficresfb); */
7458: /* printf("End of Computing Back forecasting \n"); */
7459: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7460:
1.218 brouard 7461: /* } */
1.217 brouard 7462:
1.126 brouard 7463: /************** Forecasting *****not tested NB*************/
1.227 brouard 7464: /* 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 7465:
1.227 brouard 7466: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7467: /* int *popage; */
7468: /* double calagedatem, agelim, kk1, kk2; */
7469: /* double *popeffectif,*popcount; */
7470: /* double ***p3mat,***tabpop,***tabpopprev; */
7471: /* /\* double ***mobaverage; *\/ */
7472: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7473:
1.227 brouard 7474: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7475: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7476: /* agelim=AGESUP; */
7477: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7478:
1.227 brouard 7479: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7480:
7481:
1.227 brouard 7482: /* strcpy(filerespop,"POP_"); */
7483: /* strcat(filerespop,fileresu); */
7484: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7485: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7486: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7487: /* } */
7488: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7489: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7490:
1.227 brouard 7491: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7492:
1.227 brouard 7493: /* /\* if (mobilav!=0) { *\/ */
7494: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7495: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7496: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7497: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7498: /* /\* } *\/ */
7499: /* /\* } *\/ */
1.126 brouard 7500:
1.227 brouard 7501: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7502: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7503:
1.227 brouard 7504: /* agelim=AGESUP; */
1.126 brouard 7505:
1.227 brouard 7506: /* hstepm=1; */
7507: /* hstepm=hstepm/stepm; */
1.218 brouard 7508:
1.227 brouard 7509: /* if (popforecast==1) { */
7510: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7511: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7512: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7513: /* } */
7514: /* popage=ivector(0,AGESUP); */
7515: /* popeffectif=vector(0,AGESUP); */
7516: /* popcount=vector(0,AGESUP); */
1.126 brouard 7517:
1.227 brouard 7518: /* i=1; */
7519: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7520:
1.227 brouard 7521: /* imx=i; */
7522: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7523: /* } */
1.218 brouard 7524:
1.227 brouard 7525: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7526: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7527: /* k=k+1; */
7528: /* fprintf(ficrespop,"\n#******"); */
7529: /* for(j=1;j<=cptcoveff;j++) { */
7530: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7531: /* } */
7532: /* fprintf(ficrespop,"******\n"); */
7533: /* fprintf(ficrespop,"# Age"); */
7534: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7535: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7536:
1.227 brouard 7537: /* for (cpt=0; cpt<=0;cpt++) { */
7538: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7539:
1.227 brouard 7540: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7541: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7542: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7543:
1.227 brouard 7544: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7545: /* oldm=oldms;savm=savms; */
7546: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7547:
1.227 brouard 7548: /* for (h=0; h<=nhstepm; h++){ */
7549: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7550: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7551: /* } */
7552: /* for(j=1; j<=nlstate+ndeath;j++) { */
7553: /* kk1=0.;kk2=0; */
7554: /* for(i=1; i<=nlstate;i++) { */
7555: /* if (mobilav==1) */
7556: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7557: /* else { */
7558: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7559: /* } */
7560: /* } */
7561: /* if (h==(int)(calagedatem+12*cpt)){ */
7562: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7563: /* /\*fprintf(ficrespop," %.3f", kk1); */
7564: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7565: /* } */
7566: /* } */
7567: /* for(i=1; i<=nlstate;i++){ */
7568: /* kk1=0.; */
7569: /* for(j=1; j<=nlstate;j++){ */
7570: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7571: /* } */
7572: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7573: /* } */
1.218 brouard 7574:
1.227 brouard 7575: /* if (h==(int)(calagedatem+12*cpt)) */
7576: /* for(j=1; j<=nlstate;j++) */
7577: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7578: /* } */
7579: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7580: /* } */
7581: /* } */
1.218 brouard 7582:
1.227 brouard 7583: /* /\******\/ */
1.218 brouard 7584:
1.227 brouard 7585: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7586: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7587: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7588: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7589: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7590:
1.227 brouard 7591: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7592: /* oldm=oldms;savm=savms; */
7593: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7594: /* for (h=0; h<=nhstepm; h++){ */
7595: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7596: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7597: /* } */
7598: /* for(j=1; j<=nlstate+ndeath;j++) { */
7599: /* kk1=0.;kk2=0; */
7600: /* for(i=1; i<=nlstate;i++) { */
7601: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7602: /* } */
7603: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7604: /* } */
7605: /* } */
7606: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7607: /* } */
7608: /* } */
7609: /* } */
7610: /* } */
1.218 brouard 7611:
1.227 brouard 7612: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7613:
1.227 brouard 7614: /* if (popforecast==1) { */
7615: /* free_ivector(popage,0,AGESUP); */
7616: /* free_vector(popeffectif,0,AGESUP); */
7617: /* free_vector(popcount,0,AGESUP); */
7618: /* } */
7619: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7620: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7621: /* fclose(ficrespop); */
7622: /* } /\* End of popforecast *\/ */
1.218 brouard 7623:
1.126 brouard 7624: int fileappend(FILE *fichier, char *optionfich)
7625: {
7626: if((fichier=fopen(optionfich,"a"))==NULL) {
7627: printf("Problem with file: %s\n", optionfich);
7628: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7629: return (0);
7630: }
7631: fflush(fichier);
7632: return (1);
7633: }
7634:
7635:
7636: /**************** function prwizard **********************/
7637: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7638: {
7639:
7640: /* Wizard to print covariance matrix template */
7641:
1.164 brouard 7642: char ca[32], cb[32];
7643: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7644: int numlinepar;
7645:
7646: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7647: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7648: for(i=1; i <=nlstate; i++){
7649: jj=0;
7650: for(j=1; j <=nlstate+ndeath; j++){
7651: if(j==i) continue;
7652: jj++;
7653: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7654: printf("%1d%1d",i,j);
7655: fprintf(ficparo,"%1d%1d",i,j);
7656: for(k=1; k<=ncovmodel;k++){
7657: /* printf(" %lf",param[i][j][k]); */
7658: /* fprintf(ficparo," %lf",param[i][j][k]); */
7659: printf(" 0.");
7660: fprintf(ficparo," 0.");
7661: }
7662: printf("\n");
7663: fprintf(ficparo,"\n");
7664: }
7665: }
7666: printf("# Scales (for hessian or gradient estimation)\n");
7667: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7668: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7669: for(i=1; i <=nlstate; i++){
7670: jj=0;
7671: for(j=1; j <=nlstate+ndeath; j++){
7672: if(j==i) continue;
7673: jj++;
7674: fprintf(ficparo,"%1d%1d",i,j);
7675: printf("%1d%1d",i,j);
7676: fflush(stdout);
7677: for(k=1; k<=ncovmodel;k++){
7678: /* printf(" %le",delti3[i][j][k]); */
7679: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7680: printf(" 0.");
7681: fprintf(ficparo," 0.");
7682: }
7683: numlinepar++;
7684: printf("\n");
7685: fprintf(ficparo,"\n");
7686: }
7687: }
7688: printf("# Covariance matrix\n");
7689: /* # 121 Var(a12)\n\ */
7690: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7691: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7692: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7693: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7694: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7695: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7696: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7697: fflush(stdout);
7698: fprintf(ficparo,"# Covariance matrix\n");
7699: /* # 121 Var(a12)\n\ */
7700: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7701: /* # ...\n\ */
7702: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7703:
7704: for(itimes=1;itimes<=2;itimes++){
7705: jj=0;
7706: for(i=1; i <=nlstate; i++){
7707: for(j=1; j <=nlstate+ndeath; j++){
7708: if(j==i) continue;
7709: for(k=1; k<=ncovmodel;k++){
7710: jj++;
7711: ca[0]= k+'a'-1;ca[1]='\0';
7712: if(itimes==1){
7713: printf("#%1d%1d%d",i,j,k);
7714: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7715: }else{
7716: printf("%1d%1d%d",i,j,k);
7717: fprintf(ficparo,"%1d%1d%d",i,j,k);
7718: /* printf(" %.5le",matcov[i][j]); */
7719: }
7720: ll=0;
7721: for(li=1;li <=nlstate; li++){
7722: for(lj=1;lj <=nlstate+ndeath; lj++){
7723: if(lj==li) continue;
7724: for(lk=1;lk<=ncovmodel;lk++){
7725: ll++;
7726: if(ll<=jj){
7727: cb[0]= lk +'a'-1;cb[1]='\0';
7728: if(ll<jj){
7729: if(itimes==1){
7730: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7731: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7732: }else{
7733: printf(" 0.");
7734: fprintf(ficparo," 0.");
7735: }
7736: }else{
7737: if(itimes==1){
7738: printf(" Var(%s%1d%1d)",ca,i,j);
7739: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7740: }else{
7741: printf(" 0.");
7742: fprintf(ficparo," 0.");
7743: }
7744: }
7745: }
7746: } /* end lk */
7747: } /* end lj */
7748: } /* end li */
7749: printf("\n");
7750: fprintf(ficparo,"\n");
7751: numlinepar++;
7752: } /* end k*/
7753: } /*end j */
7754: } /* end i */
7755: } /* end itimes */
7756:
7757: } /* end of prwizard */
7758: /******************* Gompertz Likelihood ******************************/
7759: double gompertz(double x[])
7760: {
7761: double A,B,L=0.0,sump=0.,num=0.;
7762: int i,n=0; /* n is the size of the sample */
7763:
1.220 brouard 7764: for (i=1;i<=imx ; i++) {
1.126 brouard 7765: sump=sump+weight[i];
7766: /* sump=sump+1;*/
7767: num=num+1;
7768: }
7769:
7770:
7771: /* for (i=0; i<=imx; i++)
7772: 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]);*/
7773:
7774: for (i=1;i<=imx ; i++)
7775: {
7776: if (cens[i] == 1 && wav[i]>1)
7777: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
7778:
7779: if (cens[i] == 0 && wav[i]>1)
7780: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
7781: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
7782:
7783: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7784: if (wav[i] > 1 ) { /* ??? */
7785: L=L+A*weight[i];
7786: /* 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]);*/
7787: }
7788: }
7789:
7790: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7791:
7792: return -2*L*num/sump;
7793: }
7794:
1.136 brouard 7795: #ifdef GSL
7796: /******************* Gompertz_f Likelihood ******************************/
7797: double gompertz_f(const gsl_vector *v, void *params)
7798: {
7799: double A,B,LL=0.0,sump=0.,num=0.;
7800: double *x= (double *) v->data;
7801: int i,n=0; /* n is the size of the sample */
7802:
7803: for (i=0;i<=imx-1 ; i++) {
7804: sump=sump+weight[i];
7805: /* sump=sump+1;*/
7806: num=num+1;
7807: }
7808:
7809:
7810: /* for (i=0; i<=imx; i++)
7811: 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]);*/
7812: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
7813: for (i=1;i<=imx ; i++)
7814: {
7815: if (cens[i] == 1 && wav[i]>1)
7816: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
7817:
7818: if (cens[i] == 0 && wav[i]>1)
7819: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
7820: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
7821:
7822: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7823: if (wav[i] > 1 ) { /* ??? */
7824: LL=LL+A*weight[i];
7825: /* 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]);*/
7826: }
7827: }
7828:
7829: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7830: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
7831:
7832: return -2*LL*num/sump;
7833: }
7834: #endif
7835:
1.126 brouard 7836: /******************* Printing html file ***********/
1.201 brouard 7837: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7838: int lastpass, int stepm, int weightopt, char model[],\
7839: int imx, double p[],double **matcov,double agemortsup){
7840: int i,k;
7841:
7842: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
7843: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
7844: for (i=1;i<=2;i++)
7845: 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 7846: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 7847: fprintf(fichtm,"</ul>");
7848:
7849: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
7850:
7851: 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>");
7852:
7853: for (k=agegomp;k<(agemortsup-2);k++)
7854: 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]);
7855:
7856:
7857: fflush(fichtm);
7858: }
7859:
7860: /******************* Gnuplot file **************/
1.201 brouard 7861: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 7862:
7863: char dirfileres[132],optfileres[132];
1.164 brouard 7864:
1.126 brouard 7865: int ng;
7866:
7867:
7868: /*#ifdef windows */
7869: fprintf(ficgp,"cd \"%s\" \n",pathc);
7870: /*#endif */
7871:
7872:
7873: strcpy(dirfileres,optionfilefiname);
7874: strcpy(optfileres,"vpl");
1.199 brouard 7875: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 7876: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 7877: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 7878: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 7879: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
7880:
7881: }
7882:
1.136 brouard 7883: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
7884: {
1.126 brouard 7885:
1.136 brouard 7886: /*-------- data file ----------*/
7887: FILE *fic;
7888: char dummy[]=" ";
1.240 brouard 7889: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 7890: int lstra;
1.136 brouard 7891: int linei, month, year,iout;
7892: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 7893: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 7894: char *stratrunc;
1.223 brouard 7895:
1.240 brouard 7896: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
7897: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 7898:
1.240 brouard 7899: for(v=1; v <=ncovcol;v++){
7900: DummyV[v]=0;
7901: FixedV[v]=0;
7902: }
7903: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
7904: DummyV[v]=1;
7905: FixedV[v]=0;
7906: }
7907: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
7908: DummyV[v]=0;
7909: FixedV[v]=1;
7910: }
7911: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
7912: DummyV[v]=1;
7913: FixedV[v]=1;
7914: }
7915: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
7916: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
7917: 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]);
7918: }
1.126 brouard 7919:
1.136 brouard 7920: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 7921: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
7922: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 7923: }
1.126 brouard 7924:
1.136 brouard 7925: i=1;
7926: linei=0;
7927: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
7928: linei=linei+1;
7929: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
7930: if(line[j] == '\t')
7931: line[j] = ' ';
7932: }
7933: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
7934: ;
7935: };
7936: line[j+1]=0; /* Trims blanks at end of line */
7937: if(line[0]=='#'){
7938: fprintf(ficlog,"Comment line\n%s\n",line);
7939: printf("Comment line\n%s\n",line);
7940: continue;
7941: }
7942: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 7943: strcpy(line, linetmp);
1.223 brouard 7944:
7945: /* Loops on waves */
7946: for (j=maxwav;j>=1;j--){
7947: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 7948: cutv(stra, strb, line, ' ');
7949: if(strb[0]=='.') { /* Missing value */
7950: lval=-1;
7951: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
7952: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
7953: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
7954: 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);
7955: 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);
7956: return 1;
7957: }
7958: }else{
7959: errno=0;
7960: /* what_kind_of_number(strb); */
7961: dval=strtod(strb,&endptr);
7962: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
7963: /* if(strb != endptr && *endptr == '\0') */
7964: /* dval=dlval; */
7965: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
7966: if( strb[0]=='\0' || (*endptr != '\0')){
7967: 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);
7968: 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);
7969: return 1;
7970: }
7971: cotqvar[j][iv][i]=dval;
7972: cotvar[j][ntv+iv][i]=dval;
7973: }
7974: strcpy(line,stra);
1.223 brouard 7975: }/* end loop ntqv */
1.225 brouard 7976:
1.223 brouard 7977: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 7978: cutv(stra, strb, line, ' ');
7979: if(strb[0]=='.') { /* Missing value */
7980: lval=-1;
7981: }else{
7982: errno=0;
7983: lval=strtol(strb,&endptr,10);
7984: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
7985: if( strb[0]=='\0' || (*endptr != '\0')){
7986: 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);
7987: 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);
7988: return 1;
7989: }
7990: }
7991: if(lval <-1 || lval >1){
7992: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 7993: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
7994: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 7995: For example, for multinomial values like 1, 2 and 3,\n \
7996: build V1=0 V2=0 for the reference value (1),\n \
7997: V1=1 V2=0 for (2) \n \
1.223 brouard 7998: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 7999: output of IMaCh is often meaningless.\n \
1.223 brouard 8000: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8001: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8002: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8003: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8004: For example, for multinomial values like 1, 2 and 3,\n \
8005: build V1=0 V2=0 for the reference value (1),\n \
8006: V1=1 V2=0 for (2) \n \
1.223 brouard 8007: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8008: output of IMaCh is often meaningless.\n \
1.223 brouard 8009: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8010: return 1;
8011: }
8012: cotvar[j][iv][i]=(double)(lval);
8013: strcpy(line,stra);
1.223 brouard 8014: }/* end loop ntv */
1.225 brouard 8015:
1.223 brouard 8016: /* Statuses at wave */
1.137 brouard 8017: cutv(stra, strb, line, ' ');
1.223 brouard 8018: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8019: lval=-1;
1.136 brouard 8020: }else{
1.238 brouard 8021: errno=0;
8022: lval=strtol(strb,&endptr,10);
8023: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8024: if( strb[0]=='\0' || (*endptr != '\0')){
8025: 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);
8026: 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);
8027: return 1;
8028: }
1.136 brouard 8029: }
1.225 brouard 8030:
1.136 brouard 8031: s[j][i]=lval;
1.225 brouard 8032:
1.223 brouard 8033: /* Date of Interview */
1.136 brouard 8034: strcpy(line,stra);
8035: cutv(stra, strb,line,' ');
1.169 brouard 8036: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8037: }
1.169 brouard 8038: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8039: month=99;
8040: year=9999;
1.136 brouard 8041: }else{
1.225 brouard 8042: 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);
8043: 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);
8044: return 1;
1.136 brouard 8045: }
8046: anint[j][i]= (double) year;
8047: mint[j][i]= (double)month;
8048: strcpy(line,stra);
1.223 brouard 8049: } /* End loop on waves */
1.225 brouard 8050:
1.223 brouard 8051: /* Date of death */
1.136 brouard 8052: cutv(stra, strb,line,' ');
1.169 brouard 8053: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8054: }
1.169 brouard 8055: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8056: month=99;
8057: year=9999;
8058: }else{
1.141 brouard 8059: 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 8060: 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);
8061: return 1;
1.136 brouard 8062: }
8063: andc[i]=(double) year;
8064: moisdc[i]=(double) month;
8065: strcpy(line,stra);
8066:
1.223 brouard 8067: /* Date of birth */
1.136 brouard 8068: cutv(stra, strb,line,' ');
1.169 brouard 8069: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8070: }
1.169 brouard 8071: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8072: month=99;
8073: year=9999;
8074: }else{
1.141 brouard 8075: 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);
8076: 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 8077: return 1;
1.136 brouard 8078: }
8079: if (year==9999) {
1.141 brouard 8080: 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);
8081: 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 8082: return 1;
8083:
1.136 brouard 8084: }
8085: annais[i]=(double)(year);
8086: moisnais[i]=(double)(month);
8087: strcpy(line,stra);
1.225 brouard 8088:
1.223 brouard 8089: /* Sample weight */
1.136 brouard 8090: cutv(stra, strb,line,' ');
8091: errno=0;
8092: dval=strtod(strb,&endptr);
8093: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8094: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8095: 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 8096: fflush(ficlog);
8097: return 1;
8098: }
8099: weight[i]=dval;
8100: strcpy(line,stra);
1.225 brouard 8101:
1.223 brouard 8102: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8103: cutv(stra, strb, line, ' ');
8104: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8105: lval=-1;
1.223 brouard 8106: }else{
1.225 brouard 8107: errno=0;
8108: /* what_kind_of_number(strb); */
8109: dval=strtod(strb,&endptr);
8110: /* if(strb != endptr && *endptr == '\0') */
8111: /* dval=dlval; */
8112: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8113: if( strb[0]=='\0' || (*endptr != '\0')){
8114: 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);
8115: 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);
8116: return 1;
8117: }
8118: coqvar[iv][i]=dval;
1.226 brouard 8119: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8120: }
8121: strcpy(line,stra);
8122: }/* end loop nqv */
1.136 brouard 8123:
1.223 brouard 8124: /* Covariate values */
1.136 brouard 8125: for (j=ncovcol;j>=1;j--){
8126: cutv(stra, strb,line,' ');
1.223 brouard 8127: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8128: lval=-1;
1.136 brouard 8129: }else{
1.225 brouard 8130: errno=0;
8131: lval=strtol(strb,&endptr,10);
8132: if( strb[0]=='\0' || (*endptr != '\0')){
8133: 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);
8134: 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);
8135: return 1;
8136: }
1.136 brouard 8137: }
8138: if(lval <-1 || lval >1){
1.225 brouard 8139: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8140: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8141: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8142: For example, for multinomial values like 1, 2 and 3,\n \
8143: build V1=0 V2=0 for the reference value (1),\n \
8144: V1=1 V2=0 for (2) \n \
1.136 brouard 8145: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8146: output of IMaCh is often meaningless.\n \
1.136 brouard 8147: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8148: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8149: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8150: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8151: For example, for multinomial values like 1, 2 and 3,\n \
8152: build V1=0 V2=0 for the reference value (1),\n \
8153: V1=1 V2=0 for (2) \n \
1.136 brouard 8154: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8155: output of IMaCh is often meaningless.\n \
1.136 brouard 8156: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8157: return 1;
1.136 brouard 8158: }
8159: covar[j][i]=(double)(lval);
8160: strcpy(line,stra);
8161: }
8162: lstra=strlen(stra);
1.225 brouard 8163:
1.136 brouard 8164: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8165: stratrunc = &(stra[lstra-9]);
8166: num[i]=atol(stratrunc);
8167: }
8168: else
8169: num[i]=atol(stra);
8170: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8171: 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;}*/
8172:
8173: i=i+1;
8174: } /* End loop reading data */
1.225 brouard 8175:
1.136 brouard 8176: *imax=i-1; /* Number of individuals */
8177: fclose(fic);
1.225 brouard 8178:
1.136 brouard 8179: return (0);
1.164 brouard 8180: /* endread: */
1.225 brouard 8181: printf("Exiting readdata: ");
8182: fclose(fic);
8183: return (1);
1.223 brouard 8184: }
1.126 brouard 8185:
1.234 brouard 8186: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8187: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8188: while (*p2 == ' ')
1.234 brouard 8189: p2++;
8190: /* while ((*p1++ = *p2++) !=0) */
8191: /* ; */
8192: /* do */
8193: /* while (*p2 == ' ') */
8194: /* p2++; */
8195: /* while (*p1++ == *p2++); */
8196: *stri=p2;
1.145 brouard 8197: }
8198:
1.235 brouard 8199: int decoderesult ( char resultline[], int nres)
1.230 brouard 8200: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8201: {
1.235 brouard 8202: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8203: char resultsav[MAXLINE];
1.234 brouard 8204: int resultmodel[MAXLINE];
8205: int modelresult[MAXLINE];
1.230 brouard 8206: char stra[80], strb[80], strc[80], strd[80],stre[80];
8207:
1.234 brouard 8208: removefirstspace(&resultline);
1.233 brouard 8209: printf("decoderesult:%s\n",resultline);
1.230 brouard 8210:
8211: if (strstr(resultline,"v") !=0){
8212: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8213: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8214: return 1;
8215: }
8216: trimbb(resultsav, resultline);
8217: if (strlen(resultsav) >1){
8218: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8219: }
1.234 brouard 8220: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8221: 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);
8222: 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);
8223: }
8224: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8225: if(nbocc(resultsav,'=') >1){
8226: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8227: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8228: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8229: }else
8230: cutl(strc,strd,resultsav,'=');
1.230 brouard 8231: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8232:
1.230 brouard 8233: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8234: Tvarsel[k]=atoi(strc);
8235: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8236: /* cptcovsel++; */
8237: if (nbocc(stra,'=') >0)
8238: strcpy(resultsav,stra); /* and analyzes it */
8239: }
1.235 brouard 8240: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8241: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8242: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8243: match=0;
1.236 brouard 8244: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8245: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8246: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8247: match=1;
8248: break;
8249: }
8250: }
8251: if(match == 0){
8252: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8253: }
8254: }
8255: }
1.235 brouard 8256: /* Checking for missing or useless values in comparison of current model needs */
8257: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8258: match=0;
1.235 brouard 8259: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8260: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8261: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8262: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8263: ++match;
8264: }
8265: }
8266: }
8267: if(match == 0){
8268: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8269: }else if(match > 1){
8270: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8271: }
8272: }
1.235 brouard 8273:
1.234 brouard 8274: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8275: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8276: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8277: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8278: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8279: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8280: /* 1 0 0 0 */
8281: /* 2 1 0 0 */
8282: /* 3 0 1 0 */
8283: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8284: /* 5 0 0 1 */
8285: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8286: /* 7 0 1 1 */
8287: /* 8 1 1 1 */
1.237 brouard 8288: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8289: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8290: /* V5*age V5 known which value for nres? */
8291: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8292: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8293: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8294: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8295: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8296: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8297: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8298: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8299: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8300: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8301: k4++;;
8302: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8303: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8304: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8305: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8306: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8307: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8308: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8309: k4q++;;
8310: }
8311: }
1.234 brouard 8312:
1.235 brouard 8313: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8314: return (0);
8315: }
1.235 brouard 8316:
1.230 brouard 8317: int decodemodel( char model[], int lastobs)
8318: /**< This routine decodes the model and returns:
1.224 brouard 8319: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8320: * - nagesqr = 1 if age*age in the model, otherwise 0.
8321: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8322: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8323: * - cptcovage number of covariates with age*products =2
8324: * - cptcovs number of simple covariates
8325: * - 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
8326: * which is a new column after the 9 (ncovcol) variables.
8327: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8328: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8329: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8330: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8331: */
1.136 brouard 8332: {
1.238 brouard 8333: int i, j, k, ks, v;
1.227 brouard 8334: int j1, k1, k2, k3, k4;
1.136 brouard 8335: char modelsav[80];
1.145 brouard 8336: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8337: char *strpt;
1.136 brouard 8338:
1.145 brouard 8339: /*removespace(model);*/
1.136 brouard 8340: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8341: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8342: if (strstr(model,"AGE") !=0){
1.192 brouard 8343: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8344: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8345: return 1;
8346: }
1.141 brouard 8347: if (strstr(model,"v") !=0){
8348: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8349: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8350: return 1;
8351: }
1.187 brouard 8352: strcpy(modelsav,model);
8353: if ((strpt=strstr(model,"age*age")) !=0){
8354: printf(" strpt=%s, model=%s\n",strpt, model);
8355: if(strpt != model){
1.234 brouard 8356: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8357: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8358: corresponding column of parameters.\n",model);
1.234 brouard 8359: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8360: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8361: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8362: return 1;
1.225 brouard 8363: }
1.187 brouard 8364: nagesqr=1;
8365: if (strstr(model,"+age*age") !=0)
1.234 brouard 8366: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8367: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8368: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8369: else
1.234 brouard 8370: substrchaine(modelsav, model, "age*age");
1.187 brouard 8371: }else
8372: nagesqr=0;
8373: if (strlen(modelsav) >1){
8374: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8375: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8376: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8377: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8378: * cst, age and age*age
8379: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8380: /* including age products which are counted in cptcovage.
8381: * but the covariates which are products must be treated
8382: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8383: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8384: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8385:
8386:
1.187 brouard 8387: /* Design
8388: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8389: * < ncovcol=8 >
8390: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8391: * k= 1 2 3 4 5 6 7 8
8392: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8393: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8394: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8395: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8396: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8397: * Tage[++cptcovage]=k
8398: * if products, new covar are created after ncovcol with k1
8399: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8400: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8401: * 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
8402: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8403: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8404: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8405: * < ncovcol=8 >
8406: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8407: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8408: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8409: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8410: * p Tprod[1]@2={ 6, 5}
8411: *p Tvard[1][1]@4= {7, 8, 5, 6}
8412: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8413: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8414: *How to reorganize?
8415: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8416: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8417: * {2, 1, 4, 8, 5, 6, 3, 7}
8418: * Struct []
8419: */
1.225 brouard 8420:
1.187 brouard 8421: /* This loop fills the array Tvar from the string 'model'.*/
8422: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8423: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8424: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8425: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8426: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8427: /* k=1 Tvar[1]=2 (from V2) */
8428: /* k=5 Tvar[5] */
8429: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8430: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8431: /* } */
1.198 brouard 8432: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8433: /*
8434: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8435: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8436: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8437: }
1.187 brouard 8438: cptcovage=0;
8439: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8440: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8441: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8442: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8443: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8444: /*scanf("%d",i);*/
8445: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8446: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8447: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8448: /* covar is not filled and then is empty */
8449: cptcovprod--;
8450: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8451: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8452: Typevar[k]=1; /* 1 for age product */
8453: cptcovage++; /* Sums the number of covariates which include age as a product */
8454: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8455: /*printf("stre=%s ", stre);*/
8456: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8457: cptcovprod--;
8458: cutl(stre,strb,strc,'V');
8459: Tvar[k]=atoi(stre);
8460: Typevar[k]=1; /* 1 for age product */
8461: cptcovage++;
8462: Tage[cptcovage]=k;
8463: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8464: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8465: cptcovn++;
8466: cptcovprodnoage++;k1++;
8467: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8468: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8469: because this model-covariate is a construction we invent a new column
8470: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8471: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8472: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8473: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8474: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8475: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8476: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8477: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8478: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8479: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8480: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8481: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8482: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8483: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8484: for (i=1; i<=lastobs;i++){
8485: /* Computes the new covariate which is a product of
8486: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8487: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8488: }
8489: } /* End age is not in the model */
8490: } /* End if model includes a product */
8491: else { /* no more sum */
8492: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8493: /* scanf("%d",i);*/
8494: cutl(strd,strc,strb,'V');
8495: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8496: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8497: Tvar[k]=atoi(strd);
8498: Typevar[k]=0; /* 0 for simple covariates */
8499: }
8500: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8501: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8502: scanf("%d",i);*/
1.187 brouard 8503: } /* end of loop + on total covariates */
8504: } /* end if strlen(modelsave == 0) age*age might exist */
8505: } /* end if strlen(model == 0) */
1.136 brouard 8506:
8507: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8508: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8509:
1.136 brouard 8510: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8511: printf("cptcovprod=%d ", cptcovprod);
8512: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8513: scanf("%d ",i);*/
8514:
8515:
1.230 brouard 8516: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8517: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8518: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8519: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8520: k = 1 2 3 4 5 6 7 8 9
8521: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8522: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8523: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8524: Dummy[k] 1 0 0 0 3 1 1 2 3
8525: Tmodelind[combination of covar]=k;
1.225 brouard 8526: */
8527: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8528: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8529: /* 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 8530: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8531: printf("Model=%s\n\
8532: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8533: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8534: 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);
8535: fprintf(ficlog,"Model=%s\n\
8536: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8537: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8538: 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 8539: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8540: 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 */
8541: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8542: Fixed[k]= 0;
8543: Dummy[k]= 0;
1.225 brouard 8544: ncoveff++;
1.232 brouard 8545: ncovf++;
1.234 brouard 8546: nsd++;
8547: modell[k].maintype= FTYPE;
8548: TvarsD[nsd]=Tvar[k];
8549: TvarsDind[nsd]=k;
8550: TvarF[ncovf]=Tvar[k];
8551: TvarFind[ncovf]=k;
8552: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8553: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8554: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8555: Fixed[k]= 0;
8556: Dummy[k]= 0;
8557: ncoveff++;
8558: ncovf++;
8559: modell[k].maintype= FTYPE;
8560: TvarF[ncovf]=Tvar[k];
8561: TvarFind[ncovf]=k;
1.230 brouard 8562: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8563: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8564: }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 8565: Fixed[k]= 0;
8566: Dummy[k]= 1;
1.230 brouard 8567: nqfveff++;
1.234 brouard 8568: modell[k].maintype= FTYPE;
8569: modell[k].subtype= FQ;
8570: nsq++;
8571: TvarsQ[nsq]=Tvar[k];
8572: TvarsQind[nsq]=k;
1.232 brouard 8573: ncovf++;
1.234 brouard 8574: TvarF[ncovf]=Tvar[k];
8575: TvarFind[ncovf]=k;
1.231 brouard 8576: 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 8577: 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 8578: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8579: Fixed[k]= 1;
8580: Dummy[k]= 0;
1.225 brouard 8581: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8582: modell[k].maintype= VTYPE;
8583: modell[k].subtype= VD;
8584: nsd++;
8585: TvarsD[nsd]=Tvar[k];
8586: TvarsDind[nsd]=k;
8587: ncovv++; /* Only simple time varying variables */
8588: TvarV[ncovv]=Tvar[k];
1.242 ! brouard 8589: 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 8590: 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 */
8591: 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 8592: 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);
8593: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8594: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8595: Fixed[k]= 1;
8596: Dummy[k]= 1;
8597: nqtveff++;
8598: modell[k].maintype= VTYPE;
8599: modell[k].subtype= VQ;
8600: ncovv++; /* Only simple time varying variables */
8601: nsq++;
8602: TvarsQ[nsq]=Tvar[k];
8603: TvarsQind[nsq]=k;
8604: TvarV[ncovv]=Tvar[k];
1.242 ! brouard 8605: 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 8606: 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 */
8607: 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 8608: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8609: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8610: 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 8611: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8612: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8613: ncova++;
8614: TvarA[ncova]=Tvar[k];
8615: TvarAind[ncova]=k;
1.231 brouard 8616: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8617: Fixed[k]= 2;
8618: Dummy[k]= 2;
8619: modell[k].maintype= ATYPE;
8620: modell[k].subtype= APFD;
8621: /* ncoveff++; */
1.227 brouard 8622: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8623: Fixed[k]= 2;
8624: Dummy[k]= 3;
8625: modell[k].maintype= ATYPE;
8626: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8627: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8628: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8629: Fixed[k]= 3;
8630: Dummy[k]= 2;
8631: modell[k].maintype= ATYPE;
8632: modell[k].subtype= APVD; /* Product age * varying dummy */
8633: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8634: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8635: Fixed[k]= 3;
8636: Dummy[k]= 3;
8637: modell[k].maintype= ATYPE;
8638: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8639: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8640: }
8641: }else if (Typevar[k] == 2) { /* product without age */
8642: k1=Tposprod[k];
8643: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8644: if(Tvard[k1][2] <=ncovcol){
8645: Fixed[k]= 1;
8646: Dummy[k]= 0;
8647: modell[k].maintype= FTYPE;
8648: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8649: ncovf++; /* Fixed variables without age */
8650: TvarF[ncovf]=Tvar[k];
8651: TvarFind[ncovf]=k;
8652: }else if(Tvard[k1][2] <=ncovcol+nqv){
8653: Fixed[k]= 0; /* or 2 ?*/
8654: Dummy[k]= 1;
8655: modell[k].maintype= FTYPE;
8656: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8657: ncovf++; /* Varying variables without age */
8658: TvarF[ncovf]=Tvar[k];
8659: TvarFind[ncovf]=k;
8660: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8661: Fixed[k]= 1;
8662: Dummy[k]= 0;
8663: modell[k].maintype= VTYPE;
8664: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8665: ncovv++; /* Varying variables without age */
8666: TvarV[ncovv]=Tvar[k];
8667: TvarVind[ncovv]=k;
8668: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8669: Fixed[k]= 1;
8670: Dummy[k]= 1;
8671: modell[k].maintype= VTYPE;
8672: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8673: ncovv++; /* Varying variables without age */
8674: TvarV[ncovv]=Tvar[k];
8675: TvarVind[ncovv]=k;
8676: }
1.227 brouard 8677: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8678: if(Tvard[k1][2] <=ncovcol){
8679: Fixed[k]= 0; /* or 2 ?*/
8680: Dummy[k]= 1;
8681: modell[k].maintype= FTYPE;
8682: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8683: ncovf++; /* Fixed variables without age */
8684: TvarF[ncovf]=Tvar[k];
8685: TvarFind[ncovf]=k;
8686: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8687: Fixed[k]= 1;
8688: Dummy[k]= 1;
8689: modell[k].maintype= VTYPE;
8690: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8691: ncovv++; /* Varying variables without age */
8692: TvarV[ncovv]=Tvar[k];
8693: TvarVind[ncovv]=k;
8694: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8695: Fixed[k]= 1;
8696: Dummy[k]= 1;
8697: modell[k].maintype= VTYPE;
8698: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8699: ncovv++; /* Varying variables without age */
8700: TvarV[ncovv]=Tvar[k];
8701: TvarVind[ncovv]=k;
8702: ncovv++; /* Varying variables without age */
8703: TvarV[ncovv]=Tvar[k];
8704: TvarVind[ncovv]=k;
8705: }
1.227 brouard 8706: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8707: if(Tvard[k1][2] <=ncovcol){
8708: Fixed[k]= 1;
8709: Dummy[k]= 1;
8710: modell[k].maintype= VTYPE;
8711: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8712: ncovv++; /* Varying variables without age */
8713: TvarV[ncovv]=Tvar[k];
8714: TvarVind[ncovv]=k;
8715: }else if(Tvard[k1][2] <=ncovcol+nqv){
8716: Fixed[k]= 1;
8717: Dummy[k]= 1;
8718: modell[k].maintype= VTYPE;
8719: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8720: ncovv++; /* Varying variables without age */
8721: TvarV[ncovv]=Tvar[k];
8722: TvarVind[ncovv]=k;
8723: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8724: Fixed[k]= 1;
8725: Dummy[k]= 0;
8726: modell[k].maintype= VTYPE;
8727: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8728: ncovv++; /* Varying variables without age */
8729: TvarV[ncovv]=Tvar[k];
8730: TvarVind[ncovv]=k;
8731: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8732: Fixed[k]= 1;
8733: Dummy[k]= 1;
8734: modell[k].maintype= VTYPE;
8735: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8736: ncovv++; /* Varying variables without age */
8737: TvarV[ncovv]=Tvar[k];
8738: TvarVind[ncovv]=k;
8739: }
1.227 brouard 8740: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8741: if(Tvard[k1][2] <=ncovcol){
8742: Fixed[k]= 1;
8743: Dummy[k]= 1;
8744: modell[k].maintype= VTYPE;
8745: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8746: ncovv++; /* Varying variables without age */
8747: TvarV[ncovv]=Tvar[k];
8748: TvarVind[ncovv]=k;
8749: }else if(Tvard[k1][2] <=ncovcol+nqv){
8750: Fixed[k]= 1;
8751: Dummy[k]= 1;
8752: modell[k].maintype= VTYPE;
8753: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
8754: ncovv++; /* Varying variables without age */
8755: TvarV[ncovv]=Tvar[k];
8756: TvarVind[ncovv]=k;
8757: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8758: Fixed[k]= 1;
8759: Dummy[k]= 1;
8760: modell[k].maintype= VTYPE;
8761: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
8762: ncovv++; /* Varying variables without age */
8763: TvarV[ncovv]=Tvar[k];
8764: TvarVind[ncovv]=k;
8765: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8766: Fixed[k]= 1;
8767: Dummy[k]= 1;
8768: modell[k].maintype= VTYPE;
8769: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
8770: ncovv++; /* Varying variables without age */
8771: TvarV[ncovv]=Tvar[k];
8772: TvarVind[ncovv]=k;
8773: }
1.227 brouard 8774: }else{
1.240 brouard 8775: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8776: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8777: } /*end k1*/
1.225 brouard 8778: }else{
1.226 brouard 8779: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
8780: 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 8781: }
1.227 brouard 8782: 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 8783: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 8784: 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]);
8785: }
8786: /* Searching for doublons in the model */
8787: for(k1=1; k1<= cptcovt;k1++){
8788: for(k2=1; k2 <k1;k2++){
8789: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 8790: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
8791: if(Tvar[k1]==Tvar[k2]){
8792: 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]]);
8793: 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);
8794: return(1);
8795: }
8796: }else if (Typevar[k1] ==2){
8797: k3=Tposprod[k1];
8798: k4=Tposprod[k2];
8799: 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])) ){
8800: 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]]);
8801: 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);
8802: return(1);
8803: }
8804: }
1.227 brouard 8805: }
8806: }
1.225 brouard 8807: }
8808: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
8809: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 8810: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
8811: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 8812: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 8813: /*endread:*/
1.225 brouard 8814: printf("Exiting decodemodel: ");
8815: return (1);
1.136 brouard 8816: }
8817:
1.169 brouard 8818: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.136 brouard 8819: {
8820: int i, m;
1.218 brouard 8821: int firstone=0;
8822:
1.136 brouard 8823: for (i=1; i<=imx; i++) {
8824: for(m=2; (m<= maxwav); m++) {
8825: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
8826: anint[m][i]=9999;
1.216 brouard 8827: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
8828: s[m][i]=-1;
1.136 brouard 8829: }
8830: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 8831: *nberr = *nberr + 1;
1.218 brouard 8832: if(firstone == 0){
8833: firstone=1;
8834: 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);
8835: }
8836: 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 8837: s[m][i]=-1;
8838: }
8839: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 8840: (*nberr)++;
1.136 brouard 8841: 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]);
8842: 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]);
8843: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
8844: }
8845: }
8846: }
8847:
8848: for (i=1; i<=imx; i++) {
8849: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
8850: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 8851: 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 8852: if (s[m][i] >= nlstate+1) {
1.169 brouard 8853: if(agedc[i]>0){
8854: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 8855: agev[m][i]=agedc[i];
1.214 brouard 8856: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 8857: }else {
1.136 brouard 8858: if ((int)andc[i]!=9999){
8859: nbwarn++;
8860: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
8861: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
8862: agev[m][i]=-1;
8863: }
8864: }
1.169 brouard 8865: } /* agedc > 0 */
1.214 brouard 8866: } /* end if */
1.136 brouard 8867: else if(s[m][i] !=9){ /* Standard case, age in fractional
8868: years but with the precision of a month */
8869: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
8870: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
8871: agev[m][i]=1;
8872: else if(agev[m][i] < *agemin){
8873: *agemin=agev[m][i];
8874: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
8875: }
8876: else if(agev[m][i] >*agemax){
8877: *agemax=agev[m][i];
1.156 brouard 8878: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 8879: }
8880: /*agev[m][i]=anint[m][i]-annais[i];*/
8881: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 8882: } /* en if 9*/
1.136 brouard 8883: else { /* =9 */
1.214 brouard 8884: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 8885: agev[m][i]=1;
8886: s[m][i]=-1;
8887: }
8888: }
1.214 brouard 8889: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 8890: agev[m][i]=1;
1.214 brouard 8891: else{
8892: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8893: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8894: agev[m][i]=0;
8895: }
8896: } /* End for lastpass */
8897: }
1.136 brouard 8898:
8899: for (i=1; i<=imx; i++) {
8900: for(m=firstpass; (m<=lastpass); m++){
8901: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 8902: (*nberr)++;
1.136 brouard 8903: 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);
8904: 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);
8905: return 1;
8906: }
8907: }
8908: }
8909:
8910: /*for (i=1; i<=imx; i++){
8911: for (m=firstpass; (m<lastpass); m++){
8912: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
8913: }
8914:
8915: }*/
8916:
8917:
1.139 brouard 8918: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
8919: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 8920:
8921: return (0);
1.164 brouard 8922: /* endread:*/
1.136 brouard 8923: printf("Exiting calandcheckages: ");
8924: return (1);
8925: }
8926:
1.172 brouard 8927: #if defined(_MSC_VER)
8928: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8929: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8930: //#include "stdafx.h"
8931: //#include <stdio.h>
8932: //#include <tchar.h>
8933: //#include <windows.h>
8934: //#include <iostream>
8935: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
8936:
8937: LPFN_ISWOW64PROCESS fnIsWow64Process;
8938:
8939: BOOL IsWow64()
8940: {
8941: BOOL bIsWow64 = FALSE;
8942:
8943: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
8944: // (HANDLE, PBOOL);
8945:
8946: //LPFN_ISWOW64PROCESS fnIsWow64Process;
8947:
8948: HMODULE module = GetModuleHandle(_T("kernel32"));
8949: const char funcName[] = "IsWow64Process";
8950: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
8951: GetProcAddress(module, funcName);
8952:
8953: if (NULL != fnIsWow64Process)
8954: {
8955: if (!fnIsWow64Process(GetCurrentProcess(),
8956: &bIsWow64))
8957: //throw std::exception("Unknown error");
8958: printf("Unknown error\n");
8959: }
8960: return bIsWow64 != FALSE;
8961: }
8962: #endif
1.177 brouard 8963:
1.191 brouard 8964: void syscompilerinfo(int logged)
1.167 brouard 8965: {
8966: /* #include "syscompilerinfo.h"*/
1.185 brouard 8967: /* command line Intel compiler 32bit windows, XP compatible:*/
8968: /* /GS /W3 /Gy
8969: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
8970: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
8971: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 8972: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
8973: */
8974: /* 64 bits */
1.185 brouard 8975: /*
8976: /GS /W3 /Gy
8977: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
8978: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
8979: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
8980: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
8981: /* Optimization are useless and O3 is slower than O2 */
8982: /*
8983: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
8984: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
8985: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
8986: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
8987: */
1.186 brouard 8988: /* Link is */ /* /OUT:"visual studio
1.185 brouard 8989: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
8990: /PDB:"visual studio
8991: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
8992: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
8993: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
8994: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
8995: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
8996: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
8997: uiAccess='false'"
8998: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
8999: /NOLOGO /TLBID:1
9000: */
1.177 brouard 9001: #if defined __INTEL_COMPILER
1.178 brouard 9002: #if defined(__GNUC__)
9003: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9004: #endif
1.177 brouard 9005: #elif defined(__GNUC__)
1.179 brouard 9006: #ifndef __APPLE__
1.174 brouard 9007: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9008: #endif
1.177 brouard 9009: struct utsname sysInfo;
1.178 brouard 9010: int cross = CROSS;
9011: if (cross){
9012: printf("Cross-");
1.191 brouard 9013: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9014: }
1.174 brouard 9015: #endif
9016:
1.171 brouard 9017: #include <stdint.h>
1.178 brouard 9018:
1.191 brouard 9019: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9020: #if defined(__clang__)
1.191 brouard 9021: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9022: #endif
9023: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9024: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9025: #endif
9026: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9027: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9028: #endif
9029: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9030: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9031: #endif
9032: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9033: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9034: #endif
9035: #if defined(_MSC_VER)
1.191 brouard 9036: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9037: #endif
9038: #if defined(__PGI)
1.191 brouard 9039: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9040: #endif
9041: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9042: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9043: #endif
1.191 brouard 9044: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9045:
1.167 brouard 9046: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9047: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9048: // Windows (x64 and x86)
1.191 brouard 9049: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9050: #elif __unix__ // all unices, not all compilers
9051: // Unix
1.191 brouard 9052: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9053: #elif __linux__
9054: // linux
1.191 brouard 9055: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9056: #elif __APPLE__
1.174 brouard 9057: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9058: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9059: #endif
9060:
9061: /* __MINGW32__ */
9062: /* __CYGWIN__ */
9063: /* __MINGW64__ */
9064: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9065: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9066: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9067: /* _WIN64 // Defined for applications for Win64. */
9068: /* _M_X64 // Defined for compilations that target x64 processors. */
9069: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9070:
1.167 brouard 9071: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9072: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9073: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9074: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9075: #else
1.191 brouard 9076: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9077: #endif
9078:
1.169 brouard 9079: #if defined(__GNUC__)
9080: # if defined(__GNUC_PATCHLEVEL__)
9081: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9082: + __GNUC_MINOR__ * 100 \
9083: + __GNUC_PATCHLEVEL__)
9084: # else
9085: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9086: + __GNUC_MINOR__ * 100)
9087: # endif
1.174 brouard 9088: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9089: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9090:
9091: if (uname(&sysInfo) != -1) {
9092: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9093: 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 9094: }
9095: else
9096: perror("uname() error");
1.179 brouard 9097: //#ifndef __INTEL_COMPILER
9098: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9099: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9100: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9101: #endif
1.169 brouard 9102: #endif
1.172 brouard 9103:
9104: // void main()
9105: // {
1.169 brouard 9106: #if defined(_MSC_VER)
1.174 brouard 9107: if (IsWow64()){
1.191 brouard 9108: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9109: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9110: }
9111: else{
1.191 brouard 9112: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9113: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9114: }
1.172 brouard 9115: // printf("\nPress Enter to continue...");
9116: // getchar();
9117: // }
9118:
1.169 brouard 9119: #endif
9120:
1.167 brouard 9121:
1.219 brouard 9122: }
1.136 brouard 9123:
1.219 brouard 9124: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9125: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9126: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9127: /* double ftolpl = 1.e-10; */
1.180 brouard 9128: double age, agebase, agelim;
1.203 brouard 9129: double tot;
1.180 brouard 9130:
1.202 brouard 9131: strcpy(filerespl,"PL_");
9132: strcat(filerespl,fileresu);
9133: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9134: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9135: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9136: }
1.227 brouard 9137: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9138: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9139: pstamp(ficrespl);
1.203 brouard 9140: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9141: fprintf(ficrespl,"#Age ");
9142: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9143: fprintf(ficrespl,"\n");
1.180 brouard 9144:
1.219 brouard 9145: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9146:
1.219 brouard 9147: agebase=ageminpar;
9148: agelim=agemaxpar;
1.180 brouard 9149:
1.227 brouard 9150: /* i1=pow(2,ncoveff); */
1.234 brouard 9151: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9152: if (cptcovn < 1){i1=1;}
1.180 brouard 9153:
1.238 brouard 9154: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9155: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9156: if(TKresult[nres]!= k)
9157: continue;
1.235 brouard 9158:
1.238 brouard 9159: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9160: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9161: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9162: /* k=k+1; */
9163: /* to clean */
9164: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9165: fprintf(ficrespl,"#******");
9166: printf("#******");
9167: fprintf(ficlog,"#******");
9168: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9169: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9170: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9171: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9172: }
9173: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9174: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9175: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9176: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9177: }
9178: fprintf(ficrespl,"******\n");
9179: printf("******\n");
9180: fprintf(ficlog,"******\n");
9181: if(invalidvarcomb[k]){
9182: printf("\nCombination (%d) ignored because no case \n",k);
9183: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9184: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9185: continue;
9186: }
1.219 brouard 9187:
1.238 brouard 9188: fprintf(ficrespl,"#Age ");
9189: for(j=1;j<=cptcoveff;j++) {
9190: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9191: }
9192: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9193: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9194:
1.238 brouard 9195: for (age=agebase; age<=agelim; age++){
9196: /* for (age=agebase; age<=agebase; age++){ */
9197: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9198: fprintf(ficrespl,"%.0f ",age );
9199: for(j=1;j<=cptcoveff;j++)
9200: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9201: tot=0.;
9202: for(i=1; i<=nlstate;i++){
9203: tot += prlim[i][i];
9204: fprintf(ficrespl," %.5f", prlim[i][i]);
9205: }
9206: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9207: } /* Age */
9208: /* was end of cptcod */
9209: } /* cptcov */
9210: } /* nres */
1.219 brouard 9211: return 0;
1.180 brouard 9212: }
9213:
1.218 brouard 9214: 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){
9215: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9216:
9217: /* Computes the back prevalence limit for any combination of covariate values
9218: * at any age between ageminpar and agemaxpar
9219: */
1.235 brouard 9220: int i, j, k, i1, nres=0 ;
1.217 brouard 9221: /* double ftolpl = 1.e-10; */
9222: double age, agebase, agelim;
9223: double tot;
1.218 brouard 9224: /* double ***mobaverage; */
9225: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9226:
9227: strcpy(fileresplb,"PLB_");
9228: strcat(fileresplb,fileresu);
9229: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9230: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9231: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9232: }
9233: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9234: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9235: pstamp(ficresplb);
9236: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9237: fprintf(ficresplb,"#Age ");
9238: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9239: fprintf(ficresplb,"\n");
9240:
1.218 brouard 9241:
9242: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9243:
9244: agebase=ageminpar;
9245: agelim=agemaxpar;
9246:
9247:
1.227 brouard 9248: i1=pow(2,cptcoveff);
1.218 brouard 9249: if (cptcovn < 1){i1=1;}
1.227 brouard 9250:
1.238 brouard 9251: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9252: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9253: if(TKresult[nres]!= k)
9254: continue;
9255: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9256: fprintf(ficresplb,"#******");
9257: printf("#******");
9258: fprintf(ficlog,"#******");
9259: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9260: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9261: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9262: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9263: }
9264: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9265: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9266: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9267: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9268: }
9269: fprintf(ficresplb,"******\n");
9270: printf("******\n");
9271: fprintf(ficlog,"******\n");
9272: if(invalidvarcomb[k]){
9273: printf("\nCombination (%d) ignored because no cases \n",k);
9274: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9275: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9276: continue;
9277: }
1.218 brouard 9278:
1.238 brouard 9279: fprintf(ficresplb,"#Age ");
9280: for(j=1;j<=cptcoveff;j++) {
9281: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9282: }
9283: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9284: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9285:
9286:
1.238 brouard 9287: for (age=agebase; age<=agelim; age++){
9288: /* for (age=agebase; age<=agebase; age++){ */
9289: if(mobilavproj > 0){
9290: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9291: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 ! brouard 9292: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9293: }else if (mobilavproj == 0){
9294: 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);
9295: 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);
9296: exit(1);
9297: }else{
9298: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 ! brouard 9299: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9300: }
9301: fprintf(ficresplb,"%.0f ",age );
9302: for(j=1;j<=cptcoveff;j++)
9303: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9304: tot=0.;
9305: for(i=1; i<=nlstate;i++){
9306: tot += bprlim[i][i];
9307: fprintf(ficresplb," %.5f", bprlim[i][i]);
9308: }
9309: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9310: } /* Age */
9311: /* was end of cptcod */
9312: } /* end of any combination */
9313: } /* end of nres */
1.218 brouard 9314: /* hBijx(p, bage, fage); */
9315: /* fclose(ficrespijb); */
9316:
9317: return 0;
1.217 brouard 9318: }
1.218 brouard 9319:
1.180 brouard 9320: int hPijx(double *p, int bage, int fage){
9321: /*------------- h Pij x at various ages ------------*/
9322:
9323: int stepsize;
9324: int agelim;
9325: int hstepm;
9326: int nhstepm;
1.235 brouard 9327: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9328:
9329: double agedeb;
9330: double ***p3mat;
9331:
1.201 brouard 9332: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9333: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9334: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9335: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9336: }
9337: printf("Computing pij: result on file '%s' \n", filerespij);
9338: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9339:
9340: stepsize=(int) (stepm+YEARM-1)/YEARM;
9341: /*if (stepm<=24) stepsize=2;*/
9342:
9343: agelim=AGESUP;
9344: hstepm=stepsize*YEARM; /* Every year of age */
9345: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9346:
1.180 brouard 9347: /* hstepm=1; aff par mois*/
9348: pstamp(ficrespij);
9349: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9350: i1= pow(2,cptcoveff);
1.218 brouard 9351: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9352: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9353: /* k=k+1; */
1.235 brouard 9354: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9355: for(k=1; k<=i1;k++){
9356: if(TKresult[nres]!= k)
9357: continue;
1.183 brouard 9358: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9359: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9360: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9361: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9362: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9363: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9364: }
1.183 brouard 9365: fprintf(ficrespij,"******\n");
9366:
9367: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9368: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9369: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9370:
9371: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9372:
1.183 brouard 9373: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9374: oldm=oldms;savm=savms;
1.235 brouard 9375: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9376: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9377: for(i=1; i<=nlstate;i++)
9378: for(j=1; j<=nlstate+ndeath;j++)
9379: fprintf(ficrespij," %1d-%1d",i,j);
9380: fprintf(ficrespij,"\n");
9381: for (h=0; h<=nhstepm; h++){
9382: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9383: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9384: for(i=1; i<=nlstate;i++)
9385: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9386: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9387: fprintf(ficrespij,"\n");
9388: }
1.183 brouard 9389: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9390: fprintf(ficrespij,"\n");
9391: }
1.180 brouard 9392: /*}*/
9393: }
1.218 brouard 9394: return 0;
1.180 brouard 9395: }
1.218 brouard 9396:
9397: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9398: /*------------- h Bij x at various ages ------------*/
9399:
9400: int stepsize;
1.218 brouard 9401: /* int agelim; */
9402: int ageminl;
1.217 brouard 9403: int hstepm;
9404: int nhstepm;
1.238 brouard 9405: int h, i, i1, j, k, nres;
1.218 brouard 9406:
1.217 brouard 9407: double agedeb;
9408: double ***p3mat;
1.218 brouard 9409:
9410: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9411: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9412: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9413: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9414: }
9415: printf("Computing pij back: result on file '%s' \n", filerespijb);
9416: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9417:
9418: stepsize=(int) (stepm+YEARM-1)/YEARM;
9419: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9420:
1.218 brouard 9421: /* agelim=AGESUP; */
9422: ageminl=30;
9423: hstepm=stepsize*YEARM; /* Every year of age */
9424: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9425:
9426: /* hstepm=1; aff par mois*/
9427: pstamp(ficrespijb);
9428: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9429: i1= pow(2,cptcoveff);
1.218 brouard 9430: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9431: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9432: /* k=k+1; */
1.238 brouard 9433: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9434: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9435: if(TKresult[nres]!= k)
9436: continue;
9437: fprintf(ficrespijb,"\n#****** ");
9438: for(j=1;j<=cptcoveff;j++)
9439: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9440: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9441: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9442: }
9443: fprintf(ficrespijb,"******\n");
9444: if(invalidvarcomb[k]){
9445: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9446: continue;
9447: }
9448:
9449: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9450: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9451: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9452: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9453: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9454:
9455: /* nhstepm=nhstepm*YEARM; aff par mois*/
9456:
9457: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9458: /* oldm=oldms;savm=savms; */
9459: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9460: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9461: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9462: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9463: for(i=1; i<=nlstate;i++)
9464: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9465: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9466: fprintf(ficrespijb,"\n");
1.238 brouard 9467: for (h=0; h<=nhstepm; h++){
9468: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9469: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9470: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9471: for(i=1; i<=nlstate;i++)
9472: for(j=1; j<=nlstate+ndeath;j++)
9473: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9474: fprintf(ficrespijb,"\n");
9475: }
9476: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9477: fprintf(ficrespijb,"\n");
9478: } /* end age deb */
9479: } /* end combination */
9480: } /* end nres */
1.218 brouard 9481: return 0;
9482: } /* hBijx */
1.217 brouard 9483:
1.180 brouard 9484:
1.136 brouard 9485: /***********************************************/
9486: /**************** Main Program *****************/
9487: /***********************************************/
9488:
9489: int main(int argc, char *argv[])
9490: {
9491: #ifdef GSL
9492: const gsl_multimin_fminimizer_type *T;
9493: size_t iteri = 0, it;
9494: int rval = GSL_CONTINUE;
9495: int status = GSL_SUCCESS;
9496: double ssval;
9497: #endif
9498: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9499: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9500: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9501: int jj, ll, li, lj, lk;
1.136 brouard 9502: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9503: int num_filled;
1.136 brouard 9504: int itimes;
9505: int NDIM=2;
9506: int vpopbased=0;
1.235 brouard 9507: int nres=0;
1.136 brouard 9508:
1.164 brouard 9509: char ca[32], cb[32];
1.136 brouard 9510: /* FILE *fichtm; *//* Html File */
9511: /* FILE *ficgp;*/ /*Gnuplot File */
9512: struct stat info;
1.191 brouard 9513: double agedeb=0.;
1.194 brouard 9514:
9515: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9516: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9517:
1.165 brouard 9518: double fret;
1.191 brouard 9519: double dum=0.; /* Dummy variable */
1.136 brouard 9520: double ***p3mat;
1.218 brouard 9521: /* double ***mobaverage; */
1.164 brouard 9522:
9523: char line[MAXLINE];
1.197 brouard 9524: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9525:
1.234 brouard 9526: char modeltemp[MAXLINE];
1.230 brouard 9527: char resultline[MAXLINE];
9528:
1.136 brouard 9529: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9530: char *tok, *val; /* pathtot */
1.136 brouard 9531: int firstobs=1, lastobs=10;
1.195 brouard 9532: int c, h , cpt, c2;
1.191 brouard 9533: int jl=0;
9534: int i1, j1, jk, stepsize=0;
1.194 brouard 9535: int count=0;
9536:
1.164 brouard 9537: int *tab;
1.136 brouard 9538: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9539: int backcast=0;
1.136 brouard 9540: int mobilav=0,popforecast=0;
1.191 brouard 9541: int hstepm=0, nhstepm=0;
1.136 brouard 9542: int agemortsup;
9543: float sumlpop=0.;
9544: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9545: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9546:
1.191 brouard 9547: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9548: double ftolpl=FTOL;
9549: double **prlim;
1.217 brouard 9550: double **bprlim;
1.136 brouard 9551: double ***param; /* Matrix of parameters */
9552: double *p;
9553: double **matcov; /* Matrix of covariance */
1.203 brouard 9554: double **hess; /* Hessian matrix */
1.136 brouard 9555: double ***delti3; /* Scale */
9556: double *delti; /* Scale */
9557: double ***eij, ***vareij;
9558: double **varpl; /* Variances of prevalence limits by age */
9559: double *epj, vepp;
1.164 brouard 9560:
1.136 brouard 9561: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9562: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9563:
1.136 brouard 9564: double **ximort;
1.145 brouard 9565: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9566: int *dcwave;
9567:
1.164 brouard 9568: char z[1]="c";
1.136 brouard 9569:
9570: /*char *strt;*/
9571: char strtend[80];
1.126 brouard 9572:
1.164 brouard 9573:
1.126 brouard 9574: /* setlocale (LC_ALL, ""); */
9575: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9576: /* textdomain (PACKAGE); */
9577: /* setlocale (LC_CTYPE, ""); */
9578: /* setlocale (LC_MESSAGES, ""); */
9579:
9580: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9581: rstart_time = time(NULL);
9582: /* (void) gettimeofday(&start_time,&tzp);*/
9583: start_time = *localtime(&rstart_time);
1.126 brouard 9584: curr_time=start_time;
1.157 brouard 9585: /*tml = *localtime(&start_time.tm_sec);*/
9586: /* strcpy(strstart,asctime(&tml)); */
9587: strcpy(strstart,asctime(&start_time));
1.126 brouard 9588:
9589: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9590: /* tp.tm_sec = tp.tm_sec +86400; */
9591: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9592: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9593: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9594: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9595: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9596: /* strt=asctime(&tmg); */
9597: /* printf("Time(after) =%s",strstart); */
9598: /* (void) time (&time_value);
9599: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9600: * tm = *localtime(&time_value);
9601: * strstart=asctime(&tm);
9602: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9603: */
9604:
9605: nberr=0; /* Number of errors and warnings */
9606: nbwarn=0;
1.184 brouard 9607: #ifdef WIN32
9608: _getcwd(pathcd, size);
9609: #else
1.126 brouard 9610: getcwd(pathcd, size);
1.184 brouard 9611: #endif
1.191 brouard 9612: syscompilerinfo(0);
1.196 brouard 9613: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9614: if(argc <=1){
9615: printf("\nEnter the parameter file name: ");
1.205 brouard 9616: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9617: printf("ERROR Empty parameter file name\n");
9618: goto end;
9619: }
1.126 brouard 9620: i=strlen(pathr);
9621: if(pathr[i-1]=='\n')
9622: pathr[i-1]='\0';
1.156 brouard 9623: i=strlen(pathr);
1.205 brouard 9624: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9625: pathr[i-1]='\0';
1.205 brouard 9626: }
9627: i=strlen(pathr);
9628: if( i==0 ){
9629: printf("ERROR Empty parameter file name\n");
9630: goto end;
9631: }
9632: for (tok = pathr; tok != NULL; ){
1.126 brouard 9633: printf("Pathr |%s|\n",pathr);
9634: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9635: printf("val= |%s| pathr=%s\n",val,pathr);
9636: strcpy (pathtot, val);
9637: if(pathr[0] == '\0') break; /* Dirty */
9638: }
9639: }
9640: else{
9641: strcpy(pathtot,argv[1]);
9642: }
9643: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9644: /*cygwin_split_path(pathtot,path,optionfile);
9645: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9646: /* cutv(path,optionfile,pathtot,'\\');*/
9647:
9648: /* Split argv[0], imach program to get pathimach */
9649: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9650: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9651: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9652: /* strcpy(pathimach,argv[0]); */
9653: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9654: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9655: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9656: #ifdef WIN32
9657: _chdir(path); /* Can be a relative path */
9658: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9659: #else
1.126 brouard 9660: chdir(path); /* Can be a relative path */
1.184 brouard 9661: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9662: #endif
9663: printf("Current directory %s!\n",pathcd);
1.126 brouard 9664: strcpy(command,"mkdir ");
9665: strcat(command,optionfilefiname);
9666: if((outcmd=system(command)) != 0){
1.169 brouard 9667: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9668: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9669: /* fclose(ficlog); */
9670: /* exit(1); */
9671: }
9672: /* if((imk=mkdir(optionfilefiname))<0){ */
9673: /* perror("mkdir"); */
9674: /* } */
9675:
9676: /*-------- arguments in the command line --------*/
9677:
1.186 brouard 9678: /* Main Log file */
1.126 brouard 9679: strcat(filelog, optionfilefiname);
9680: strcat(filelog,".log"); /* */
9681: if((ficlog=fopen(filelog,"w"))==NULL) {
9682: printf("Problem with logfile %s\n",filelog);
9683: goto end;
9684: }
9685: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9686: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9687: fprintf(ficlog,"\nEnter the parameter file name: \n");
9688: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9689: path=%s \n\
9690: optionfile=%s\n\
9691: optionfilext=%s\n\
1.156 brouard 9692: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9693:
1.197 brouard 9694: syscompilerinfo(1);
1.167 brouard 9695:
1.126 brouard 9696: printf("Local time (at start):%s",strstart);
9697: fprintf(ficlog,"Local time (at start): %s",strstart);
9698: fflush(ficlog);
9699: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9700: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9701:
9702: /* */
9703: strcpy(fileres,"r");
9704: strcat(fileres, optionfilefiname);
1.201 brouard 9705: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9706: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9707: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9708:
1.186 brouard 9709: /* Main ---------arguments file --------*/
1.126 brouard 9710:
9711: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9712: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9713: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9714: fflush(ficlog);
1.149 brouard 9715: /* goto end; */
9716: exit(70);
1.126 brouard 9717: }
9718:
9719:
9720:
9721: strcpy(filereso,"o");
1.201 brouard 9722: strcat(filereso,fileresu);
1.126 brouard 9723: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9724: printf("Problem with Output resultfile: %s\n", filereso);
9725: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9726: fflush(ficlog);
9727: goto end;
9728: }
9729:
9730: /* Reads comments: lines beginning with '#' */
9731: numlinepar=0;
1.197 brouard 9732:
9733: /* First parameter line */
9734: while(fgets(line, MAXLINE, ficpar)) {
9735: /* If line starts with a # it is a comment */
9736: if (line[0] == '#') {
9737: numlinepar++;
9738: fputs(line,stdout);
9739: fputs(line,ficparo);
9740: fputs(line,ficlog);
9741: continue;
9742: }else
9743: break;
9744: }
9745: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9746: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9747: if (num_filled != 5) {
9748: printf("Should be 5 parameters\n");
9749: }
1.126 brouard 9750: numlinepar++;
1.197 brouard 9751: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
9752: }
9753: /* Second parameter line */
9754: while(fgets(line, MAXLINE, ficpar)) {
9755: /* If line starts with a # it is a comment */
9756: if (line[0] == '#') {
9757: numlinepar++;
9758: fputs(line,stdout);
9759: fputs(line,ficparo);
9760: fputs(line,ficlog);
9761: continue;
9762: }else
9763: break;
9764: }
1.223 brouard 9765: 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", \
9766: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
9767: if (num_filled != 11) {
9768: 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 9769: printf("but line=%s\n",line);
1.197 brouard 9770: }
1.223 brouard 9771: 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 9772: }
1.203 brouard 9773: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 9774: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 9775: /* Third parameter line */
9776: while(fgets(line, MAXLINE, ficpar)) {
9777: /* If line starts with a # it is a comment */
9778: if (line[0] == '#') {
9779: numlinepar++;
9780: fputs(line,stdout);
9781: fputs(line,ficparo);
9782: fputs(line,ficlog);
9783: continue;
9784: }else
9785: break;
9786: }
1.201 brouard 9787: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
9788: if (num_filled == 0)
9789: model[0]='\0';
9790: else if (num_filled != 1){
1.197 brouard 9791: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9792: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9793: model[0]='\0';
9794: goto end;
9795: }
9796: else{
9797: if (model[0]=='+'){
9798: for(i=1; i<=strlen(model);i++)
9799: modeltemp[i-1]=model[i];
1.201 brouard 9800: strcpy(model,modeltemp);
1.197 brouard 9801: }
9802: }
1.199 brouard 9803: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 9804: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 9805: }
9806: /* 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); */
9807: /* numlinepar=numlinepar+3; /\* In general *\/ */
9808: /* 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 9809: 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);
9810: 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 9811: fflush(ficlog);
1.190 brouard 9812: /* if(model[0]=='#'|| model[0]== '\0'){ */
9813: if(model[0]=='#'){
1.187 brouard 9814: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
9815: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
9816: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
9817: if(mle != -1){
9818: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
9819: exit(1);
9820: }
9821: }
1.126 brouard 9822: while((c=getc(ficpar))=='#' && c!= EOF){
9823: ungetc(c,ficpar);
9824: fgets(line, MAXLINE, ficpar);
9825: numlinepar++;
1.195 brouard 9826: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
9827: z[0]=line[1];
9828: }
9829: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 9830: fputs(line, stdout);
9831: //puts(line);
1.126 brouard 9832: fputs(line,ficparo);
9833: fputs(line,ficlog);
9834: }
9835: ungetc(c,ficpar);
9836:
9837:
1.145 brouard 9838: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 9839: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 9840: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 9841: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 9842: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
9843: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
9844: v1+v2*age+v2*v3 makes cptcovn = 3
9845: */
9846: if (strlen(model)>1)
1.187 brouard 9847: 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 9848: else
1.187 brouard 9849: ncovmodel=2; /* Constant and age */
1.133 brouard 9850: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
9851: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 9852: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
9853: 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);
9854: 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);
9855: fflush(stdout);
9856: fclose (ficlog);
9857: goto end;
9858: }
1.126 brouard 9859: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9860: delti=delti3[1][1];
9861: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
9862: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
9863: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 9864: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
9865: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9866: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
9867: fclose (ficparo);
9868: fclose (ficlog);
9869: goto end;
9870: exit(0);
1.220 brouard 9871: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 9872: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 9873: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
9874: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9875: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9876: matcov=matrix(1,npar,1,npar);
1.203 brouard 9877: hess=matrix(1,npar,1,npar);
1.220 brouard 9878: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 9879: /* Read guessed parameters */
1.126 brouard 9880: /* Reads comments: lines beginning with '#' */
9881: while((c=getc(ficpar))=='#' && c!= EOF){
9882: ungetc(c,ficpar);
9883: fgets(line, MAXLINE, ficpar);
9884: numlinepar++;
1.141 brouard 9885: fputs(line,stdout);
1.126 brouard 9886: fputs(line,ficparo);
9887: fputs(line,ficlog);
9888: }
9889: ungetc(c,ficpar);
9890:
9891: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9892: for(i=1; i <=nlstate; i++){
1.234 brouard 9893: j=0;
1.126 brouard 9894: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 9895: if(jj==i) continue;
9896: j++;
9897: fscanf(ficpar,"%1d%1d",&i1,&j1);
9898: if ((i1 != i) || (j1 != jj)){
9899: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 9900: It might be a problem of design; if ncovcol and the model are correct\n \
9901: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 9902: exit(1);
9903: }
9904: fprintf(ficparo,"%1d%1d",i1,j1);
9905: if(mle==1)
9906: printf("%1d%1d",i,jj);
9907: fprintf(ficlog,"%1d%1d",i,jj);
9908: for(k=1; k<=ncovmodel;k++){
9909: fscanf(ficpar," %lf",¶m[i][j][k]);
9910: if(mle==1){
9911: printf(" %lf",param[i][j][k]);
9912: fprintf(ficlog," %lf",param[i][j][k]);
9913: }
9914: else
9915: fprintf(ficlog," %lf",param[i][j][k]);
9916: fprintf(ficparo," %lf",param[i][j][k]);
9917: }
9918: fscanf(ficpar,"\n");
9919: numlinepar++;
9920: if(mle==1)
9921: printf("\n");
9922: fprintf(ficlog,"\n");
9923: fprintf(ficparo,"\n");
1.126 brouard 9924: }
9925: }
9926: fflush(ficlog);
1.234 brouard 9927:
1.145 brouard 9928: /* Reads scales values */
1.126 brouard 9929: p=param[1][1];
9930:
9931: /* Reads comments: lines beginning with '#' */
9932: while((c=getc(ficpar))=='#' && c!= EOF){
9933: ungetc(c,ficpar);
9934: fgets(line, MAXLINE, ficpar);
9935: numlinepar++;
1.141 brouard 9936: fputs(line,stdout);
1.126 brouard 9937: fputs(line,ficparo);
9938: fputs(line,ficlog);
9939: }
9940: ungetc(c,ficpar);
9941:
9942: for(i=1; i <=nlstate; i++){
9943: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 9944: fscanf(ficpar,"%1d%1d",&i1,&j1);
9945: if ( (i1-i) * (j1-j) != 0){
9946: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
9947: exit(1);
9948: }
9949: printf("%1d%1d",i,j);
9950: fprintf(ficparo,"%1d%1d",i1,j1);
9951: fprintf(ficlog,"%1d%1d",i1,j1);
9952: for(k=1; k<=ncovmodel;k++){
9953: fscanf(ficpar,"%le",&delti3[i][j][k]);
9954: printf(" %le",delti3[i][j][k]);
9955: fprintf(ficparo," %le",delti3[i][j][k]);
9956: fprintf(ficlog," %le",delti3[i][j][k]);
9957: }
9958: fscanf(ficpar,"\n");
9959: numlinepar++;
9960: printf("\n");
9961: fprintf(ficparo,"\n");
9962: fprintf(ficlog,"\n");
1.126 brouard 9963: }
9964: }
9965: fflush(ficlog);
1.234 brouard 9966:
1.145 brouard 9967: /* Reads covariance matrix */
1.126 brouard 9968: delti=delti3[1][1];
1.220 brouard 9969:
9970:
1.126 brouard 9971: /* 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 9972:
1.126 brouard 9973: /* Reads comments: lines beginning with '#' */
9974: while((c=getc(ficpar))=='#' && c!= EOF){
9975: ungetc(c,ficpar);
9976: fgets(line, MAXLINE, ficpar);
9977: numlinepar++;
1.141 brouard 9978: fputs(line,stdout);
1.126 brouard 9979: fputs(line,ficparo);
9980: fputs(line,ficlog);
9981: }
9982: ungetc(c,ficpar);
1.220 brouard 9983:
1.126 brouard 9984: matcov=matrix(1,npar,1,npar);
1.203 brouard 9985: hess=matrix(1,npar,1,npar);
1.131 brouard 9986: for(i=1; i <=npar; i++)
9987: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 9988:
1.194 brouard 9989: /* Scans npar lines */
1.126 brouard 9990: for(i=1; i <=npar; i++){
1.226 brouard 9991: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 9992: if(count != 3){
1.226 brouard 9993: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 9994: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
9995: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 9996: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 9997: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
9998: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 9999: exit(1);
1.220 brouard 10000: }else{
1.226 brouard 10001: if(mle==1)
10002: printf("%1d%1d%d",i1,j1,jk);
10003: }
10004: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10005: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10006: for(j=1; j <=i; j++){
1.226 brouard 10007: fscanf(ficpar," %le",&matcov[i][j]);
10008: if(mle==1){
10009: printf(" %.5le",matcov[i][j]);
10010: }
10011: fprintf(ficlog," %.5le",matcov[i][j]);
10012: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10013: }
10014: fscanf(ficpar,"\n");
10015: numlinepar++;
10016: if(mle==1)
1.220 brouard 10017: printf("\n");
1.126 brouard 10018: fprintf(ficlog,"\n");
10019: fprintf(ficparo,"\n");
10020: }
1.194 brouard 10021: /* End of read covariance matrix npar lines */
1.126 brouard 10022: for(i=1; i <=npar; i++)
10023: for(j=i+1;j<=npar;j++)
1.226 brouard 10024: matcov[i][j]=matcov[j][i];
1.126 brouard 10025:
10026: if(mle==1)
10027: printf("\n");
10028: fprintf(ficlog,"\n");
10029:
10030: fflush(ficlog);
10031:
10032: /*-------- Rewriting parameter file ----------*/
10033: strcpy(rfileres,"r"); /* "Rparameterfile */
10034: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10035: strcat(rfileres,"."); /* */
10036: strcat(rfileres,optionfilext); /* Other files have txt extension */
10037: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10038: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10039: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10040: }
10041: fprintf(ficres,"#%s\n",version);
10042: } /* End of mle != -3 */
1.218 brouard 10043:
1.186 brouard 10044: /* Main data
10045: */
1.126 brouard 10046: n= lastobs;
10047: num=lvector(1,n);
10048: moisnais=vector(1,n);
10049: annais=vector(1,n);
10050: moisdc=vector(1,n);
10051: andc=vector(1,n);
1.220 brouard 10052: weight=vector(1,n);
1.126 brouard 10053: agedc=vector(1,n);
10054: cod=ivector(1,n);
1.220 brouard 10055: for(i=1;i<=n;i++){
1.234 brouard 10056: num[i]=0;
10057: moisnais[i]=0;
10058: annais[i]=0;
10059: moisdc[i]=0;
10060: andc[i]=0;
10061: agedc[i]=0;
10062: cod[i]=0;
10063: weight[i]=1.0; /* Equal weights, 1 by default */
10064: }
1.126 brouard 10065: mint=matrix(1,maxwav,1,n);
10066: anint=matrix(1,maxwav,1,n);
1.131 brouard 10067: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10068: tab=ivector(1,NCOVMAX);
1.144 brouard 10069: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10070: 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 10071:
1.136 brouard 10072: /* Reads data from file datafile */
10073: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10074: goto end;
10075:
10076: /* Calculation of the number of parameters from char model */
1.234 brouard 10077: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10078: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10079: k=3 V4 Tvar[k=3]= 4 (from V4)
10080: k=2 V1 Tvar[k=2]= 1 (from V1)
10081: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10082: */
10083:
10084: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10085: TvarsDind=ivector(1,NCOVMAX); /* */
10086: TvarsD=ivector(1,NCOVMAX); /* */
10087: TvarsQind=ivector(1,NCOVMAX); /* */
10088: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10089: TvarF=ivector(1,NCOVMAX); /* */
10090: TvarFind=ivector(1,NCOVMAX); /* */
10091: TvarV=ivector(1,NCOVMAX); /* */
10092: TvarVind=ivector(1,NCOVMAX); /* */
10093: TvarA=ivector(1,NCOVMAX); /* */
10094: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10095: TvarFD=ivector(1,NCOVMAX); /* */
10096: TvarFDind=ivector(1,NCOVMAX); /* */
10097: TvarFQ=ivector(1,NCOVMAX); /* */
10098: TvarFQind=ivector(1,NCOVMAX); /* */
10099: TvarVD=ivector(1,NCOVMAX); /* */
10100: TvarVDind=ivector(1,NCOVMAX); /* */
10101: TvarVQ=ivector(1,NCOVMAX); /* */
10102: TvarVQind=ivector(1,NCOVMAX); /* */
10103:
1.230 brouard 10104: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10105: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10106: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10107: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10108: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10109: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10110: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10111: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10112: */
10113: /* For model-covariate k tells which data-covariate to use but
10114: because this model-covariate is a construction we invent a new column
10115: ncovcol + k1
10116: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10117: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10118: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10119: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10120: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10121: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10122: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10123: */
1.145 brouard 10124: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10125: 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 10126: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10127: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10128: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10129: 4 covariates (3 plus signs)
10130: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10131: */
1.230 brouard 10132: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10133: * individual dummy, fixed or varying:
10134: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10135: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10136: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10137: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10138: * Tmodelind[1]@9={9,0,3,2,}*/
10139: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10140: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10141: * individual quantitative, fixed or varying:
10142: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10143: * 3, 1, 0, 0, 0, 0, 0, 0},
10144: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10145: /* Main decodemodel */
10146:
1.187 brouard 10147:
1.223 brouard 10148: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10149: goto end;
10150:
1.137 brouard 10151: if((double)(lastobs-imx)/(double)imx > 1.10){
10152: nbwarn++;
10153: 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);
10154: 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);
10155: }
1.136 brouard 10156: /* if(mle==1){*/
1.137 brouard 10157: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10158: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10159: }
10160:
10161: /*-calculation of age at interview from date of interview and age at death -*/
10162: agev=matrix(1,maxwav,1,imx);
10163:
10164: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10165: goto end;
10166:
1.126 brouard 10167:
1.136 brouard 10168: agegomp=(int)agemin;
10169: free_vector(moisnais,1,n);
10170: free_vector(annais,1,n);
1.126 brouard 10171: /* free_matrix(mint,1,maxwav,1,n);
10172: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10173: /* free_vector(moisdc,1,n); */
10174: /* free_vector(andc,1,n); */
1.145 brouard 10175: /* */
10176:
1.126 brouard 10177: wav=ivector(1,imx);
1.214 brouard 10178: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10179: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10180: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10181: 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.*/
10182: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10183: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10184:
10185: /* Concatenates waves */
1.214 brouard 10186: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10187: Death is a valid wave (if date is known).
10188: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10189: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10190: and mw[mi+1][i]. dh depends on stepm.
10191: */
10192:
1.126 brouard 10193: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.145 brouard 10194: /* */
10195:
1.215 brouard 10196: free_vector(moisdc,1,n);
10197: free_vector(andc,1,n);
10198:
1.126 brouard 10199: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10200: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10201: ncodemax[1]=1;
1.145 brouard 10202: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10203: cptcoveff=0;
1.220 brouard 10204: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10205: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10206: }
10207:
10208: ncovcombmax=pow(2,cptcoveff);
10209: invalidvarcomb=ivector(1, ncovcombmax);
10210: for(i=1;i<ncovcombmax;i++)
10211: invalidvarcomb[i]=0;
10212:
1.211 brouard 10213: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10214: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10215: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10216:
1.200 brouard 10217: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10218: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10219: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10220: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10221: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10222: * (currently 0 or 1) in the data.
10223: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10224: * corresponding modality (h,j).
10225: */
10226:
1.145 brouard 10227: h=0;
10228: /*if (cptcovn > 0) */
1.126 brouard 10229: m=pow(2,cptcoveff);
10230:
1.144 brouard 10231: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10232: * For k=4 covariates, h goes from 1 to m=2**k
10233: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10234: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10235: * h\k 1 2 3 4
1.143 brouard 10236: *______________________________
10237: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10238: * 2 2 1 1 1
10239: * 3 i=2 1 2 1 1
10240: * 4 2 2 1 1
10241: * 5 i=3 1 i=2 1 2 1
10242: * 6 2 1 2 1
10243: * 7 i=4 1 2 2 1
10244: * 8 2 2 2 1
1.197 brouard 10245: * 9 i=5 1 i=3 1 i=2 1 2
10246: * 10 2 1 1 2
10247: * 11 i=6 1 2 1 2
10248: * 12 2 2 1 2
10249: * 13 i=7 1 i=4 1 2 2
10250: * 14 2 1 2 2
10251: * 15 i=8 1 2 2 2
10252: * 16 2 2 2 2
1.143 brouard 10253: */
1.212 brouard 10254: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10255: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10256: * and the value of each covariate?
10257: * V1=1, V2=1, V3=2, V4=1 ?
10258: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10259: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10260: * In order to get the real value in the data, we use nbcode
10261: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10262: * We are keeping this crazy system in order to be able (in the future?)
10263: * to have more than 2 values (0 or 1) for a covariate.
10264: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10265: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10266: * bbbbbbbb
10267: * 76543210
10268: * h-1 00000101 (6-1=5)
1.219 brouard 10269: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10270: * &
10271: * 1 00000001 (1)
1.219 brouard 10272: * 00000000 = 1 & ((h-1) >> (k-1))
10273: * +1= 00000001 =1
1.211 brouard 10274: *
10275: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10276: * h' 1101 =2^3+2^2+0x2^1+2^0
10277: * >>k' 11
10278: * & 00000001
10279: * = 00000001
10280: * +1 = 00000010=2 = codtabm(14,3)
10281: * Reverse h=6 and m=16?
10282: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10283: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10284: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10285: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10286: * V3=decodtabm(14,3,2**4)=2
10287: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10288: *(h-1) >> (j-1) 0011 =13 >> 2
10289: * &1 000000001
10290: * = 000000001
10291: * +1= 000000010 =2
10292: * 2211
10293: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10294: * V3=2
1.220 brouard 10295: * codtabm and decodtabm are identical
1.211 brouard 10296: */
10297:
1.145 brouard 10298:
10299: free_ivector(Ndum,-1,NCOVMAX);
10300:
10301:
1.126 brouard 10302:
1.186 brouard 10303: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10304: strcpy(optionfilegnuplot,optionfilefiname);
10305: if(mle==-3)
1.201 brouard 10306: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10307: strcat(optionfilegnuplot,".gp");
10308:
10309: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10310: printf("Problem with file %s",optionfilegnuplot);
10311: }
10312: else{
1.204 brouard 10313: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10314: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10315: //fprintf(ficgp,"set missing 'NaNq'\n");
10316: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10317: }
10318: /* fclose(ficgp);*/
1.186 brouard 10319:
10320:
10321: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10322:
10323: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10324: if(mle==-3)
1.201 brouard 10325: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10326: strcat(optionfilehtm,".htm");
10327: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10328: printf("Problem with %s \n",optionfilehtm);
10329: exit(0);
1.126 brouard 10330: }
10331:
10332: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10333: strcat(optionfilehtmcov,"-cov.htm");
10334: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10335: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10336: }
10337: else{
10338: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10339: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10340: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10341: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10342: }
10343:
1.213 brouard 10344: 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 10345: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10346: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10347: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10348: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10349: \n\
10350: <hr size=\"2\" color=\"#EC5E5E\">\
10351: <ul><li><h4>Parameter files</h4>\n\
10352: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10353: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10354: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10355: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10356: - Date and time at start: %s</ul>\n",\
10357: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10358: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10359: fileres,fileres,\
10360: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10361: fflush(fichtm);
10362:
10363: strcpy(pathr,path);
10364: strcat(pathr,optionfilefiname);
1.184 brouard 10365: #ifdef WIN32
10366: _chdir(optionfilefiname); /* Move to directory named optionfile */
10367: #else
1.126 brouard 10368: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10369: #endif
10370:
1.126 brouard 10371:
1.220 brouard 10372: /* Calculates basic frequencies. Computes observed prevalence at single age
10373: and for any valid combination of covariates
1.126 brouard 10374: and prints on file fileres'p'. */
1.227 brouard 10375: freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
10376: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10377:
10378: fprintf(fichtm,"\n");
10379: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10380: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10381: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10382: imx,agemin,agemax,jmin,jmax,jmean);
10383: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10384: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10385: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10386: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10387: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10388:
1.126 brouard 10389: /* For Powell, parameters are in a vector p[] starting at p[1]
10390: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10391: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10392:
10393: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10394: /* For mortality only */
1.126 brouard 10395: if (mle==-3){
1.136 brouard 10396: ximort=matrix(1,NDIM,1,NDIM);
1.220 brouard 10397: for(i=1;i<=NDIM;i++)
10398: for(j=1;j<=NDIM;j++)
10399: ximort[i][j]=0.;
1.186 brouard 10400: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10401: cens=ivector(1,n);
10402: ageexmed=vector(1,n);
10403: agecens=vector(1,n);
10404: dcwave=ivector(1,n);
1.223 brouard 10405:
1.126 brouard 10406: for (i=1; i<=imx; i++){
10407: dcwave[i]=-1;
10408: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10409: if (s[m][i]>nlstate) {
10410: dcwave[i]=m;
10411: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10412: break;
10413: }
1.126 brouard 10414: }
1.226 brouard 10415:
1.126 brouard 10416: for (i=1; i<=imx; i++) {
10417: if (wav[i]>0){
1.226 brouard 10418: ageexmed[i]=agev[mw[1][i]][i];
10419: j=wav[i];
10420: agecens[i]=1.;
10421:
10422: if (ageexmed[i]> 1 && wav[i] > 0){
10423: agecens[i]=agev[mw[j][i]][i];
10424: cens[i]= 1;
10425: }else if (ageexmed[i]< 1)
10426: cens[i]= -1;
10427: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10428: cens[i]=0 ;
1.126 brouard 10429: }
10430: else cens[i]=-1;
10431: }
10432:
10433: for (i=1;i<=NDIM;i++) {
10434: for (j=1;j<=NDIM;j++)
1.226 brouard 10435: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10436: }
10437:
1.145 brouard 10438: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10439: /*printf("%lf %lf", p[1], p[2]);*/
10440:
10441:
1.136 brouard 10442: #ifdef GSL
10443: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10444: #else
1.126 brouard 10445: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10446: #endif
1.201 brouard 10447: strcpy(filerespow,"POW-MORT_");
10448: strcat(filerespow,fileresu);
1.126 brouard 10449: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10450: printf("Problem with resultfile: %s\n", filerespow);
10451: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10452: }
1.136 brouard 10453: #ifdef GSL
10454: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10455: #else
1.126 brouard 10456: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10457: #endif
1.126 brouard 10458: /* for (i=1;i<=nlstate;i++)
10459: for(j=1;j<=nlstate+ndeath;j++)
10460: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10461: */
10462: fprintf(ficrespow,"\n");
1.136 brouard 10463: #ifdef GSL
10464: /* gsl starts here */
10465: T = gsl_multimin_fminimizer_nmsimplex;
10466: gsl_multimin_fminimizer *sfm = NULL;
10467: gsl_vector *ss, *x;
10468: gsl_multimin_function minex_func;
10469:
10470: /* Initial vertex size vector */
10471: ss = gsl_vector_alloc (NDIM);
10472:
10473: if (ss == NULL){
10474: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10475: }
10476: /* Set all step sizes to 1 */
10477: gsl_vector_set_all (ss, 0.001);
10478:
10479: /* Starting point */
1.126 brouard 10480:
1.136 brouard 10481: x = gsl_vector_alloc (NDIM);
10482:
10483: if (x == NULL){
10484: gsl_vector_free(ss);
10485: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10486: }
10487:
10488: /* Initialize method and iterate */
10489: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10490: /* gsl_vector_set(x, 0, 0.0268); */
10491: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10492: gsl_vector_set(x, 0, p[1]);
10493: gsl_vector_set(x, 1, p[2]);
10494:
10495: minex_func.f = &gompertz_f;
10496: minex_func.n = NDIM;
10497: minex_func.params = (void *)&p; /* ??? */
10498:
10499: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10500: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10501:
10502: printf("Iterations beginning .....\n\n");
10503: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10504:
10505: iteri=0;
10506: while (rval == GSL_CONTINUE){
10507: iteri++;
10508: status = gsl_multimin_fminimizer_iterate(sfm);
10509:
10510: if (status) printf("error: %s\n", gsl_strerror (status));
10511: fflush(0);
10512:
10513: if (status)
10514: break;
10515:
10516: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10517: ssval = gsl_multimin_fminimizer_size (sfm);
10518:
10519: if (rval == GSL_SUCCESS)
10520: printf ("converged to a local maximum at\n");
10521:
10522: printf("%5d ", iteri);
10523: for (it = 0; it < NDIM; it++){
10524: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10525: }
10526: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10527: }
10528:
10529: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10530:
10531: gsl_vector_free(x); /* initial values */
10532: gsl_vector_free(ss); /* inital step size */
10533: for (it=0; it<NDIM; it++){
10534: p[it+1]=gsl_vector_get(sfm->x,it);
10535: fprintf(ficrespow," %.12lf", p[it]);
10536: }
10537: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10538: #endif
10539: #ifdef POWELL
10540: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10541: #endif
1.126 brouard 10542: fclose(ficrespow);
10543:
1.203 brouard 10544: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10545:
10546: for(i=1; i <=NDIM; i++)
10547: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10548: matcov[i][j]=matcov[j][i];
1.126 brouard 10549:
10550: printf("\nCovariance matrix\n ");
1.203 brouard 10551: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10552: for(i=1; i <=NDIM; i++) {
10553: for(j=1;j<=NDIM;j++){
1.220 brouard 10554: printf("%f ",matcov[i][j]);
10555: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10556: }
1.203 brouard 10557: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10558: }
10559:
10560: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10561: for (i=1;i<=NDIM;i++) {
1.126 brouard 10562: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10563: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10564: }
1.126 brouard 10565: lsurv=vector(1,AGESUP);
10566: lpop=vector(1,AGESUP);
10567: tpop=vector(1,AGESUP);
10568: lsurv[agegomp]=100000;
10569:
10570: for (k=agegomp;k<=AGESUP;k++) {
10571: agemortsup=k;
10572: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10573: }
10574:
10575: for (k=agegomp;k<agemortsup;k++)
10576: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10577:
10578: for (k=agegomp;k<agemortsup;k++){
10579: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10580: sumlpop=sumlpop+lpop[k];
10581: }
10582:
10583: tpop[agegomp]=sumlpop;
10584: for (k=agegomp;k<(agemortsup-3);k++){
10585: /* tpop[k+1]=2;*/
10586: tpop[k+1]=tpop[k]-lpop[k];
10587: }
10588:
10589:
10590: printf("\nAge lx qx dx Lx Tx e(x)\n");
10591: for (k=agegomp;k<(agemortsup-2);k++)
10592: 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]);
10593:
10594:
10595: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10596: ageminpar=50;
10597: agemaxpar=100;
1.194 brouard 10598: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10599: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10600: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10601: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10602: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10603: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10604: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10605: }else{
10606: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10607: 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 10608: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10609: }
1.201 brouard 10610: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10611: stepm, weightopt,\
10612: model,imx,p,matcov,agemortsup);
10613:
10614: free_vector(lsurv,1,AGESUP);
10615: free_vector(lpop,1,AGESUP);
10616: free_vector(tpop,1,AGESUP);
1.220 brouard 10617: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10618: free_ivector(cens,1,n);
10619: free_vector(agecens,1,n);
10620: free_ivector(dcwave,1,n);
1.220 brouard 10621: #ifdef GSL
1.136 brouard 10622: #endif
1.186 brouard 10623: } /* Endof if mle==-3 mortality only */
1.205 brouard 10624: /* Standard */
10625: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10626: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10627: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10628: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10629: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10630: for (k=1; k<=npar;k++)
10631: printf(" %d %8.5f",k,p[k]);
10632: printf("\n");
1.205 brouard 10633: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10634: /* mlikeli uses func not funcone */
10635: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10636: }
10637: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10638: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10639: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10640: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10641: }
10642: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10643: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10644: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10645: for (k=1; k<=npar;k++)
10646: printf(" %d %8.5f",k,p[k]);
10647: printf("\n");
10648:
10649: /*--------- results files --------------*/
1.224 brouard 10650: 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 10651:
10652:
10653: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10654: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10655: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10656: for(i=1,jk=1; i <=nlstate; i++){
10657: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10658: if (k != i) {
10659: printf("%d%d ",i,k);
10660: fprintf(ficlog,"%d%d ",i,k);
10661: fprintf(ficres,"%1d%1d ",i,k);
10662: for(j=1; j <=ncovmodel; j++){
10663: printf("%12.7f ",p[jk]);
10664: fprintf(ficlog,"%12.7f ",p[jk]);
10665: fprintf(ficres,"%12.7f ",p[jk]);
10666: jk++;
10667: }
10668: printf("\n");
10669: fprintf(ficlog,"\n");
10670: fprintf(ficres,"\n");
10671: }
1.126 brouard 10672: }
10673: }
1.203 brouard 10674: if(mle != 0){
10675: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10676: ftolhess=ftol; /* Usually correct */
1.203 brouard 10677: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10678: 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");
10679: 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");
10680: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10681: for(k=1; k <=(nlstate+ndeath); k++){
10682: if (k != i) {
10683: printf("%d%d ",i,k);
10684: fprintf(ficlog,"%d%d ",i,k);
10685: for(j=1; j <=ncovmodel; j++){
10686: 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]));
10687: 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]));
10688: jk++;
10689: }
10690: printf("\n");
10691: fprintf(ficlog,"\n");
10692: }
10693: }
1.193 brouard 10694: }
1.203 brouard 10695: } /* end of hesscov and Wald tests */
1.225 brouard 10696:
1.203 brouard 10697: /* */
1.126 brouard 10698: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10699: printf("# Scales (for hessian or gradient estimation)\n");
10700: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10701: for(i=1,jk=1; i <=nlstate; i++){
10702: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10703: if (j!=i) {
10704: fprintf(ficres,"%1d%1d",i,j);
10705: printf("%1d%1d",i,j);
10706: fprintf(ficlog,"%1d%1d",i,j);
10707: for(k=1; k<=ncovmodel;k++){
10708: printf(" %.5e",delti[jk]);
10709: fprintf(ficlog," %.5e",delti[jk]);
10710: fprintf(ficres," %.5e",delti[jk]);
10711: jk++;
10712: }
10713: printf("\n");
10714: fprintf(ficlog,"\n");
10715: fprintf(ficres,"\n");
10716: }
1.126 brouard 10717: }
10718: }
10719:
10720: 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 10721: if(mle >= 1) /* To big for the screen */
1.126 brouard 10722: 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");
10723: 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");
10724: /* # 121 Var(a12)\n\ */
10725: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10726: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10727: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10728: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10729: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10730: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10731: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10732:
10733:
10734: /* Just to have a covariance matrix which will be more understandable
10735: even is we still don't want to manage dictionary of variables
10736: */
10737: for(itimes=1;itimes<=2;itimes++){
10738: jj=0;
10739: for(i=1; i <=nlstate; i++){
1.225 brouard 10740: for(j=1; j <=nlstate+ndeath; j++){
10741: if(j==i) continue;
10742: for(k=1; k<=ncovmodel;k++){
10743: jj++;
10744: ca[0]= k+'a'-1;ca[1]='\0';
10745: if(itimes==1){
10746: if(mle>=1)
10747: printf("#%1d%1d%d",i,j,k);
10748: fprintf(ficlog,"#%1d%1d%d",i,j,k);
10749: fprintf(ficres,"#%1d%1d%d",i,j,k);
10750: }else{
10751: if(mle>=1)
10752: printf("%1d%1d%d",i,j,k);
10753: fprintf(ficlog,"%1d%1d%d",i,j,k);
10754: fprintf(ficres,"%1d%1d%d",i,j,k);
10755: }
10756: ll=0;
10757: for(li=1;li <=nlstate; li++){
10758: for(lj=1;lj <=nlstate+ndeath; lj++){
10759: if(lj==li) continue;
10760: for(lk=1;lk<=ncovmodel;lk++){
10761: ll++;
10762: if(ll<=jj){
10763: cb[0]= lk +'a'-1;cb[1]='\0';
10764: if(ll<jj){
10765: if(itimes==1){
10766: if(mle>=1)
10767: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10768: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10769: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10770: }else{
10771: if(mle>=1)
10772: printf(" %.5e",matcov[jj][ll]);
10773: fprintf(ficlog," %.5e",matcov[jj][ll]);
10774: fprintf(ficres," %.5e",matcov[jj][ll]);
10775: }
10776: }else{
10777: if(itimes==1){
10778: if(mle>=1)
10779: printf(" Var(%s%1d%1d)",ca,i,j);
10780: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
10781: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
10782: }else{
10783: if(mle>=1)
10784: printf(" %.7e",matcov[jj][ll]);
10785: fprintf(ficlog," %.7e",matcov[jj][ll]);
10786: fprintf(ficres," %.7e",matcov[jj][ll]);
10787: }
10788: }
10789: }
10790: } /* end lk */
10791: } /* end lj */
10792: } /* end li */
10793: if(mle>=1)
10794: printf("\n");
10795: fprintf(ficlog,"\n");
10796: fprintf(ficres,"\n");
10797: numlinepar++;
10798: } /* end k*/
10799: } /*end j */
1.126 brouard 10800: } /* end i */
10801: } /* end itimes */
10802:
10803: fflush(ficlog);
10804: fflush(ficres);
1.225 brouard 10805: while(fgets(line, MAXLINE, ficpar)) {
10806: /* If line starts with a # it is a comment */
10807: if (line[0] == '#') {
10808: numlinepar++;
10809: fputs(line,stdout);
10810: fputs(line,ficparo);
10811: fputs(line,ficlog);
10812: continue;
10813: }else
10814: break;
10815: }
10816:
1.209 brouard 10817: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
10818: /* ungetc(c,ficpar); */
10819: /* fgets(line, MAXLINE, ficpar); */
10820: /* fputs(line,stdout); */
10821: /* fputs(line,ficparo); */
10822: /* } */
10823: /* ungetc(c,ficpar); */
1.126 brouard 10824:
10825: estepm=0;
1.209 brouard 10826: 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 10827:
10828: if (num_filled != 6) {
10829: 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);
10830: 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);
10831: goto end;
10832: }
10833: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
10834: }
10835: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
10836: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
10837:
1.209 brouard 10838: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 10839: if (estepm==0 || estepm < stepm) estepm=stepm;
10840: if (fage <= 2) {
10841: bage = ageminpar;
10842: fage = agemaxpar;
10843: }
10844:
10845: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 10846: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
10847: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 10848:
1.186 brouard 10849: /* Other stuffs, more or less useful */
1.126 brouard 10850: while((c=getc(ficpar))=='#' && c!= EOF){
10851: ungetc(c,ficpar);
10852: fgets(line, MAXLINE, ficpar);
1.141 brouard 10853: fputs(line,stdout);
1.126 brouard 10854: fputs(line,ficparo);
10855: }
10856: ungetc(c,ficpar);
10857:
10858: 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);
10859: 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);
10860: 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);
10861: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
10862: 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);
10863:
10864: while((c=getc(ficpar))=='#' && c!= EOF){
10865: ungetc(c,ficpar);
10866: fgets(line, MAXLINE, ficpar);
1.141 brouard 10867: fputs(line,stdout);
1.126 brouard 10868: fputs(line,ficparo);
10869: }
10870: ungetc(c,ficpar);
10871:
10872:
10873: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
10874: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
10875:
10876: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 10877: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 10878: fprintf(ficparo,"pop_based=%d\n",popbased);
10879: fprintf(ficres,"pop_based=%d\n",popbased);
10880:
10881: while((c=getc(ficpar))=='#' && c!= EOF){
10882: ungetc(c,ficpar);
10883: fgets(line, MAXLINE, ficpar);
1.141 brouard 10884: fputs(line,stdout);
1.238 brouard 10885: fputs(line,ficres);
1.126 brouard 10886: fputs(line,ficparo);
10887: }
10888: ungetc(c,ficpar);
10889:
10890: 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);
10891: 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);
10892: 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);
10893: 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);
10894: 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);
10895: /* day and month of proj2 are not used but only year anproj2.*/
10896:
1.217 brouard 10897: while((c=getc(ficpar))=='#' && c!= EOF){
10898: ungetc(c,ficpar);
10899: fgets(line, MAXLINE, ficpar);
10900: fputs(line,stdout);
10901: fputs(line,ficparo);
1.238 brouard 10902: fputs(line,ficres);
1.217 brouard 10903: }
10904: ungetc(c,ficpar);
10905:
10906: 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);
1.223 brouard 10907: 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);
10908: 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);
10909: 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 10910: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 10911:
1.230 brouard 10912: /* Results */
1.235 brouard 10913: nresult=0;
1.230 brouard 10914: while(fgets(line, MAXLINE, ficpar)) {
10915: /* If line starts with a # it is a comment */
10916: if (line[0] == '#') {
10917: numlinepar++;
10918: fputs(line,stdout);
10919: fputs(line,ficparo);
10920: fputs(line,ficlog);
1.238 brouard 10921: fputs(line,ficres);
1.230 brouard 10922: continue;
10923: }else
10924: break;
10925: }
1.240 brouard 10926: if (!feof(ficpar))
1.230 brouard 10927: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 10928: if (num_filled == 0){
1.230 brouard 10929: resultline[0]='\0';
1.240 brouard 10930: break;
10931: } else if (num_filled != 1){
1.230 brouard 10932: printf("ERROR %d: result line should be at minimum 'result=' %s\n",num_filled, line);
10933: }
1.235 brouard 10934: nresult++; /* Sum of resultlines */
10935: printf("Result %d: result=%s\n",nresult, resultline);
10936: if(nresult > MAXRESULTLINES){
10937: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10938: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10939: goto end;
10940: }
10941: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 10942: fprintf(ficparo,"result: %s\n",resultline);
10943: fprintf(ficres,"result: %s\n",resultline);
10944: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 10945: while(fgets(line, MAXLINE, ficpar)) {
10946: /* If line starts with a # it is a comment */
10947: if (line[0] == '#') {
10948: numlinepar++;
10949: fputs(line,stdout);
10950: fputs(line,ficparo);
1.238 brouard 10951: fputs(line,ficres);
1.230 brouard 10952: fputs(line,ficlog);
10953: continue;
10954: }else
10955: break;
10956: }
10957: if (feof(ficpar))
10958: break;
10959: else{ /* Processess output results for this combination of covariate values */
10960: }
1.240 brouard 10961: } /* end while */
1.230 brouard 10962:
10963:
1.126 brouard 10964:
1.230 brouard 10965: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 10966: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 10967:
10968: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 10969: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 10970: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 10971: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10972: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 10973: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 10974: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10975: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10976: }else{
1.218 brouard 10977: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 10978: }
10979: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 10980: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
10981: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 10982:
1.225 brouard 10983: /*------------ free_vector -------------*/
10984: /* chdir(path); */
1.220 brouard 10985:
1.215 brouard 10986: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
10987: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
10988: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
10989: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 10990: free_lvector(num,1,n);
10991: free_vector(agedc,1,n);
10992: /*free_matrix(covar,0,NCOVMAX,1,n);*/
10993: /*free_matrix(covar,1,NCOVMAX,1,n);*/
10994: fclose(ficparo);
10995: fclose(ficres);
1.220 brouard 10996:
10997:
1.186 brouard 10998: /* Other results (useful)*/
1.220 brouard 10999:
11000:
1.126 brouard 11001: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11002: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11003: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11004: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11005: fclose(ficrespl);
11006:
11007: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11008: /*#include "hpijx.h"*/
11009: hPijx(p, bage, fage);
1.145 brouard 11010: fclose(ficrespij);
1.227 brouard 11011:
1.220 brouard 11012: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11013: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11014: k=1;
1.126 brouard 11015: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11016:
1.219 brouard 11017: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11018: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11019: for(i=1;i<=AGESUP;i++)
1.219 brouard 11020: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11021: for(k=1;k<=ncovcombmax;k++)
11022: probs[i][j][k]=0.;
1.219 brouard 11023: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11024: if (mobilav!=0 ||mobilavproj !=0 ) {
11025: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11026: for(i=1;i<=AGESUP;i++)
11027: for(j=1;j<=nlstate;j++)
11028: for(k=1;k<=ncovcombmax;k++)
11029: mobaverages[i][j][k]=0.;
1.219 brouard 11030: mobaverage=mobaverages;
11031: if (mobilav!=0) {
1.235 brouard 11032: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11033: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11034: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11035: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11036: }
1.219 brouard 11037: }
11038: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11039: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11040: else if (mobilavproj !=0) {
1.235 brouard 11041: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11042: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11043: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11044: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11045: }
1.219 brouard 11046: }
11047: }/* end if moving average */
1.227 brouard 11048:
1.126 brouard 11049: /*---------- Forecasting ------------------*/
11050: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11051: if(prevfcast==1){
11052: /* if(stepm ==1){*/
1.225 brouard 11053: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11054: }
1.217 brouard 11055: if(backcast==1){
1.219 brouard 11056: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11057: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11058: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11059:
11060: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11061:
11062: bprlim=matrix(1,nlstate,1,nlstate);
11063: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11064: fclose(ficresplb);
11065:
1.222 brouard 11066: hBijx(p, bage, fage, mobaverage);
11067: fclose(ficrespijb);
1.219 brouard 11068: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11069:
11070: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11071: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11072: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11073: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11074: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11075: }
1.217 brouard 11076:
1.186 brouard 11077:
11078: /* ------ Other prevalence ratios------------ */
1.126 brouard 11079:
1.215 brouard 11080: free_ivector(wav,1,imx);
11081: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11082: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11083: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11084:
11085:
1.127 brouard 11086: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11087:
1.201 brouard 11088: strcpy(filerese,"E_");
11089: strcat(filerese,fileresu);
1.126 brouard 11090: if((ficreseij=fopen(filerese,"w"))==NULL) {
11091: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11092: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11093: }
1.208 brouard 11094: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11095: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11096:
11097: pstamp(ficreseij);
1.219 brouard 11098:
1.235 brouard 11099: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11100: if (cptcovn < 1){i1=1;}
11101:
11102: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11103: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11104: if(TKresult[nres]!= k)
11105: continue;
1.219 brouard 11106: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11107: printf("\n#****** ");
1.225 brouard 11108: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11109: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11110: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11111: }
11112: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11113: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11114: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11115: }
11116: fprintf(ficreseij,"******\n");
1.235 brouard 11117: printf("******\n");
1.219 brouard 11118:
11119: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11120: oldm=oldms;savm=savms;
1.235 brouard 11121: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11122:
1.219 brouard 11123: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11124: }
11125: fclose(ficreseij);
1.208 brouard 11126: printf("done evsij\n");fflush(stdout);
11127: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11128:
1.227 brouard 11129: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11130:
11131:
1.201 brouard 11132: strcpy(filerest,"T_");
11133: strcat(filerest,fileresu);
1.127 brouard 11134: if((ficrest=fopen(filerest,"w"))==NULL) {
11135: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11136: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11137: }
1.208 brouard 11138: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11139: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11140:
1.126 brouard 11141:
1.201 brouard 11142: strcpy(fileresstde,"STDE_");
11143: strcat(fileresstde,fileresu);
1.126 brouard 11144: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11145: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11146: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11147: }
1.227 brouard 11148: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11149: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11150:
1.201 brouard 11151: strcpy(filerescve,"CVE_");
11152: strcat(filerescve,fileresu);
1.126 brouard 11153: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11154: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11155: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11156: }
1.227 brouard 11157: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11158: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11159:
1.201 brouard 11160: strcpy(fileresv,"V_");
11161: strcat(fileresv,fileresu);
1.126 brouard 11162: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11163: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11164: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11165: }
1.227 brouard 11166: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11167: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11168:
1.145 brouard 11169: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11170: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11171:
1.235 brouard 11172: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11173: if (cptcovn < 1){i1=1;}
11174:
11175: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11176: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11177: if(TKresult[nres]!= k)
11178: continue;
1.242 ! brouard 11179: printf("\n#****** Result for:");
! 11180: fprintf(ficrest,"\n#****** Result for:");
! 11181: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11182: for(j=1;j<=cptcoveff;j++){
11183: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11184: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11185: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11186: }
1.235 brouard 11187: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11188: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11189: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11190: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11191: }
1.208 brouard 11192: fprintf(ficrest,"******\n");
1.227 brouard 11193: fprintf(ficlog,"******\n");
11194: printf("******\n");
1.208 brouard 11195:
11196: fprintf(ficresstdeij,"\n#****** ");
11197: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11198: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11199: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11200: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11201: }
1.235 brouard 11202: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11203: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11204: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11205: }
1.208 brouard 11206: fprintf(ficresstdeij,"******\n");
11207: fprintf(ficrescveij,"******\n");
11208:
11209: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11210: /* pstamp(ficresvij); */
1.225 brouard 11211: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11212: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11213: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11214: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11215: }
1.208 brouard 11216: fprintf(ficresvij,"******\n");
11217:
11218: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11219: oldm=oldms;savm=savms;
1.235 brouard 11220: printf(" cvevsij ");
11221: fprintf(ficlog, " cvevsij ");
11222: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11223: printf(" end cvevsij \n ");
11224: fprintf(ficlog, " end cvevsij \n ");
11225:
11226: /*
11227: */
11228: /* goto endfree; */
11229:
11230: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11231: pstamp(ficrest);
11232:
11233:
11234: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11235: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11236: cptcod= 0; /* To be deleted */
11237: printf("varevsij vpopbased=%d \n",vpopbased);
11238: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11239: 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 11240: 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 ");
11241: if(vpopbased==1)
11242: 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);
11243: else
11244: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11245: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11246: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11247: fprintf(ficrest,"\n");
11248: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11249: epj=vector(1,nlstate+1);
11250: printf("Computing age specific period (stable) prevalences in each health state \n");
11251: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11252: for(age=bage; age <=fage ;age++){
1.235 brouard 11253: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11254: if (vpopbased==1) {
11255: if(mobilav ==0){
11256: for(i=1; i<=nlstate;i++)
11257: prlim[i][i]=probs[(int)age][i][k];
11258: }else{ /* mobilav */
11259: for(i=1; i<=nlstate;i++)
11260: prlim[i][i]=mobaverage[(int)age][i][k];
11261: }
11262: }
1.219 brouard 11263:
1.227 brouard 11264: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11265: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11266: /* printf(" age %4.0f ",age); */
11267: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11268: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11269: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11270: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11271: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11272: }
11273: epj[nlstate+1] +=epj[j];
11274: }
11275: /* printf(" age %4.0f \n",age); */
1.219 brouard 11276:
1.227 brouard 11277: for(i=1, vepp=0.;i <=nlstate;i++)
11278: for(j=1;j <=nlstate;j++)
11279: vepp += vareij[i][j][(int)age];
11280: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11281: for(j=1;j <=nlstate;j++){
11282: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11283: }
11284: fprintf(ficrest,"\n");
11285: }
1.208 brouard 11286: } /* End vpopbased */
11287: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11288: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11289: free_vector(epj,1,nlstate+1);
1.235 brouard 11290: printf("done selection\n");fflush(stdout);
11291: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11292:
1.145 brouard 11293: /*}*/
1.235 brouard 11294: } /* End k selection */
1.227 brouard 11295:
11296: printf("done State-specific expectancies\n");fflush(stdout);
11297: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11298:
1.126 brouard 11299: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11300:
1.201 brouard 11301: strcpy(fileresvpl,"VPL_");
11302: strcat(fileresvpl,fileresu);
1.126 brouard 11303: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11304: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11305: exit(0);
11306: }
1.208 brouard 11307: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11308: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11309:
1.145 brouard 11310: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11311: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11312:
1.235 brouard 11313: i1=pow(2,cptcoveff);
11314: if (cptcovn < 1){i1=1;}
11315:
11316: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11317: for(k=1; k<=i1;k++){
11318: if(TKresult[nres]!= k)
11319: continue;
1.227 brouard 11320: fprintf(ficresvpl,"\n#****** ");
11321: printf("\n#****** ");
11322: fprintf(ficlog,"\n#****** ");
11323: for(j=1;j<=cptcoveff;j++) {
11324: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11325: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11326: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11327: }
1.235 brouard 11328: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11329: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11330: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11331: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11332: }
1.227 brouard 11333: fprintf(ficresvpl,"******\n");
11334: printf("******\n");
11335: fprintf(ficlog,"******\n");
11336:
11337: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11338: oldm=oldms;savm=savms;
1.235 brouard 11339: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11340: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11341: /*}*/
1.126 brouard 11342: }
1.227 brouard 11343:
1.126 brouard 11344: fclose(ficresvpl);
1.208 brouard 11345: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11346: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11347:
11348: free_vector(weight,1,n);
11349: free_imatrix(Tvard,1,NCOVMAX,1,2);
11350: free_imatrix(s,1,maxwav+1,1,n);
11351: free_matrix(anint,1,maxwav,1,n);
11352: free_matrix(mint,1,maxwav,1,n);
11353: free_ivector(cod,1,n);
11354: free_ivector(tab,1,NCOVMAX);
11355: fclose(ficresstdeij);
11356: fclose(ficrescveij);
11357: fclose(ficresvij);
11358: fclose(ficrest);
11359: fclose(ficpar);
11360:
11361:
1.126 brouard 11362: /*---------- End : free ----------------*/
1.219 brouard 11363: if (mobilav!=0 ||mobilavproj !=0)
11364: 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 11365: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11366: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11367: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11368: } /* mle==-3 arrives here for freeing */
1.227 brouard 11369: /* endfree:*/
11370: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11371: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11372: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11373: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11374: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11375: free_matrix(coqvar,1,maxwav,1,n);
11376: free_matrix(covar,0,NCOVMAX,1,n);
11377: free_matrix(matcov,1,npar,1,npar);
11378: free_matrix(hess,1,npar,1,npar);
11379: /*free_vector(delti,1,npar);*/
11380: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11381: free_matrix(agev,1,maxwav,1,imx);
11382: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11383:
11384: free_ivector(ncodemax,1,NCOVMAX);
11385: free_ivector(ncodemaxwundef,1,NCOVMAX);
11386: free_ivector(Dummy,-1,NCOVMAX);
11387: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11388: free_ivector(DummyV,1,NCOVMAX);
11389: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11390: free_ivector(Typevar,-1,NCOVMAX);
11391: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11392: free_ivector(TvarsQ,1,NCOVMAX);
11393: free_ivector(TvarsQind,1,NCOVMAX);
11394: free_ivector(TvarsD,1,NCOVMAX);
11395: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11396: free_ivector(TvarFD,1,NCOVMAX);
11397: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11398: free_ivector(TvarF,1,NCOVMAX);
11399: free_ivector(TvarFind,1,NCOVMAX);
11400: free_ivector(TvarV,1,NCOVMAX);
11401: free_ivector(TvarVind,1,NCOVMAX);
11402: free_ivector(TvarA,1,NCOVMAX);
11403: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11404: free_ivector(TvarFQ,1,NCOVMAX);
11405: free_ivector(TvarFQind,1,NCOVMAX);
11406: free_ivector(TvarVD,1,NCOVMAX);
11407: free_ivector(TvarVDind,1,NCOVMAX);
11408: free_ivector(TvarVQ,1,NCOVMAX);
11409: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11410: free_ivector(Tvarsel,1,NCOVMAX);
11411: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11412: free_ivector(Tposprod,1,NCOVMAX);
11413: free_ivector(Tprod,1,NCOVMAX);
11414: free_ivector(Tvaraff,1,NCOVMAX);
11415: free_ivector(invalidvarcomb,1,ncovcombmax);
11416: free_ivector(Tage,1,NCOVMAX);
11417: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11418: free_ivector(TmodelInvind,1,NCOVMAX);
11419: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11420:
11421: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11422: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11423: fflush(fichtm);
11424: fflush(ficgp);
11425:
1.227 brouard 11426:
1.126 brouard 11427: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11428: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11429: 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 11430: }else{
11431: printf("End of Imach\n");
11432: fprintf(ficlog,"End of Imach\n");
11433: }
11434: printf("See log file on %s\n",filelog);
11435: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11436: /*(void) gettimeofday(&end_time,&tzp);*/
11437: rend_time = time(NULL);
11438: end_time = *localtime(&rend_time);
11439: /* tml = *localtime(&end_time.tm_sec); */
11440: strcpy(strtend,asctime(&end_time));
1.126 brouard 11441: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11442: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11443: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11444:
1.157 brouard 11445: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11446: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11447: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11448: /* printf("Total time was %d uSec.\n", total_usecs);*/
11449: /* if(fileappend(fichtm,optionfilehtm)){ */
11450: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11451: fclose(fichtm);
11452: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11453: fclose(fichtmcov);
11454: fclose(ficgp);
11455: fclose(ficlog);
11456: /*------ End -----------*/
1.227 brouard 11457:
11458:
11459: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11460: #ifdef WIN32
1.227 brouard 11461: if (_chdir(pathcd) != 0)
11462: printf("Can't move to directory %s!\n",path);
11463: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11464: #else
1.227 brouard 11465: if(chdir(pathcd) != 0)
11466: printf("Can't move to directory %s!\n", path);
11467: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11468: #endif
1.126 brouard 11469: printf("Current directory %s!\n",pathcd);
11470: /*strcat(plotcmd,CHARSEPARATOR);*/
11471: sprintf(plotcmd,"gnuplot");
1.157 brouard 11472: #ifdef _WIN32
1.126 brouard 11473: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11474: #endif
11475: if(!stat(plotcmd,&info)){
1.158 brouard 11476: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11477: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11478: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11479: }else
11480: strcpy(pplotcmd,plotcmd);
1.157 brouard 11481: #ifdef __unix
1.126 brouard 11482: strcpy(plotcmd,GNUPLOTPROGRAM);
11483: if(!stat(plotcmd,&info)){
1.158 brouard 11484: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11485: }else
11486: strcpy(pplotcmd,plotcmd);
11487: #endif
11488: }else
11489: strcpy(pplotcmd,plotcmd);
11490:
11491: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11492: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11493:
1.126 brouard 11494: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11495: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11496: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11497: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11498: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11499: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11500: }
1.158 brouard 11501: printf(" Successful, please wait...");
1.126 brouard 11502: while (z[0] != 'q') {
11503: /* chdir(path); */
1.154 brouard 11504: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11505: scanf("%s",z);
11506: /* if (z[0] == 'c') system("./imach"); */
11507: if (z[0] == 'e') {
1.158 brouard 11508: #ifdef __APPLE__
1.152 brouard 11509: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11510: #elif __linux
11511: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11512: #else
1.152 brouard 11513: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11514: #endif
11515: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11516: system(pplotcmd);
1.126 brouard 11517: }
11518: else if (z[0] == 'g') system(plotcmd);
11519: else if (z[0] == 'q') exit(0);
11520: }
1.227 brouard 11521: end:
1.126 brouard 11522: while (z[0] != 'q') {
1.195 brouard 11523: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11524: scanf("%s",z);
11525: }
11526: }
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