Annotation of imach/src/imach.c, revision 1.252
1.252 ! brouard 1: /* $Id: imach.c,v 1.251 2016/09/15 15:01:13 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.252 ! brouard 4: Revision 1.251 2016/09/15 15:01:13 brouard
! 5: Summary: not working
! 6:
1.251 brouard 7: Revision 1.250 2016/09/08 16:07:27 brouard
8: Summary: continue
9:
1.250 brouard 10: Revision 1.249 2016/09/07 17:14:18 brouard
11: Summary: Starting values from frequencies
12:
1.249 brouard 13: Revision 1.248 2016/09/07 14:10:18 brouard
14: *** empty log message ***
15:
1.248 brouard 16: Revision 1.247 2016/09/02 11:11:21 brouard
17: *** empty log message ***
18:
1.247 brouard 19: Revision 1.246 2016/09/02 08:49:22 brouard
20: *** empty log message ***
21:
1.246 brouard 22: Revision 1.245 2016/09/02 07:25:01 brouard
23: *** empty log message ***
24:
1.245 brouard 25: Revision 1.244 2016/09/02 07:17:34 brouard
26: *** empty log message ***
27:
1.244 brouard 28: Revision 1.243 2016/09/02 06:45:35 brouard
29: *** empty log message ***
30:
1.243 brouard 31: Revision 1.242 2016/08/30 15:01:20 brouard
32: Summary: Fixing a lots
33:
1.242 brouard 34: Revision 1.241 2016/08/29 17:17:25 brouard
35: Summary: gnuplot problem in Back projection to fix
36:
1.241 brouard 37: Revision 1.240 2016/08/29 07:53:18 brouard
38: Summary: Better
39:
1.240 brouard 40: Revision 1.239 2016/08/26 15:51:03 brouard
41: Summary: Improvement in Powell output in order to copy and paste
42:
43: Author:
44:
1.239 brouard 45: Revision 1.238 2016/08/26 14:23:35 brouard
46: Summary: Starting tests of 0.99
47:
1.238 brouard 48: Revision 1.237 2016/08/26 09:20:19 brouard
49: Summary: to valgrind
50:
1.237 brouard 51: Revision 1.236 2016/08/25 10:50:18 brouard
52: *** empty log message ***
53:
1.236 brouard 54: Revision 1.235 2016/08/25 06:59:23 brouard
55: *** empty log message ***
56:
1.235 brouard 57: Revision 1.234 2016/08/23 16:51:20 brouard
58: *** empty log message ***
59:
1.234 brouard 60: Revision 1.233 2016/08/23 07:40:50 brouard
61: Summary: not working
62:
1.233 brouard 63: Revision 1.232 2016/08/22 14:20:21 brouard
64: Summary: not working
65:
1.232 brouard 66: Revision 1.231 2016/08/22 07:17:15 brouard
67: Summary: not working
68:
1.231 brouard 69: Revision 1.230 2016/08/22 06:55:53 brouard
70: Summary: Not working
71:
1.230 brouard 72: Revision 1.229 2016/07/23 09:45:53 brouard
73: Summary: Completing for func too
74:
1.229 brouard 75: Revision 1.228 2016/07/22 17:45:30 brouard
76: Summary: Fixing some arrays, still debugging
77:
1.227 brouard 78: Revision 1.226 2016/07/12 18:42:34 brouard
79: Summary: temp
80:
1.226 brouard 81: Revision 1.225 2016/07/12 08:40:03 brouard
82: Summary: saving but not running
83:
1.225 brouard 84: Revision 1.224 2016/07/01 13:16:01 brouard
85: Summary: Fixes
86:
1.224 brouard 87: Revision 1.223 2016/02/19 09:23:35 brouard
88: Summary: temporary
89:
1.223 brouard 90: Revision 1.222 2016/02/17 08:14:50 brouard
91: Summary: Probably last 0.98 stable version 0.98r6
92:
1.222 brouard 93: Revision 1.221 2016/02/15 23:35:36 brouard
94: Summary: minor bug
95:
1.220 brouard 96: Revision 1.219 2016/02/15 00:48:12 brouard
97: *** empty log message ***
98:
1.219 brouard 99: Revision 1.218 2016/02/12 11:29:23 brouard
100: Summary: 0.99 Back projections
101:
1.218 brouard 102: Revision 1.217 2015/12/23 17:18:31 brouard
103: Summary: Experimental backcast
104:
1.217 brouard 105: Revision 1.216 2015/12/18 17:32:11 brouard
106: Summary: 0.98r4 Warning and status=-2
107:
108: Version 0.98r4 is now:
109: - displaying an error when status is -1, date of interview unknown and date of death known;
110: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
111: Older changes concerning s=-2, dating from 2005 have been supersed.
112:
1.216 brouard 113: Revision 1.215 2015/12/16 08:52:24 brouard
114: Summary: 0.98r4 working
115:
1.215 brouard 116: Revision 1.214 2015/12/16 06:57:54 brouard
117: Summary: temporary not working
118:
1.214 brouard 119: Revision 1.213 2015/12/11 18:22:17 brouard
120: Summary: 0.98r4
121:
1.213 brouard 122: Revision 1.212 2015/11/21 12:47:24 brouard
123: Summary: minor typo
124:
1.212 brouard 125: Revision 1.211 2015/11/21 12:41:11 brouard
126: Summary: 0.98r3 with some graph of projected cross-sectional
127:
128: Author: Nicolas Brouard
129:
1.211 brouard 130: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 ! brouard 131: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 132: Summary: Adding ftolpl parameter
133: Author: N Brouard
134:
135: We had difficulties to get smoothed confidence intervals. It was due
136: to the period prevalence which wasn't computed accurately. The inner
137: parameter ftolpl is now an outer parameter of the .imach parameter
138: file after estepm. If ftolpl is small 1.e-4 and estepm too,
139: computation are long.
140:
1.209 brouard 141: Revision 1.208 2015/11/17 14:31:57 brouard
142: Summary: temporary
143:
1.208 brouard 144: Revision 1.207 2015/10/27 17:36:57 brouard
145: *** empty log message ***
146:
1.207 brouard 147: Revision 1.206 2015/10/24 07:14:11 brouard
148: *** empty log message ***
149:
1.206 brouard 150: Revision 1.205 2015/10/23 15:50:53 brouard
151: Summary: 0.98r3 some clarification for graphs on likelihood contributions
152:
1.205 brouard 153: Revision 1.204 2015/10/01 16:20:26 brouard
154: Summary: Some new graphs of contribution to likelihood
155:
1.204 brouard 156: Revision 1.203 2015/09/30 17:45:14 brouard
157: Summary: looking at better estimation of the hessian
158:
159: Also a better criteria for convergence to the period prevalence And
160: therefore adding the number of years needed to converge. (The
161: prevalence in any alive state shold sum to one
162:
1.203 brouard 163: Revision 1.202 2015/09/22 19:45:16 brouard
164: Summary: Adding some overall graph on contribution to likelihood. Might change
165:
1.202 brouard 166: Revision 1.201 2015/09/15 17:34:58 brouard
167: Summary: 0.98r0
168:
169: - Some new graphs like suvival functions
170: - Some bugs fixed like model=1+age+V2.
171:
1.201 brouard 172: Revision 1.200 2015/09/09 16:53:55 brouard
173: Summary: Big bug thanks to Flavia
174:
175: Even model=1+age+V2. did not work anymore
176:
1.200 brouard 177: Revision 1.199 2015/09/07 14:09:23 brouard
178: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
179:
1.199 brouard 180: Revision 1.198 2015/09/03 07:14:39 brouard
181: Summary: 0.98q5 Flavia
182:
1.198 brouard 183: Revision 1.197 2015/09/01 18:24:39 brouard
184: *** empty log message ***
185:
1.197 brouard 186: Revision 1.196 2015/08/18 23:17:52 brouard
187: Summary: 0.98q5
188:
1.196 brouard 189: Revision 1.195 2015/08/18 16:28:39 brouard
190: Summary: Adding a hack for testing purpose
191:
192: After reading the title, ftol and model lines, if the comment line has
193: a q, starting with #q, the answer at the end of the run is quit. It
194: permits to run test files in batch with ctest. The former workaround was
195: $ echo q | imach foo.imach
196:
1.195 brouard 197: Revision 1.194 2015/08/18 13:32:00 brouard
198: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
199:
1.194 brouard 200: Revision 1.193 2015/08/04 07:17:42 brouard
201: Summary: 0.98q4
202:
1.193 brouard 203: Revision 1.192 2015/07/16 16:49:02 brouard
204: Summary: Fixing some outputs
205:
1.192 brouard 206: Revision 1.191 2015/07/14 10:00:33 brouard
207: Summary: Some fixes
208:
1.191 brouard 209: Revision 1.190 2015/05/05 08:51:13 brouard
210: Summary: Adding digits in output parameters (7 digits instead of 6)
211:
212: Fix 1+age+.
213:
1.190 brouard 214: Revision 1.189 2015/04/30 14:45:16 brouard
215: Summary: 0.98q2
216:
1.189 brouard 217: Revision 1.188 2015/04/30 08:27:53 brouard
218: *** empty log message ***
219:
1.188 brouard 220: Revision 1.187 2015/04/29 09:11:15 brouard
221: *** empty log message ***
222:
1.187 brouard 223: Revision 1.186 2015/04/23 12:01:52 brouard
224: Summary: V1*age is working now, version 0.98q1
225:
226: Some codes had been disabled in order to simplify and Vn*age was
227: working in the optimization phase, ie, giving correct MLE parameters,
228: but, as usual, outputs were not correct and program core dumped.
229:
1.186 brouard 230: Revision 1.185 2015/03/11 13:26:42 brouard
231: Summary: Inclusion of compile and links command line for Intel Compiler
232:
1.185 brouard 233: Revision 1.184 2015/03/11 11:52:39 brouard
234: Summary: Back from Windows 8. Intel Compiler
235:
1.184 brouard 236: Revision 1.183 2015/03/10 20:34:32 brouard
237: Summary: 0.98q0, trying with directest, mnbrak fixed
238:
239: We use directest instead of original Powell test; probably no
240: incidence on the results, but better justifications;
241: We fixed Numerical Recipes mnbrak routine which was wrong and gave
242: wrong results.
243:
1.183 brouard 244: Revision 1.182 2015/02/12 08:19:57 brouard
245: Summary: Trying to keep directest which seems simpler and more general
246: Author: Nicolas Brouard
247:
1.182 brouard 248: Revision 1.181 2015/02/11 23:22:24 brouard
249: Summary: Comments on Powell added
250:
251: Author:
252:
1.181 brouard 253: Revision 1.180 2015/02/11 17:33:45 brouard
254: Summary: Finishing move from main to function (hpijx and prevalence_limit)
255:
1.180 brouard 256: Revision 1.179 2015/01/04 09:57:06 brouard
257: Summary: back to OS/X
258:
1.179 brouard 259: Revision 1.178 2015/01/04 09:35:48 brouard
260: *** empty log message ***
261:
1.178 brouard 262: Revision 1.177 2015/01/03 18:40:56 brouard
263: Summary: Still testing ilc32 on OSX
264:
1.177 brouard 265: Revision 1.176 2015/01/03 16:45:04 brouard
266: *** empty log message ***
267:
1.176 brouard 268: Revision 1.175 2015/01/03 16:33:42 brouard
269: *** empty log message ***
270:
1.175 brouard 271: Revision 1.174 2015/01/03 16:15:49 brouard
272: Summary: Still in cross-compilation
273:
1.174 brouard 274: Revision 1.173 2015/01/03 12:06:26 brouard
275: Summary: trying to detect cross-compilation
276:
1.173 brouard 277: Revision 1.172 2014/12/27 12:07:47 brouard
278: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
279:
1.172 brouard 280: Revision 1.171 2014/12/23 13:26:59 brouard
281: Summary: Back from Visual C
282:
283: Still problem with utsname.h on Windows
284:
1.171 brouard 285: Revision 1.170 2014/12/23 11:17:12 brouard
286: Summary: Cleaning some \%% back to %%
287:
288: The escape was mandatory for a specific compiler (which one?), but too many warnings.
289:
1.170 brouard 290: Revision 1.169 2014/12/22 23:08:31 brouard
291: Summary: 0.98p
292:
293: Outputs some informations on compiler used, OS etc. Testing on different platforms.
294:
1.169 brouard 295: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 296: Summary: update
1.169 brouard 297:
1.168 brouard 298: Revision 1.167 2014/12/22 13:50:56 brouard
299: Summary: Testing uname and compiler version and if compiled 32 or 64
300:
301: Testing on Linux 64
302:
1.167 brouard 303: Revision 1.166 2014/12/22 11:40:47 brouard
304: *** empty log message ***
305:
1.166 brouard 306: Revision 1.165 2014/12/16 11:20:36 brouard
307: Summary: After compiling on Visual C
308:
309: * imach.c (Module): Merging 1.61 to 1.162
310:
1.165 brouard 311: Revision 1.164 2014/12/16 10:52:11 brouard
312: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
313:
314: * imach.c (Module): Merging 1.61 to 1.162
315:
1.164 brouard 316: Revision 1.163 2014/12/16 10:30:11 brouard
317: * imach.c (Module): Merging 1.61 to 1.162
318:
1.163 brouard 319: Revision 1.162 2014/09/25 11:43:39 brouard
320: Summary: temporary backup 0.99!
321:
1.162 brouard 322: Revision 1.1 2014/09/16 11:06:58 brouard
323: Summary: With some code (wrong) for nlopt
324:
325: Author:
326:
327: Revision 1.161 2014/09/15 20:41:41 brouard
328: Summary: Problem with macro SQR on Intel compiler
329:
1.161 brouard 330: Revision 1.160 2014/09/02 09:24:05 brouard
331: *** empty log message ***
332:
1.160 brouard 333: Revision 1.159 2014/09/01 10:34:10 brouard
334: Summary: WIN32
335: Author: Brouard
336:
1.159 brouard 337: Revision 1.158 2014/08/27 17:11:51 brouard
338: *** empty log message ***
339:
1.158 brouard 340: Revision 1.157 2014/08/27 16:26:55 brouard
341: Summary: Preparing windows Visual studio version
342: Author: Brouard
343:
344: In order to compile on Visual studio, time.h is now correct and time_t
345: and tm struct should be used. difftime should be used but sometimes I
346: just make the differences in raw time format (time(&now).
347: Trying to suppress #ifdef LINUX
348: Add xdg-open for __linux in order to open default browser.
349:
1.157 brouard 350: Revision 1.156 2014/08/25 20:10:10 brouard
351: *** empty log message ***
352:
1.156 brouard 353: Revision 1.155 2014/08/25 18:32:34 brouard
354: Summary: New compile, minor changes
355: Author: Brouard
356:
1.155 brouard 357: Revision 1.154 2014/06/20 17:32:08 brouard
358: Summary: Outputs now all graphs of convergence to period prevalence
359:
1.154 brouard 360: Revision 1.153 2014/06/20 16:45:46 brouard
361: Summary: If 3 live state, convergence to period prevalence on same graph
362: Author: Brouard
363:
1.153 brouard 364: Revision 1.152 2014/06/18 17:54:09 brouard
365: Summary: open browser, use gnuplot on same dir than imach if not found in the path
366:
1.152 brouard 367: Revision 1.151 2014/06/18 16:43:30 brouard
368: *** empty log message ***
369:
1.151 brouard 370: Revision 1.150 2014/06/18 16:42:35 brouard
371: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
372: Author: brouard
373:
1.150 brouard 374: Revision 1.149 2014/06/18 15:51:14 brouard
375: Summary: Some fixes in parameter files errors
376: Author: Nicolas Brouard
377:
1.149 brouard 378: Revision 1.148 2014/06/17 17:38:48 brouard
379: Summary: Nothing new
380: Author: Brouard
381:
382: Just a new packaging for OS/X version 0.98nS
383:
1.148 brouard 384: Revision 1.147 2014/06/16 10:33:11 brouard
385: *** empty log message ***
386:
1.147 brouard 387: Revision 1.146 2014/06/16 10:20:28 brouard
388: Summary: Merge
389: Author: Brouard
390:
391: Merge, before building revised version.
392:
1.146 brouard 393: Revision 1.145 2014/06/10 21:23:15 brouard
394: Summary: Debugging with valgrind
395: Author: Nicolas Brouard
396:
397: Lot of changes in order to output the results with some covariates
398: After the Edimburgh REVES conference 2014, it seems mandatory to
399: improve the code.
400: No more memory valgrind error but a lot has to be done in order to
401: continue the work of splitting the code into subroutines.
402: Also, decodemodel has been improved. Tricode is still not
403: optimal. nbcode should be improved. Documentation has been added in
404: the source code.
405:
1.144 brouard 406: Revision 1.143 2014/01/26 09:45:38 brouard
407: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
408:
409: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
410: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
411:
1.143 brouard 412: Revision 1.142 2014/01/26 03:57:36 brouard
413: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
414:
415: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
416:
1.142 brouard 417: Revision 1.141 2014/01/26 02:42:01 brouard
418: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
419:
1.141 brouard 420: Revision 1.140 2011/09/02 10:37:54 brouard
421: Summary: times.h is ok with mingw32 now.
422:
1.140 brouard 423: Revision 1.139 2010/06/14 07:50:17 brouard
424: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
425: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
426:
1.139 brouard 427: Revision 1.138 2010/04/30 18:19:40 brouard
428: *** empty log message ***
429:
1.138 brouard 430: Revision 1.137 2010/04/29 18:11:38 brouard
431: (Module): Checking covariates for more complex models
432: than V1+V2. A lot of change to be done. Unstable.
433:
1.137 brouard 434: Revision 1.136 2010/04/26 20:30:53 brouard
435: (Module): merging some libgsl code. Fixing computation
436: of likelione (using inter/intrapolation if mle = 0) in order to
437: get same likelihood as if mle=1.
438: Some cleaning of code and comments added.
439:
1.136 brouard 440: Revision 1.135 2009/10/29 15:33:14 brouard
441: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
442:
1.135 brouard 443: Revision 1.134 2009/10/29 13:18:53 brouard
444: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
445:
1.134 brouard 446: Revision 1.133 2009/07/06 10:21:25 brouard
447: just nforces
448:
1.133 brouard 449: Revision 1.132 2009/07/06 08:22:05 brouard
450: Many tings
451:
1.132 brouard 452: Revision 1.131 2009/06/20 16:22:47 brouard
453: Some dimensions resccaled
454:
1.131 brouard 455: Revision 1.130 2009/05/26 06:44:34 brouard
456: (Module): Max Covariate is now set to 20 instead of 8. A
457: lot of cleaning with variables initialized to 0. Trying to make
458: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
459:
1.130 brouard 460: Revision 1.129 2007/08/31 13:49:27 lievre
461: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
462:
1.129 lievre 463: Revision 1.128 2006/06/30 13:02:05 brouard
464: (Module): Clarifications on computing e.j
465:
1.128 brouard 466: Revision 1.127 2006/04/28 18:11:50 brouard
467: (Module): Yes the sum of survivors was wrong since
468: imach-114 because nhstepm was no more computed in the age
469: loop. Now we define nhstepma in the age loop.
470: (Module): In order to speed up (in case of numerous covariates) we
471: compute health expectancies (without variances) in a first step
472: and then all the health expectancies with variances or standard
473: deviation (needs data from the Hessian matrices) which slows the
474: computation.
475: In the future we should be able to stop the program is only health
476: expectancies and graph are needed without standard deviations.
477:
1.127 brouard 478: Revision 1.126 2006/04/28 17:23:28 brouard
479: (Module): Yes the sum of survivors was wrong since
480: imach-114 because nhstepm was no more computed in the age
481: loop. Now we define nhstepma in the age loop.
482: Version 0.98h
483:
1.126 brouard 484: Revision 1.125 2006/04/04 15:20:31 lievre
485: Errors in calculation of health expectancies. Age was not initialized.
486: Forecasting file added.
487:
488: Revision 1.124 2006/03/22 17:13:53 lievre
489: Parameters are printed with %lf instead of %f (more numbers after the comma).
490: The log-likelihood is printed in the log file
491:
492: Revision 1.123 2006/03/20 10:52:43 brouard
493: * imach.c (Module): <title> changed, corresponds to .htm file
494: name. <head> headers where missing.
495:
496: * imach.c (Module): Weights can have a decimal point as for
497: English (a comma might work with a correct LC_NUMERIC environment,
498: otherwise the weight is truncated).
499: Modification of warning when the covariates values are not 0 or
500: 1.
501: Version 0.98g
502:
503: Revision 1.122 2006/03/20 09:45:41 brouard
504: (Module): Weights can have a decimal point as for
505: English (a comma might work with a correct LC_NUMERIC environment,
506: otherwise the weight is truncated).
507: Modification of warning when the covariates values are not 0 or
508: 1.
509: Version 0.98g
510:
511: Revision 1.121 2006/03/16 17:45:01 lievre
512: * imach.c (Module): Comments concerning covariates added
513:
514: * imach.c (Module): refinements in the computation of lli if
515: status=-2 in order to have more reliable computation if stepm is
516: not 1 month. Version 0.98f
517:
518: Revision 1.120 2006/03/16 15:10:38 lievre
519: (Module): refinements in the computation of lli if
520: status=-2 in order to have more reliable computation if stepm is
521: not 1 month. Version 0.98f
522:
523: Revision 1.119 2006/03/15 17:42:26 brouard
524: (Module): Bug if status = -2, the loglikelihood was
525: computed as likelihood omitting the logarithm. Version O.98e
526:
527: Revision 1.118 2006/03/14 18:20:07 brouard
528: (Module): varevsij Comments added explaining the second
529: table of variances if popbased=1 .
530: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
531: (Module): Function pstamp added
532: (Module): Version 0.98d
533:
534: Revision 1.117 2006/03/14 17:16:22 brouard
535: (Module): varevsij Comments added explaining the second
536: table of variances if popbased=1 .
537: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
538: (Module): Function pstamp added
539: (Module): Version 0.98d
540:
541: Revision 1.116 2006/03/06 10:29:27 brouard
542: (Module): Variance-covariance wrong links and
543: varian-covariance of ej. is needed (Saito).
544:
545: Revision 1.115 2006/02/27 12:17:45 brouard
546: (Module): One freematrix added in mlikeli! 0.98c
547:
548: Revision 1.114 2006/02/26 12:57:58 brouard
549: (Module): Some improvements in processing parameter
550: filename with strsep.
551:
552: Revision 1.113 2006/02/24 14:20:24 brouard
553: (Module): Memory leaks checks with valgrind and:
554: datafile was not closed, some imatrix were not freed and on matrix
555: allocation too.
556:
557: Revision 1.112 2006/01/30 09:55:26 brouard
558: (Module): Back to gnuplot.exe instead of wgnuplot.exe
559:
560: Revision 1.111 2006/01/25 20:38:18 brouard
561: (Module): Lots of cleaning and bugs added (Gompertz)
562: (Module): Comments can be added in data file. Missing date values
563: can be a simple dot '.'.
564:
565: Revision 1.110 2006/01/25 00:51:50 brouard
566: (Module): Lots of cleaning and bugs added (Gompertz)
567:
568: Revision 1.109 2006/01/24 19:37:15 brouard
569: (Module): Comments (lines starting with a #) are allowed in data.
570:
571: Revision 1.108 2006/01/19 18:05:42 lievre
572: Gnuplot problem appeared...
573: To be fixed
574:
575: Revision 1.107 2006/01/19 16:20:37 brouard
576: Test existence of gnuplot in imach path
577:
578: Revision 1.106 2006/01/19 13:24:36 brouard
579: Some cleaning and links added in html output
580:
581: Revision 1.105 2006/01/05 20:23:19 lievre
582: *** empty log message ***
583:
584: Revision 1.104 2005/09/30 16:11:43 lievre
585: (Module): sump fixed, loop imx fixed, and simplifications.
586: (Module): If the status is missing at the last wave but we know
587: that the person is alive, then we can code his/her status as -2
588: (instead of missing=-1 in earlier versions) and his/her
589: contributions to the likelihood is 1 - Prob of dying from last
590: health status (= 1-p13= p11+p12 in the easiest case of somebody in
591: the healthy state at last known wave). Version is 0.98
592:
593: Revision 1.103 2005/09/30 15:54:49 lievre
594: (Module): sump fixed, loop imx fixed, and simplifications.
595:
596: Revision 1.102 2004/09/15 17:31:30 brouard
597: Add the possibility to read data file including tab characters.
598:
599: Revision 1.101 2004/09/15 10:38:38 brouard
600: Fix on curr_time
601:
602: Revision 1.100 2004/07/12 18:29:06 brouard
603: Add version for Mac OS X. Just define UNIX in Makefile
604:
605: Revision 1.99 2004/06/05 08:57:40 brouard
606: *** empty log message ***
607:
608: Revision 1.98 2004/05/16 15:05:56 brouard
609: New version 0.97 . First attempt to estimate force of mortality
610: directly from the data i.e. without the need of knowing the health
611: state at each age, but using a Gompertz model: log u =a + b*age .
612: This is the basic analysis of mortality and should be done before any
613: other analysis, in order to test if the mortality estimated from the
614: cross-longitudinal survey is different from the mortality estimated
615: from other sources like vital statistic data.
616:
617: The same imach parameter file can be used but the option for mle should be -3.
618:
1.133 brouard 619: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 620: former routines in order to include the new code within the former code.
621:
622: The output is very simple: only an estimate of the intercept and of
623: the slope with 95% confident intervals.
624:
625: Current limitations:
626: A) Even if you enter covariates, i.e. with the
627: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
628: B) There is no computation of Life Expectancy nor Life Table.
629:
630: Revision 1.97 2004/02/20 13:25:42 lievre
631: Version 0.96d. Population forecasting command line is (temporarily)
632: suppressed.
633:
634: Revision 1.96 2003/07/15 15:38:55 brouard
635: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
636: rewritten within the same printf. Workaround: many printfs.
637:
638: Revision 1.95 2003/07/08 07:54:34 brouard
639: * imach.c (Repository):
640: (Repository): Using imachwizard code to output a more meaningful covariance
641: matrix (cov(a12,c31) instead of numbers.
642:
643: Revision 1.94 2003/06/27 13:00:02 brouard
644: Just cleaning
645:
646: Revision 1.93 2003/06/25 16:33:55 brouard
647: (Module): On windows (cygwin) function asctime_r doesn't
648: exist so I changed back to asctime which exists.
649: (Module): Version 0.96b
650:
651: Revision 1.92 2003/06/25 16:30:45 brouard
652: (Module): On windows (cygwin) function asctime_r doesn't
653: exist so I changed back to asctime which exists.
654:
655: Revision 1.91 2003/06/25 15:30:29 brouard
656: * imach.c (Repository): Duplicated warning errors corrected.
657: (Repository): Elapsed time after each iteration is now output. It
658: helps to forecast when convergence will be reached. Elapsed time
659: is stamped in powell. We created a new html file for the graphs
660: concerning matrix of covariance. It has extension -cov.htm.
661:
662: Revision 1.90 2003/06/24 12:34:15 brouard
663: (Module): Some bugs corrected for windows. Also, when
664: mle=-1 a template is output in file "or"mypar.txt with the design
665: of the covariance matrix to be input.
666:
667: Revision 1.89 2003/06/24 12:30:52 brouard
668: (Module): Some bugs corrected for windows. Also, when
669: mle=-1 a template is output in file "or"mypar.txt with the design
670: of the covariance matrix to be input.
671:
672: Revision 1.88 2003/06/23 17:54:56 brouard
673: * 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.
674:
675: Revision 1.87 2003/06/18 12:26:01 brouard
676: Version 0.96
677:
678: Revision 1.86 2003/06/17 20:04:08 brouard
679: (Module): Change position of html and gnuplot routines and added
680: routine fileappend.
681:
682: Revision 1.85 2003/06/17 13:12:43 brouard
683: * imach.c (Repository): Check when date of death was earlier that
684: current date of interview. It may happen when the death was just
685: prior to the death. In this case, dh was negative and likelihood
686: was wrong (infinity). We still send an "Error" but patch by
687: assuming that the date of death was just one stepm after the
688: interview.
689: (Repository): Because some people have very long ID (first column)
690: we changed int to long in num[] and we added a new lvector for
691: memory allocation. But we also truncated to 8 characters (left
692: truncation)
693: (Repository): No more line truncation errors.
694:
695: Revision 1.84 2003/06/13 21:44:43 brouard
696: * imach.c (Repository): Replace "freqsummary" at a correct
697: place. It differs from routine "prevalence" which may be called
698: many times. Probs is memory consuming and must be used with
699: parcimony.
700: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
701:
702: Revision 1.83 2003/06/10 13:39:11 lievre
703: *** empty log message ***
704:
705: Revision 1.82 2003/06/05 15:57:20 brouard
706: Add log in imach.c and fullversion number is now printed.
707:
708: */
709: /*
710: Interpolated Markov Chain
711:
712: Short summary of the programme:
713:
1.227 brouard 714: This program computes Healthy Life Expectancies or State-specific
715: (if states aren't health statuses) Expectancies from
716: cross-longitudinal data. Cross-longitudinal data consist in:
717:
718: -1- a first survey ("cross") where individuals from different ages
719: are interviewed on their health status or degree of disability (in
720: the case of a health survey which is our main interest)
721:
722: -2- at least a second wave of interviews ("longitudinal") which
723: measure each change (if any) in individual health status. Health
724: expectancies are computed from the time spent in each health state
725: according to a model. More health states you consider, more time is
726: necessary to reach the Maximum Likelihood of the parameters involved
727: in the model. The simplest model is the multinomial logistic model
728: where pij is the probability to be observed in state j at the second
729: wave conditional to be observed in state i at the first
730: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
731: etc , where 'age' is age and 'sex' is a covariate. If you want to
732: have a more complex model than "constant and age", you should modify
733: the program where the markup *Covariates have to be included here
734: again* invites you to do it. More covariates you add, slower the
1.126 brouard 735: convergence.
736:
737: The advantage of this computer programme, compared to a simple
738: multinomial logistic model, is clear when the delay between waves is not
739: identical for each individual. Also, if a individual missed an
740: intermediate interview, the information is lost, but taken into
741: account using an interpolation or extrapolation.
742:
743: hPijx is the probability to be observed in state i at age x+h
744: conditional to the observed state i at age x. The delay 'h' can be
745: split into an exact number (nh*stepm) of unobserved intermediate
746: states. This elementary transition (by month, quarter,
747: semester or year) is modelled as a multinomial logistic. The hPx
748: matrix is simply the matrix product of nh*stepm elementary matrices
749: and the contribution of each individual to the likelihood is simply
750: hPijx.
751:
752: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 753: of the life expectancies. It also computes the period (stable) prevalence.
754:
755: Back prevalence and projections:
1.227 brouard 756:
757: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
758: double agemaxpar, double ftolpl, int *ncvyearp, double
759: dateprev1,double dateprev2, int firstpass, int lastpass, int
760: mobilavproj)
761:
762: Computes the back prevalence limit for any combination of
763: covariate values k at any age between ageminpar and agemaxpar and
764: returns it in **bprlim. In the loops,
765:
766: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
767: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
768:
769: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 770: Computes for any combination of covariates k and any age between bage and fage
771: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
772: oldm=oldms;savm=savms;
1.227 brouard 773:
774: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 775: Computes the transition matrix starting at age 'age' over
776: 'nhstepm*hstepm*stepm' months (i.e. until
777: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 778: nhstepm*hstepm matrices.
779:
780: Returns p3mat[i][j][h] after calling
781: p3mat[i][j][h]=matprod2(newm,
782: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
783: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
784: oldm);
1.226 brouard 785:
786: Important routines
787:
788: - func (or funcone), computes logit (pij) distinguishing
789: o fixed variables (single or product dummies or quantitative);
790: o varying variables by:
791: (1) wave (single, product dummies, quantitative),
792: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
793: % fixed dummy (treated) or quantitative (not done because time-consuming);
794: % varying dummy (not done) or quantitative (not done);
795: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
796: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
797: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
798: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
799: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 800:
1.226 brouard 801:
802:
1.133 brouard 803: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
804: Institut national d'études démographiques, Paris.
1.126 brouard 805: This software have been partly granted by Euro-REVES, a concerted action
806: from the European Union.
807: It is copyrighted identically to a GNU software product, ie programme and
808: software can be distributed freely for non commercial use. Latest version
809: can be accessed at http://euroreves.ined.fr/imach .
810:
811: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
812: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
813:
814: **********************************************************************/
815: /*
816: main
817: read parameterfile
818: read datafile
819: concatwav
820: freqsummary
821: if (mle >= 1)
822: mlikeli
823: print results files
824: if mle==1
825: computes hessian
826: read end of parameter file: agemin, agemax, bage, fage, estepm
827: begin-prev-date,...
828: open gnuplot file
829: open html file
1.145 brouard 830: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
831: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
832: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
833: freexexit2 possible for memory heap.
834:
835: h Pij x | pij_nom ficrestpij
836: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
837: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
838: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
839:
840: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
841: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
842: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
843: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
844: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
845:
1.126 brouard 846: forecasting if prevfcast==1 prevforecast call prevalence()
847: health expectancies
848: Variance-covariance of DFLE
849: prevalence()
850: movingaverage()
851: varevsij()
852: if popbased==1 varevsij(,popbased)
853: total life expectancies
854: Variance of period (stable) prevalence
855: end
856: */
857:
1.187 brouard 858: /* #define DEBUG */
859: /* #define DEBUGBRENT */
1.203 brouard 860: /* #define DEBUGLINMIN */
861: /* #define DEBUGHESS */
862: #define DEBUGHESSIJ
1.224 brouard 863: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 864: #define POWELL /* Instead of NLOPT */
1.224 brouard 865: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 866: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
867: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 868:
869: #include <math.h>
870: #include <stdio.h>
871: #include <stdlib.h>
872: #include <string.h>
1.226 brouard 873: #include <ctype.h>
1.159 brouard 874:
875: #ifdef _WIN32
876: #include <io.h>
1.172 brouard 877: #include <windows.h>
878: #include <tchar.h>
1.159 brouard 879: #else
1.126 brouard 880: #include <unistd.h>
1.159 brouard 881: #endif
1.126 brouard 882:
883: #include <limits.h>
884: #include <sys/types.h>
1.171 brouard 885:
886: #if defined(__GNUC__)
887: #include <sys/utsname.h> /* Doesn't work on Windows */
888: #endif
889:
1.126 brouard 890: #include <sys/stat.h>
891: #include <errno.h>
1.159 brouard 892: /* extern int errno; */
1.126 brouard 893:
1.157 brouard 894: /* #ifdef LINUX */
895: /* #include <time.h> */
896: /* #include "timeval.h" */
897: /* #else */
898: /* #include <sys/time.h> */
899: /* #endif */
900:
1.126 brouard 901: #include <time.h>
902:
1.136 brouard 903: #ifdef GSL
904: #include <gsl/gsl_errno.h>
905: #include <gsl/gsl_multimin.h>
906: #endif
907:
1.167 brouard 908:
1.162 brouard 909: #ifdef NLOPT
910: #include <nlopt.h>
911: typedef struct {
912: double (* function)(double [] );
913: } myfunc_data ;
914: #endif
915:
1.126 brouard 916: /* #include <libintl.h> */
917: /* #define _(String) gettext (String) */
918:
1.251 brouard 919: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 920:
921: #define GNUPLOTPROGRAM "gnuplot"
922: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
923: #define FILENAMELENGTH 132
924:
925: #define GLOCK_ERROR_NOPATH -1 /* empty path */
926: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
927:
1.144 brouard 928: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
929: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 930:
931: #define NINTERVMAX 8
1.144 brouard 932: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
933: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
934: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 935: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 936: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
937: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 938: #define MAXN 20000
1.144 brouard 939: #define YEARM 12. /**< Number of months per year */
1.218 brouard 940: /* #define AGESUP 130 */
941: #define AGESUP 150
942: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 943: #define AGEBASE 40
1.194 brouard 944: #define AGEOVERFLOW 1.e20
1.164 brouard 945: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 946: #ifdef _WIN32
947: #define DIRSEPARATOR '\\'
948: #define CHARSEPARATOR "\\"
949: #define ODIRSEPARATOR '/'
950: #else
1.126 brouard 951: #define DIRSEPARATOR '/'
952: #define CHARSEPARATOR "/"
953: #define ODIRSEPARATOR '\\'
954: #endif
955:
1.252 ! brouard 956: /* $Id: imach.c,v 1.251 2016/09/15 15:01:13 brouard Exp $ */
1.126 brouard 957: /* $State: Exp $ */
1.196 brouard 958: #include "version.h"
959: char version[]=__IMACH_VERSION__;
1.224 brouard 960: 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";
1.252 ! brouard 961: char fullversion[]="$Revision: 1.251 $ $Date: 2016/09/15 15:01:13 $";
1.126 brouard 962: char strstart[80];
963: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 964: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 965: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 966: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
967: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
968: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 969: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
970: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 971: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
972: int cptcovprodnoage=0; /**< Number of covariate products without age */
973: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 974: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
975: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 976: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 977: int nsd=0; /**< Total number of single dummy variables (output) */
978: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 979: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 980: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 981: int ntveff=0; /**< ntveff number of effective time varying variables */
982: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 983: int cptcov=0; /* Working variable */
1.218 brouard 984: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 985: int npar=NPARMAX;
986: int nlstate=2; /* Number of live states */
987: int ndeath=1; /* Number of dead states */
1.130 brouard 988: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 989: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 990: int popbased=0;
991:
992: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 993: int maxwav=0; /* Maxim number of waves */
994: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
995: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
996: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 997: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 998: int mle=1, weightopt=0;
1.126 brouard 999: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1000: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1001: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1002: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1003: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1004: int selected(int kvar); /* Is covariate kvar selected for printing results */
1005:
1.130 brouard 1006: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1007: double **matprod2(); /* test */
1.126 brouard 1008: double **oldm, **newm, **savm; /* Working pointers to matrices */
1009: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1010: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1011:
1.136 brouard 1012: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1013: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1014: FILE *ficlog, *ficrespow;
1.130 brouard 1015: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1016: double fretone; /* Only one call to likelihood */
1.130 brouard 1017: long ipmx=0; /* Number of contributions */
1.126 brouard 1018: double sw; /* Sum of weights */
1019: char filerespow[FILENAMELENGTH];
1020: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1021: FILE *ficresilk;
1022: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1023: FILE *ficresprobmorprev;
1024: FILE *fichtm, *fichtmcov; /* Html File */
1025: FILE *ficreseij;
1026: char filerese[FILENAMELENGTH];
1027: FILE *ficresstdeij;
1028: char fileresstde[FILENAMELENGTH];
1029: FILE *ficrescveij;
1030: char filerescve[FILENAMELENGTH];
1031: FILE *ficresvij;
1032: char fileresv[FILENAMELENGTH];
1033: FILE *ficresvpl;
1034: char fileresvpl[FILENAMELENGTH];
1035: char title[MAXLINE];
1.234 brouard 1036: char model[MAXLINE]; /**< The model line */
1.217 brouard 1037: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1038: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1039: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1040: char command[FILENAMELENGTH];
1041: int outcmd=0;
1042:
1.217 brouard 1043: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1044: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1045: char filelog[FILENAMELENGTH]; /* Log file */
1046: char filerest[FILENAMELENGTH];
1047: char fileregp[FILENAMELENGTH];
1048: char popfile[FILENAMELENGTH];
1049:
1050: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1051:
1.157 brouard 1052: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1053: /* struct timezone tzp; */
1054: /* extern int gettimeofday(); */
1055: struct tm tml, *gmtime(), *localtime();
1056:
1057: extern time_t time();
1058:
1059: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1060: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1061: struct tm tm;
1062:
1.126 brouard 1063: char strcurr[80], strfor[80];
1064:
1065: char *endptr;
1066: long lval;
1067: double dval;
1068:
1069: #define NR_END 1
1070: #define FREE_ARG char*
1071: #define FTOL 1.0e-10
1072:
1073: #define NRANSI
1.240 brouard 1074: #define ITMAX 200
1075: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1076:
1077: #define TOL 2.0e-4
1078:
1079: #define CGOLD 0.3819660
1080: #define ZEPS 1.0e-10
1081: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1082:
1083: #define GOLD 1.618034
1084: #define GLIMIT 100.0
1085: #define TINY 1.0e-20
1086:
1087: static double maxarg1,maxarg2;
1088: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1089: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1090:
1091: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1092: #define rint(a) floor(a+0.5)
1.166 brouard 1093: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1094: #define mytinydouble 1.0e-16
1.166 brouard 1095: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1096: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1097: /* static double dsqrarg; */
1098: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1099: static double sqrarg;
1100: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1101: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1102: int agegomp= AGEGOMP;
1103:
1104: int imx;
1105: int stepm=1;
1106: /* Stepm, step in month: minimum step interpolation*/
1107:
1108: int estepm;
1109: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1110:
1111: int m,nb;
1112: long *num;
1.197 brouard 1113: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1114: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1115: covariate for which somebody answered excluding
1116: undefined. Usually 2: 0 and 1. */
1117: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1118: covariate for which somebody answered including
1119: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1120: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1121: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1122: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1123: double *ageexmed,*agecens;
1124: double dateintmean=0;
1125:
1126: double *weight;
1127: int **s; /* Status */
1.141 brouard 1128: double *agedc;
1.145 brouard 1129: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1130: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1131: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1132: double **coqvar; /* Fixed quantitative covariate iqv */
1133: double ***cotvar; /* Time varying covariate itv */
1134: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1135: double idx;
1136: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1137: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1138: /*k 1 2 3 4 5 6 7 8 9 */
1139: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1140: /* Tndvar[k] 1 2 3 4 5 */
1141: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1142: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1143: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1144: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1145: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1146: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1147: /* Tprod[i]=k 4 7 */
1148: /* Tage[i]=k 5 8 */
1149: /* */
1150: /* Type */
1151: /* V 1 2 3 4 5 */
1152: /* F F V V V */
1153: /* D Q D D Q */
1154: /* */
1155: int *TvarsD;
1156: int *TvarsDind;
1157: int *TvarsQ;
1158: int *TvarsQind;
1159:
1.235 brouard 1160: #define MAXRESULTLINES 10
1161: int nresult=0;
1162: int TKresult[MAXRESULTLINES];
1.237 brouard 1163: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1164: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1165: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1166: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1167: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1168: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1169:
1.234 brouard 1170: /* 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 1171: 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 */
1172: 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 */
1173: 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 */
1174: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1175: 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 */
1176: 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 1177: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1178: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1179: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1180: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1181: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1182: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1183: 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 */
1184: 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 */
1185:
1.230 brouard 1186: int *Tvarsel; /**< Selected covariates for output */
1187: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1188: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1189: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1190: 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 1191: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1192: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1193: int *Tage;
1.227 brouard 1194: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1195: 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 1196: 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*/
1197: 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 1198: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1199: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1200: int **Tvard;
1201: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1202: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1203: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1204: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1205: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1206: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1207: double *lsurv, *lpop, *tpop;
1208:
1.231 brouard 1209: #define FD 1; /* Fixed dummy covariate */
1210: #define FQ 2; /* Fixed quantitative covariate */
1211: #define FP 3; /* Fixed product covariate */
1212: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1213: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1214: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1215: #define VD 10; /* Varying dummy covariate */
1216: #define VQ 11; /* Varying quantitative covariate */
1217: #define VP 12; /* Varying product covariate */
1218: #define VPDD 13; /* Varying product dummy*dummy covariate */
1219: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1220: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1221: #define APFD 16; /* Age product * fixed dummy covariate */
1222: #define APFQ 17; /* Age product * fixed quantitative covariate */
1223: #define APVD 18; /* Age product * varying dummy covariate */
1224: #define APVQ 19; /* Age product * varying quantitative covariate */
1225:
1226: #define FTYPE 1; /* Fixed covariate */
1227: #define VTYPE 2; /* Varying covariate (loop in wave) */
1228: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1229:
1230: struct kmodel{
1231: int maintype; /* main type */
1232: int subtype; /* subtype */
1233: };
1234: struct kmodel modell[NCOVMAX];
1235:
1.143 brouard 1236: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1237: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1238:
1239: /**************** split *************************/
1240: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1241: {
1242: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1243: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1244: */
1245: char *ss; /* pointer */
1.186 brouard 1246: int l1=0, l2=0; /* length counters */
1.126 brouard 1247:
1248: l1 = strlen(path ); /* length of path */
1249: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1250: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1251: if ( ss == NULL ) { /* no directory, so determine current directory */
1252: strcpy( name, path ); /* we got the fullname name because no directory */
1253: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1254: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1255: /* get current working directory */
1256: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1257: #ifdef WIN32
1258: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1259: #else
1260: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1261: #endif
1.126 brouard 1262: return( GLOCK_ERROR_GETCWD );
1263: }
1264: /* got dirc from getcwd*/
1265: printf(" DIRC = %s \n",dirc);
1.205 brouard 1266: } else { /* strip directory from path */
1.126 brouard 1267: ss++; /* after this, the filename */
1268: l2 = strlen( ss ); /* length of filename */
1269: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1270: strcpy( name, ss ); /* save file name */
1271: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1272: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1273: printf(" DIRC2 = %s \n",dirc);
1274: }
1275: /* We add a separator at the end of dirc if not exists */
1276: l1 = strlen( dirc ); /* length of directory */
1277: if( dirc[l1-1] != DIRSEPARATOR ){
1278: dirc[l1] = DIRSEPARATOR;
1279: dirc[l1+1] = 0;
1280: printf(" DIRC3 = %s \n",dirc);
1281: }
1282: ss = strrchr( name, '.' ); /* find last / */
1283: if (ss >0){
1284: ss++;
1285: strcpy(ext,ss); /* save extension */
1286: l1= strlen( name);
1287: l2= strlen(ss)+1;
1288: strncpy( finame, name, l1-l2);
1289: finame[l1-l2]= 0;
1290: }
1291:
1292: return( 0 ); /* we're done */
1293: }
1294:
1295:
1296: /******************************************/
1297:
1298: void replace_back_to_slash(char *s, char*t)
1299: {
1300: int i;
1301: int lg=0;
1302: i=0;
1303: lg=strlen(t);
1304: for(i=0; i<= lg; i++) {
1305: (s[i] = t[i]);
1306: if (t[i]== '\\') s[i]='/';
1307: }
1308: }
1309:
1.132 brouard 1310: char *trimbb(char *out, char *in)
1.137 brouard 1311: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1312: char *s;
1313: s=out;
1314: while (*in != '\0'){
1.137 brouard 1315: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1316: in++;
1317: }
1318: *out++ = *in++;
1319: }
1320: *out='\0';
1321: return s;
1322: }
1323:
1.187 brouard 1324: /* char *substrchaine(char *out, char *in, char *chain) */
1325: /* { */
1326: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1327: /* char *s, *t; */
1328: /* t=in;s=out; */
1329: /* while ((*in != *chain) && (*in != '\0')){ */
1330: /* *out++ = *in++; */
1331: /* } */
1332:
1333: /* /\* *in matches *chain *\/ */
1334: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1335: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1336: /* } */
1337: /* in--; chain--; */
1338: /* while ( (*in != '\0')){ */
1339: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1340: /* *out++ = *in++; */
1341: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1342: /* } */
1343: /* *out='\0'; */
1344: /* out=s; */
1345: /* return out; */
1346: /* } */
1347: char *substrchaine(char *out, char *in, char *chain)
1348: {
1349: /* Substract chain 'chain' from 'in', return and output 'out' */
1350: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1351:
1352: char *strloc;
1353:
1354: strcpy (out, in);
1355: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1356: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1357: if(strloc != NULL){
1358: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1359: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1360: /* strcpy (strloc, strloc +strlen(chain));*/
1361: }
1362: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1363: return out;
1364: }
1365:
1366:
1.145 brouard 1367: char *cutl(char *blocc, char *alocc, char *in, char occ)
1368: {
1.187 brouard 1369: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1370: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1371: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1372: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1373: */
1.160 brouard 1374: char *s, *t;
1.145 brouard 1375: t=in;s=in;
1376: while ((*in != occ) && (*in != '\0')){
1377: *alocc++ = *in++;
1378: }
1379: if( *in == occ){
1380: *(alocc)='\0';
1381: s=++in;
1382: }
1383:
1384: if (s == t) {/* occ not found */
1385: *(alocc-(in-s))='\0';
1386: in=s;
1387: }
1388: while ( *in != '\0'){
1389: *blocc++ = *in++;
1390: }
1391:
1392: *blocc='\0';
1393: return t;
1394: }
1.137 brouard 1395: char *cutv(char *blocc, char *alocc, char *in, char occ)
1396: {
1.187 brouard 1397: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1398: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1399: gives blocc="abcdef2ghi" and alocc="j".
1400: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1401: */
1402: char *s, *t;
1403: t=in;s=in;
1404: while (*in != '\0'){
1405: while( *in == occ){
1406: *blocc++ = *in++;
1407: s=in;
1408: }
1409: *blocc++ = *in++;
1410: }
1411: if (s == t) /* occ not found */
1412: *(blocc-(in-s))='\0';
1413: else
1414: *(blocc-(in-s)-1)='\0';
1415: in=s;
1416: while ( *in != '\0'){
1417: *alocc++ = *in++;
1418: }
1419:
1420: *alocc='\0';
1421: return s;
1422: }
1423:
1.126 brouard 1424: int nbocc(char *s, char occ)
1425: {
1426: int i,j=0;
1427: int lg=20;
1428: i=0;
1429: lg=strlen(s);
1430: for(i=0; i<= lg; i++) {
1.234 brouard 1431: if (s[i] == occ ) j++;
1.126 brouard 1432: }
1433: return j;
1434: }
1435:
1.137 brouard 1436: /* void cutv(char *u,char *v, char*t, char occ) */
1437: /* { */
1438: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1439: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1440: /* gives u="abcdef2ghi" and v="j" *\/ */
1441: /* int i,lg,j,p=0; */
1442: /* i=0; */
1443: /* lg=strlen(t); */
1444: /* for(j=0; j<=lg-1; j++) { */
1445: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1446: /* } */
1.126 brouard 1447:
1.137 brouard 1448: /* for(j=0; j<p; j++) { */
1449: /* (u[j] = t[j]); */
1450: /* } */
1451: /* u[p]='\0'; */
1.126 brouard 1452:
1.137 brouard 1453: /* for(j=0; j<= lg; j++) { */
1454: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1455: /* } */
1456: /* } */
1.126 brouard 1457:
1.160 brouard 1458: #ifdef _WIN32
1459: char * strsep(char **pp, const char *delim)
1460: {
1461: char *p, *q;
1462:
1463: if ((p = *pp) == NULL)
1464: return 0;
1465: if ((q = strpbrk (p, delim)) != NULL)
1466: {
1467: *pp = q + 1;
1468: *q = '\0';
1469: }
1470: else
1471: *pp = 0;
1472: return p;
1473: }
1474: #endif
1475:
1.126 brouard 1476: /********************** nrerror ********************/
1477:
1478: void nrerror(char error_text[])
1479: {
1480: fprintf(stderr,"ERREUR ...\n");
1481: fprintf(stderr,"%s\n",error_text);
1482: exit(EXIT_FAILURE);
1483: }
1484: /*********************** vector *******************/
1485: double *vector(int nl, int nh)
1486: {
1487: double *v;
1488: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1489: if (!v) nrerror("allocation failure in vector");
1490: return v-nl+NR_END;
1491: }
1492:
1493: /************************ free vector ******************/
1494: void free_vector(double*v, int nl, int nh)
1495: {
1496: free((FREE_ARG)(v+nl-NR_END));
1497: }
1498:
1499: /************************ivector *******************************/
1500: int *ivector(long nl,long nh)
1501: {
1502: int *v;
1503: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1504: if (!v) nrerror("allocation failure in ivector");
1505: return v-nl+NR_END;
1506: }
1507:
1508: /******************free ivector **************************/
1509: void free_ivector(int *v, long nl, long nh)
1510: {
1511: free((FREE_ARG)(v+nl-NR_END));
1512: }
1513:
1514: /************************lvector *******************************/
1515: long *lvector(long nl,long nh)
1516: {
1517: long *v;
1518: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1519: if (!v) nrerror("allocation failure in ivector");
1520: return v-nl+NR_END;
1521: }
1522:
1523: /******************free lvector **************************/
1524: void free_lvector(long *v, long nl, long nh)
1525: {
1526: free((FREE_ARG)(v+nl-NR_END));
1527: }
1528:
1529: /******************* imatrix *******************************/
1530: int **imatrix(long nrl, long nrh, long ncl, long nch)
1531: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1532: {
1533: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1534: int **m;
1535:
1536: /* allocate pointers to rows */
1537: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1538: if (!m) nrerror("allocation failure 1 in matrix()");
1539: m += NR_END;
1540: m -= nrl;
1541:
1542:
1543: /* allocate rows and set pointers to them */
1544: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1545: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1546: m[nrl] += NR_END;
1547: m[nrl] -= ncl;
1548:
1549: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1550:
1551: /* return pointer to array of pointers to rows */
1552: return m;
1553: }
1554:
1555: /****************** free_imatrix *************************/
1556: void free_imatrix(m,nrl,nrh,ncl,nch)
1557: int **m;
1558: long nch,ncl,nrh,nrl;
1559: /* free an int matrix allocated by imatrix() */
1560: {
1561: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1562: free((FREE_ARG) (m+nrl-NR_END));
1563: }
1564:
1565: /******************* matrix *******************************/
1566: double **matrix(long nrl, long nrh, long ncl, long nch)
1567: {
1568: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1569: double **m;
1570:
1571: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1572: if (!m) nrerror("allocation failure 1 in matrix()");
1573: m += NR_END;
1574: m -= nrl;
1575:
1576: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1577: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1578: m[nrl] += NR_END;
1579: m[nrl] -= ncl;
1580:
1581: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1582: return m;
1.145 brouard 1583: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1584: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1585: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1586: */
1587: }
1588:
1589: /*************************free matrix ************************/
1590: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1591: {
1592: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1593: free((FREE_ARG)(m+nrl-NR_END));
1594: }
1595:
1596: /******************* ma3x *******************************/
1597: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1598: {
1599: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1600: double ***m;
1601:
1602: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1603: if (!m) nrerror("allocation failure 1 in matrix()");
1604: m += NR_END;
1605: m -= nrl;
1606:
1607: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1608: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1609: m[nrl] += NR_END;
1610: m[nrl] -= ncl;
1611:
1612: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1613:
1614: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1615: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1616: m[nrl][ncl] += NR_END;
1617: m[nrl][ncl] -= nll;
1618: for (j=ncl+1; j<=nch; j++)
1619: m[nrl][j]=m[nrl][j-1]+nlay;
1620:
1621: for (i=nrl+1; i<=nrh; i++) {
1622: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1623: for (j=ncl+1; j<=nch; j++)
1624: m[i][j]=m[i][j-1]+nlay;
1625: }
1626: return m;
1627: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1628: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1629: */
1630: }
1631:
1632: /*************************free ma3x ************************/
1633: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1634: {
1635: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1636: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1637: free((FREE_ARG)(m+nrl-NR_END));
1638: }
1639:
1640: /*************** function subdirf ***********/
1641: char *subdirf(char fileres[])
1642: {
1643: /* Caution optionfilefiname is hidden */
1644: strcpy(tmpout,optionfilefiname);
1645: strcat(tmpout,"/"); /* Add to the right */
1646: strcat(tmpout,fileres);
1647: return tmpout;
1648: }
1649:
1650: /*************** function subdirf2 ***********/
1651: char *subdirf2(char fileres[], char *preop)
1652: {
1653:
1654: /* Caution optionfilefiname is hidden */
1655: strcpy(tmpout,optionfilefiname);
1656: strcat(tmpout,"/");
1657: strcat(tmpout,preop);
1658: strcat(tmpout,fileres);
1659: return tmpout;
1660: }
1661:
1662: /*************** function subdirf3 ***********/
1663: char *subdirf3(char fileres[], char *preop, char *preop2)
1664: {
1665:
1666: /* Caution optionfilefiname is hidden */
1667: strcpy(tmpout,optionfilefiname);
1668: strcat(tmpout,"/");
1669: strcat(tmpout,preop);
1670: strcat(tmpout,preop2);
1671: strcat(tmpout,fileres);
1672: return tmpout;
1673: }
1.213 brouard 1674:
1675: /*************** function subdirfext ***********/
1676: char *subdirfext(char fileres[], char *preop, char *postop)
1677: {
1678:
1679: strcpy(tmpout,preop);
1680: strcat(tmpout,fileres);
1681: strcat(tmpout,postop);
1682: return tmpout;
1683: }
1.126 brouard 1684:
1.213 brouard 1685: /*************** function subdirfext3 ***********/
1686: char *subdirfext3(char fileres[], char *preop, char *postop)
1687: {
1688:
1689: /* Caution optionfilefiname is hidden */
1690: strcpy(tmpout,optionfilefiname);
1691: strcat(tmpout,"/");
1692: strcat(tmpout,preop);
1693: strcat(tmpout,fileres);
1694: strcat(tmpout,postop);
1695: return tmpout;
1696: }
1697:
1.162 brouard 1698: char *asc_diff_time(long time_sec, char ascdiff[])
1699: {
1700: long sec_left, days, hours, minutes;
1701: days = (time_sec) / (60*60*24);
1702: sec_left = (time_sec) % (60*60*24);
1703: hours = (sec_left) / (60*60) ;
1704: sec_left = (sec_left) %(60*60);
1705: minutes = (sec_left) /60;
1706: sec_left = (sec_left) % (60);
1707: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1708: return ascdiff;
1709: }
1710:
1.126 brouard 1711: /***************** f1dim *************************/
1712: extern int ncom;
1713: extern double *pcom,*xicom;
1714: extern double (*nrfunc)(double []);
1715:
1716: double f1dim(double x)
1717: {
1718: int j;
1719: double f;
1720: double *xt;
1721:
1722: xt=vector(1,ncom);
1723: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1724: f=(*nrfunc)(xt);
1725: free_vector(xt,1,ncom);
1726: return f;
1727: }
1728:
1729: /*****************brent *************************/
1730: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1731: {
1732: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1733: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1734: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1735: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1736: * returned function value.
1737: */
1.126 brouard 1738: int iter;
1739: double a,b,d,etemp;
1.159 brouard 1740: double fu=0,fv,fw,fx;
1.164 brouard 1741: double ftemp=0.;
1.126 brouard 1742: double p,q,r,tol1,tol2,u,v,w,x,xm;
1743: double e=0.0;
1744:
1745: a=(ax < cx ? ax : cx);
1746: b=(ax > cx ? ax : cx);
1747: x=w=v=bx;
1748: fw=fv=fx=(*f)(x);
1749: for (iter=1;iter<=ITMAX;iter++) {
1750: xm=0.5*(a+b);
1751: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1752: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1753: printf(".");fflush(stdout);
1754: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1755: #ifdef DEBUGBRENT
1.126 brouard 1756: 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);
1757: 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);
1758: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1759: #endif
1760: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1761: *xmin=x;
1762: return fx;
1763: }
1764: ftemp=fu;
1765: if (fabs(e) > tol1) {
1766: r=(x-w)*(fx-fv);
1767: q=(x-v)*(fx-fw);
1768: p=(x-v)*q-(x-w)*r;
1769: q=2.0*(q-r);
1770: if (q > 0.0) p = -p;
1771: q=fabs(q);
1772: etemp=e;
1773: e=d;
1774: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1775: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1776: else {
1.224 brouard 1777: d=p/q;
1778: u=x+d;
1779: if (u-a < tol2 || b-u < tol2)
1780: d=SIGN(tol1,xm-x);
1.126 brouard 1781: }
1782: } else {
1783: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1784: }
1785: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1786: fu=(*f)(u);
1787: if (fu <= fx) {
1788: if (u >= x) a=x; else b=x;
1789: SHFT(v,w,x,u)
1.183 brouard 1790: SHFT(fv,fw,fx,fu)
1791: } else {
1792: if (u < x) a=u; else b=u;
1793: if (fu <= fw || w == x) {
1.224 brouard 1794: v=w;
1795: w=u;
1796: fv=fw;
1797: fw=fu;
1.183 brouard 1798: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1799: v=u;
1800: fv=fu;
1.183 brouard 1801: }
1802: }
1.126 brouard 1803: }
1804: nrerror("Too many iterations in brent");
1805: *xmin=x;
1806: return fx;
1807: }
1808:
1809: /****************** mnbrak ***********************/
1810:
1811: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1812: double (*func)(double))
1.183 brouard 1813: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1814: the downhill direction (defined by the function as evaluated at the initial points) and returns
1815: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1816: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1817: */
1.126 brouard 1818: double ulim,u,r,q, dum;
1819: double fu;
1.187 brouard 1820:
1821: double scale=10.;
1822: int iterscale=0;
1823:
1824: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1825: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1826:
1827:
1828: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1829: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1830: /* *bx = *ax - (*ax - *bx)/scale; */
1831: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1832: /* } */
1833:
1.126 brouard 1834: if (*fb > *fa) {
1835: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1836: SHFT(dum,*fb,*fa,dum)
1837: }
1.126 brouard 1838: *cx=(*bx)+GOLD*(*bx-*ax);
1839: *fc=(*func)(*cx);
1.183 brouard 1840: #ifdef DEBUG
1.224 brouard 1841: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1842: 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 1843: #endif
1.224 brouard 1844: 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 1845: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1846: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1847: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1848: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1849: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1850: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1851: fu=(*func)(u);
1.163 brouard 1852: #ifdef DEBUG
1853: /* f(x)=A(x-u)**2+f(u) */
1854: double A, fparabu;
1855: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1856: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1857: 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);
1858: 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 1859: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1860: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1861: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1862: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1863: #endif
1.184 brouard 1864: #ifdef MNBRAKORIGINAL
1.183 brouard 1865: #else
1.191 brouard 1866: /* if (fu > *fc) { */
1867: /* #ifdef DEBUG */
1868: /* printf("mnbrak4 fu > fc \n"); */
1869: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1870: /* #endif */
1871: /* /\* 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 *\\/ *\/ */
1872: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1873: /* dum=u; /\* Shifting c and u *\/ */
1874: /* u = *cx; */
1875: /* *cx = dum; */
1876: /* dum = fu; */
1877: /* fu = *fc; */
1878: /* *fc =dum; */
1879: /* } else { /\* end *\/ */
1880: /* #ifdef DEBUG */
1881: /* printf("mnbrak3 fu < fc \n"); */
1882: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1883: /* #endif */
1884: /* dum=u; /\* Shifting c and u *\/ */
1885: /* u = *cx; */
1886: /* *cx = dum; */
1887: /* dum = fu; */
1888: /* fu = *fc; */
1889: /* *fc =dum; */
1890: /* } */
1.224 brouard 1891: #ifdef DEBUGMNBRAK
1892: double A, fparabu;
1893: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1894: fparabu= *fa - A*(*ax-u)*(*ax-u);
1895: 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);
1896: 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 1897: #endif
1.191 brouard 1898: dum=u; /* Shifting c and u */
1899: u = *cx;
1900: *cx = dum;
1901: dum = fu;
1902: fu = *fc;
1903: *fc =dum;
1.183 brouard 1904: #endif
1.162 brouard 1905: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1906: #ifdef DEBUG
1.224 brouard 1907: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1908: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1909: #endif
1.126 brouard 1910: fu=(*func)(u);
1911: if (fu < *fc) {
1.183 brouard 1912: #ifdef DEBUG
1.224 brouard 1913: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1914: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1915: #endif
1916: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1917: SHFT(*fb,*fc,fu,(*func)(u))
1918: #ifdef DEBUG
1919: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1920: #endif
1921: }
1.162 brouard 1922: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1923: #ifdef DEBUG
1.224 brouard 1924: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1925: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1926: #endif
1.126 brouard 1927: u=ulim;
1928: fu=(*func)(u);
1.183 brouard 1929: } else { /* u could be left to b (if r > q parabola has a maximum) */
1930: #ifdef DEBUG
1.224 brouard 1931: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1932: 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 1933: #endif
1.126 brouard 1934: u=(*cx)+GOLD*(*cx-*bx);
1935: fu=(*func)(u);
1.224 brouard 1936: #ifdef DEBUG
1937: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1938: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1939: #endif
1.183 brouard 1940: } /* end tests */
1.126 brouard 1941: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1942: SHFT(*fa,*fb,*fc,fu)
1943: #ifdef DEBUG
1.224 brouard 1944: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1945: 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 1946: #endif
1947: } /* 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 1948: }
1949:
1950: /*************** linmin ************************/
1.162 brouard 1951: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1952: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1953: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1954: the value of func at the returned location p . This is actually all accomplished by calling the
1955: routines mnbrak and brent .*/
1.126 brouard 1956: int ncom;
1957: double *pcom,*xicom;
1958: double (*nrfunc)(double []);
1959:
1.224 brouard 1960: #ifdef LINMINORIGINAL
1.126 brouard 1961: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1962: #else
1963: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1964: #endif
1.126 brouard 1965: {
1966: double brent(double ax, double bx, double cx,
1967: double (*f)(double), double tol, double *xmin);
1968: double f1dim(double x);
1969: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1970: double *fc, double (*func)(double));
1971: int j;
1972: double xx,xmin,bx,ax;
1973: double fx,fb,fa;
1.187 brouard 1974:
1.203 brouard 1975: #ifdef LINMINORIGINAL
1976: #else
1977: double scale=10., axs, xxs; /* Scale added for infinity */
1978: #endif
1979:
1.126 brouard 1980: ncom=n;
1981: pcom=vector(1,n);
1982: xicom=vector(1,n);
1983: nrfunc=func;
1984: for (j=1;j<=n;j++) {
1985: pcom[j]=p[j];
1.202 brouard 1986: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1987: }
1.187 brouard 1988:
1.203 brouard 1989: #ifdef LINMINORIGINAL
1990: xx=1.;
1991: #else
1992: axs=0.0;
1993: xxs=1.;
1994: do{
1995: xx= xxs;
1996: #endif
1.187 brouard 1997: ax=0.;
1998: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
1999: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2000: /* 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)) */
2001: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2002: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2003: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2004: /* 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 2005: #ifdef LINMINORIGINAL
2006: #else
2007: if (fx != fx){
1.224 brouard 2008: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2009: printf("|");
2010: fprintf(ficlog,"|");
1.203 brouard 2011: #ifdef DEBUGLINMIN
1.224 brouard 2012: 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 2013: #endif
2014: }
1.224 brouard 2015: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2016: #endif
2017:
1.191 brouard 2018: #ifdef DEBUGLINMIN
2019: 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 2020: 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 2021: #endif
1.224 brouard 2022: #ifdef LINMINORIGINAL
2023: #else
2024: if(fb == fx){ /* Flat function in the direction */
2025: xmin=xx;
2026: *flat=1;
2027: }else{
2028: *flat=0;
2029: #endif
2030: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2031: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2032: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2033: /* fmin = f(p[j] + xmin * xi[j]) */
2034: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2035: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2036: #ifdef DEBUG
1.224 brouard 2037: 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);
2038: 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);
2039: #endif
2040: #ifdef LINMINORIGINAL
2041: #else
2042: }
1.126 brouard 2043: #endif
1.191 brouard 2044: #ifdef DEBUGLINMIN
2045: printf("linmin end ");
1.202 brouard 2046: fprintf(ficlog,"linmin end ");
1.191 brouard 2047: #endif
1.126 brouard 2048: for (j=1;j<=n;j++) {
1.203 brouard 2049: #ifdef LINMINORIGINAL
2050: xi[j] *= xmin;
2051: #else
2052: #ifdef DEBUGLINMIN
2053: if(xxs <1.0)
2054: printf(" before xi[%d]=%12.8f", j,xi[j]);
2055: #endif
2056: 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) */
2057: #ifdef DEBUGLINMIN
2058: if(xxs <1.0)
2059: 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 );
2060: #endif
2061: #endif
1.187 brouard 2062: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2063: }
1.191 brouard 2064: #ifdef DEBUGLINMIN
1.203 brouard 2065: printf("\n");
1.191 brouard 2066: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2067: 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 2068: for (j=1;j<=n;j++) {
1.202 brouard 2069: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2070: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2071: if(j % ncovmodel == 0){
1.191 brouard 2072: printf("\n");
1.202 brouard 2073: fprintf(ficlog,"\n");
2074: }
1.191 brouard 2075: }
1.203 brouard 2076: #else
1.191 brouard 2077: #endif
1.126 brouard 2078: free_vector(xicom,1,n);
2079: free_vector(pcom,1,n);
2080: }
2081:
2082:
2083: /*************** powell ************************/
1.162 brouard 2084: /*
2085: Minimization of a function func of n variables. Input consists of an initial starting point
2086: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2087: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2088: such that failure to decrease by more than this amount on one iteration signals doneness. On
2089: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2090: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2091: */
1.224 brouard 2092: #ifdef LINMINORIGINAL
2093: #else
2094: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2095: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2096: #endif
1.126 brouard 2097: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2098: double (*func)(double []))
2099: {
1.224 brouard 2100: #ifdef LINMINORIGINAL
2101: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2102: double (*func)(double []));
1.224 brouard 2103: #else
1.241 brouard 2104: void linmin(double p[], double xi[], int n, double *fret,
2105: double (*func)(double []),int *flat);
1.224 brouard 2106: #endif
1.239 brouard 2107: int i,ibig,j,jk,k;
1.126 brouard 2108: double del,t,*pt,*ptt,*xit;
1.181 brouard 2109: double directest;
1.126 brouard 2110: double fp,fptt;
2111: double *xits;
2112: int niterf, itmp;
1.224 brouard 2113: #ifdef LINMINORIGINAL
2114: #else
2115:
2116: flatdir=ivector(1,n);
2117: for (j=1;j<=n;j++) flatdir[j]=0;
2118: #endif
1.126 brouard 2119:
2120: pt=vector(1,n);
2121: ptt=vector(1,n);
2122: xit=vector(1,n);
2123: xits=vector(1,n);
2124: *fret=(*func)(p);
2125: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2126: rcurr_time = time(NULL);
1.126 brouard 2127: for (*iter=1;;++(*iter)) {
1.187 brouard 2128: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2129: ibig=0;
2130: del=0.0;
1.157 brouard 2131: rlast_time=rcurr_time;
2132: /* (void) gettimeofday(&curr_time,&tzp); */
2133: rcurr_time = time(NULL);
2134: curr_time = *localtime(&rcurr_time);
2135: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2136: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2137: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2138: for (i=1;i<=n;i++) {
1.126 brouard 2139: fprintf(ficrespow," %.12lf", p[i]);
2140: }
1.239 brouard 2141: fprintf(ficrespow,"\n");fflush(ficrespow);
2142: printf("\n#model= 1 + age ");
2143: fprintf(ficlog,"\n#model= 1 + age ");
2144: if(nagesqr==1){
1.241 brouard 2145: printf(" + age*age ");
2146: fprintf(ficlog," + age*age ");
1.239 brouard 2147: }
2148: for(j=1;j <=ncovmodel-2;j++){
2149: if(Typevar[j]==0) {
2150: printf(" + V%d ",Tvar[j]);
2151: fprintf(ficlog," + V%d ",Tvar[j]);
2152: }else if(Typevar[j]==1) {
2153: printf(" + V%d*age ",Tvar[j]);
2154: fprintf(ficlog," + V%d*age ",Tvar[j]);
2155: }else if(Typevar[j]==2) {
2156: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2157: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2158: }
2159: }
1.126 brouard 2160: printf("\n");
1.239 brouard 2161: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2162: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2163: fprintf(ficlog,"\n");
1.239 brouard 2164: for(i=1,jk=1; i <=nlstate; i++){
2165: for(k=1; k <=(nlstate+ndeath); k++){
2166: if (k != i) {
2167: printf("%d%d ",i,k);
2168: fprintf(ficlog,"%d%d ",i,k);
2169: for(j=1; j <=ncovmodel; j++){
2170: printf("%12.7f ",p[jk]);
2171: fprintf(ficlog,"%12.7f ",p[jk]);
2172: jk++;
2173: }
2174: printf("\n");
2175: fprintf(ficlog,"\n");
2176: }
2177: }
2178: }
1.241 brouard 2179: if(*iter <=3 && *iter >1){
1.157 brouard 2180: tml = *localtime(&rcurr_time);
2181: strcpy(strcurr,asctime(&tml));
2182: rforecast_time=rcurr_time;
1.126 brouard 2183: itmp = strlen(strcurr);
2184: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2185: strcurr[itmp-1]='\0';
1.162 brouard 2186: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2187: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2188: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2189: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2190: forecast_time = *localtime(&rforecast_time);
2191: strcpy(strfor,asctime(&forecast_time));
2192: itmp = strlen(strfor);
2193: if(strfor[itmp-1]=='\n')
2194: strfor[itmp-1]='\0';
2195: 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);
2196: 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 2197: }
2198: }
1.187 brouard 2199: for (i=1;i<=n;i++) { /* For each direction i */
2200: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2201: fptt=(*fret);
2202: #ifdef DEBUG
1.203 brouard 2203: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2204: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2205: #endif
1.203 brouard 2206: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2207: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2208: #ifdef LINMINORIGINAL
1.188 brouard 2209: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2210: #else
2211: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2212: flatdir[i]=flat; /* Function is vanishing in that direction i */
2213: #endif
2214: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2215: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2216: /* because that direction will be replaced unless the gain del is small */
2217: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2218: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2219: /* with the new direction. */
2220: del=fabs(fptt-(*fret));
2221: ibig=i;
1.126 brouard 2222: }
2223: #ifdef DEBUG
2224: printf("%d %.12e",i,(*fret));
2225: fprintf(ficlog,"%d %.12e",i,(*fret));
2226: for (j=1;j<=n;j++) {
1.224 brouard 2227: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2228: printf(" x(%d)=%.12e",j,xit[j]);
2229: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2230: }
2231: for(j=1;j<=n;j++) {
1.225 brouard 2232: printf(" p(%d)=%.12e",j,p[j]);
2233: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2234: }
2235: printf("\n");
2236: fprintf(ficlog,"\n");
2237: #endif
1.187 brouard 2238: } /* end loop on each direction i */
2239: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2240: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2241: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2242: for(j=1;j<=n;j++) {
1.225 brouard 2243: if(flatdir[j] >0){
2244: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2245: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2246: }
2247: /* printf("\n"); */
2248: /* fprintf(ficlog,"\n"); */
2249: }
1.243 brouard 2250: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2251: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2252: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2253: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2254: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2255: /* decreased of more than 3.84 */
2256: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2257: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2258: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2259:
1.188 brouard 2260: /* Starting the program with initial values given by a former maximization will simply change */
2261: /* the scales of the directions and the directions, because the are reset to canonical directions */
2262: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2263: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2264: #ifdef DEBUG
2265: int k[2],l;
2266: k[0]=1;
2267: k[1]=-1;
2268: printf("Max: %.12e",(*func)(p));
2269: fprintf(ficlog,"Max: %.12e",(*func)(p));
2270: for (j=1;j<=n;j++) {
2271: printf(" %.12e",p[j]);
2272: fprintf(ficlog," %.12e",p[j]);
2273: }
2274: printf("\n");
2275: fprintf(ficlog,"\n");
2276: for(l=0;l<=1;l++) {
2277: for (j=1;j<=n;j++) {
2278: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2279: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2280: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2281: }
2282: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2283: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2284: }
2285: #endif
2286:
1.224 brouard 2287: #ifdef LINMINORIGINAL
2288: #else
2289: free_ivector(flatdir,1,n);
2290: #endif
1.126 brouard 2291: free_vector(xit,1,n);
2292: free_vector(xits,1,n);
2293: free_vector(ptt,1,n);
2294: free_vector(pt,1,n);
2295: return;
1.192 brouard 2296: } /* enough precision */
1.240 brouard 2297: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2298: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2299: ptt[j]=2.0*p[j]-pt[j];
2300: xit[j]=p[j]-pt[j];
2301: pt[j]=p[j];
2302: }
1.181 brouard 2303: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2304: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2305: if (*iter <=4) {
1.225 brouard 2306: #else
2307: #endif
1.224 brouard 2308: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2309: #else
1.161 brouard 2310: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2311: #endif
1.162 brouard 2312: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2313: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2314: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2315: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2316: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2317: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2318: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2319: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2320: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2321: /* Even if f3 <f1, directest can be negative and t >0 */
2322: /* mu² and del² are equal when f3=f1 */
2323: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2324: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2325: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2326: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2327: #ifdef NRCORIGINAL
2328: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2329: #else
2330: 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 2331: t= t- del*SQR(fp-fptt);
1.183 brouard 2332: #endif
1.202 brouard 2333: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2334: #ifdef DEBUG
1.181 brouard 2335: 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);
2336: 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 2337: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2338: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2339: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2340: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2341: 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);
2342: 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);
2343: #endif
1.183 brouard 2344: #ifdef POWELLORIGINAL
2345: if (t < 0.0) { /* Then we use it for new direction */
2346: #else
1.182 brouard 2347: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2348: 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 2349: 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 2350: 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 2351: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2352: }
1.181 brouard 2353: if (directest < 0.0) { /* Then we use it for new direction */
2354: #endif
1.191 brouard 2355: #ifdef DEBUGLINMIN
1.234 brouard 2356: printf("Before linmin in direction P%d-P0\n",n);
2357: for (j=1;j<=n;j++) {
2358: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2359: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2360: if(j % ncovmodel == 0){
2361: printf("\n");
2362: fprintf(ficlog,"\n");
2363: }
2364: }
1.224 brouard 2365: #endif
2366: #ifdef LINMINORIGINAL
1.234 brouard 2367: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2368: #else
1.234 brouard 2369: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2370: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2371: #endif
1.234 brouard 2372:
1.191 brouard 2373: #ifdef DEBUGLINMIN
1.234 brouard 2374: for (j=1;j<=n;j++) {
2375: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2376: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2377: if(j % ncovmodel == 0){
2378: printf("\n");
2379: fprintf(ficlog,"\n");
2380: }
2381: }
1.224 brouard 2382: #endif
1.234 brouard 2383: for (j=1;j<=n;j++) {
2384: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2385: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2386: }
1.224 brouard 2387: #ifdef LINMINORIGINAL
2388: #else
1.234 brouard 2389: for (j=1, flatd=0;j<=n;j++) {
2390: if(flatdir[j]>0)
2391: flatd++;
2392: }
2393: if(flatd >0){
2394: printf("%d flat directions\n",flatd);
2395: fprintf(ficlog,"%d flat directions\n",flatd);
2396: for (j=1;j<=n;j++) {
2397: if(flatdir[j]>0){
2398: printf("%d ",j);
2399: fprintf(ficlog,"%d ",j);
2400: }
2401: }
2402: printf("\n");
2403: fprintf(ficlog,"\n");
2404: }
1.191 brouard 2405: #endif
1.234 brouard 2406: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2407: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2408:
1.126 brouard 2409: #ifdef DEBUG
1.234 brouard 2410: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2411: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2412: for(j=1;j<=n;j++){
2413: printf(" %lf",xit[j]);
2414: fprintf(ficlog," %lf",xit[j]);
2415: }
2416: printf("\n");
2417: fprintf(ficlog,"\n");
1.126 brouard 2418: #endif
1.192 brouard 2419: } /* end of t or directest negative */
1.224 brouard 2420: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2421: #else
1.234 brouard 2422: } /* end if (fptt < fp) */
1.192 brouard 2423: #endif
1.225 brouard 2424: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2425: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2426: #else
1.224 brouard 2427: #endif
1.234 brouard 2428: } /* loop iteration */
1.126 brouard 2429: }
1.234 brouard 2430:
1.126 brouard 2431: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2432:
1.235 brouard 2433: 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 2434: {
1.235 brouard 2435: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2436: (and selected quantitative values in nres)
2437: by left multiplying the unit
1.234 brouard 2438: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2439: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2440: /* Wx is row vector: population in state 1, population in state 2, population dead */
2441: /* or prevalence in state 1, prevalence in state 2, 0 */
2442: /* newm is the matrix after multiplications, its rows are identical at a factor */
2443: /* Initial matrix pimij */
2444: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2445: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2446: /* 0, 0 , 1} */
2447: /*
2448: * and after some iteration: */
2449: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2450: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2451: /* 0, 0 , 1} */
2452: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2453: /* {0.51571254859325999, 0.4842874514067399, */
2454: /* 0.51326036147820708, 0.48673963852179264} */
2455: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2456:
1.126 brouard 2457: int i, ii,j,k;
1.209 brouard 2458: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2459: /* double **matprod2(); */ /* test */
1.218 brouard 2460: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2461: double **newm;
1.209 brouard 2462: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2463: int ncvloop=0;
1.169 brouard 2464:
1.209 brouard 2465: min=vector(1,nlstate);
2466: max=vector(1,nlstate);
2467: meandiff=vector(1,nlstate);
2468:
1.218 brouard 2469: /* Starting with matrix unity */
1.126 brouard 2470: for (ii=1;ii<=nlstate+ndeath;ii++)
2471: for (j=1;j<=nlstate+ndeath;j++){
2472: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2473: }
1.169 brouard 2474:
2475: cov[1]=1.;
2476:
2477: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2478: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2479: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2480: ncvloop++;
1.126 brouard 2481: newm=savm;
2482: /* Covariates have to be included here again */
1.138 brouard 2483: cov[2]=agefin;
1.187 brouard 2484: if(nagesqr==1)
2485: cov[3]= agefin*agefin;;
1.234 brouard 2486: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2487: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2488: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2489: /* 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 2490: }
2491: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2492: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2493: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2494: /* 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 2495: }
1.237 brouard 2496: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2497: if(Dummy[Tvar[Tage[k]]]){
2498: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2499: } else{
1.235 brouard 2500: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2501: }
1.235 brouard 2502: /* 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 2503: }
1.237 brouard 2504: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2505: /* 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 2506: if(Dummy[Tvard[k][1]==0]){
2507: if(Dummy[Tvard[k][2]==0]){
2508: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2509: }else{
2510: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2511: }
2512: }else{
2513: if(Dummy[Tvard[k][2]==0]){
2514: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2515: }else{
2516: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2517: }
2518: }
1.234 brouard 2519: }
1.138 brouard 2520: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2521: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2522: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2523: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2524: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2525: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2526: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2527:
1.126 brouard 2528: savm=oldm;
2529: oldm=newm;
1.209 brouard 2530:
2531: for(j=1; j<=nlstate; j++){
2532: max[j]=0.;
2533: min[j]=1.;
2534: }
2535: for(i=1;i<=nlstate;i++){
2536: sumnew=0;
2537: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2538: for(j=1; j<=nlstate; j++){
2539: prlim[i][j]= newm[i][j]/(1-sumnew);
2540: max[j]=FMAX(max[j],prlim[i][j]);
2541: min[j]=FMIN(min[j],prlim[i][j]);
2542: }
2543: }
2544:
1.126 brouard 2545: maxmax=0.;
1.209 brouard 2546: for(j=1; j<=nlstate; j++){
2547: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2548: maxmax=FMAX(maxmax,meandiff[j]);
2549: /* 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 2550: } /* j loop */
1.203 brouard 2551: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2552: /* 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 2553: if(maxmax < ftolpl){
1.209 brouard 2554: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2555: free_vector(min,1,nlstate);
2556: free_vector(max,1,nlstate);
2557: free_vector(meandiff,1,nlstate);
1.126 brouard 2558: return prlim;
2559: }
1.169 brouard 2560: } /* age loop */
1.208 brouard 2561: /* After some age loop it doesn't converge */
1.209 brouard 2562: 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 2563: 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 2564: /* 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); */
2565: free_vector(min,1,nlstate);
2566: free_vector(max,1,nlstate);
2567: free_vector(meandiff,1,nlstate);
1.208 brouard 2568:
1.169 brouard 2569: return prlim; /* should not reach here */
1.126 brouard 2570: }
2571:
1.217 brouard 2572:
2573: /**** Back Prevalence limit (stable or period prevalence) ****************/
2574:
1.218 brouard 2575: /* 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) */
2576: /* 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 2577: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2578: {
1.218 brouard 2579: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2580: matrix by transitions matrix until convergence is reached with precision ftolpl */
2581: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2582: /* Wx is row vector: population in state 1, population in state 2, population dead */
2583: /* or prevalence in state 1, prevalence in state 2, 0 */
2584: /* newm is the matrix after multiplications, its rows are identical at a factor */
2585: /* Initial matrix pimij */
2586: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2587: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2588: /* 0, 0 , 1} */
2589: /*
2590: * and after some iteration: */
2591: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2592: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2593: /* 0, 0 , 1} */
2594: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2595: /* {0.51571254859325999, 0.4842874514067399, */
2596: /* 0.51326036147820708, 0.48673963852179264} */
2597: /* If we start from prlim again, prlim tends to a constant matrix */
2598:
2599: int i, ii,j,k;
1.247 brouard 2600: int first=0;
1.217 brouard 2601: double *min, *max, *meandiff, maxmax,sumnew=0.;
2602: /* double **matprod2(); */ /* test */
2603: double **out, cov[NCOVMAX+1], **bmij();
2604: double **newm;
1.218 brouard 2605: double **dnewm, **doldm, **dsavm; /* for use */
2606: double **oldm, **savm; /* for use */
2607:
1.217 brouard 2608: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2609: int ncvloop=0;
2610:
2611: min=vector(1,nlstate);
2612: max=vector(1,nlstate);
2613: meandiff=vector(1,nlstate);
2614:
1.218 brouard 2615: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2616: oldm=oldms; savm=savms;
2617:
2618: /* Starting with matrix unity */
2619: for (ii=1;ii<=nlstate+ndeath;ii++)
2620: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2621: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2622: }
2623:
2624: cov[1]=1.;
2625:
2626: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2627: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2628: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2629: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2630: ncvloop++;
1.218 brouard 2631: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2632: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2633: /* Covariates have to be included here again */
2634: cov[2]=agefin;
2635: if(nagesqr==1)
2636: cov[3]= agefin*agefin;;
1.242 brouard 2637: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2638: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2639: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2640: /* 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)); */
2641: }
2642: /* for (k=1; k<=cptcovn;k++) { */
2643: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2644: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2645: /* /\* 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])]); *\/ */
2646: /* } */
2647: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2648: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2649: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2650: /* 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]); */
2651: }
2652: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2653: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2654: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2655: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2656: for (k=1; k<=cptcovage;k++){ /* For product with age */
2657: if(Dummy[Tvar[Tage[k]]]){
2658: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2659: } else{
2660: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2661: }
2662: /* 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]); */
2663: }
2664: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2665: /* 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]); */
2666: if(Dummy[Tvard[k][1]==0]){
2667: if(Dummy[Tvard[k][2]==0]){
2668: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2669: }else{
2670: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2671: }
2672: }else{
2673: if(Dummy[Tvard[k][2]==0]){
2674: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2675: }else{
2676: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2677: }
2678: }
1.217 brouard 2679: }
2680:
2681: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2682: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2683: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2684: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2685: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2686: /* ij should be linked to the correct index of cov */
2687: /* age and covariate values ij are in 'cov', but we need to pass
2688: * ij for the observed prevalence at age and status and covariate
2689: * number: prevacurrent[(int)agefin][ii][ij]
2690: */
2691: /* 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 *\/ */
2692: /* 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 *\/ */
2693: 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 2694: savm=oldm;
2695: oldm=newm;
2696: for(j=1; j<=nlstate; j++){
2697: max[j]=0.;
2698: min[j]=1.;
2699: }
2700: for(j=1; j<=nlstate; j++){
2701: for(i=1;i<=nlstate;i++){
1.234 brouard 2702: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2703: bprlim[i][j]= newm[i][j];
2704: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2705: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2706: }
2707: }
1.218 brouard 2708:
1.217 brouard 2709: maxmax=0.;
2710: for(i=1; i<=nlstate; i++){
2711: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2712: maxmax=FMAX(maxmax,meandiff[i]);
2713: /* 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); */
2714: } /* j loop */
2715: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2716: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2717: if(maxmax < ftolpl){
1.220 brouard 2718: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2719: free_vector(min,1,nlstate);
2720: free_vector(max,1,nlstate);
2721: free_vector(meandiff,1,nlstate);
2722: return bprlim;
2723: }
2724: } /* age loop */
2725: /* After some age loop it doesn't converge */
1.247 brouard 2726: if(first){
2727: first=1;
2728: printf("Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. Others in log file only...\n\
2729: 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);
2730: }
2731: fprintf(ficlog,"Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. \n\
1.217 brouard 2732: 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);
2733: /* 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); */
2734: free_vector(min,1,nlstate);
2735: free_vector(max,1,nlstate);
2736: free_vector(meandiff,1,nlstate);
2737:
2738: return bprlim; /* should not reach here */
2739: }
2740:
1.126 brouard 2741: /*************** transition probabilities ***************/
2742:
2743: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2744: {
1.138 brouard 2745: /* According to parameters values stored in x and the covariate's values stored in cov,
2746: computes the probability to be observed in state j being in state i by appying the
2747: model to the ncovmodel covariates (including constant and age).
2748: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2749: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2750: ncth covariate in the global vector x is given by the formula:
2751: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2752: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2753: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2754: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2755: Outputs ps[i][j] the probability to be observed in j being in j according to
2756: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2757: */
2758: double s1, lnpijopii;
1.126 brouard 2759: /*double t34;*/
1.164 brouard 2760: int i,j, nc, ii, jj;
1.126 brouard 2761:
1.223 brouard 2762: for(i=1; i<= nlstate; i++){
2763: for(j=1; j<i;j++){
2764: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2765: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2766: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2767: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2768: }
2769: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2770: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2771: }
2772: for(j=i+1; j<=nlstate+ndeath;j++){
2773: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2774: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2775: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2776: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2777: }
2778: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2779: }
2780: }
1.218 brouard 2781:
1.223 brouard 2782: for(i=1; i<= nlstate; i++){
2783: s1=0;
2784: for(j=1; j<i; j++){
2785: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2786: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2787: }
2788: for(j=i+1; j<=nlstate+ndeath; j++){
2789: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2790: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2791: }
2792: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2793: ps[i][i]=1./(s1+1.);
2794: /* Computing other pijs */
2795: for(j=1; j<i; j++)
2796: ps[i][j]= exp(ps[i][j])*ps[i][i];
2797: for(j=i+1; j<=nlstate+ndeath; j++)
2798: ps[i][j]= exp(ps[i][j])*ps[i][i];
2799: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2800: } /* end i */
1.218 brouard 2801:
1.223 brouard 2802: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2803: for(jj=1; jj<= nlstate+ndeath; jj++){
2804: ps[ii][jj]=0;
2805: ps[ii][ii]=1;
2806: }
2807: }
1.218 brouard 2808:
2809:
1.223 brouard 2810: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2811: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2812: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2813: /* } */
2814: /* printf("\n "); */
2815: /* } */
2816: /* printf("\n ");printf("%lf ",cov[2]);*/
2817: /*
2818: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2819: goto end;*/
1.223 brouard 2820: return ps;
1.126 brouard 2821: }
2822:
1.218 brouard 2823: /*************** backward transition probabilities ***************/
2824:
2825: /* 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 ) */
2826: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2827: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2828: {
1.222 brouard 2829: /* Computes the backward probability at age agefin and covariate ij
2830: * and returns in **ps as well as **bmij.
2831: */
1.218 brouard 2832: int i, ii, j,k;
1.222 brouard 2833:
2834: double **out, **pmij();
2835: double sumnew=0.;
1.218 brouard 2836: double agefin;
1.222 brouard 2837:
2838: double **dnewm, **dsavm, **doldm;
2839: double **bbmij;
2840:
1.218 brouard 2841: doldm=ddoldms; /* global pointers */
1.222 brouard 2842: dnewm=ddnewms;
2843: dsavm=ddsavms;
2844:
2845: agefin=cov[2];
2846: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2847: the observed prevalence (with this covariate ij) */
2848: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2849: /* We do have the matrix Px in savm and we need pij */
2850: for (j=1;j<=nlstate+ndeath;j++){
2851: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2852: for (ii=1;ii<=nlstate;ii++){
2853: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2854: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2855: for (ii=1;ii<=nlstate+ndeath;ii++){
2856: if(sumnew >= 1.e-10){
2857: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2858: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2859: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2860: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2861: /* }else */
2862: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2863: }else{
1.242 brouard 2864: ;
2865: /* 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 2866: }
2867: } /*End ii */
2868: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2869: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2870: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2871: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2872: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2873: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2874: /* left Product of this matrix by diag matrix of prevalences (savm) */
2875: for (j=1;j<=nlstate+ndeath;j++){
2876: for (ii=1;ii<=nlstate+ndeath;ii++){
2877: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2878: }
2879: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2880: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2881: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2882: /* end bmij */
2883: return ps;
1.218 brouard 2884: }
1.217 brouard 2885: /*************** transition probabilities ***************/
2886:
1.218 brouard 2887: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2888: {
2889: /* According to parameters values stored in x and the covariate's values stored in cov,
2890: computes the probability to be observed in state j being in state i by appying the
2891: model to the ncovmodel covariates (including constant and age).
2892: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2893: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2894: ncth covariate in the global vector x is given by the formula:
2895: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2896: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2897: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2898: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2899: Outputs ps[i][j] the probability to be observed in j being in j according to
2900: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2901: */
2902: double s1, lnpijopii;
2903: /*double t34;*/
2904: int i,j, nc, ii, jj;
2905:
1.234 brouard 2906: for(i=1; i<= nlstate; i++){
2907: for(j=1; j<i;j++){
2908: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2909: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2910: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2911: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2912: }
2913: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2914: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2915: }
2916: for(j=i+1; j<=nlstate+ndeath;j++){
2917: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2918: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2919: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2920: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2921: }
2922: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2923: }
2924: }
2925:
2926: for(i=1; i<= nlstate; i++){
2927: s1=0;
2928: for(j=1; j<i; j++){
2929: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2930: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2931: }
2932: for(j=i+1; j<=nlstate+ndeath; j++){
2933: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2934: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2935: }
2936: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2937: ps[i][i]=1./(s1+1.);
2938: /* Computing other pijs */
2939: for(j=1; j<i; j++)
2940: ps[i][j]= exp(ps[i][j])*ps[i][i];
2941: for(j=i+1; j<=nlstate+ndeath; j++)
2942: ps[i][j]= exp(ps[i][j])*ps[i][i];
2943: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2944: } /* end i */
2945:
2946: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2947: for(jj=1; jj<= nlstate+ndeath; jj++){
2948: ps[ii][jj]=0;
2949: ps[ii][ii]=1;
2950: }
2951: }
2952: /* Added for backcast */ /* Transposed matrix too */
2953: for(jj=1; jj<= nlstate+ndeath; jj++){
2954: s1=0.;
2955: for(ii=1; ii<= nlstate+ndeath; ii++){
2956: s1+=ps[ii][jj];
2957: }
2958: for(ii=1; ii<= nlstate; ii++){
2959: ps[ii][jj]=ps[ii][jj]/s1;
2960: }
2961: }
2962: /* Transposition */
2963: for(jj=1; jj<= nlstate+ndeath; jj++){
2964: for(ii=jj; ii<= nlstate+ndeath; ii++){
2965: s1=ps[ii][jj];
2966: ps[ii][jj]=ps[jj][ii];
2967: ps[jj][ii]=s1;
2968: }
2969: }
2970: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2971: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2972: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2973: /* } */
2974: /* printf("\n "); */
2975: /* } */
2976: /* printf("\n ");printf("%lf ",cov[2]);*/
2977: /*
2978: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2979: goto end;*/
2980: return ps;
1.217 brouard 2981: }
2982:
2983:
1.126 brouard 2984: /**************** Product of 2 matrices ******************/
2985:
1.145 brouard 2986: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2987: {
2988: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2989: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2990: /* in, b, out are matrice of pointers which should have been initialized
2991: before: only the contents of out is modified. The function returns
2992: a pointer to pointers identical to out */
1.145 brouard 2993: int i, j, k;
1.126 brouard 2994: for(i=nrl; i<= nrh; i++)
1.145 brouard 2995: for(k=ncolol; k<=ncoloh; k++){
2996: out[i][k]=0.;
2997: for(j=ncl; j<=nch; j++)
2998: out[i][k] +=in[i][j]*b[j][k];
2999: }
1.126 brouard 3000: return out;
3001: }
3002:
3003:
3004: /************* Higher Matrix Product ***************/
3005:
1.235 brouard 3006: 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 3007: {
1.218 brouard 3008: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3009: 'nhstepm*hstepm*stepm' months (i.e. until
3010: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3011: nhstepm*hstepm matrices.
3012: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3013: (typically every 2 years instead of every month which is too big
3014: for the memory).
3015: Model is determined by parameters x and covariates have to be
3016: included manually here.
3017:
3018: */
3019:
3020: int i, j, d, h, k;
1.131 brouard 3021: double **out, cov[NCOVMAX+1];
1.126 brouard 3022: double **newm;
1.187 brouard 3023: double agexact;
1.214 brouard 3024: double agebegin, ageend;
1.126 brouard 3025:
3026: /* Hstepm could be zero and should return the unit matrix */
3027: for (i=1;i<=nlstate+ndeath;i++)
3028: for (j=1;j<=nlstate+ndeath;j++){
3029: oldm[i][j]=(i==j ? 1.0 : 0.0);
3030: po[i][j][0]=(i==j ? 1.0 : 0.0);
3031: }
3032: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3033: for(h=1; h <=nhstepm; h++){
3034: for(d=1; d <=hstepm; d++){
3035: newm=savm;
3036: /* Covariates have to be included here again */
3037: cov[1]=1.;
1.214 brouard 3038: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3039: cov[2]=agexact;
3040: if(nagesqr==1)
1.227 brouard 3041: cov[3]= agexact*agexact;
1.235 brouard 3042: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3043: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3044: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3045: /* 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)); */
3046: }
3047: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3048: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3049: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3050: /* 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]); */
3051: }
3052: for (k=1; k<=cptcovage;k++){
3053: if(Dummy[Tvar[Tage[k]]]){
3054: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3055: } else{
3056: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3057: }
3058: /* 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]); */
3059: }
3060: for (k=1; k<=cptcovprod;k++){ /* */
3061: /* 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]); */
3062: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3063: }
3064: /* for (k=1; k<=cptcovn;k++) */
3065: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3066: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3067: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3068: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3069: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3070:
3071:
1.126 brouard 3072: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3073: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3074: /* right multiplication of oldm by the current matrix */
1.126 brouard 3075: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3076: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3077: /* if((int)age == 70){ */
3078: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3079: /* for(i=1; i<=nlstate+ndeath; i++) { */
3080: /* printf("%d pmmij ",i); */
3081: /* for(j=1;j<=nlstate+ndeath;j++) { */
3082: /* printf("%f ",pmmij[i][j]); */
3083: /* } */
3084: /* printf(" oldm "); */
3085: /* for(j=1;j<=nlstate+ndeath;j++) { */
3086: /* printf("%f ",oldm[i][j]); */
3087: /* } */
3088: /* printf("\n"); */
3089: /* } */
3090: /* } */
1.126 brouard 3091: savm=oldm;
3092: oldm=newm;
3093: }
3094: for(i=1; i<=nlstate+ndeath; i++)
3095: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3096: po[i][j][h]=newm[i][j];
3097: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3098: }
1.128 brouard 3099: /*printf("h=%d ",h);*/
1.126 brouard 3100: } /* end h */
1.218 brouard 3101: /* printf("\n H=%d \n",h); */
1.126 brouard 3102: return po;
3103: }
3104:
1.217 brouard 3105: /************* Higher Back Matrix Product ***************/
1.218 brouard 3106: /* 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 3107: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3108: {
1.218 brouard 3109: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3110: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3111: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3112: nhstepm*hstepm matrices.
3113: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3114: (typically every 2 years instead of every month which is too big
1.217 brouard 3115: for the memory).
1.218 brouard 3116: Model is determined by parameters x and covariates have to be
3117: included manually here.
1.217 brouard 3118:
1.222 brouard 3119: */
1.217 brouard 3120:
3121: int i, j, d, h, k;
3122: double **out, cov[NCOVMAX+1];
3123: double **newm;
3124: double agexact;
3125: double agebegin, ageend;
1.222 brouard 3126: double **oldm, **savm;
1.217 brouard 3127:
1.222 brouard 3128: oldm=oldms;savm=savms;
1.217 brouard 3129: /* Hstepm could be zero and should return the unit matrix */
3130: for (i=1;i<=nlstate+ndeath;i++)
3131: for (j=1;j<=nlstate+ndeath;j++){
3132: oldm[i][j]=(i==j ? 1.0 : 0.0);
3133: po[i][j][0]=(i==j ? 1.0 : 0.0);
3134: }
3135: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3136: for(h=1; h <=nhstepm; h++){
3137: for(d=1; d <=hstepm; d++){
3138: newm=savm;
3139: /* Covariates have to be included here again */
3140: cov[1]=1.;
3141: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3142: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3143: cov[2]=agexact;
3144: if(nagesqr==1)
1.222 brouard 3145: cov[3]= agexact*agexact;
1.218 brouard 3146: for (k=1; k<=cptcovn;k++)
1.222 brouard 3147: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3148: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3149: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3150: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3151: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3152: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3153: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3154: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3155: /* 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 3156:
3157:
1.217 brouard 3158: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3159: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3160: /* Careful transposed matrix */
1.222 brouard 3161: /* age is in cov[2] */
1.218 brouard 3162: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3163: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3164: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3165: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3166: /* if((int)age == 70){ */
3167: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3168: /* for(i=1; i<=nlstate+ndeath; i++) { */
3169: /* printf("%d pmmij ",i); */
3170: /* for(j=1;j<=nlstate+ndeath;j++) { */
3171: /* printf("%f ",pmmij[i][j]); */
3172: /* } */
3173: /* printf(" oldm "); */
3174: /* for(j=1;j<=nlstate+ndeath;j++) { */
3175: /* printf("%f ",oldm[i][j]); */
3176: /* } */
3177: /* printf("\n"); */
3178: /* } */
3179: /* } */
3180: savm=oldm;
3181: oldm=newm;
3182: }
3183: for(i=1; i<=nlstate+ndeath; i++)
3184: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3185: po[i][j][h]=newm[i][j];
3186: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3187: }
3188: /*printf("h=%d ",h);*/
3189: } /* end h */
1.222 brouard 3190: /* printf("\n H=%d \n",h); */
1.217 brouard 3191: return po;
3192: }
3193:
3194:
1.162 brouard 3195: #ifdef NLOPT
3196: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3197: double fret;
3198: double *xt;
3199: int j;
3200: myfunc_data *d2 = (myfunc_data *) pd;
3201: /* xt = (p1-1); */
3202: xt=vector(1,n);
3203: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3204:
3205: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3206: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3207: printf("Function = %.12lf ",fret);
3208: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3209: printf("\n");
3210: free_vector(xt,1,n);
3211: return fret;
3212: }
3213: #endif
1.126 brouard 3214:
3215: /*************** log-likelihood *************/
3216: double func( double *x)
3217: {
1.226 brouard 3218: int i, ii, j, k, mi, d, kk;
3219: int ioffset=0;
3220: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3221: double **out;
3222: double lli; /* Individual log likelihood */
3223: int s1, s2;
1.228 brouard 3224: 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 3225: double bbh, survp;
3226: long ipmx;
3227: double agexact;
3228: /*extern weight */
3229: /* We are differentiating ll according to initial status */
3230: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3231: /*for(i=1;i<imx;i++)
3232: printf(" %d\n",s[4][i]);
3233: */
1.162 brouard 3234:
1.226 brouard 3235: ++countcallfunc;
1.162 brouard 3236:
1.226 brouard 3237: cov[1]=1.;
1.126 brouard 3238:
1.226 brouard 3239: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3240: ioffset=0;
1.226 brouard 3241: if(mle==1){
3242: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3243: /* Computes the values of the ncovmodel covariates of the model
3244: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3245: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3246: to be observed in j being in i according to the model.
3247: */
1.243 brouard 3248: ioffset=2+nagesqr ;
1.233 brouard 3249: /* Fixed */
1.234 brouard 3250: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3251: 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)*/
3252: }
1.226 brouard 3253: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3254: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3255: has been calculated etc */
3256: /* For an individual i, wav[i] gives the number of effective waves */
3257: /* We compute the contribution to Likelihood of each effective transition
3258: mw[mi][i] is real wave of the mi th effectve wave */
3259: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3260: s2=s[mw[mi+1][i]][i];
3261: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3262: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3263: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3264: */
3265: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3266: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3267: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3268: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3269: }
3270: for (ii=1;ii<=nlstate+ndeath;ii++)
3271: for (j=1;j<=nlstate+ndeath;j++){
3272: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3273: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3274: }
3275: for(d=0; d<dh[mi][i]; d++){
3276: newm=savm;
3277: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3278: cov[2]=agexact;
3279: if(nagesqr==1)
3280: cov[3]= agexact*agexact; /* Should be changed here */
3281: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3282: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3283: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3284: else
3285: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3286: }
3287: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3288: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3289: savm=oldm;
3290: oldm=newm;
3291: } /* end mult */
3292:
3293: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3294: /* But now since version 0.9 we anticipate for bias at large stepm.
3295: * If stepm is larger than one month (smallest stepm) and if the exact delay
3296: * (in months) between two waves is not a multiple of stepm, we rounded to
3297: * the nearest (and in case of equal distance, to the lowest) interval but now
3298: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3299: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3300: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3301: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3302: * -stepm/2 to stepm/2 .
3303: * For stepm=1 the results are the same as for previous versions of Imach.
3304: * For stepm > 1 the results are less biased than in previous versions.
3305: */
1.234 brouard 3306: s1=s[mw[mi][i]][i];
3307: s2=s[mw[mi+1][i]][i];
3308: bbh=(double)bh[mi][i]/(double)stepm;
3309: /* bias bh is positive if real duration
3310: * is higher than the multiple of stepm and negative otherwise.
3311: */
3312: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3313: if( s2 > nlstate){
3314: /* i.e. if s2 is a death state and if the date of death is known
3315: then the contribution to the likelihood is the probability to
3316: die between last step unit time and current step unit time,
3317: which is also equal to probability to die before dh
3318: minus probability to die before dh-stepm .
3319: In version up to 0.92 likelihood was computed
3320: as if date of death was unknown. Death was treated as any other
3321: health state: the date of the interview describes the actual state
3322: and not the date of a change in health state. The former idea was
3323: to consider that at each interview the state was recorded
3324: (healthy, disable or death) and IMaCh was corrected; but when we
3325: introduced the exact date of death then we should have modified
3326: the contribution of an exact death to the likelihood. This new
3327: contribution is smaller and very dependent of the step unit
3328: stepm. It is no more the probability to die between last interview
3329: and month of death but the probability to survive from last
3330: interview up to one month before death multiplied by the
3331: probability to die within a month. Thanks to Chris
3332: Jackson for correcting this bug. Former versions increased
3333: mortality artificially. The bad side is that we add another loop
3334: which slows down the processing. The difference can be up to 10%
3335: lower mortality.
3336: */
3337: /* If, at the beginning of the maximization mostly, the
3338: cumulative probability or probability to be dead is
3339: constant (ie = 1) over time d, the difference is equal to
3340: 0. out[s1][3] = savm[s1][3]: probability, being at state
3341: s1 at precedent wave, to be dead a month before current
3342: wave is equal to probability, being at state s1 at
3343: precedent wave, to be dead at mont of the current
3344: wave. Then the observed probability (that this person died)
3345: is null according to current estimated parameter. In fact,
3346: it should be very low but not zero otherwise the log go to
3347: infinity.
3348: */
1.183 brouard 3349: /* #ifdef INFINITYORIGINAL */
3350: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3351: /* #else */
3352: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3353: /* lli=log(mytinydouble); */
3354: /* else */
3355: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3356: /* #endif */
1.226 brouard 3357: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3358:
1.226 brouard 3359: } else if ( s2==-1 ) { /* alive */
3360: for (j=1,survp=0. ; j<=nlstate; j++)
3361: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3362: /*survp += out[s1][j]; */
3363: lli= log(survp);
3364: }
3365: else if (s2==-4) {
3366: for (j=3,survp=0. ; j<=nlstate; j++)
3367: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3368: lli= log(survp);
3369: }
3370: else if (s2==-5) {
3371: for (j=1,survp=0. ; j<=2; j++)
3372: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3373: lli= log(survp);
3374: }
3375: else{
3376: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3377: /* 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 */
3378: }
3379: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3380: /*if(lli ==000.0)*/
3381: /*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); */
3382: ipmx +=1;
3383: sw += weight[i];
3384: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3385: /* if (lli < log(mytinydouble)){ */
3386: /* 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); */
3387: /* 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]); */
3388: /* } */
3389: } /* end of wave */
3390: } /* end of individual */
3391: } else if(mle==2){
3392: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3393: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3394: for(mi=1; mi<= wav[i]-1; mi++){
3395: for (ii=1;ii<=nlstate+ndeath;ii++)
3396: for (j=1;j<=nlstate+ndeath;j++){
3397: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3398: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3399: }
3400: for(d=0; d<=dh[mi][i]; d++){
3401: newm=savm;
3402: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3403: cov[2]=agexact;
3404: if(nagesqr==1)
3405: cov[3]= agexact*agexact;
3406: for (kk=1; kk<=cptcovage;kk++) {
3407: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3408: }
3409: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3410: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3411: savm=oldm;
3412: oldm=newm;
3413: } /* end mult */
3414:
3415: s1=s[mw[mi][i]][i];
3416: s2=s[mw[mi+1][i]][i];
3417: bbh=(double)bh[mi][i]/(double)stepm;
3418: 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 */
3419: ipmx +=1;
3420: sw += weight[i];
3421: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3422: } /* end of wave */
3423: } /* end of individual */
3424: } else if(mle==3){ /* exponential inter-extrapolation */
3425: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3426: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3427: for(mi=1; mi<= wav[i]-1; mi++){
3428: for (ii=1;ii<=nlstate+ndeath;ii++)
3429: for (j=1;j<=nlstate+ndeath;j++){
3430: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3431: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3432: }
3433: for(d=0; d<dh[mi][i]; d++){
3434: newm=savm;
3435: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3436: cov[2]=agexact;
3437: if(nagesqr==1)
3438: cov[3]= agexact*agexact;
3439: for (kk=1; kk<=cptcovage;kk++) {
3440: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3441: }
3442: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3443: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3444: savm=oldm;
3445: oldm=newm;
3446: } /* end mult */
3447:
3448: s1=s[mw[mi][i]][i];
3449: s2=s[mw[mi+1][i]][i];
3450: bbh=(double)bh[mi][i]/(double)stepm;
3451: 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 */
3452: ipmx +=1;
3453: sw += weight[i];
3454: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3455: } /* end of wave */
3456: } /* end of individual */
3457: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3458: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3459: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3460: for(mi=1; mi<= wav[i]-1; mi++){
3461: for (ii=1;ii<=nlstate+ndeath;ii++)
3462: for (j=1;j<=nlstate+ndeath;j++){
3463: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3464: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3465: }
3466: for(d=0; d<dh[mi][i]; d++){
3467: newm=savm;
3468: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3469: cov[2]=agexact;
3470: if(nagesqr==1)
3471: cov[3]= agexact*agexact;
3472: for (kk=1; kk<=cptcovage;kk++) {
3473: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3474: }
1.126 brouard 3475:
1.226 brouard 3476: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3477: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3478: savm=oldm;
3479: oldm=newm;
3480: } /* end mult */
3481:
3482: s1=s[mw[mi][i]][i];
3483: s2=s[mw[mi+1][i]][i];
3484: if( s2 > nlstate){
3485: lli=log(out[s1][s2] - savm[s1][s2]);
3486: } else if ( s2==-1 ) { /* alive */
3487: for (j=1,survp=0. ; j<=nlstate; j++)
3488: survp += out[s1][j];
3489: lli= log(survp);
3490: }else{
3491: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3492: }
3493: ipmx +=1;
3494: sw += weight[i];
3495: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3496: /* 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 3497: } /* end of wave */
3498: } /* end of individual */
3499: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3500: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3501: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3502: for(mi=1; mi<= wav[i]-1; mi++){
3503: for (ii=1;ii<=nlstate+ndeath;ii++)
3504: for (j=1;j<=nlstate+ndeath;j++){
3505: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3506: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3507: }
3508: for(d=0; d<dh[mi][i]; d++){
3509: newm=savm;
3510: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3511: cov[2]=agexact;
3512: if(nagesqr==1)
3513: cov[3]= agexact*agexact;
3514: for (kk=1; kk<=cptcovage;kk++) {
3515: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3516: }
1.126 brouard 3517:
1.226 brouard 3518: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3519: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3520: savm=oldm;
3521: oldm=newm;
3522: } /* end mult */
3523:
3524: s1=s[mw[mi][i]][i];
3525: s2=s[mw[mi+1][i]][i];
3526: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3527: ipmx +=1;
3528: sw += weight[i];
3529: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3530: /*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]);*/
3531: } /* end of wave */
3532: } /* end of individual */
3533: } /* End of if */
3534: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3535: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3536: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3537: return -l;
1.126 brouard 3538: }
3539:
3540: /*************** log-likelihood *************/
3541: double funcone( double *x)
3542: {
1.228 brouard 3543: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3544: int i, ii, j, k, mi, d, kk;
1.228 brouard 3545: int ioffset=0;
1.131 brouard 3546: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3547: double **out;
3548: double lli; /* Individual log likelihood */
3549: double llt;
3550: int s1, s2;
1.228 brouard 3551: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3552:
1.126 brouard 3553: double bbh, survp;
1.187 brouard 3554: double agexact;
1.214 brouard 3555: double agebegin, ageend;
1.126 brouard 3556: /*extern weight */
3557: /* We are differentiating ll according to initial status */
3558: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3559: /*for(i=1;i<imx;i++)
3560: printf(" %d\n",s[4][i]);
3561: */
3562: cov[1]=1.;
3563:
3564: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3565: ioffset=0;
3566: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3567: /* ioffset=2+nagesqr+cptcovage; */
3568: ioffset=2+nagesqr;
1.232 brouard 3569: /* Fixed */
1.224 brouard 3570: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3571: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3572: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3573: 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)*/
3574: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3575: /* cov[2+6]=covar[Tvar[6]][i]; */
3576: /* cov[2+6]=covar[2][i]; V2 */
3577: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3578: /* cov[2+7]=covar[Tvar[7]][i]; */
3579: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3580: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3581: /* cov[2+9]=covar[Tvar[9]][i]; */
3582: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3583: }
1.232 brouard 3584: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3585: /* 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?)*\/ */
3586: /* } */
1.231 brouard 3587: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3588: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3589: /* } */
1.225 brouard 3590:
1.233 brouard 3591:
3592: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3593: /* Wave varying (but not age varying) */
3594: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3595: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3596: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3597: }
1.232 brouard 3598: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3599: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3600: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3601: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3602: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3603: /* 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 3604: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3605: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3606: /* /\* 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]); *\/ */
3607: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3608: /* } */
1.126 brouard 3609: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3610: for (j=1;j<=nlstate+ndeath;j++){
3611: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3612: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3613: }
1.214 brouard 3614:
3615: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3616: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3617: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3618: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3619: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3620: and mw[mi+1][i]. dh depends on stepm.*/
3621: newm=savm;
1.247 brouard 3622: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3623: cov[2]=agexact;
3624: if(nagesqr==1)
3625: cov[3]= agexact*agexact;
3626: for (kk=1; kk<=cptcovage;kk++) {
3627: if(!FixedV[Tvar[Tage[kk]]])
3628: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3629: else
3630: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3631: }
3632: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3633: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3634: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3635: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3636: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3637: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3638: savm=oldm;
3639: oldm=newm;
1.126 brouard 3640: } /* end mult */
3641:
3642: s1=s[mw[mi][i]][i];
3643: s2=s[mw[mi+1][i]][i];
1.217 brouard 3644: /* if(s2==-1){ */
3645: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3646: /* /\* exit(1); *\/ */
3647: /* } */
1.126 brouard 3648: bbh=(double)bh[mi][i]/(double)stepm;
3649: /* bias is positive if real duration
3650: * is higher than the multiple of stepm and negative otherwise.
3651: */
3652: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3653: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3654: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3655: for (j=1,survp=0. ; j<=nlstate; j++)
3656: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3657: lli= log(survp);
1.126 brouard 3658: }else if (mle==1){
1.242 brouard 3659: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3660: } else if(mle==2){
1.242 brouard 3661: 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 3662: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3663: 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 3664: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3665: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3666: } else{ /* mle=0 back to 1 */
1.242 brouard 3667: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3668: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3669: } /* End of if */
3670: ipmx +=1;
3671: sw += weight[i];
3672: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3673: /*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 3674: if(globpr){
1.246 brouard 3675: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3676: %11.6f %11.6f %11.6f ", \
1.242 brouard 3677: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3678: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3679: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3680: llt +=ll[k]*gipmx/gsw;
3681: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3682: }
3683: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3684: }
1.232 brouard 3685: } /* end of wave */
3686: } /* end of individual */
3687: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3688: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3689: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3690: if(globpr==0){ /* First time we count the contributions and weights */
3691: gipmx=ipmx;
3692: gsw=sw;
3693: }
3694: return -l;
1.126 brouard 3695: }
3696:
3697:
3698: /*************** function likelione ***********/
3699: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3700: {
3701: /* This routine should help understanding what is done with
3702: the selection of individuals/waves and
3703: to check the exact contribution to the likelihood.
3704: Plotting could be done.
3705: */
3706: int k;
3707:
3708: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3709: strcpy(fileresilk,"ILK_");
1.202 brouard 3710: strcat(fileresilk,fileresu);
1.126 brouard 3711: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3712: printf("Problem with resultfile: %s\n", fileresilk);
3713: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3714: }
1.214 brouard 3715: 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");
3716: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3717: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3718: for(k=1; k<=nlstate; k++)
3719: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3720: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3721: }
3722:
3723: *fretone=(*funcone)(p);
3724: if(*globpri !=0){
3725: fclose(ficresilk);
1.205 brouard 3726: if (mle ==0)
3727: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3728: else if(mle >=1)
3729: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3730: 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 3731:
1.208 brouard 3732:
3733: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3734: 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 3735: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3736: }
1.207 brouard 3737: 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 3738: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3739: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3740: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3741: fflush(fichtm);
1.205 brouard 3742: }
1.126 brouard 3743: return;
3744: }
3745:
3746:
3747: /*********** Maximum Likelihood Estimation ***************/
3748:
3749: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3750: {
1.165 brouard 3751: int i,j, iter=0;
1.126 brouard 3752: double **xi;
3753: double fret;
3754: double fretone; /* Only one call to likelihood */
3755: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3756:
3757: #ifdef NLOPT
3758: int creturn;
3759: nlopt_opt opt;
3760: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3761: double *lb;
3762: double minf; /* the minimum objective value, upon return */
3763: double * p1; /* Shifted parameters from 0 instead of 1 */
3764: myfunc_data dinst, *d = &dinst;
3765: #endif
3766:
3767:
1.126 brouard 3768: xi=matrix(1,npar,1,npar);
3769: for (i=1;i<=npar;i++)
3770: for (j=1;j<=npar;j++)
3771: xi[i][j]=(i==j ? 1.0 : 0.0);
3772: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3773: strcpy(filerespow,"POW_");
1.126 brouard 3774: strcat(filerespow,fileres);
3775: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3776: printf("Problem with resultfile: %s\n", filerespow);
3777: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3778: }
3779: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3780: for (i=1;i<=nlstate;i++)
3781: for(j=1;j<=nlstate+ndeath;j++)
3782: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3783: fprintf(ficrespow,"\n");
1.162 brouard 3784: #ifdef POWELL
1.126 brouard 3785: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3786: #endif
1.126 brouard 3787:
1.162 brouard 3788: #ifdef NLOPT
3789: #ifdef NEWUOA
3790: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3791: #else
3792: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3793: #endif
3794: lb=vector(0,npar-1);
3795: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3796: nlopt_set_lower_bounds(opt, lb);
3797: nlopt_set_initial_step1(opt, 0.1);
3798:
3799: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3800: d->function = func;
3801: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3802: nlopt_set_min_objective(opt, myfunc, d);
3803: nlopt_set_xtol_rel(opt, ftol);
3804: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3805: printf("nlopt failed! %d\n",creturn);
3806: }
3807: else {
3808: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3809: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3810: iter=1; /* not equal */
3811: }
3812: nlopt_destroy(opt);
3813: #endif
1.126 brouard 3814: free_matrix(xi,1,npar,1,npar);
3815: fclose(ficrespow);
1.203 brouard 3816: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3817: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3818: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3819:
3820: }
3821:
3822: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3823: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3824: {
3825: double **a,**y,*x,pd;
1.203 brouard 3826: /* double **hess; */
1.164 brouard 3827: int i, j;
1.126 brouard 3828: int *indx;
3829:
3830: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3831: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3832: void lubksb(double **a, int npar, int *indx, double b[]) ;
3833: void ludcmp(double **a, int npar, int *indx, double *d) ;
3834: double gompertz(double p[]);
1.203 brouard 3835: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3836:
3837: printf("\nCalculation of the hessian matrix. Wait...\n");
3838: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3839: for (i=1;i<=npar;i++){
1.203 brouard 3840: printf("%d-",i);fflush(stdout);
3841: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3842:
3843: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3844:
3845: /* printf(" %f ",p[i]);
3846: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3847: }
3848:
3849: for (i=1;i<=npar;i++) {
3850: for (j=1;j<=npar;j++) {
3851: if (j>i) {
1.203 brouard 3852: printf(".%d-%d",i,j);fflush(stdout);
3853: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3854: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3855:
3856: hess[j][i]=hess[i][j];
3857: /*printf(" %lf ",hess[i][j]);*/
3858: }
3859: }
3860: }
3861: printf("\n");
3862: fprintf(ficlog,"\n");
3863:
3864: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3865: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3866:
3867: a=matrix(1,npar,1,npar);
3868: y=matrix(1,npar,1,npar);
3869: x=vector(1,npar);
3870: indx=ivector(1,npar);
3871: for (i=1;i<=npar;i++)
3872: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3873: ludcmp(a,npar,indx,&pd);
3874:
3875: for (j=1;j<=npar;j++) {
3876: for (i=1;i<=npar;i++) x[i]=0;
3877: x[j]=1;
3878: lubksb(a,npar,indx,x);
3879: for (i=1;i<=npar;i++){
3880: matcov[i][j]=x[i];
3881: }
3882: }
3883:
3884: printf("\n#Hessian matrix#\n");
3885: fprintf(ficlog,"\n#Hessian matrix#\n");
3886: for (i=1;i<=npar;i++) {
3887: for (j=1;j<=npar;j++) {
1.203 brouard 3888: printf("%.6e ",hess[i][j]);
3889: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3890: }
3891: printf("\n");
3892: fprintf(ficlog,"\n");
3893: }
3894:
1.203 brouard 3895: /* printf("\n#Covariance matrix#\n"); */
3896: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3897: /* for (i=1;i<=npar;i++) { */
3898: /* for (j=1;j<=npar;j++) { */
3899: /* printf("%.6e ",matcov[i][j]); */
3900: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3901: /* } */
3902: /* printf("\n"); */
3903: /* fprintf(ficlog,"\n"); */
3904: /* } */
3905:
1.126 brouard 3906: /* Recompute Inverse */
1.203 brouard 3907: /* for (i=1;i<=npar;i++) */
3908: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3909: /* ludcmp(a,npar,indx,&pd); */
3910:
3911: /* printf("\n#Hessian matrix recomputed#\n"); */
3912:
3913: /* for (j=1;j<=npar;j++) { */
3914: /* for (i=1;i<=npar;i++) x[i]=0; */
3915: /* x[j]=1; */
3916: /* lubksb(a,npar,indx,x); */
3917: /* for (i=1;i<=npar;i++){ */
3918: /* y[i][j]=x[i]; */
3919: /* printf("%.3e ",y[i][j]); */
3920: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3921: /* } */
3922: /* printf("\n"); */
3923: /* fprintf(ficlog,"\n"); */
3924: /* } */
3925:
3926: /* Verifying the inverse matrix */
3927: #ifdef DEBUGHESS
3928: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3929:
1.203 brouard 3930: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3931: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3932:
3933: for (j=1;j<=npar;j++) {
3934: for (i=1;i<=npar;i++){
1.203 brouard 3935: printf("%.2f ",y[i][j]);
3936: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3937: }
3938: printf("\n");
3939: fprintf(ficlog,"\n");
3940: }
1.203 brouard 3941: #endif
1.126 brouard 3942:
3943: free_matrix(a,1,npar,1,npar);
3944: free_matrix(y,1,npar,1,npar);
3945: free_vector(x,1,npar);
3946: free_ivector(indx,1,npar);
1.203 brouard 3947: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3948:
3949:
3950: }
3951:
3952: /*************** hessian matrix ****************/
3953: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3954: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3955: int i;
3956: int l=1, lmax=20;
1.203 brouard 3957: double k1,k2, res, fx;
1.132 brouard 3958: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3959: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3960: int k=0,kmax=10;
3961: double l1;
3962:
3963: fx=func(x);
3964: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3965: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3966: l1=pow(10,l);
3967: delts=delt;
3968: for(k=1 ; k <kmax; k=k+1){
3969: delt = delta*(l1*k);
3970: p2[theta]=x[theta] +delt;
1.145 brouard 3971: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3972: p2[theta]=x[theta]-delt;
3973: k2=func(p2)-fx;
3974: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3975: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3976:
1.203 brouard 3977: #ifdef DEBUGHESSII
1.126 brouard 3978: 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);
3979: 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);
3980: #endif
3981: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3982: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3983: k=kmax;
3984: }
3985: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3986: k=kmax; l=lmax*10;
1.126 brouard 3987: }
3988: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3989: delts=delt;
3990: }
1.203 brouard 3991: } /* End loop k */
1.126 brouard 3992: }
3993: delti[theta]=delts;
3994: return res;
3995:
3996: }
3997:
1.203 brouard 3998: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 3999: {
4000: int i;
1.164 brouard 4001: int l=1, lmax=20;
1.126 brouard 4002: double k1,k2,k3,k4,res,fx;
1.132 brouard 4003: double p2[MAXPARM+1];
1.203 brouard 4004: int k, kmax=1;
4005: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4006:
4007: int firstime=0;
1.203 brouard 4008:
1.126 brouard 4009: fx=func(x);
1.203 brouard 4010: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4011: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4012: p2[thetai]=x[thetai]+delti[thetai]*k;
4013: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4014: k1=func(p2)-fx;
4015:
1.203 brouard 4016: p2[thetai]=x[thetai]+delti[thetai]*k;
4017: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4018: k2=func(p2)-fx;
4019:
1.203 brouard 4020: p2[thetai]=x[thetai]-delti[thetai]*k;
4021: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4022: k3=func(p2)-fx;
4023:
1.203 brouard 4024: p2[thetai]=x[thetai]-delti[thetai]*k;
4025: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4026: k4=func(p2)-fx;
1.203 brouard 4027: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4028: if(k1*k2*k3*k4 <0.){
1.208 brouard 4029: firstime=1;
1.203 brouard 4030: kmax=kmax+10;
1.208 brouard 4031: }
4032: if(kmax >=10 || firstime ==1){
1.246 brouard 4033: printf("Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you could increase ftol=%.2e\n",thetai,thetaj, ftol);
4034: fprintf(ficlog,"Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you could increase ftol=%.2e\n",thetai,thetaj, ftol);
1.203 brouard 4035: 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);
4036: 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);
4037: }
4038: #ifdef DEBUGHESSIJ
4039: v1=hess[thetai][thetai];
4040: v2=hess[thetaj][thetaj];
4041: cv12=res;
4042: /* Computing eigen value of Hessian matrix */
4043: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4044: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4045: if ((lc2 <0) || (lc1 <0) ){
4046: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4047: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4048: 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);
4049: 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);
4050: }
1.126 brouard 4051: #endif
4052: }
4053: return res;
4054: }
4055:
1.203 brouard 4056: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4057: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4058: /* { */
4059: /* int i; */
4060: /* int l=1, lmax=20; */
4061: /* double k1,k2,k3,k4,res,fx; */
4062: /* double p2[MAXPARM+1]; */
4063: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4064: /* int k=0,kmax=10; */
4065: /* double l1; */
4066:
4067: /* fx=func(x); */
4068: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4069: /* l1=pow(10,l); */
4070: /* delts=delt; */
4071: /* for(k=1 ; k <kmax; k=k+1){ */
4072: /* delt = delti*(l1*k); */
4073: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4074: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4075: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4076: /* k1=func(p2)-fx; */
4077:
4078: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4079: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4080: /* k2=func(p2)-fx; */
4081:
4082: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4083: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4084: /* k3=func(p2)-fx; */
4085:
4086: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4087: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4088: /* k4=func(p2)-fx; */
4089: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4090: /* #ifdef DEBUGHESSIJ */
4091: /* 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); */
4092: /* 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); */
4093: /* #endif */
4094: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4095: /* k=kmax; */
4096: /* } */
4097: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4098: /* k=kmax; l=lmax*10; */
4099: /* } */
4100: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4101: /* delts=delt; */
4102: /* } */
4103: /* } /\* End loop k *\/ */
4104: /* } */
4105: /* delti[theta]=delts; */
4106: /* return res; */
4107: /* } */
4108:
4109:
1.126 brouard 4110: /************** Inverse of matrix **************/
4111: void ludcmp(double **a, int n, int *indx, double *d)
4112: {
4113: int i,imax,j,k;
4114: double big,dum,sum,temp;
4115: double *vv;
4116:
4117: vv=vector(1,n);
4118: *d=1.0;
4119: for (i=1;i<=n;i++) {
4120: big=0.0;
4121: for (j=1;j<=n;j++)
4122: if ((temp=fabs(a[i][j])) > big) big=temp;
4123: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4124: vv[i]=1.0/big;
4125: }
4126: for (j=1;j<=n;j++) {
4127: for (i=1;i<j;i++) {
4128: sum=a[i][j];
4129: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4130: a[i][j]=sum;
4131: }
4132: big=0.0;
4133: for (i=j;i<=n;i++) {
4134: sum=a[i][j];
4135: for (k=1;k<j;k++)
4136: sum -= a[i][k]*a[k][j];
4137: a[i][j]=sum;
4138: if ( (dum=vv[i]*fabs(sum)) >= big) {
4139: big=dum;
4140: imax=i;
4141: }
4142: }
4143: if (j != imax) {
4144: for (k=1;k<=n;k++) {
4145: dum=a[imax][k];
4146: a[imax][k]=a[j][k];
4147: a[j][k]=dum;
4148: }
4149: *d = -(*d);
4150: vv[imax]=vv[j];
4151: }
4152: indx[j]=imax;
4153: if (a[j][j] == 0.0) a[j][j]=TINY;
4154: if (j != n) {
4155: dum=1.0/(a[j][j]);
4156: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4157: }
4158: }
4159: free_vector(vv,1,n); /* Doesn't work */
4160: ;
4161: }
4162:
4163: void lubksb(double **a, int n, int *indx, double b[])
4164: {
4165: int i,ii=0,ip,j;
4166: double sum;
4167:
4168: for (i=1;i<=n;i++) {
4169: ip=indx[i];
4170: sum=b[ip];
4171: b[ip]=b[i];
4172: if (ii)
4173: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4174: else if (sum) ii=i;
4175: b[i]=sum;
4176: }
4177: for (i=n;i>=1;i--) {
4178: sum=b[i];
4179: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4180: b[i]=sum/a[i][i];
4181: }
4182: }
4183:
4184: void pstamp(FILE *fichier)
4185: {
1.196 brouard 4186: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4187: }
4188:
4189: /************ Frequencies ********************/
1.251 brouard 4190: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4191: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4192: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4193: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4194:
1.250 brouard 4195: int i, m, jk, j1, bool, z1,j, k, iv, jj=0;
1.226 brouard 4196: int iind=0, iage=0;
4197: int mi; /* Effective wave */
4198: int first;
4199: double ***freq; /* Frequencies */
4200: double *meanq;
4201: double **meanqt;
4202: double *pp, **prop, *posprop, *pospropt;
4203: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4204: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4205: double agebegin, ageend;
4206:
4207: pp=vector(1,nlstate);
1.251 brouard 4208: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4209: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4210: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4211: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4212: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4213: meanqt=matrix(1,lastpass,1,nqtveff);
4214: strcpy(fileresp,"P_");
4215: strcat(fileresp,fileresu);
4216: /*strcat(fileresphtm,fileresu);*/
4217: if((ficresp=fopen(fileresp,"w"))==NULL) {
4218: printf("Problem with prevalence resultfile: %s\n", fileresp);
4219: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4220: exit(0);
4221: }
1.240 brouard 4222:
1.226 brouard 4223: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4224: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4225: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4226: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4227: fflush(ficlog);
4228: exit(70);
4229: }
4230: else{
4231: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4232: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4233: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4234: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4235: }
1.237 brouard 4236: 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 4237:
1.226 brouard 4238: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4239: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4240: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4241: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4242: fflush(ficlog);
4243: exit(70);
1.240 brouard 4244: } else{
1.226 brouard 4245: 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 4246: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4247: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4248: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4249: }
1.240 brouard 4250: 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);
4251:
1.251 brouard 4252: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4253: j1=0;
1.126 brouard 4254:
1.227 brouard 4255: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4256: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4257: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4258:
4259:
1.226 brouard 4260: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4261: reference=low_education V1=0,V2=0
4262: med_educ V1=1 V2=0,
4263: high_educ V1=0 V2=1
4264: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4265: */
1.249 brouard 4266: dateintsum=0;
4267: k2cpt=0;
4268:
1.251 brouard 4269: for (j = 0; j <= cptcoveff; j+=cptcoveff){ /* j= 0 constant model */
4270: first=1;
4271: 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 */
4272: posproptt=0.;
4273: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4274: scanf("%d", i);*/
4275: for (i=-5; i<=nlstate+ndeath; i++)
4276: for (jk=-5; jk<=nlstate+ndeath; jk++)
4277: for(m=iagemin; m <= iagemax+3; m++)
4278: freq[i][jk][m]=0;
4279:
4280: for (i=1; i<=nlstate; i++) {
1.240 brouard 4281: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4282: prop[i][m]=0;
4283: posprop[i]=0;
4284: pospropt[i]=0;
4285: }
4286: /* for (z1=1; z1<= nqfveff; z1++) { */
4287: /* meanq[z1]+=0.; */
4288: /* for(m=1;m<=lastpass;m++){ */
4289: /* meanqt[m][z1]=0.; */
4290: /* } */
4291: /* } */
4292:
4293: /* dateintsum=0; */
4294: /* k2cpt=0; */
4295:
4296: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4297: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4298: bool=1;
4299: if(j !=0){
4300: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4301: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4302: /* for (z1=1; z1<= nqfveff; z1++) { */
4303: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4304: /* } */
4305: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4306: /* if(Tvaraff[z1] ==-20){ */
4307: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4308: /* }else if(Tvaraff[z1] ==-10){ */
4309: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4310: /* }else */
4311: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4312: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4313: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4314: /* 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",
4315: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4316: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4317: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4318: } /* Onlyf fixed */
4319: } /* end z1 */
4320: } /* cptcovn > 0 */
4321: } /* end any */
4322: }/* end j==0 */
4323: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4324: /* for(m=firstpass; m<=lastpass; m++){ */
4325: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4326: m=mw[mi][iind];
4327: if(j!=0){
4328: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4329: for (z1=1; z1<=cptcoveff; z1++) {
4330: if( Fixed[Tmodelind[z1]]==1){
4331: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4332: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4333: value is -1, we don't select. It differs from the
4334: constant and age model which counts them. */
4335: bool=0; /* not selected */
4336: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4337: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4338: bool=0;
4339: }
4340: }
4341: }
4342: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4343: } /* end j==0 */
4344: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4345: if(bool==1){
4346: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4347: and mw[mi+1][iind]. dh depends on stepm. */
4348: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4349: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4350: if(m >=firstpass && m <=lastpass){
4351: k2=anint[m][iind]+(mint[m][iind]/12.);
4352: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4353: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4354: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4355: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4356: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4357: if (m<lastpass) {
4358: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4359: /* 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]); */
4360: if(s[m][iind]==-1)
4361: 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.));
4362: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4363: /* if((int)agev[m][iind] == 55) */
4364: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4365: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4366: freq[s[m][iind]][s[m+1][iind]][iagemax+3] += weight[iind]; /* Total is in iagemax+3 *//* At age of beginning of transition, where status is known */
1.234 brouard 4367: }
1.251 brouard 4368: } /* end if between passes */
4369: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4370: dateintsum=dateintsum+k2; /* on all covariates ?*/
4371: k2cpt++;
4372: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4373: }
1.251 brouard 4374: }else{
4375: bool=1;
4376: }/* end bool 2 */
4377: } /* end m */
4378: } /* end bool */
4379: } /* end iind = 1 to imx */
4380: /* prop[s][age] is feeded for any initial and valid live state as well as
4381: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4382:
4383:
4384: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4385: pstamp(ficresp);
4386: if (cptcoveff>0 && j!=0){
4387: printf( "\n#********** Variable ");
4388: fprintf(ficresp, "\n#********** Variable ");
4389: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4390: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4391: fprintf(ficlog, "\n#********** Variable ");
4392: for (z1=1; z1<=cptcoveff; z1++){
4393: if(!FixedV[Tvaraff[z1]]){
4394: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4395: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4396: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4397: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4398: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4399: }else{
1.251 brouard 4400: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4401: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4402: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4403: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4404: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4405: }
4406: }
4407: printf( "**********\n#");
4408: fprintf(ficresp, "**********\n#");
4409: fprintf(ficresphtm, "**********</h3>\n");
4410: fprintf(ficresphtmfr, "**********</h3>\n");
4411: fprintf(ficlog, "**********\n");
4412: }
4413: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4414: for(i=1; i<=nlstate;i++) {
4415: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4416: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4417: }
4418: fprintf(ficresp, "\n");
4419: fprintf(ficresphtm, "\n");
4420:
4421: /* Header of frequency table by age */
4422: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4423: fprintf(ficresphtmfr,"<th>Age</th> ");
4424: for(jk=-1; jk <=nlstate+ndeath; jk++){
4425: for(m=-1; m <=nlstate+ndeath; m++){
4426: if(jk!=0 && m!=0)
4427: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4428: }
1.226 brouard 4429: }
1.251 brouard 4430: fprintf(ficresphtmfr, "\n");
4431:
4432: /* For each age */
4433: for(iage=iagemin; iage <= iagemax+3; iage++){
4434: fprintf(ficresphtm,"<tr>");
4435: if(iage==iagemax+1){
4436: fprintf(ficlog,"1");
4437: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4438: }else if(iage==iagemax+2){
4439: fprintf(ficlog,"0");
4440: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4441: }else if(iage==iagemax+3){
4442: fprintf(ficlog,"Total");
4443: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4444: }else{
1.240 brouard 4445: if(first==1){
1.251 brouard 4446: first=0;
4447: printf("See log file for details...\n");
4448: }
4449: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4450: fprintf(ficlog,"Age %d", iage);
4451: }
4452: for(jk=1; jk <=nlstate ; jk++){
4453: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4454: pp[jk] += freq[jk][m][iage];
4455: }
4456: for(jk=1; jk <=nlstate ; jk++){
4457: for(m=-1, pos=0; m <=0 ; m++)
4458: pos += freq[jk][m][iage];
4459: if(pp[jk]>=1.e-10){
4460: if(first==1){
4461: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4462: }
4463: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4464: }else{
4465: if(first==1)
4466: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4467: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4468: }
4469: }
4470:
1.251 brouard 4471: for(jk=1; jk <=nlstate ; jk++){
4472: /* posprop[jk]=0; */
4473: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4474: pp[jk] += freq[jk][m][iage];
4475: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4476:
4477: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4478: pos += pp[jk]; /* pos is the total number of transitions until this age */
4479: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4480: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4481: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4482: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4483: }
1.251 brouard 4484: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4485: if(pos>=1.e-5){
1.251 brouard 4486: if(first==1)
4487: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4488: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4489: }else{
4490: if(first==1)
4491: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4492: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4493: }
4494: if( iage <= iagemax){
4495: if(pos>=1.e-5){
4496: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4497: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4498: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4499: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4500: }
4501: else{
4502: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4503: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4504: }
1.240 brouard 4505: }
1.251 brouard 4506: pospropt[jk] +=posprop[jk];
4507: } /* end loop jk */
4508: /* pospropt=0.; */
4509: for(jk=-1; jk <=nlstate+ndeath; jk++){
4510: for(m=-1; m <=nlstate+ndeath; m++){
4511: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4512: if(first==1){
4513: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4514: }
4515: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4516: }
4517: if(jk!=0 && m!=0)
4518: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4519: }
1.251 brouard 4520: } /* end loop jk */
4521: posproptt=0.;
4522: for(jk=1; jk <=nlstate; jk++){
4523: posproptt += pospropt[jk];
4524: }
4525: fprintf(ficresphtmfr,"</tr>\n ");
4526: if(iage <= iagemax){
4527: fprintf(ficresp,"\n");
4528: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4529: }
1.251 brouard 4530: if(first==1)
4531: printf("Others in log...\n");
4532: fprintf(ficlog,"\n");
4533: } /* end loop age iage */
4534: fprintf(ficresphtm,"<tr><th>Tot</th>");
4535: for(jk=1; jk <=nlstate ; jk++){
4536: if(posproptt < 1.e-5){
4537: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4538: }else{
4539: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4540: }
1.226 brouard 4541: }
1.251 brouard 4542: fprintf(ficresphtm,"</tr>\n");
4543: fprintf(ficresphtm,"</table>\n");
4544: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4545: if(posproptt < 1.e-5){
1.251 brouard 4546: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4547: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4548: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4549: invalidvarcomb[j1]=1;
1.226 brouard 4550: }else{
1.251 brouard 4551: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4552: invalidvarcomb[j1]=0;
1.226 brouard 4553: }
1.251 brouard 4554: fprintf(ficresphtmfr,"</table>\n");
4555: fprintf(ficlog,"\n");
4556: if(j!=0){
4557: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4558: for(i=1,jk=1; i <=nlstate; i++){
4559: for(k=1; k <=(nlstate+ndeath); k++){
4560: if (k != i) {
4561: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
4562: if(jj==1){ /* Constant case */
4563: if(j1==1){ /* All dummy covariates to zero */
4564: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4565: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 ! brouard 4566: printf("%d%d ",i,k);
! 4567: fprintf(ficlog,"%d%d ",i,k);
! 4568: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]),freq[i][k][iagemax+3]/freq[i][i][iagemax+3], sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]));
! 4569: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
! 4570: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4571: }
1.252 ! brouard 4572: }else if(jj==2 || nagesqr==1){ /* age or age*age parameter */
! 4573: ;
! 4574: }else if( j1!=1 && (j1==2 || (log(j1-1.)/log(2.)-(int)(log(j1-1.)/log(2.))) <0.010) && ( TvarsDind[(int)(log(j1-1.)/log(2.))+1]+2+nagesqr == jj) && Dummy[jj-2-nagesqr]==0){ /* We want only if the position, jj, in model corresponds to unique covariate equal to 1 in j1 combination */
! 4575: printf("j1=%d, jj=%d, (int)(log(j1-1.)/log(2.))+1=%d, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(int)(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
! 4576: printf("j1=%d, jj=%d, (log(j1-1.)/log(2.))+1=%f, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
1.251 brouard 4577: pstart[jk]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 ! brouard 4578: printf("%d%d ",i,k);
! 4579: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4580: printf("jk=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",jk,i,k,jk,p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3],freq[i][k][iagemax+4],freq[i][i][iagemax+4], log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4])),(freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]), sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]+1/freq[i][k][iagemax+4]+1/freq[i][i][iagemax+4]));
4581: }else{ /* Other cases, like quantitative fixed or varying covariates */
4582: ;
4583: }
4584: /* printf("%12.7f )", param[i][jj][k]); */
4585: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4586: jk++;
4587: } /* end jj */
4588: } /* end k!= i */
4589: } /* end k */
4590: } /* end i, jk */
4591: } /* end j !=0 */
4592: } /* end selected combination of covariate j1 */
4593: if(j==0){ /* We can estimate starting values from the occurences in each case */
4594: printf("#Freqsummary: Starting values for the constants:\n");
4595: fprintf(ficlog,"\n");
4596: for(i=1,jk=1; i <=nlstate; i++){
4597: for(k=1; k <=(nlstate+ndeath); k++){
4598: if (k != i) {
4599: printf("%d%d ",i,k);
4600: fprintf(ficlog,"%d%d ",i,k);
4601: for(jj=1; jj <=ncovmodel; jj++){
4602: if(jj==1){
4603: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4604: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4605: }
4606: /* printf("%12.7f )", param[i][jj][k]); */
4607: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4608: jk++;
1.250 brouard 4609: }
1.251 brouard 4610: printf("\n");
4611: fprintf(ficlog,"\n");
1.250 brouard 4612: }
4613: }
4614: }
1.251 brouard 4615: printf("#Freqsummary\n");
4616: fprintf(ficlog,"\n");
4617: for(jk=-1; jk <=nlstate+ndeath; jk++){
4618: for(m=-1; m <=nlstate+ndeath; m++){
4619: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4620: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4621: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4622: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4623: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4624: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4625: /* } */
4626: }
4627: } /* end loop jk */
4628:
4629: printf("\n");
4630: fprintf(ficlog,"\n");
4631: } /* end j=0 */
1.249 brouard 4632: } /* end j */
1.252 ! brouard 4633:
! 4634: if(mle == -2){
! 4635: for(i=1, jk=1; i <=nlstate; i++){
! 4636: for(j=1; j <=nlstate+ndeath; j++){
! 4637: if(j!=i){
! 4638: /*ca[0]= k+'a'-1;ca[1]='\0';*/
! 4639: printf("%1d%1d",i,j);
! 4640: fprintf(ficparo,"%1d%1d",i,j);
! 4641: for(k=1; k<=ncovmodel;k++){
! 4642: /* printf(" %lf",param[i][j][k]); */
! 4643: /* fprintf(ficparo," %lf",param[i][j][k]); */
! 4644: p[jk]=pstart[jk];
! 4645: printf(" %f ",pstart[jk]);
! 4646: fprintf(ficparo," %f ",pstart[jk]);
! 4647: jk++;
! 4648: }
! 4649: printf("\n");
! 4650: fprintf(ficparo,"\n");
! 4651: }
! 4652: }
! 4653: }
! 4654: } /* end mle=-2 */
1.226 brouard 4655: dateintmean=dateintsum/k2cpt;
1.240 brouard 4656:
1.226 brouard 4657: fclose(ficresp);
4658: fclose(ficresphtm);
4659: fclose(ficresphtmfr);
4660: free_vector(meanq,1,nqfveff);
4661: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.251 brouard 4662: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4663: free_vector(pospropt,1,nlstate);
4664: free_vector(posprop,1,nlstate);
1.251 brouard 4665: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4666: free_vector(pp,1,nlstate);
4667: /* End of freqsummary */
4668: }
1.126 brouard 4669:
4670: /************ Prevalence ********************/
1.227 brouard 4671: 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)
4672: {
4673: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4674: in each health status at the date of interview (if between dateprev1 and dateprev2).
4675: We still use firstpass and lastpass as another selection.
4676: */
1.126 brouard 4677:
1.227 brouard 4678: int i, m, jk, j1, bool, z1,j, iv;
4679: int mi; /* Effective wave */
4680: int iage;
4681: double agebegin, ageend;
4682:
4683: double **prop;
4684: double posprop;
4685: double y2; /* in fractional years */
4686: int iagemin, iagemax;
4687: int first; /** to stop verbosity which is redirected to log file */
4688:
4689: iagemin= (int) agemin;
4690: iagemax= (int) agemax;
4691: /*pp=vector(1,nlstate);*/
1.251 brouard 4692: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4693: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4694: j1=0;
1.222 brouard 4695:
1.227 brouard 4696: /*j=cptcoveff;*/
4697: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4698:
1.227 brouard 4699: first=1;
4700: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4701: for (i=1; i<=nlstate; i++)
1.251 brouard 4702: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4703: prop[i][iage]=0.0;
4704: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4705: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4706: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4707:
4708: for (i=1; i<=imx; i++) { /* Each individual */
4709: bool=1;
4710: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4711: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4712: m=mw[mi][i];
4713: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4714: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4715: for (z1=1; z1<=cptcoveff; z1++){
4716: if( Fixed[Tmodelind[z1]]==1){
4717: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4718: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4719: bool=0;
4720: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4721: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4722: bool=0;
4723: }
4724: }
4725: if(bool==1){ /* Otherwise we skip that wave/person */
4726: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4727: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4728: if(m >=firstpass && m <=lastpass){
4729: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4730: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4731: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4732: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4733: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4734: 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);
4735: exit(1);
4736: }
4737: if (s[m][i]>0 && s[m][i]<=nlstate) {
4738: /*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]]);*/
4739: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4740: prop[s[m][i]][iagemax+3] += weight[i];
4741: } /* end valid statuses */
4742: } /* end selection of dates */
4743: } /* end selection of waves */
4744: } /* end bool */
4745: } /* end wave */
4746: } /* end individual */
4747: for(i=iagemin; i <= iagemax+3; i++){
4748: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4749: posprop += prop[jk][i];
4750: }
4751:
4752: for(jk=1; jk <=nlstate ; jk++){
4753: if( i <= iagemax){
4754: if(posprop>=1.e-5){
4755: probs[i][jk][j1]= prop[jk][i]/posprop;
4756: } else{
4757: if(first==1){
4758: first=0;
4759: 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]);
4760: }
4761: }
4762: }
4763: }/* end jk */
4764: }/* end i */
1.222 brouard 4765: /*} *//* end i1 */
1.227 brouard 4766: } /* end j1 */
1.222 brouard 4767:
1.227 brouard 4768: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4769: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4770: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4771: } /* End of prevalence */
1.126 brouard 4772:
4773: /************* Waves Concatenation ***************/
4774:
4775: 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)
4776: {
4777: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4778: Death is a valid wave (if date is known).
4779: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4780: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4781: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4782: */
1.126 brouard 4783:
1.224 brouard 4784: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4785: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4786: double sum=0., jmean=0.;*/
1.224 brouard 4787: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4788: int j, k=0,jk, ju, jl;
4789: double sum=0.;
4790: first=0;
1.214 brouard 4791: firstwo=0;
1.217 brouard 4792: firsthree=0;
1.218 brouard 4793: firstfour=0;
1.164 brouard 4794: jmin=100000;
1.126 brouard 4795: jmax=-1;
4796: jmean=0.;
1.224 brouard 4797:
4798: /* Treating live states */
1.214 brouard 4799: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4800: mi=0; /* First valid wave */
1.227 brouard 4801: mli=0; /* Last valid wave */
1.126 brouard 4802: m=firstpass;
1.214 brouard 4803: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4804: 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 */
4805: mli=m-1;/* mw[++mi][i]=m-1; */
4806: }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 */
4807: mw[++mi][i]=m;
4808: mli=m;
1.224 brouard 4809: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4810: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4811: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4812: }
1.227 brouard 4813: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4814: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4815: break;
1.224 brouard 4816: #else
1.227 brouard 4817: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4818: if(firsthree == 0){
4819: 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);
4820: firsthree=1;
4821: }
4822: 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);
4823: mw[++mi][i]=m;
4824: mli=m;
4825: }
4826: if(s[m][i]==-2){ /* Vital status is really unknown */
4827: nbwarn++;
4828: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4829: 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);
4830: 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);
4831: }
4832: break;
4833: }
4834: break;
1.224 brouard 4835: #endif
1.227 brouard 4836: }/* End m >= lastpass */
1.126 brouard 4837: }/* end while */
1.224 brouard 4838:
1.227 brouard 4839: /* 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 4840: /* After last pass */
1.224 brouard 4841: /* Treating death states */
1.214 brouard 4842: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4843: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4844: /* } */
1.126 brouard 4845: mi++; /* Death is another wave */
4846: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4847: /* Only death is a correct wave */
1.126 brouard 4848: mw[mi][i]=m;
1.224 brouard 4849: }
4850: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4851: 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 4852: /* m++; */
4853: /* mi++; */
4854: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4855: /* mw[mi][i]=m; */
1.218 brouard 4856: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4857: 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 */
4858: nbwarn++;
4859: if(firstfiv==0){
4860: 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 );
4861: firstfiv=1;
4862: }else{
4863: 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 );
4864: }
4865: }else{ /* Death occured afer last wave potential bias */
4866: nberr++;
4867: if(firstwo==0){
4868: 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 );
4869: firstwo=1;
4870: }
4871: 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 );
4872: }
1.218 brouard 4873: }else{ /* end date of interview is known */
1.227 brouard 4874: /* death is known but not confirmed by death status at any wave */
4875: if(firstfour==0){
4876: 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 );
4877: firstfour=1;
4878: }
4879: 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 4880: }
1.224 brouard 4881: } /* end if date of death is known */
4882: #endif
4883: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4884: /* wav[i]=mw[mi][i]; */
1.126 brouard 4885: if(mi==0){
4886: nbwarn++;
4887: if(first==0){
1.227 brouard 4888: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4889: first=1;
1.126 brouard 4890: }
4891: if(first==1){
1.227 brouard 4892: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4893: }
4894: } /* end mi==0 */
4895: } /* End individuals */
1.214 brouard 4896: /* wav and mw are no more changed */
1.223 brouard 4897:
1.214 brouard 4898:
1.126 brouard 4899: for(i=1; i<=imx; i++){
4900: for(mi=1; mi<wav[i];mi++){
4901: if (stepm <=0)
1.227 brouard 4902: dh[mi][i]=1;
1.126 brouard 4903: else{
1.227 brouard 4904: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4905: if (agedc[i] < 2*AGESUP) {
4906: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4907: if(j==0) j=1; /* Survives at least one month after exam */
4908: else if(j<0){
4909: nberr++;
4910: 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]);
4911: j=1; /* Temporary Dangerous patch */
4912: 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);
4913: 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]);
4914: 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);
4915: }
4916: k=k+1;
4917: if (j >= jmax){
4918: jmax=j;
4919: ijmax=i;
4920: }
4921: if (j <= jmin){
4922: jmin=j;
4923: ijmin=i;
4924: }
4925: sum=sum+j;
4926: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
4927: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
4928: }
4929: }
4930: else{
4931: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 4932: /* 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 4933:
1.227 brouard 4934: k=k+1;
4935: if (j >= jmax) {
4936: jmax=j;
4937: ijmax=i;
4938: }
4939: else if (j <= jmin){
4940: jmin=j;
4941: ijmin=i;
4942: }
4943: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
4944: /*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]);*/
4945: if(j<0){
4946: nberr++;
4947: 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]);
4948: 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]);
4949: }
4950: sum=sum+j;
4951: }
4952: jk= j/stepm;
4953: jl= j -jk*stepm;
4954: ju= j -(jk+1)*stepm;
4955: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
4956: if(jl==0){
4957: dh[mi][i]=jk;
4958: bh[mi][i]=0;
4959: }else{ /* We want a negative bias in order to only have interpolation ie
4960: * to avoid the price of an extra matrix product in likelihood */
4961: dh[mi][i]=jk+1;
4962: bh[mi][i]=ju;
4963: }
4964: }else{
4965: if(jl <= -ju){
4966: dh[mi][i]=jk;
4967: bh[mi][i]=jl; /* bias is positive if real duration
4968: * is higher than the multiple of stepm and negative otherwise.
4969: */
4970: }
4971: else{
4972: dh[mi][i]=jk+1;
4973: bh[mi][i]=ju;
4974: }
4975: if(dh[mi][i]==0){
4976: dh[mi][i]=1; /* At least one step */
4977: bh[mi][i]=ju; /* At least one step */
4978: /* 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);*/
4979: }
4980: } /* end if mle */
1.126 brouard 4981: }
4982: } /* end wave */
4983: }
4984: jmean=sum/k;
4985: 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 4986: 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 4987: }
1.126 brouard 4988:
4989: /*********** Tricode ****************************/
1.220 brouard 4990: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 4991: {
4992: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
4993: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
4994: * Boring subroutine which should only output nbcode[Tvar[j]][k]
4995: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
4996: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
4997: */
1.130 brouard 4998:
1.242 brouard 4999: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5000: int modmaxcovj=0; /* Modality max of covariates j */
5001: int cptcode=0; /* Modality max of covariates j */
5002: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5003:
5004:
1.242 brouard 5005: /* cptcoveff=0; */
5006: /* *cptcov=0; */
1.126 brouard 5007:
1.242 brouard 5008: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5009:
1.242 brouard 5010: /* Loop on covariates without age and products and no quantitative variable */
5011: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5012: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5013: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5014: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5015: switch(Fixed[k]) {
5016: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5017: 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*/
5018: ij=(int)(covar[Tvar[k]][i]);
5019: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5020: * If product of Vn*Vm, still boolean *:
5021: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5022: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5023: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5024: modality of the nth covariate of individual i. */
5025: if (ij > modmaxcovj)
5026: modmaxcovj=ij;
5027: else if (ij < modmincovj)
5028: modmincovj=ij;
5029: if ((ij < -1) && (ij > NCOVMAX)){
5030: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5031: exit(1);
5032: }else
5033: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5034: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5035: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5036: /* getting the maximum value of the modality of the covariate
5037: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5038: female ies 1, then modmaxcovj=1.
5039: */
5040: } /* end for loop on individuals i */
5041: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5042: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5043: cptcode=modmaxcovj;
5044: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5045: /*for (i=0; i<=cptcode; i++) {*/
5046: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5047: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5048: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5049: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5050: if( j != -1){
5051: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5052: covariate for which somebody answered excluding
5053: undefined. Usually 2: 0 and 1. */
5054: }
5055: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5056: covariate for which somebody answered including
5057: undefined. Usually 3: -1, 0 and 1. */
5058: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5059: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5060: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5061:
1.242 brouard 5062: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5063: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5064: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5065: /* modmincovj=3; modmaxcovj = 7; */
5066: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5067: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5068: /* defining two dummy variables: variables V1_1 and V1_2.*/
5069: /* nbcode[Tvar[j]][ij]=k; */
5070: /* nbcode[Tvar[j]][1]=0; */
5071: /* nbcode[Tvar[j]][2]=1; */
5072: /* nbcode[Tvar[j]][3]=2; */
5073: /* To be continued (not working yet). */
5074: ij=0; /* ij is similar to i but can jump over null modalities */
5075: 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*/
5076: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5077: break;
5078: }
5079: ij++;
5080: 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*/
5081: cptcode = ij; /* New max modality for covar j */
5082: } /* end of loop on modality i=-1 to 1 or more */
5083: break;
5084: case 1: /* Testing on varying covariate, could be simple and
5085: * should look at waves or product of fixed *
5086: * varying. No time to test -1, assuming 0 and 1 only */
5087: ij=0;
5088: for(i=0; i<=1;i++){
5089: nbcode[Tvar[k]][++ij]=i;
5090: }
5091: break;
5092: default:
5093: break;
5094: } /* end switch */
5095: } /* end dummy test */
5096:
5097: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5098: /* /\*recode from 0 *\/ */
5099: /* k is a modality. If we have model=V1+V1*sex */
5100: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5101: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5102: /* } */
5103: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5104: /* if (ij > ncodemax[j]) { */
5105: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5106: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5107: /* break; */
5108: /* } */
5109: /* } /\* end of loop on modality k *\/ */
5110: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5111:
5112: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5113: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5114: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5115: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5116: 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 */
5117: 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 */
5118: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5119: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5120:
5121: ij=0;
5122: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5123: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5124: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5125: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5126: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5127: /* If product not in single variable we don't print results */
5128: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5129: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5130: 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*/
5131: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5132: 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 */
5133: if(Fixed[k]!=0)
5134: anyvaryingduminmodel=1;
5135: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5136: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5137: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5138: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5139: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5140: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5141: }
5142: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5143: /* ij--; */
5144: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5145: *cptcov=ij; /*Number of total real effective covariates: effective
5146: * because they can be excluded from the model and real
5147: * if in the model but excluded because missing values, but how to get k from ij?*/
5148: for(j=ij+1; j<= cptcovt; j++){
5149: Tvaraff[j]=0;
5150: Tmodelind[j]=0;
5151: }
5152: for(j=ntveff+1; j<= cptcovt; j++){
5153: TmodelInvind[j]=0;
5154: }
5155: /* To be sorted */
5156: ;
5157: }
1.126 brouard 5158:
1.145 brouard 5159:
1.126 brouard 5160: /*********** Health Expectancies ****************/
5161:
1.235 brouard 5162: 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 5163:
5164: {
5165: /* Health expectancies, no variances */
1.164 brouard 5166: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5167: int nhstepma, nstepma; /* Decreasing with age */
5168: double age, agelim, hf;
5169: double ***p3mat;
5170: double eip;
5171:
1.238 brouard 5172: /* pstamp(ficreseij); */
1.126 brouard 5173: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5174: fprintf(ficreseij,"# Age");
5175: for(i=1; i<=nlstate;i++){
5176: for(j=1; j<=nlstate;j++){
5177: fprintf(ficreseij," e%1d%1d ",i,j);
5178: }
5179: fprintf(ficreseij," e%1d. ",i);
5180: }
5181: fprintf(ficreseij,"\n");
5182:
5183:
5184: if(estepm < stepm){
5185: printf ("Problem %d lower than %d\n",estepm, stepm);
5186: }
5187: else hstepm=estepm;
5188: /* We compute the life expectancy from trapezoids spaced every estepm months
5189: * This is mainly to measure the difference between two models: for example
5190: * if stepm=24 months pijx are given only every 2 years and by summing them
5191: * we are calculating an estimate of the Life Expectancy assuming a linear
5192: * progression in between and thus overestimating or underestimating according
5193: * to the curvature of the survival function. If, for the same date, we
5194: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5195: * to compare the new estimate of Life expectancy with the same linear
5196: * hypothesis. A more precise result, taking into account a more precise
5197: * curvature will be obtained if estepm is as small as stepm. */
5198:
5199: /* For example we decided to compute the life expectancy with the smallest unit */
5200: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5201: nhstepm is the number of hstepm from age to agelim
5202: nstepm is the number of stepm from age to agelin.
5203: Look at hpijx to understand the reason of that which relies in memory size
5204: and note for a fixed period like estepm months */
5205: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5206: survival function given by stepm (the optimization length). Unfortunately it
5207: means that if the survival funtion is printed only each two years of age and if
5208: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5209: results. So we changed our mind and took the option of the best precision.
5210: */
5211: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5212:
5213: agelim=AGESUP;
5214: /* If stepm=6 months */
5215: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5216: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5217:
5218: /* nhstepm age range expressed in number of stepm */
5219: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5220: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5221: /* if (stepm >= YEARM) hstepm=1;*/
5222: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5223: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5224:
5225: for (age=bage; age<=fage; age ++){
5226: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5227: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5228: /* if (stepm >= YEARM) hstepm=1;*/
5229: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5230:
5231: /* If stepm=6 months */
5232: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5233: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5234:
1.235 brouard 5235: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5236:
5237: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5238:
5239: printf("%d|",(int)age);fflush(stdout);
5240: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5241:
5242: /* Computing expectancies */
5243: for(i=1; i<=nlstate;i++)
5244: for(j=1; j<=nlstate;j++)
5245: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5246: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5247:
5248: /* 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]);*/
5249:
5250: }
5251:
5252: fprintf(ficreseij,"%3.0f",age );
5253: for(i=1; i<=nlstate;i++){
5254: eip=0;
5255: for(j=1; j<=nlstate;j++){
5256: eip +=eij[i][j][(int)age];
5257: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5258: }
5259: fprintf(ficreseij,"%9.4f", eip );
5260: }
5261: fprintf(ficreseij,"\n");
5262:
5263: }
5264: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5265: printf("\n");
5266: fprintf(ficlog,"\n");
5267:
5268: }
5269:
1.235 brouard 5270: 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 5271:
5272: {
5273: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5274: to initial status i, ei. .
1.126 brouard 5275: */
5276: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5277: int nhstepma, nstepma; /* Decreasing with age */
5278: double age, agelim, hf;
5279: double ***p3matp, ***p3matm, ***varhe;
5280: double **dnewm,**doldm;
5281: double *xp, *xm;
5282: double **gp, **gm;
5283: double ***gradg, ***trgradg;
5284: int theta;
5285:
5286: double eip, vip;
5287:
5288: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5289: xp=vector(1,npar);
5290: xm=vector(1,npar);
5291: dnewm=matrix(1,nlstate*nlstate,1,npar);
5292: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5293:
5294: pstamp(ficresstdeij);
5295: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5296: fprintf(ficresstdeij,"# Age");
5297: for(i=1; i<=nlstate;i++){
5298: for(j=1; j<=nlstate;j++)
5299: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5300: fprintf(ficresstdeij," e%1d. ",i);
5301: }
5302: fprintf(ficresstdeij,"\n");
5303:
5304: pstamp(ficrescveij);
5305: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5306: fprintf(ficrescveij,"# Age");
5307: for(i=1; i<=nlstate;i++)
5308: for(j=1; j<=nlstate;j++){
5309: cptj= (j-1)*nlstate+i;
5310: for(i2=1; i2<=nlstate;i2++)
5311: for(j2=1; j2<=nlstate;j2++){
5312: cptj2= (j2-1)*nlstate+i2;
5313: if(cptj2 <= cptj)
5314: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5315: }
5316: }
5317: fprintf(ficrescveij,"\n");
5318:
5319: if(estepm < stepm){
5320: printf ("Problem %d lower than %d\n",estepm, stepm);
5321: }
5322: else hstepm=estepm;
5323: /* We compute the life expectancy from trapezoids spaced every estepm months
5324: * This is mainly to measure the difference between two models: for example
5325: * if stepm=24 months pijx are given only every 2 years and by summing them
5326: * we are calculating an estimate of the Life Expectancy assuming a linear
5327: * progression in between and thus overestimating or underestimating according
5328: * to the curvature of the survival function. If, for the same date, we
5329: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5330: * to compare the new estimate of Life expectancy with the same linear
5331: * hypothesis. A more precise result, taking into account a more precise
5332: * curvature will be obtained if estepm is as small as stepm. */
5333:
5334: /* For example we decided to compute the life expectancy with the smallest unit */
5335: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5336: nhstepm is the number of hstepm from age to agelim
5337: nstepm is the number of stepm from age to agelin.
5338: Look at hpijx to understand the reason of that which relies in memory size
5339: and note for a fixed period like estepm months */
5340: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5341: survival function given by stepm (the optimization length). Unfortunately it
5342: means that if the survival funtion is printed only each two years of age and if
5343: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5344: results. So we changed our mind and took the option of the best precision.
5345: */
5346: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5347:
5348: /* If stepm=6 months */
5349: /* nhstepm age range expressed in number of stepm */
5350: agelim=AGESUP;
5351: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5352: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5353: /* if (stepm >= YEARM) hstepm=1;*/
5354: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5355:
5356: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5357: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5358: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5359: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5360: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5361: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5362:
5363: for (age=bage; age<=fage; age ++){
5364: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5365: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5366: /* if (stepm >= YEARM) hstepm=1;*/
5367: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5368:
1.126 brouard 5369: /* If stepm=6 months */
5370: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5371: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5372:
5373: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5374:
1.126 brouard 5375: /* Computing Variances of health expectancies */
5376: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5377: decrease memory allocation */
5378: for(theta=1; theta <=npar; theta++){
5379: for(i=1; i<=npar; i++){
1.222 brouard 5380: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5381: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5382: }
1.235 brouard 5383: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5384: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5385:
1.126 brouard 5386: for(j=1; j<= nlstate; j++){
1.222 brouard 5387: for(i=1; i<=nlstate; i++){
5388: for(h=0; h<=nhstepm-1; h++){
5389: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5390: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5391: }
5392: }
1.126 brouard 5393: }
1.218 brouard 5394:
1.126 brouard 5395: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5396: for(h=0; h<=nhstepm-1; h++){
5397: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5398: }
1.126 brouard 5399: }/* End theta */
5400:
5401:
5402: for(h=0; h<=nhstepm-1; h++)
5403: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5404: for(theta=1; theta <=npar; theta++)
5405: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5406:
1.218 brouard 5407:
1.222 brouard 5408: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5409: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5410: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5411:
1.222 brouard 5412: printf("%d|",(int)age);fflush(stdout);
5413: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5414: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5415: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5416: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5417: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5418: for(ij=1;ij<=nlstate*nlstate;ij++)
5419: for(ji=1;ji<=nlstate*nlstate;ji++)
5420: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5421: }
5422: }
1.218 brouard 5423:
1.126 brouard 5424: /* Computing expectancies */
1.235 brouard 5425: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5426: for(i=1; i<=nlstate;i++)
5427: for(j=1; j<=nlstate;j++)
1.222 brouard 5428: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5429: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5430:
1.222 brouard 5431: /* 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 5432:
1.222 brouard 5433: }
1.218 brouard 5434:
1.126 brouard 5435: fprintf(ficresstdeij,"%3.0f",age );
5436: for(i=1; i<=nlstate;i++){
5437: eip=0.;
5438: vip=0.;
5439: for(j=1; j<=nlstate;j++){
1.222 brouard 5440: eip += eij[i][j][(int)age];
5441: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5442: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5443: 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 5444: }
5445: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5446: }
5447: fprintf(ficresstdeij,"\n");
1.218 brouard 5448:
1.126 brouard 5449: fprintf(ficrescveij,"%3.0f",age );
5450: for(i=1; i<=nlstate;i++)
5451: for(j=1; j<=nlstate;j++){
1.222 brouard 5452: cptj= (j-1)*nlstate+i;
5453: for(i2=1; i2<=nlstate;i2++)
5454: for(j2=1; j2<=nlstate;j2++){
5455: cptj2= (j2-1)*nlstate+i2;
5456: if(cptj2 <= cptj)
5457: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5458: }
1.126 brouard 5459: }
5460: fprintf(ficrescveij,"\n");
1.218 brouard 5461:
1.126 brouard 5462: }
5463: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5464: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5465: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5466: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5467: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5468: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5469: printf("\n");
5470: fprintf(ficlog,"\n");
1.218 brouard 5471:
1.126 brouard 5472: free_vector(xm,1,npar);
5473: free_vector(xp,1,npar);
5474: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5475: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5476: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5477: }
1.218 brouard 5478:
1.126 brouard 5479: /************ Variance ******************/
1.235 brouard 5480: 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 5481: {
5482: /* Variance of health expectancies */
5483: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5484: /* double **newm;*/
5485: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5486:
5487: /* int movingaverage(); */
5488: double **dnewm,**doldm;
5489: double **dnewmp,**doldmp;
5490: int i, j, nhstepm, hstepm, h, nstepm ;
5491: int k;
5492: double *xp;
5493: double **gp, **gm; /* for var eij */
5494: double ***gradg, ***trgradg; /*for var eij */
5495: double **gradgp, **trgradgp; /* for var p point j */
5496: double *gpp, *gmp; /* for var p point j */
5497: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5498: double ***p3mat;
5499: double age,agelim, hf;
5500: /* double ***mobaverage; */
5501: int theta;
5502: char digit[4];
5503: char digitp[25];
5504:
5505: char fileresprobmorprev[FILENAMELENGTH];
5506:
5507: if(popbased==1){
5508: if(mobilav!=0)
5509: strcpy(digitp,"-POPULBASED-MOBILAV_");
5510: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5511: }
5512: else
5513: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5514:
1.218 brouard 5515: /* if (mobilav!=0) { */
5516: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5517: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5518: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5519: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5520: /* } */
5521: /* } */
5522:
5523: strcpy(fileresprobmorprev,"PRMORPREV-");
5524: sprintf(digit,"%-d",ij);
5525: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5526: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5527: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5528: strcat(fileresprobmorprev,fileresu);
5529: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5530: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5531: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5532: }
5533: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5534: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5535: pstamp(ficresprobmorprev);
5536: 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 5537: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5538: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5539: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5540: }
5541: for(j=1;j<=cptcoveff;j++)
5542: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5543: fprintf(ficresprobmorprev,"\n");
5544:
1.218 brouard 5545: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5546: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5547: fprintf(ficresprobmorprev," p.%-d SE",j);
5548: for(i=1; i<=nlstate;i++)
5549: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5550: }
5551: fprintf(ficresprobmorprev,"\n");
5552:
5553: fprintf(ficgp,"\n# Routine varevsij");
5554: fprintf(ficgp,"\nunset title \n");
5555: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5556: 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");
5557: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5558: /* } */
5559: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5560: pstamp(ficresvij);
5561: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5562: if(popbased==1)
5563: 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);
5564: else
5565: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5566: fprintf(ficresvij,"# Age");
5567: for(i=1; i<=nlstate;i++)
5568: for(j=1; j<=nlstate;j++)
5569: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5570: fprintf(ficresvij,"\n");
5571:
5572: xp=vector(1,npar);
5573: dnewm=matrix(1,nlstate,1,npar);
5574: doldm=matrix(1,nlstate,1,nlstate);
5575: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5576: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5577:
5578: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5579: gpp=vector(nlstate+1,nlstate+ndeath);
5580: gmp=vector(nlstate+1,nlstate+ndeath);
5581: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5582:
1.218 brouard 5583: if(estepm < stepm){
5584: printf ("Problem %d lower than %d\n",estepm, stepm);
5585: }
5586: else hstepm=estepm;
5587: /* For example we decided to compute the life expectancy with the smallest unit */
5588: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5589: nhstepm is the number of hstepm from age to agelim
5590: nstepm is the number of stepm from age to agelim.
5591: Look at function hpijx to understand why because of memory size limitations,
5592: we decided (b) to get a life expectancy respecting the most precise curvature of the
5593: survival function given by stepm (the optimization length). Unfortunately it
5594: means that if the survival funtion is printed every two years of age and if
5595: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5596: results. So we changed our mind and took the option of the best precision.
5597: */
5598: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5599: agelim = AGESUP;
5600: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5601: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5602: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5603: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5604: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5605: gp=matrix(0,nhstepm,1,nlstate);
5606: gm=matrix(0,nhstepm,1,nlstate);
5607:
5608:
5609: for(theta=1; theta <=npar; theta++){
5610: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5611: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5612: }
5613:
1.242 brouard 5614: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5615:
5616: if (popbased==1) {
5617: if(mobilav ==0){
5618: for(i=1; i<=nlstate;i++)
5619: prlim[i][i]=probs[(int)age][i][ij];
5620: }else{ /* mobilav */
5621: for(i=1; i<=nlstate;i++)
5622: prlim[i][i]=mobaverage[(int)age][i][ij];
5623: }
5624: }
5625:
1.235 brouard 5626: 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 5627: for(j=1; j<= nlstate; j++){
5628: for(h=0; h<=nhstepm; h++){
5629: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5630: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5631: }
5632: }
5633: /* Next for computing probability of death (h=1 means
5634: computed over hstepm matrices product = hstepm*stepm months)
5635: as a weighted average of prlim.
5636: */
5637: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5638: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5639: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5640: }
5641: /* end probability of death */
5642:
5643: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5644: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5645:
1.242 brouard 5646: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5647:
5648: if (popbased==1) {
5649: if(mobilav ==0){
5650: for(i=1; i<=nlstate;i++)
5651: prlim[i][i]=probs[(int)age][i][ij];
5652: }else{ /* mobilav */
5653: for(i=1; i<=nlstate;i++)
5654: prlim[i][i]=mobaverage[(int)age][i][ij];
5655: }
5656: }
5657:
1.235 brouard 5658: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5659:
5660: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5661: for(h=0; h<=nhstepm; h++){
5662: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5663: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5664: }
5665: }
5666: /* This for computing probability of death (h=1 means
5667: computed over hstepm matrices product = hstepm*stepm months)
5668: as a weighted average of prlim.
5669: */
5670: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5671: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5672: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5673: }
5674: /* end probability of death */
5675:
5676: for(j=1; j<= nlstate; j++) /* vareij */
5677: for(h=0; h<=nhstepm; h++){
5678: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5679: }
5680:
5681: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5682: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5683: }
5684:
5685: } /* End theta */
5686:
5687: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5688:
5689: for(h=0; h<=nhstepm; h++) /* veij */
5690: for(j=1; j<=nlstate;j++)
5691: for(theta=1; theta <=npar; theta++)
5692: trgradg[h][j][theta]=gradg[h][theta][j];
5693:
5694: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5695: for(theta=1; theta <=npar; theta++)
5696: trgradgp[j][theta]=gradgp[theta][j];
5697:
5698:
5699: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5700: for(i=1;i<=nlstate;i++)
5701: for(j=1;j<=nlstate;j++)
5702: vareij[i][j][(int)age] =0.;
5703:
5704: for(h=0;h<=nhstepm;h++){
5705: for(k=0;k<=nhstepm;k++){
5706: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5707: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5708: for(i=1;i<=nlstate;i++)
5709: for(j=1;j<=nlstate;j++)
5710: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5711: }
5712: }
5713:
5714: /* pptj */
5715: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5716: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5717: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5718: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5719: varppt[j][i]=doldmp[j][i];
5720: /* end ppptj */
5721: /* x centered again */
5722:
1.242 brouard 5723: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5724:
5725: if (popbased==1) {
5726: if(mobilav ==0){
5727: for(i=1; i<=nlstate;i++)
5728: prlim[i][i]=probs[(int)age][i][ij];
5729: }else{ /* mobilav */
5730: for(i=1; i<=nlstate;i++)
5731: prlim[i][i]=mobaverage[(int)age][i][ij];
5732: }
5733: }
5734:
5735: /* This for computing probability of death (h=1 means
5736: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5737: as a weighted average of prlim.
5738: */
1.235 brouard 5739: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5740: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5741: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5742: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5743: }
5744: /* end probability of death */
5745:
5746: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5747: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5748: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5749: for(i=1; i<=nlstate;i++){
5750: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5751: }
5752: }
5753: fprintf(ficresprobmorprev,"\n");
5754:
5755: fprintf(ficresvij,"%.0f ",age );
5756: for(i=1; i<=nlstate;i++)
5757: for(j=1; j<=nlstate;j++){
5758: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5759: }
5760: fprintf(ficresvij,"\n");
5761: free_matrix(gp,0,nhstepm,1,nlstate);
5762: free_matrix(gm,0,nhstepm,1,nlstate);
5763: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5764: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5765: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5766: } /* End age */
5767: free_vector(gpp,nlstate+1,nlstate+ndeath);
5768: free_vector(gmp,nlstate+1,nlstate+ndeath);
5769: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5770: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5771: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5772: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5773: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5774: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5775: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5776: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5777: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5778: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5779: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5780: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5781: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5782: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5783: 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);
5784: /* 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 5785: */
1.218 brouard 5786: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5787: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5788:
1.218 brouard 5789: free_vector(xp,1,npar);
5790: free_matrix(doldm,1,nlstate,1,nlstate);
5791: free_matrix(dnewm,1,nlstate,1,npar);
5792: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5793: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5794: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5795: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5796: fclose(ficresprobmorprev);
5797: fflush(ficgp);
5798: fflush(fichtm);
5799: } /* end varevsij */
1.126 brouard 5800:
5801: /************ Variance of prevlim ******************/
1.235 brouard 5802: 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 5803: {
1.205 brouard 5804: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5805: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5806:
1.126 brouard 5807: double **dnewm,**doldm;
5808: int i, j, nhstepm, hstepm;
5809: double *xp;
5810: double *gp, *gm;
5811: double **gradg, **trgradg;
1.208 brouard 5812: double **mgm, **mgp;
1.126 brouard 5813: double age,agelim;
5814: int theta;
5815:
5816: pstamp(ficresvpl);
5817: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5818: fprintf(ficresvpl,"# Age ");
5819: if(nresult >=1)
5820: fprintf(ficresvpl," Result# ");
1.126 brouard 5821: for(i=1; i<=nlstate;i++)
5822: fprintf(ficresvpl," %1d-%1d",i,i);
5823: fprintf(ficresvpl,"\n");
5824:
5825: xp=vector(1,npar);
5826: dnewm=matrix(1,nlstate,1,npar);
5827: doldm=matrix(1,nlstate,1,nlstate);
5828:
5829: hstepm=1*YEARM; /* Every year of age */
5830: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5831: agelim = AGESUP;
5832: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5833: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5834: if (stepm >= YEARM) hstepm=1;
5835: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5836: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5837: mgp=matrix(1,npar,1,nlstate);
5838: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5839: gp=vector(1,nlstate);
5840: gm=vector(1,nlstate);
5841:
5842: for(theta=1; theta <=npar; theta++){
5843: for(i=1; i<=npar; i++){ /* Computes gradient */
5844: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5845: }
1.209 brouard 5846: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5847: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5848: else
1.235 brouard 5849: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5850: for(i=1;i<=nlstate;i++){
1.126 brouard 5851: gp[i] = prlim[i][i];
1.208 brouard 5852: mgp[theta][i] = prlim[i][i];
5853: }
1.126 brouard 5854: for(i=1; i<=npar; i++) /* Computes gradient */
5855: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5856: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5857: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5858: else
1.235 brouard 5859: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5860: for(i=1;i<=nlstate;i++){
1.126 brouard 5861: gm[i] = prlim[i][i];
1.208 brouard 5862: mgm[theta][i] = prlim[i][i];
5863: }
1.126 brouard 5864: for(i=1;i<=nlstate;i++)
5865: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5866: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5867: } /* End theta */
5868:
5869: trgradg =matrix(1,nlstate,1,npar);
5870:
5871: for(j=1; j<=nlstate;j++)
5872: for(theta=1; theta <=npar; theta++)
5873: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5874: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5875: /* printf("\nmgm mgp %d ",(int)age); */
5876: /* for(j=1; j<=nlstate;j++){ */
5877: /* printf(" %d ",j); */
5878: /* for(theta=1; theta <=npar; theta++) */
5879: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5880: /* printf("\n "); */
5881: /* } */
5882: /* } */
5883: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5884: /* printf("\n gradg %d ",(int)age); */
5885: /* for(j=1; j<=nlstate;j++){ */
5886: /* printf("%d ",j); */
5887: /* for(theta=1; theta <=npar; theta++) */
5888: /* printf("%d %lf ",theta,gradg[theta][j]); */
5889: /* printf("\n "); */
5890: /* } */
5891: /* } */
1.126 brouard 5892:
5893: for(i=1;i<=nlstate;i++)
5894: varpl[i][(int)age] =0.;
1.209 brouard 5895: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5896: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5897: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5898: }else{
1.126 brouard 5899: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5900: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5901: }
1.126 brouard 5902: for(i=1;i<=nlstate;i++)
5903: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5904:
5905: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 5906: if(nresult >=1)
5907: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 5908: for(i=1; i<=nlstate;i++)
5909: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
5910: fprintf(ficresvpl,"\n");
5911: free_vector(gp,1,nlstate);
5912: free_vector(gm,1,nlstate);
1.208 brouard 5913: free_matrix(mgm,1,npar,1,nlstate);
5914: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 5915: free_matrix(gradg,1,npar,1,nlstate);
5916: free_matrix(trgradg,1,nlstate,1,npar);
5917: } /* End age */
5918:
5919: free_vector(xp,1,npar);
5920: free_matrix(doldm,1,nlstate,1,npar);
5921: free_matrix(dnewm,1,nlstate,1,nlstate);
5922:
5923: }
5924:
5925: /************ Variance of one-step probabilities ******************/
5926: 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 5927: {
5928: int i, j=0, k1, l1, tj;
5929: int k2, l2, j1, z1;
5930: int k=0, l;
5931: int first=1, first1, first2;
5932: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
5933: double **dnewm,**doldm;
5934: double *xp;
5935: double *gp, *gm;
5936: double **gradg, **trgradg;
5937: double **mu;
5938: double age, cov[NCOVMAX+1];
5939: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
5940: int theta;
5941: char fileresprob[FILENAMELENGTH];
5942: char fileresprobcov[FILENAMELENGTH];
5943: char fileresprobcor[FILENAMELENGTH];
5944: double ***varpij;
5945:
5946: strcpy(fileresprob,"PROB_");
5947: strcat(fileresprob,fileres);
5948: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
5949: printf("Problem with resultfile: %s\n", fileresprob);
5950: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
5951: }
5952: strcpy(fileresprobcov,"PROBCOV_");
5953: strcat(fileresprobcov,fileresu);
5954: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
5955: printf("Problem with resultfile: %s\n", fileresprobcov);
5956: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
5957: }
5958: strcpy(fileresprobcor,"PROBCOR_");
5959: strcat(fileresprobcor,fileresu);
5960: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
5961: printf("Problem with resultfile: %s\n", fileresprobcor);
5962: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
5963: }
5964: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5965: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5966: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5967: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5968: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5969: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5970: pstamp(ficresprob);
5971: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
5972: fprintf(ficresprob,"# Age");
5973: pstamp(ficresprobcov);
5974: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
5975: fprintf(ficresprobcov,"# Age");
5976: pstamp(ficresprobcor);
5977: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
5978: fprintf(ficresprobcor,"# Age");
1.126 brouard 5979:
5980:
1.222 brouard 5981: for(i=1; i<=nlstate;i++)
5982: for(j=1; j<=(nlstate+ndeath);j++){
5983: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
5984: fprintf(ficresprobcov," p%1d-%1d ",i,j);
5985: fprintf(ficresprobcor," p%1d-%1d ",i,j);
5986: }
5987: /* fprintf(ficresprob,"\n");
5988: fprintf(ficresprobcov,"\n");
5989: fprintf(ficresprobcor,"\n");
5990: */
5991: xp=vector(1,npar);
5992: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5993: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
5994: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
5995: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
5996: first=1;
5997: fprintf(ficgp,"\n# Routine varprob");
5998: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
5999: fprintf(fichtm,"\n");
6000:
6001: 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);
6002: 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);
6003: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6004: and drawn. It helps understanding how is the covariance between two incidences.\
6005: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6006: 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 6007: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6008: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6009: standard deviations wide on each axis. <br>\
6010: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6011: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6012: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6013:
1.222 brouard 6014: cov[1]=1;
6015: /* tj=cptcoveff; */
1.225 brouard 6016: tj = (int) pow(2,cptcoveff);
1.222 brouard 6017: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6018: j1=0;
1.224 brouard 6019: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6020: if (cptcovn>0) {
6021: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6022: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6023: fprintf(ficresprob, "**********\n#\n");
6024: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6025: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6026: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6027:
1.222 brouard 6028: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6029: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6030: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6031:
6032:
1.222 brouard 6033: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6034: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6035: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6036:
1.222 brouard 6037: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6038: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6039: fprintf(ficresprobcor, "**********\n#");
6040: if(invalidvarcomb[j1]){
6041: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6042: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6043: continue;
6044: }
6045: }
6046: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6047: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6048: gp=vector(1,(nlstate)*(nlstate+ndeath));
6049: gm=vector(1,(nlstate)*(nlstate+ndeath));
6050: for (age=bage; age<=fage; age ++){
6051: cov[2]=age;
6052: if(nagesqr==1)
6053: cov[3]= age*age;
6054: for (k=1; k<=cptcovn;k++) {
6055: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6056: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6057: * 1 1 1 1 1
6058: * 2 2 1 1 1
6059: * 3 1 2 1 1
6060: */
6061: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6062: }
6063: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6064: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6065: for (k=1; k<=cptcovprod;k++)
6066: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6067:
6068:
1.222 brouard 6069: for(theta=1; theta <=npar; theta++){
6070: for(i=1; i<=npar; i++)
6071: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6072:
1.222 brouard 6073: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6074:
1.222 brouard 6075: k=0;
6076: for(i=1; i<= (nlstate); i++){
6077: for(j=1; j<=(nlstate+ndeath);j++){
6078: k=k+1;
6079: gp[k]=pmmij[i][j];
6080: }
6081: }
1.220 brouard 6082:
1.222 brouard 6083: for(i=1; i<=npar; i++)
6084: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6085:
1.222 brouard 6086: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6087: k=0;
6088: for(i=1; i<=(nlstate); i++){
6089: for(j=1; j<=(nlstate+ndeath);j++){
6090: k=k+1;
6091: gm[k]=pmmij[i][j];
6092: }
6093: }
1.220 brouard 6094:
1.222 brouard 6095: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6096: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6097: }
1.126 brouard 6098:
1.222 brouard 6099: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6100: for(theta=1; theta <=npar; theta++)
6101: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6102:
1.222 brouard 6103: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6104: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6105:
1.222 brouard 6106: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6107:
1.222 brouard 6108: k=0;
6109: for(i=1; i<=(nlstate); i++){
6110: for(j=1; j<=(nlstate+ndeath);j++){
6111: k=k+1;
6112: mu[k][(int) age]=pmmij[i][j];
6113: }
6114: }
6115: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6116: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6117: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6118:
1.222 brouard 6119: /*printf("\n%d ",(int)age);
6120: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6121: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6122: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6123: }*/
1.220 brouard 6124:
1.222 brouard 6125: fprintf(ficresprob,"\n%d ",(int)age);
6126: fprintf(ficresprobcov,"\n%d ",(int)age);
6127: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6128:
1.222 brouard 6129: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6130: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6131: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6132: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6133: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6134: }
6135: i=0;
6136: for (k=1; k<=(nlstate);k++){
6137: for (l=1; l<=(nlstate+ndeath);l++){
6138: i++;
6139: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6140: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6141: for (j=1; j<=i;j++){
6142: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6143: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6144: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6145: }
6146: }
6147: }/* end of loop for state */
6148: } /* end of loop for age */
6149: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6150: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6151: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6152: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6153:
6154: /* Confidence intervalle of pij */
6155: /*
6156: fprintf(ficgp,"\nunset parametric;unset label");
6157: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6158: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6159: 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);
6160: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6161: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6162: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6163: */
6164:
6165: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6166: first1=1;first2=2;
6167: for (k2=1; k2<=(nlstate);k2++){
6168: for (l2=1; l2<=(nlstate+ndeath);l2++){
6169: if(l2==k2) continue;
6170: j=(k2-1)*(nlstate+ndeath)+l2;
6171: for (k1=1; k1<=(nlstate);k1++){
6172: for (l1=1; l1<=(nlstate+ndeath);l1++){
6173: if(l1==k1) continue;
6174: i=(k1-1)*(nlstate+ndeath)+l1;
6175: if(i<=j) continue;
6176: for (age=bage; age<=fage; age ++){
6177: if ((int)age %5==0){
6178: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6179: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6180: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6181: mu1=mu[i][(int) age]/stepm*YEARM ;
6182: mu2=mu[j][(int) age]/stepm*YEARM;
6183: c12=cv12/sqrt(v1*v2);
6184: /* Computing eigen value of matrix of covariance */
6185: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6186: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6187: if ((lc2 <0) || (lc1 <0) ){
6188: if(first2==1){
6189: first1=0;
6190: 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);
6191: }
6192: 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);
6193: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6194: /* lc2=fabs(lc2); */
6195: }
1.220 brouard 6196:
1.222 brouard 6197: /* Eigen vectors */
6198: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6199: /*v21=sqrt(1.-v11*v11); *//* error */
6200: v21=(lc1-v1)/cv12*v11;
6201: v12=-v21;
6202: v22=v11;
6203: tnalp=v21/v11;
6204: if(first1==1){
6205: first1=0;
6206: 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);
6207: }
6208: 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);
6209: /*printf(fignu*/
6210: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6211: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6212: if(first==1){
6213: first=0;
6214: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6215: fprintf(ficgp,"\nset parametric;unset label");
6216: 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);
6217: fprintf(ficgp,"\nset ter svg size 640, 480");
6218: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6219: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6220: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6221: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6222: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6223: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6224: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6225: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6226: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6227: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6228: 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", \
6229: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6230: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6231: }else{
6232: first=0;
6233: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6234: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6235: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6236: 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", \
6237: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6238: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6239: }/* if first */
6240: } /* age mod 5 */
6241: } /* end loop age */
6242: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6243: first=1;
6244: } /*l12 */
6245: } /* k12 */
6246: } /*l1 */
6247: }/* k1 */
6248: } /* loop on combination of covariates j1 */
6249: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6250: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6251: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6252: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6253: free_vector(xp,1,npar);
6254: fclose(ficresprob);
6255: fclose(ficresprobcov);
6256: fclose(ficresprobcor);
6257: fflush(ficgp);
6258: fflush(fichtmcov);
6259: }
1.126 brouard 6260:
6261:
6262: /******************* Printing html file ***********/
1.201 brouard 6263: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6264: int lastpass, int stepm, int weightopt, char model[],\
6265: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6266: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6267: double jprev1, double mprev1,double anprev1, double dateprev1, \
6268: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6269: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6270:
6271: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6272: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6273: </ul>");
1.237 brouard 6274: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6275: </ul>", model);
1.214 brouard 6276: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6277: 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",
6278: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6279: 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 6280: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6281: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6282: fprintf(fichtm,"\
6283: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6284: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6285: fprintf(fichtm,"\
1.217 brouard 6286: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6287: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6288: fprintf(fichtm,"\
1.126 brouard 6289: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6290: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6291: fprintf(fichtm,"\
1.217 brouard 6292: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6293: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6294: fprintf(fichtm,"\
1.211 brouard 6295: - (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 6296: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6297: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6298: if(prevfcast==1){
6299: fprintf(fichtm,"\
6300: - Prevalence projections by age and states: \
1.201 brouard 6301: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6302: }
1.126 brouard 6303:
1.222 brouard 6304: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6305:
1.225 brouard 6306: m=pow(2,cptcoveff);
1.222 brouard 6307: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6308:
1.222 brouard 6309: jj1=0;
1.237 brouard 6310:
6311: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6312: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.237 brouard 6313: if(TKresult[nres]!= k1)
6314: continue;
1.220 brouard 6315:
1.222 brouard 6316: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6317: jj1++;
6318: if (cptcovn > 0) {
6319: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6320: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6321: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6322: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6323: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6324: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6325: }
1.237 brouard 6326: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6327: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6328: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6329: }
6330:
1.230 brouard 6331: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6332: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6333: if(invalidvarcomb[k1]){
6334: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6335: printf("\nCombination (%d) ignored because no cases \n",k1);
6336: continue;
6337: }
6338: }
6339: /* aij, bij */
1.241 brouard 6340: 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> \
6341: <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 6342: /* Pij */
1.241 brouard 6343: 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> \
6344: <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 6345: /* Quasi-incidences */
6346: 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 6347: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6348: 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 6349: 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> \
6350: <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 6351: /* Survival functions (period) in state j */
6352: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6353: 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> \
6354: <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 6355: }
6356: /* State specific survival functions (period) */
6357: for(cpt=1; cpt<=nlstate;cpt++){
6358: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6359: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6360: <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 6361: }
6362: /* Period (stable) prevalence in each health state */
6363: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6364: 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> \
6365: <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 6366: }
6367: if(backcast==1){
6368: /* Period (stable) back prevalence in each health state */
6369: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6370: 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> \
6371: <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 6372: }
1.217 brouard 6373: }
1.222 brouard 6374: if(prevfcast==1){
6375: /* Projection of prevalence up to period (stable) prevalence in each health state */
6376: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6377: 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> \
6378: <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 6379: }
6380: }
1.220 brouard 6381:
1.222 brouard 6382: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6383: 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> \
6384: <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 6385: }
6386: /* } /\* end i1 *\/ */
6387: }/* End k1 */
6388: fprintf(fichtm,"</ul>");
1.126 brouard 6389:
1.222 brouard 6390: fprintf(fichtm,"\
1.126 brouard 6391: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6392: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6393: - 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 6394: But because parameters are usually highly correlated (a higher incidence of disability \
6395: and a higher incidence of recovery can give very close observed transition) it might \
6396: be very useful to look not only at linear confidence intervals estimated from the \
6397: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6398: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6399: covariance matrix of the one-step probabilities. \
6400: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6401:
1.222 brouard 6402: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6403: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6404: fprintf(fichtm,"\
1.126 brouard 6405: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6406: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6407:
1.222 brouard 6408: fprintf(fichtm,"\
1.126 brouard 6409: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6410: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6411: fprintf(fichtm,"\
1.126 brouard 6412: - 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): \
6413: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6414: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6415: fprintf(fichtm,"\
1.126 brouard 6416: - (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): \
6417: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6418: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6419: fprintf(fichtm,"\
1.128 brouard 6420: - 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 6421: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6422: fprintf(fichtm,"\
1.128 brouard 6423: - 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 6424: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6425: fprintf(fichtm,"\
1.126 brouard 6426: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6427: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6428:
6429: /* if(popforecast==1) fprintf(fichtm,"\n */
6430: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6431: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6432: /* <br>",fileres,fileres,fileres,fileres); */
6433: /* else */
6434: /* 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 6435: fflush(fichtm);
6436: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6437:
1.225 brouard 6438: m=pow(2,cptcoveff);
1.222 brouard 6439: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6440:
1.222 brouard 6441: jj1=0;
1.237 brouard 6442:
1.241 brouard 6443: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6444: for(k1=1; k1<=m;k1++){
1.237 brouard 6445: if(TKresult[nres]!= k1)
6446: continue;
1.222 brouard 6447: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6448: jj1++;
1.126 brouard 6449: if (cptcovn > 0) {
6450: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6451: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6452: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6453: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6454: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6455: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6456: }
6457:
1.126 brouard 6458: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6459:
1.222 brouard 6460: if(invalidvarcomb[k1]){
6461: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6462: continue;
6463: }
1.126 brouard 6464: }
6465: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6466: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6467: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6468: <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 6469: }
6470: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6471: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6472: true period expectancies (those weighted with period prevalences are also\
6473: drawn in addition to the population based expectancies computed using\
1.241 brouard 6474: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6475: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6476: /* } /\* end i1 *\/ */
6477: }/* End k1 */
1.241 brouard 6478: }/* End nres */
1.222 brouard 6479: fprintf(fichtm,"</ul>");
6480: fflush(fichtm);
1.126 brouard 6481: }
6482:
6483: /******************* Gnuplot file **************/
1.223 brouard 6484: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6485:
6486: char dirfileres[132],optfileres[132];
1.223 brouard 6487: char gplotcondition[132];
1.237 brouard 6488: 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 6489: int lv=0, vlv=0, kl=0;
1.130 brouard 6490: int ng=0;
1.201 brouard 6491: int vpopbased;
1.223 brouard 6492: int ioffset; /* variable offset for columns */
1.235 brouard 6493: int nres=0; /* Index of resultline */
1.219 brouard 6494:
1.126 brouard 6495: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6496: /* printf("Problem with file %s",optionfilegnuplot); */
6497: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6498: /* } */
6499:
6500: /*#ifdef windows */
6501: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6502: /*#endif */
1.225 brouard 6503: m=pow(2,cptcoveff);
1.126 brouard 6504:
1.202 brouard 6505: /* Contribution to likelihood */
6506: /* Plot the probability implied in the likelihood */
1.223 brouard 6507: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6508: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6509: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6510: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6511: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6512: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6513: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6514: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6515: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6516: 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));
6517: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6518: 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));
6519: for (i=1; i<= nlstate ; i ++) {
6520: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6521: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6522: 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);
6523: for (j=2; j<= nlstate+ndeath ; j ++) {
6524: 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);
6525: }
6526: fprintf(ficgp,";\nset out; unset ylabel;\n");
6527: }
6528: /* 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 */
6529: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6530: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6531: fprintf(ficgp,"\nset out;unset log\n");
6532: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6533:
1.126 brouard 6534: strcpy(dirfileres,optionfilefiname);
6535: strcpy(optfileres,"vpl");
1.223 brouard 6536: /* 1eme*/
1.238 brouard 6537: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6538: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6539: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6540: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
6541: if(TKresult[nres]!= k1)
6542: continue;
6543: /* We are interested in selected combination by the resultline */
1.246 brouard 6544: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6545: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6546: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6547: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6548: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6549: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6550: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6551: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6552: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6553: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6554: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6555: }
6556: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6557: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6558: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6559: }
1.246 brouard 6560: /* printf("\n#\n"); */
1.238 brouard 6561: fprintf(ficgp,"\n#\n");
6562: if(invalidvarcomb[k1]){
6563: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6564: continue;
6565: }
1.235 brouard 6566:
1.241 brouard 6567: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6568: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6569: 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 6570:
1.238 brouard 6571: for (i=1; i<= nlstate ; i ++) {
6572: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6573: else fprintf(ficgp," %%*lf (%%*lf)");
6574: }
1.242 brouard 6575: 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 6576: for (i=1; i<= nlstate ; i ++) {
6577: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6578: else fprintf(ficgp," %%*lf (%%*lf)");
6579: }
1.242 brouard 6580: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6581: for (i=1; i<= nlstate ; i ++) {
6582: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6583: else fprintf(ficgp," %%*lf (%%*lf)");
6584: }
6585: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6586: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6587: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6588: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6589: if(cptcoveff ==0){
1.245 brouard 6590: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6591: }else{
6592: kl=0;
6593: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6594: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6595: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6596: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6597: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6598: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6599: kl++;
1.238 brouard 6600: /* 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 *\/ */
6601: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6602: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6603: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6604: if(k==cptcoveff){
1.245 brouard 6605: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6606: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6607: }else{
6608: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6609: kl++;
6610: }
6611: } /* end covariate */
6612: } /* end if no covariate */
6613: } /* end if backcast */
6614: fprintf(ficgp,"\nset out \n");
6615: } /* nres */
1.201 brouard 6616: } /* k1 */
6617: } /* cpt */
1.235 brouard 6618:
6619:
1.126 brouard 6620: /*2 eme*/
1.238 brouard 6621: for (k1=1; k1<= m ; k1 ++){
6622: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6623: if(TKresult[nres]!= k1)
6624: continue;
6625: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6626: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6627: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6628: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6629: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6630: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6631: vlv= nbcode[Tvaraff[k]][lv];
6632: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6633: }
1.237 brouard 6634: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6635: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6636: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6637: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6638: }
1.211 brouard 6639: fprintf(ficgp,"\n#\n");
1.223 brouard 6640: if(invalidvarcomb[k1]){
6641: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6642: continue;
6643: }
1.219 brouard 6644:
1.241 brouard 6645: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6646: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6647: if(vpopbased==0)
6648: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6649: else
6650: fprintf(ficgp,"\nreplot ");
6651: for (i=1; i<= nlstate+1 ; i ++) {
6652: k=2*i;
6653: 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);
6654: for (j=1; j<= nlstate+1 ; j ++) {
6655: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6656: else fprintf(ficgp," %%*lf (%%*lf)");
6657: }
6658: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6659: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6660: 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);
6661: for (j=1; j<= nlstate+1 ; j ++) {
6662: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6663: else fprintf(ficgp," %%*lf (%%*lf)");
6664: }
6665: fprintf(ficgp,"\" t\"\" w l lt 0,");
6666: 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);
6667: for (j=1; j<= nlstate+1 ; j ++) {
6668: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6669: else fprintf(ficgp," %%*lf (%%*lf)");
6670: }
6671: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6672: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6673: } /* state */
6674: } /* vpopbased */
1.244 brouard 6675: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6676: } /* end nres */
6677: } /* k1 end 2 eme*/
6678:
6679:
6680: /*3eme*/
6681: for (k1=1; k1<= m ; k1 ++){
6682: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.240 brouard 6683: if(TKresult[nres]!= k1)
1.238 brouard 6684: continue;
6685:
6686: for (cpt=1; cpt<= nlstate ; cpt ++) {
6687: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6688: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6689: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6690: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6691: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6692: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6693: vlv= nbcode[Tvaraff[k]][lv];
6694: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6695: }
6696: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6697: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6698: }
6699: fprintf(ficgp,"\n#\n");
6700: if(invalidvarcomb[k1]){
6701: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6702: continue;
6703: }
6704:
6705: /* k=2+nlstate*(2*cpt-2); */
6706: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6707: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6708: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6709: 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 6710: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6711: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6712: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6713: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6714: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6715: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6716:
1.238 brouard 6717: */
6718: for (i=1; i< nlstate ; i ++) {
6719: 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);
6720: /* 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 6721:
1.238 brouard 6722: }
6723: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6724: }
6725: } /* end nres */
6726: } /* end kl 3eme */
1.126 brouard 6727:
1.223 brouard 6728: /* 4eme */
1.201 brouard 6729: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6730: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6731: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6732: if(TKresult[nres]!= k1)
1.223 brouard 6733: continue;
1.238 brouard 6734: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6735: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6736: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6737: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6738: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6739: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6740: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6741: vlv= nbcode[Tvaraff[k]][lv];
6742: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6743: }
6744: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6745: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6746: }
6747: fprintf(ficgp,"\n#\n");
6748: if(invalidvarcomb[k1]){
6749: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6750: continue;
1.223 brouard 6751: }
1.238 brouard 6752:
1.241 brouard 6753: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6754: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6755: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6756: k=3;
6757: for (i=1; i<= nlstate ; i ++){
6758: if(i==1){
6759: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6760: }else{
6761: fprintf(ficgp,", '' ");
6762: }
6763: l=(nlstate+ndeath)*(i-1)+1;
6764: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6765: for (j=2; j<= nlstate+ndeath ; j ++)
6766: fprintf(ficgp,"+$%d",k+l+j-1);
6767: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6768: } /* nlstate */
6769: fprintf(ficgp,"\nset out\n");
6770: } /* end cpt state*/
6771: } /* end nres */
6772: } /* end covariate k1 */
6773:
1.220 brouard 6774: /* 5eme */
1.201 brouard 6775: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6776: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6777: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6778: if(TKresult[nres]!= k1)
1.227 brouard 6779: continue;
1.238 brouard 6780: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6781: 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);
6782: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6783: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6784: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6785: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6786: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6787: vlv= nbcode[Tvaraff[k]][lv];
6788: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6789: }
6790: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6791: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6792: }
6793: fprintf(ficgp,"\n#\n");
6794: if(invalidvarcomb[k1]){
6795: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6796: continue;
6797: }
1.227 brouard 6798:
1.241 brouard 6799: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6800: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6801: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6802: k=3;
6803: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6804: if(j==1)
6805: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6806: else
6807: fprintf(ficgp,", '' ");
6808: l=(nlstate+ndeath)*(cpt-1) +j;
6809: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6810: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6811: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6812: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6813: } /* nlstate */
6814: fprintf(ficgp,", '' ");
6815: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6816: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6817: l=(nlstate+ndeath)*(cpt-1) +j;
6818: if(j < nlstate)
6819: fprintf(ficgp,"$%d +",k+l);
6820: else
6821: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6822: }
6823: fprintf(ficgp,"\nset out\n");
6824: } /* end cpt state*/
6825: } /* end covariate */
6826: } /* end nres */
1.227 brouard 6827:
1.220 brouard 6828: /* 6eme */
1.202 brouard 6829: /* CV preval stable (period) for each covariate */
1.237 brouard 6830: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6831: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6832: if(TKresult[nres]!= k1)
6833: continue;
1.153 brouard 6834: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6835:
1.211 brouard 6836: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6837: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6838: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6839: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6840: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6841: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6842: vlv= nbcode[Tvaraff[k]][lv];
6843: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6844: }
1.237 brouard 6845: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6846: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6847: }
1.211 brouard 6848: fprintf(ficgp,"\n#\n");
1.223 brouard 6849: if(invalidvarcomb[k1]){
1.227 brouard 6850: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6851: continue;
1.223 brouard 6852: }
1.227 brouard 6853:
1.241 brouard 6854: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6855: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6856: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6857: k=3; /* Offset */
1.153 brouard 6858: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6859: if(i==1)
6860: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6861: else
6862: fprintf(ficgp,", '' ");
6863: l=(nlstate+ndeath)*(i-1)+1;
6864: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6865: for (j=2; j<= nlstate ; j ++)
6866: fprintf(ficgp,"+$%d",k+l+j-1);
6867: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6868: } /* nlstate */
1.201 brouard 6869: fprintf(ficgp,"\nset out\n");
1.153 brouard 6870: } /* end cpt state*/
6871: } /* end covariate */
1.227 brouard 6872:
6873:
1.220 brouard 6874: /* 7eme */
1.218 brouard 6875: if(backcast == 1){
1.217 brouard 6876: /* CV back preval stable (period) for each covariate */
1.237 brouard 6877: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6878: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6879: if(TKresult[nres]!= k1)
6880: continue;
1.218 brouard 6881: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6882: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6883: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6884: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6885: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6886: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6887: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6888: vlv= nbcode[Tvaraff[k]][lv];
6889: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6890: }
1.237 brouard 6891: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6892: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6893: }
1.227 brouard 6894: fprintf(ficgp,"\n#\n");
6895: if(invalidvarcomb[k1]){
6896: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6897: continue;
6898: }
6899:
1.241 brouard 6900: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6901: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6902: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6903: k=3; /* Offset */
6904: for (i=1; i<= nlstate ; i ++){
6905: if(i==1)
6906: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6907: else
6908: fprintf(ficgp,", '' ");
6909: /* l=(nlstate+ndeath)*(i-1)+1; */
6910: l=(nlstate+ndeath)*(cpt-1)+1;
6911: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
6912: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
6913: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
6914: /* for (j=2; j<= nlstate ; j ++) */
6915: /* fprintf(ficgp,"+$%d",k+l+j-1); */
6916: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
6917: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
6918: } /* nlstate */
6919: fprintf(ficgp,"\nset out\n");
1.218 brouard 6920: } /* end cpt state*/
6921: } /* end covariate */
6922: } /* End if backcast */
6923:
1.223 brouard 6924: /* 8eme */
1.218 brouard 6925: if(prevfcast==1){
6926: /* Projection from cross-sectional to stable (period) for each covariate */
6927:
1.237 brouard 6928: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6929: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6930: if(TKresult[nres]!= k1)
6931: continue;
1.211 brouard 6932: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6933: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
6934: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
6935: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6936: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6937: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6938: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6939: vlv= nbcode[Tvaraff[k]][lv];
6940: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6941: }
1.237 brouard 6942: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6943: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6944: }
1.227 brouard 6945: fprintf(ficgp,"\n#\n");
6946: if(invalidvarcomb[k1]){
6947: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6948: continue;
6949: }
6950:
6951: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 6952: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 6953: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 6954: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6955: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
6956: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6957: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6958: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6959: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6960: if(i==1){
6961: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
6962: }else{
6963: fprintf(ficgp,",\\\n '' ");
6964: }
6965: if(cptcoveff ==0){ /* No covariate */
6966: ioffset=2; /* Age is in 2 */
6967: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6968: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6969: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6970: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6971: fprintf(ficgp," u %d:(", ioffset);
6972: if(i==nlstate+1)
6973: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
6974: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6975: else
6976: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
6977: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6978: }else{ /* more than 2 covariates */
6979: if(cptcoveff ==1){
6980: ioffset=4; /* Age is in 4 */
6981: }else{
6982: ioffset=6; /* Age is in 6 */
6983: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6984: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6985: }
6986: fprintf(ficgp," u %d:(",ioffset);
6987: kl=0;
6988: strcpy(gplotcondition,"(");
6989: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
6990: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
6991: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6992: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6993: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6994: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
6995: kl++;
6996: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
6997: kl++;
6998: if(k <cptcoveff && cptcoveff>1)
6999: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7000: }
7001: strcpy(gplotcondition+strlen(gplotcondition),")");
7002: /* 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 *\/ */
7003: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7004: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7005: /* '' 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*/
7006: if(i==nlstate+1){
7007: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7008: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7009: }else{
7010: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7011: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7012: }
7013: } /* end if covariate */
7014: } /* nlstate */
7015: fprintf(ficgp,"\nset out\n");
1.223 brouard 7016: } /* end cpt state*/
7017: } /* end covariate */
7018: } /* End if prevfcast */
1.227 brouard 7019:
7020:
1.238 brouard 7021: /* 9eme writing MLE parameters */
7022: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7023: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7024: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7025: for(k=1; k <=(nlstate+ndeath); k++){
7026: if (k != i) {
1.227 brouard 7027: fprintf(ficgp,"# current state %d\n",k);
7028: for(j=1; j <=ncovmodel; j++){
7029: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7030: jk++;
7031: }
7032: fprintf(ficgp,"\n");
1.126 brouard 7033: }
7034: }
1.223 brouard 7035: }
1.187 brouard 7036: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7037:
1.145 brouard 7038: /*goto avoid;*/
1.238 brouard 7039: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7040: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7041: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7042: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7043: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7044: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7045: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7046: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7047: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7048: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7049: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7050: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7051: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7052: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7053: fprintf(ficgp,"#\n");
1.223 brouard 7054: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7055: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7056: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7057: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7058: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7059: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7060: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7061: if(TKresult[nres]!= jk)
7062: continue;
7063: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7064: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7065: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7066: }
7067: fprintf(ficgp,"\n#\n");
1.241 brouard 7068: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7069: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7070: if (ng==1){
7071: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7072: fprintf(ficgp,"\nunset log y");
7073: }else if (ng==2){
7074: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7075: fprintf(ficgp,"\nset log y");
7076: }else if (ng==3){
7077: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7078: fprintf(ficgp,"\nset log y");
7079: }else
7080: fprintf(ficgp,"\nunset title ");
7081: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7082: i=1;
7083: for(k2=1; k2<=nlstate; k2++) {
7084: k3=i;
7085: for(k=1; k<=(nlstate+ndeath); k++) {
7086: if (k != k2){
7087: switch( ng) {
7088: case 1:
7089: if(nagesqr==0)
7090: fprintf(ficgp," p%d+p%d*x",i,i+1);
7091: else /* nagesqr =1 */
7092: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7093: break;
7094: case 2: /* ng=2 */
7095: if(nagesqr==0)
7096: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7097: else /* nagesqr =1 */
7098: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7099: break;
7100: case 3:
7101: if(nagesqr==0)
7102: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7103: else /* nagesqr =1 */
7104: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7105: break;
7106: }
7107: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7108: ijp=1; /* product no age */
7109: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7110: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7111: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7112: if(j==Tage[ij]) { /* Product by age */
7113: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7114: if(DummyV[j]==0){
1.237 brouard 7115: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7116: }else{ /* quantitative */
7117: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7118: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7119: }
7120: ij++;
7121: }
7122: }else if(j==Tprod[ijp]) { /* */
7123: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7124: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7125: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7126: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7127: /* 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)]); */
7128: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7129: }else{ /* Vn is dummy and Vm is quanti */
7130: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7131: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7132: }
7133: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7134: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7135: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7136: }else{ /* Both quanti */
7137: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7138: }
7139: }
1.238 brouard 7140: ijp++;
1.237 brouard 7141: }
7142: } else{ /* simple covariate */
7143: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7144: if(Dummy[j]==0){
7145: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7146: }else{ /* quantitative */
7147: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7148: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7149: }
1.237 brouard 7150: } /* end simple */
7151: } /* end j */
1.223 brouard 7152: }else{
7153: i=i-ncovmodel;
7154: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7155: fprintf(ficgp," (1.");
7156: }
1.227 brouard 7157:
1.223 brouard 7158: if(ng != 1){
7159: fprintf(ficgp,")/(1");
1.227 brouard 7160:
1.223 brouard 7161: for(k1=1; k1 <=nlstate; k1++){
7162: if(nagesqr==0)
7163: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7164: else /* nagesqr =1 */
7165: 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 7166:
1.223 brouard 7167: ij=1;
7168: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7169: if((j-2)==Tage[ij]) { /* Bug valgrind */
7170: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7171: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7172: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7173: ij++;
7174: }
7175: }
7176: else
1.225 brouard 7177: 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 7178: }
7179: fprintf(ficgp,")");
7180: }
7181: fprintf(ficgp,")");
7182: if(ng ==2)
7183: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7184: else /* ng= 3 */
7185: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7186: }else{ /* end ng <> 1 */
7187: if( k !=k2) /* logit p11 is hard to draw */
7188: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7189: }
7190: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7191: fprintf(ficgp,",");
7192: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7193: fprintf(ficgp,",");
7194: i=i+ncovmodel;
7195: } /* end k */
7196: } /* end k2 */
7197: fprintf(ficgp,"\n set out\n");
7198: } /* end jk */
7199: } /* end ng */
7200: /* avoid: */
7201: fflush(ficgp);
1.126 brouard 7202: } /* end gnuplot */
7203:
7204:
7205: /*************** Moving average **************/
1.219 brouard 7206: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7207: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7208:
1.222 brouard 7209: int i, cpt, cptcod;
7210: int modcovmax =1;
7211: int mobilavrange, mob;
7212: int iage=0;
7213:
7214: double sum=0.;
7215: double age;
7216: double *sumnewp, *sumnewm;
7217: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7218:
7219:
1.225 brouard 7220: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7221: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7222:
7223: sumnewp = vector(1,ncovcombmax);
7224: sumnewm = vector(1,ncovcombmax);
7225: agemingood = vector(1,ncovcombmax);
7226: agemaxgood = vector(1,ncovcombmax);
7227:
7228: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7229: sumnewm[cptcod]=0.;
7230: sumnewp[cptcod]=0.;
7231: agemingood[cptcod]=0;
7232: agemaxgood[cptcod]=0;
7233: }
7234: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7235:
7236: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7237: if(mobilav==1) mobilavrange=5; /* default */
7238: else mobilavrange=mobilav;
7239: for (age=bage; age<=fage; age++)
7240: for (i=1; i<=nlstate;i++)
7241: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7242: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7243: /* We keep the original values on the extreme ages bage, fage and for
7244: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7245: we use a 5 terms etc. until the borders are no more concerned.
7246: */
7247: for (mob=3;mob <=mobilavrange;mob=mob+2){
7248: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7249: for (i=1; i<=nlstate;i++){
7250: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7251: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7252: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7253: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7254: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7255: }
7256: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7257: }
7258: }
7259: }/* end age */
7260: }/* end mob */
7261: }else
7262: return -1;
7263: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7264: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7265: if(invalidvarcomb[cptcod]){
7266: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7267: continue;
7268: }
1.219 brouard 7269:
1.222 brouard 7270: agemingood[cptcod]=fage-(mob-1)/2;
7271: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7272: sumnewm[cptcod]=0.;
7273: for (i=1; i<=nlstate;i++){
7274: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7275: }
7276: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7277: agemingood[cptcod]=age;
7278: }else{ /* bad */
7279: for (i=1; i<=nlstate;i++){
7280: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7281: } /* i */
7282: } /* end bad */
7283: }/* age */
7284: sum=0.;
7285: for (i=1; i<=nlstate;i++){
7286: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7287: }
7288: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7289: 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);
7290: /* for (i=1; i<=nlstate;i++){ */
7291: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7292: /* } /\* i *\/ */
7293: } /* end bad */
7294: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7295: /* From youngest, finding the oldest wrong */
7296: agemaxgood[cptcod]=bage+(mob-1)/2;
7297: for (age=bage+(mob-1)/2; age<=fage; age++){
7298: sumnewm[cptcod]=0.;
7299: for (i=1; i<=nlstate;i++){
7300: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7301: }
7302: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7303: agemaxgood[cptcod]=age;
7304: }else{ /* bad */
7305: for (i=1; i<=nlstate;i++){
7306: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7307: } /* i */
7308: } /* end bad */
7309: }/* age */
7310: sum=0.;
7311: for (i=1; i<=nlstate;i++){
7312: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7313: }
7314: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7315: 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);
7316: /* for (i=1; i<=nlstate;i++){ */
7317: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7318: /* } /\* i *\/ */
7319: } /* end bad */
7320:
7321: for (age=bage; age<=fage; age++){
1.235 brouard 7322: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7323: sumnewp[cptcod]=0.;
7324: sumnewm[cptcod]=0.;
7325: for (i=1; i<=nlstate;i++){
7326: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7327: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7328: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7329: }
7330: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7331: }
7332: /* printf("\n"); */
7333: /* } */
7334: /* brutal averaging */
7335: for (i=1; i<=nlstate;i++){
7336: for (age=1; age<=bage; age++){
7337: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7338: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7339: }
7340: for (age=fage; age<=AGESUP; age++){
7341: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7342: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7343: }
7344: } /* end i status */
7345: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7346: for (age=1; age<=AGESUP; age++){
7347: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7348: mobaverage[(int)age][i][cptcod]=0.;
7349: }
7350: }
7351: }/* end cptcod */
7352: free_vector(sumnewm,1, ncovcombmax);
7353: free_vector(sumnewp,1, ncovcombmax);
7354: free_vector(agemaxgood,1, ncovcombmax);
7355: free_vector(agemingood,1, ncovcombmax);
7356: return 0;
7357: }/* End movingaverage */
1.218 brouard 7358:
1.126 brouard 7359:
7360: /************** Forecasting ******************/
1.235 brouard 7361: 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 7362: /* proj1, year, month, day of starting projection
7363: agemin, agemax range of age
7364: dateprev1 dateprev2 range of dates during which prevalence is computed
7365: anproj2 year of en of projection (same day and month as proj1).
7366: */
1.235 brouard 7367: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7368: double agec; /* generic age */
7369: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7370: double *popeffectif,*popcount;
7371: double ***p3mat;
1.218 brouard 7372: /* double ***mobaverage; */
1.126 brouard 7373: char fileresf[FILENAMELENGTH];
7374:
7375: agelim=AGESUP;
1.211 brouard 7376: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7377: in each health status at the date of interview (if between dateprev1 and dateprev2).
7378: We still use firstpass and lastpass as another selection.
7379: */
1.214 brouard 7380: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7381: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7382:
1.201 brouard 7383: strcpy(fileresf,"F_");
7384: strcat(fileresf,fileresu);
1.126 brouard 7385: if((ficresf=fopen(fileresf,"w"))==NULL) {
7386: printf("Problem with forecast resultfile: %s\n", fileresf);
7387: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7388: }
1.235 brouard 7389: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7390: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7391:
1.225 brouard 7392: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7393:
7394:
7395: stepsize=(int) (stepm+YEARM-1)/YEARM;
7396: if (stepm<=12) stepsize=1;
7397: if(estepm < stepm){
7398: printf ("Problem %d lower than %d\n",estepm, stepm);
7399: }
7400: else hstepm=estepm;
7401:
7402: hstepm=hstepm/stepm;
7403: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7404: fractional in yp1 */
7405: anprojmean=yp;
7406: yp2=modf((yp1*12),&yp);
7407: mprojmean=yp;
7408: yp1=modf((yp2*30.5),&yp);
7409: jprojmean=yp;
7410: if(jprojmean==0) jprojmean=1;
7411: if(mprojmean==0) jprojmean=1;
7412:
1.227 brouard 7413: i1=pow(2,cptcoveff);
1.126 brouard 7414: if (cptcovn < 1){i1=1;}
7415:
7416: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7417:
7418: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7419:
1.126 brouard 7420: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7421: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7422: for(k=1; k<=i1;k++){
7423: if(TKresult[nres]!= k)
7424: continue;
1.227 brouard 7425: if(invalidvarcomb[k]){
7426: printf("\nCombination (%d) projection ignored because no cases \n",k);
7427: continue;
7428: }
7429: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7430: for(j=1;j<=cptcoveff;j++) {
7431: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7432: }
1.235 brouard 7433: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7434: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7435: }
1.227 brouard 7436: fprintf(ficresf," yearproj age");
7437: for(j=1; j<=nlstate+ndeath;j++){
7438: for(i=1; i<=nlstate;i++)
7439: fprintf(ficresf," p%d%d",i,j);
7440: fprintf(ficresf," wp.%d",j);
7441: }
7442: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7443: fprintf(ficresf,"\n");
7444: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7445: for (agec=fage; agec>=(ageminpar-1); agec--){
7446: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7447: nhstepm = nhstepm/hstepm;
7448: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7449: oldm=oldms;savm=savms;
1.235 brouard 7450: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7451:
7452: for (h=0; h<=nhstepm; h++){
7453: if (h*hstepm/YEARM*stepm ==yearp) {
7454: fprintf(ficresf,"\n");
7455: for(j=1;j<=cptcoveff;j++)
7456: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7457: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7458: }
7459: for(j=1; j<=nlstate+ndeath;j++) {
7460: ppij=0.;
7461: for(i=1; i<=nlstate;i++) {
7462: if (mobilav==1)
7463: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7464: else {
7465: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7466: }
7467: if (h*hstepm/YEARM*stepm== yearp) {
7468: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7469: }
7470: } /* end i */
7471: if (h*hstepm/YEARM*stepm==yearp) {
7472: fprintf(ficresf," %.3f", ppij);
7473: }
7474: }/* end j */
7475: } /* end h */
7476: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7477: } /* end agec */
7478: } /* end yearp */
7479: } /* end k */
1.219 brouard 7480:
1.126 brouard 7481: fclose(ficresf);
1.215 brouard 7482: printf("End of Computing forecasting \n");
7483: fprintf(ficlog,"End of Computing forecasting\n");
7484:
1.126 brouard 7485: }
7486:
1.218 brouard 7487: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7488: /* 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 7489: /* /\* back1, year, month, day of starting backection */
7490: /* agemin, agemax range of age */
7491: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7492: /* anback2 year of en of backection (same day and month as back1). */
7493: /* *\/ */
7494: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7495: /* double agec; /\* generic age *\/ */
7496: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7497: /* double *popeffectif,*popcount; */
7498: /* double ***p3mat; */
7499: /* /\* double ***mobaverage; *\/ */
7500: /* char fileresfb[FILENAMELENGTH]; */
7501:
7502: /* agelim=AGESUP; */
7503: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7504: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7505: /* We still use firstpass and lastpass as another selection. */
7506: /* *\/ */
7507: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7508: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7509: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7510:
7511: /* strcpy(fileresfb,"FB_"); */
7512: /* strcat(fileresfb,fileresu); */
7513: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7514: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7515: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7516: /* } */
7517: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7518: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7519:
1.225 brouard 7520: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7521:
7522: /* /\* if (mobilav!=0) { *\/ */
7523: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7524: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7525: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7526: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7527: /* /\* } *\/ */
7528: /* /\* } *\/ */
7529:
7530: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7531: /* if (stepm<=12) stepsize=1; */
7532: /* if(estepm < stepm){ */
7533: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7534: /* } */
7535: /* else hstepm=estepm; */
7536:
7537: /* hstepm=hstepm/stepm; */
7538: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7539: /* fractional in yp1 *\/ */
7540: /* anprojmean=yp; */
7541: /* yp2=modf((yp1*12),&yp); */
7542: /* mprojmean=yp; */
7543: /* yp1=modf((yp2*30.5),&yp); */
7544: /* jprojmean=yp; */
7545: /* if(jprojmean==0) jprojmean=1; */
7546: /* if(mprojmean==0) jprojmean=1; */
7547:
1.225 brouard 7548: /* i1=cptcoveff; */
1.218 brouard 7549: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7550:
1.218 brouard 7551: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7552:
1.218 brouard 7553: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7554:
7555: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7556: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7557: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7558: /* k=k+1; */
7559: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7560: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7561: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7562: /* } */
7563: /* fprintf(ficresfb," yearbproj age"); */
7564: /* for(j=1; j<=nlstate+ndeath;j++){ */
7565: /* for(i=1; i<=nlstate;i++) */
7566: /* fprintf(ficresfb," p%d%d",i,j); */
7567: /* fprintf(ficresfb," p.%d",j); */
7568: /* } */
7569: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7570: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7571: /* fprintf(ficresfb,"\n"); */
7572: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7573: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7574: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7575: /* nhstepm = nhstepm/hstepm; */
7576: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7577: /* oldm=oldms;savm=savms; */
7578: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7579: /* for (h=0; h<=nhstepm; h++){ */
7580: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7581: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7582: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7583: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7584: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7585: /* } */
7586: /* for(j=1; j<=nlstate+ndeath;j++) { */
7587: /* ppij=0.; */
7588: /* for(i=1; i<=nlstate;i++) { */
7589: /* if (mobilav==1) */
7590: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7591: /* else { */
7592: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7593: /* } */
7594: /* if (h*hstepm/YEARM*stepm== yearp) { */
7595: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7596: /* } */
7597: /* } /\* end i *\/ */
7598: /* if (h*hstepm/YEARM*stepm==yearp) { */
7599: /* fprintf(ficresfb," %.3f", ppij); */
7600: /* } */
7601: /* }/\* end j *\/ */
7602: /* } /\* end h *\/ */
7603: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7604: /* } /\* end agec *\/ */
7605: /* } /\* end yearp *\/ */
7606: /* } /\* end cptcod *\/ */
7607: /* } /\* end cptcov *\/ */
7608:
7609: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7610:
7611: /* fclose(ficresfb); */
7612: /* printf("End of Computing Back forecasting \n"); */
7613: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7614:
1.218 brouard 7615: /* } */
1.217 brouard 7616:
1.126 brouard 7617: /************** Forecasting *****not tested NB*************/
1.227 brouard 7618: /* 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 7619:
1.227 brouard 7620: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7621: /* int *popage; */
7622: /* double calagedatem, agelim, kk1, kk2; */
7623: /* double *popeffectif,*popcount; */
7624: /* double ***p3mat,***tabpop,***tabpopprev; */
7625: /* /\* double ***mobaverage; *\/ */
7626: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7627:
1.227 brouard 7628: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7629: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7630: /* agelim=AGESUP; */
7631: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7632:
1.227 brouard 7633: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7634:
7635:
1.227 brouard 7636: /* strcpy(filerespop,"POP_"); */
7637: /* strcat(filerespop,fileresu); */
7638: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7639: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7640: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7641: /* } */
7642: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7643: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7644:
1.227 brouard 7645: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7646:
1.227 brouard 7647: /* /\* if (mobilav!=0) { *\/ */
7648: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7649: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7650: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7651: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7652: /* /\* } *\/ */
7653: /* /\* } *\/ */
1.126 brouard 7654:
1.227 brouard 7655: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7656: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7657:
1.227 brouard 7658: /* agelim=AGESUP; */
1.126 brouard 7659:
1.227 brouard 7660: /* hstepm=1; */
7661: /* hstepm=hstepm/stepm; */
1.218 brouard 7662:
1.227 brouard 7663: /* if (popforecast==1) { */
7664: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7665: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7666: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7667: /* } */
7668: /* popage=ivector(0,AGESUP); */
7669: /* popeffectif=vector(0,AGESUP); */
7670: /* popcount=vector(0,AGESUP); */
1.126 brouard 7671:
1.227 brouard 7672: /* i=1; */
7673: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7674:
1.227 brouard 7675: /* imx=i; */
7676: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7677: /* } */
1.218 brouard 7678:
1.227 brouard 7679: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7680: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7681: /* k=k+1; */
7682: /* fprintf(ficrespop,"\n#******"); */
7683: /* for(j=1;j<=cptcoveff;j++) { */
7684: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7685: /* } */
7686: /* fprintf(ficrespop,"******\n"); */
7687: /* fprintf(ficrespop,"# Age"); */
7688: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7689: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7690:
1.227 brouard 7691: /* for (cpt=0; cpt<=0;cpt++) { */
7692: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7693:
1.227 brouard 7694: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7695: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7696: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7697:
1.227 brouard 7698: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7699: /* oldm=oldms;savm=savms; */
7700: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7701:
1.227 brouard 7702: /* for (h=0; h<=nhstepm; h++){ */
7703: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7704: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7705: /* } */
7706: /* for(j=1; j<=nlstate+ndeath;j++) { */
7707: /* kk1=0.;kk2=0; */
7708: /* for(i=1; i<=nlstate;i++) { */
7709: /* if (mobilav==1) */
7710: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7711: /* else { */
7712: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7713: /* } */
7714: /* } */
7715: /* if (h==(int)(calagedatem+12*cpt)){ */
7716: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7717: /* /\*fprintf(ficrespop," %.3f", kk1); */
7718: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7719: /* } */
7720: /* } */
7721: /* for(i=1; i<=nlstate;i++){ */
7722: /* kk1=0.; */
7723: /* for(j=1; j<=nlstate;j++){ */
7724: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7725: /* } */
7726: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7727: /* } */
1.218 brouard 7728:
1.227 brouard 7729: /* if (h==(int)(calagedatem+12*cpt)) */
7730: /* for(j=1; j<=nlstate;j++) */
7731: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7732: /* } */
7733: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7734: /* } */
7735: /* } */
1.218 brouard 7736:
1.227 brouard 7737: /* /\******\/ */
1.218 brouard 7738:
1.227 brouard 7739: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7740: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7741: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7742: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7743: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7744:
1.227 brouard 7745: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7746: /* oldm=oldms;savm=savms; */
7747: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7748: /* for (h=0; h<=nhstepm; h++){ */
7749: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7750: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7751: /* } */
7752: /* for(j=1; j<=nlstate+ndeath;j++) { */
7753: /* kk1=0.;kk2=0; */
7754: /* for(i=1; i<=nlstate;i++) { */
7755: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7756: /* } */
7757: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7758: /* } */
7759: /* } */
7760: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7761: /* } */
7762: /* } */
7763: /* } */
7764: /* } */
1.218 brouard 7765:
1.227 brouard 7766: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7767:
1.227 brouard 7768: /* if (popforecast==1) { */
7769: /* free_ivector(popage,0,AGESUP); */
7770: /* free_vector(popeffectif,0,AGESUP); */
7771: /* free_vector(popcount,0,AGESUP); */
7772: /* } */
7773: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7774: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7775: /* fclose(ficrespop); */
7776: /* } /\* End of popforecast *\/ */
1.218 brouard 7777:
1.126 brouard 7778: int fileappend(FILE *fichier, char *optionfich)
7779: {
7780: if((fichier=fopen(optionfich,"a"))==NULL) {
7781: printf("Problem with file: %s\n", optionfich);
7782: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7783: return (0);
7784: }
7785: fflush(fichier);
7786: return (1);
7787: }
7788:
7789:
7790: /**************** function prwizard **********************/
7791: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7792: {
7793:
7794: /* Wizard to print covariance matrix template */
7795:
1.164 brouard 7796: char ca[32], cb[32];
7797: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7798: int numlinepar;
7799:
7800: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7801: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7802: for(i=1; i <=nlstate; i++){
7803: jj=0;
7804: for(j=1; j <=nlstate+ndeath; j++){
7805: if(j==i) continue;
7806: jj++;
7807: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7808: printf("%1d%1d",i,j);
7809: fprintf(ficparo,"%1d%1d",i,j);
7810: for(k=1; k<=ncovmodel;k++){
7811: /* printf(" %lf",param[i][j][k]); */
7812: /* fprintf(ficparo," %lf",param[i][j][k]); */
7813: printf(" 0.");
7814: fprintf(ficparo," 0.");
7815: }
7816: printf("\n");
7817: fprintf(ficparo,"\n");
7818: }
7819: }
7820: printf("# Scales (for hessian or gradient estimation)\n");
7821: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7822: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7823: for(i=1; i <=nlstate; i++){
7824: jj=0;
7825: for(j=1; j <=nlstate+ndeath; j++){
7826: if(j==i) continue;
7827: jj++;
7828: fprintf(ficparo,"%1d%1d",i,j);
7829: printf("%1d%1d",i,j);
7830: fflush(stdout);
7831: for(k=1; k<=ncovmodel;k++){
7832: /* printf(" %le",delti3[i][j][k]); */
7833: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7834: printf(" 0.");
7835: fprintf(ficparo," 0.");
7836: }
7837: numlinepar++;
7838: printf("\n");
7839: fprintf(ficparo,"\n");
7840: }
7841: }
7842: printf("# Covariance matrix\n");
7843: /* # 121 Var(a12)\n\ */
7844: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7845: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7846: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7847: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7848: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7849: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7850: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7851: fflush(stdout);
7852: fprintf(ficparo,"# Covariance matrix\n");
7853: /* # 121 Var(a12)\n\ */
7854: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7855: /* # ...\n\ */
7856: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7857:
7858: for(itimes=1;itimes<=2;itimes++){
7859: jj=0;
7860: for(i=1; i <=nlstate; i++){
7861: for(j=1; j <=nlstate+ndeath; j++){
7862: if(j==i) continue;
7863: for(k=1; k<=ncovmodel;k++){
7864: jj++;
7865: ca[0]= k+'a'-1;ca[1]='\0';
7866: if(itimes==1){
7867: printf("#%1d%1d%d",i,j,k);
7868: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7869: }else{
7870: printf("%1d%1d%d",i,j,k);
7871: fprintf(ficparo,"%1d%1d%d",i,j,k);
7872: /* printf(" %.5le",matcov[i][j]); */
7873: }
7874: ll=0;
7875: for(li=1;li <=nlstate; li++){
7876: for(lj=1;lj <=nlstate+ndeath; lj++){
7877: if(lj==li) continue;
7878: for(lk=1;lk<=ncovmodel;lk++){
7879: ll++;
7880: if(ll<=jj){
7881: cb[0]= lk +'a'-1;cb[1]='\0';
7882: if(ll<jj){
7883: if(itimes==1){
7884: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7885: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7886: }else{
7887: printf(" 0.");
7888: fprintf(ficparo," 0.");
7889: }
7890: }else{
7891: if(itimes==1){
7892: printf(" Var(%s%1d%1d)",ca,i,j);
7893: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7894: }else{
7895: printf(" 0.");
7896: fprintf(ficparo," 0.");
7897: }
7898: }
7899: }
7900: } /* end lk */
7901: } /* end lj */
7902: } /* end li */
7903: printf("\n");
7904: fprintf(ficparo,"\n");
7905: numlinepar++;
7906: } /* end k*/
7907: } /*end j */
7908: } /* end i */
7909: } /* end itimes */
7910:
7911: } /* end of prwizard */
7912: /******************* Gompertz Likelihood ******************************/
7913: double gompertz(double x[])
7914: {
7915: double A,B,L=0.0,sump=0.,num=0.;
7916: int i,n=0; /* n is the size of the sample */
7917:
1.220 brouard 7918: for (i=1;i<=imx ; i++) {
1.126 brouard 7919: sump=sump+weight[i];
7920: /* sump=sump+1;*/
7921: num=num+1;
7922: }
7923:
7924:
7925: /* for (i=0; i<=imx; i++)
7926: 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]);*/
7927:
7928: for (i=1;i<=imx ; i++)
7929: {
7930: if (cens[i] == 1 && wav[i]>1)
7931: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
7932:
7933: if (cens[i] == 0 && wav[i]>1)
7934: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
7935: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
7936:
7937: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7938: if (wav[i] > 1 ) { /* ??? */
7939: L=L+A*weight[i];
7940: /* 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]);*/
7941: }
7942: }
7943:
7944: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7945:
7946: return -2*L*num/sump;
7947: }
7948:
1.136 brouard 7949: #ifdef GSL
7950: /******************* Gompertz_f Likelihood ******************************/
7951: double gompertz_f(const gsl_vector *v, void *params)
7952: {
7953: double A,B,LL=0.0,sump=0.,num=0.;
7954: double *x= (double *) v->data;
7955: int i,n=0; /* n is the size of the sample */
7956:
7957: for (i=0;i<=imx-1 ; i++) {
7958: sump=sump+weight[i];
7959: /* sump=sump+1;*/
7960: num=num+1;
7961: }
7962:
7963:
7964: /* for (i=0; i<=imx; i++)
7965: 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]);*/
7966: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
7967: for (i=1;i<=imx ; i++)
7968: {
7969: if (cens[i] == 1 && wav[i]>1)
7970: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
7971:
7972: if (cens[i] == 0 && wav[i]>1)
7973: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
7974: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
7975:
7976: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7977: if (wav[i] > 1 ) { /* ??? */
7978: LL=LL+A*weight[i];
7979: /* 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]);*/
7980: }
7981: }
7982:
7983: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7984: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
7985:
7986: return -2*LL*num/sump;
7987: }
7988: #endif
7989:
1.126 brouard 7990: /******************* Printing html file ***********/
1.201 brouard 7991: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7992: int lastpass, int stepm, int weightopt, char model[],\
7993: int imx, double p[],double **matcov,double agemortsup){
7994: int i,k;
7995:
7996: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
7997: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
7998: for (i=1;i<=2;i++)
7999: 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 8000: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8001: fprintf(fichtm,"</ul>");
8002:
8003: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8004:
8005: 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>");
8006:
8007: for (k=agegomp;k<(agemortsup-2);k++)
8008: 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]);
8009:
8010:
8011: fflush(fichtm);
8012: }
8013:
8014: /******************* Gnuplot file **************/
1.201 brouard 8015: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8016:
8017: char dirfileres[132],optfileres[132];
1.164 brouard 8018:
1.126 brouard 8019: int ng;
8020:
8021:
8022: /*#ifdef windows */
8023: fprintf(ficgp,"cd \"%s\" \n",pathc);
8024: /*#endif */
8025:
8026:
8027: strcpy(dirfileres,optionfilefiname);
8028: strcpy(optfileres,"vpl");
1.199 brouard 8029: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8030: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8031: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8032: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8033: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8034:
8035: }
8036:
1.136 brouard 8037: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8038: {
1.126 brouard 8039:
1.136 brouard 8040: /*-------- data file ----------*/
8041: FILE *fic;
8042: char dummy[]=" ";
1.240 brouard 8043: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8044: int lstra;
1.136 brouard 8045: int linei, month, year,iout;
8046: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8047: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8048: char *stratrunc;
1.223 brouard 8049:
1.240 brouard 8050: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8051: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8052:
1.240 brouard 8053: for(v=1; v <=ncovcol;v++){
8054: DummyV[v]=0;
8055: FixedV[v]=0;
8056: }
8057: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8058: DummyV[v]=1;
8059: FixedV[v]=0;
8060: }
8061: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8062: DummyV[v]=0;
8063: FixedV[v]=1;
8064: }
8065: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8066: DummyV[v]=1;
8067: FixedV[v]=1;
8068: }
8069: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8070: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8071: 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]);
8072: }
1.126 brouard 8073:
1.136 brouard 8074: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8075: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8076: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8077: }
1.126 brouard 8078:
1.136 brouard 8079: i=1;
8080: linei=0;
8081: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8082: linei=linei+1;
8083: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8084: if(line[j] == '\t')
8085: line[j] = ' ';
8086: }
8087: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8088: ;
8089: };
8090: line[j+1]=0; /* Trims blanks at end of line */
8091: if(line[0]=='#'){
8092: fprintf(ficlog,"Comment line\n%s\n",line);
8093: printf("Comment line\n%s\n",line);
8094: continue;
8095: }
8096: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8097: strcpy(line, linetmp);
1.223 brouard 8098:
8099: /* Loops on waves */
8100: for (j=maxwav;j>=1;j--){
8101: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8102: cutv(stra, strb, line, ' ');
8103: if(strb[0]=='.') { /* Missing value */
8104: lval=-1;
8105: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8106: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8107: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8108: 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);
8109: 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);
8110: return 1;
8111: }
8112: }else{
8113: errno=0;
8114: /* what_kind_of_number(strb); */
8115: dval=strtod(strb,&endptr);
8116: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8117: /* if(strb != endptr && *endptr == '\0') */
8118: /* dval=dlval; */
8119: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8120: if( strb[0]=='\0' || (*endptr != '\0')){
8121: 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);
8122: 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);
8123: return 1;
8124: }
8125: cotqvar[j][iv][i]=dval;
8126: cotvar[j][ntv+iv][i]=dval;
8127: }
8128: strcpy(line,stra);
1.223 brouard 8129: }/* end loop ntqv */
1.225 brouard 8130:
1.223 brouard 8131: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8132: cutv(stra, strb, line, ' ');
8133: if(strb[0]=='.') { /* Missing value */
8134: lval=-1;
8135: }else{
8136: errno=0;
8137: lval=strtol(strb,&endptr,10);
8138: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8139: if( strb[0]=='\0' || (*endptr != '\0')){
8140: 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);
8141: 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);
8142: return 1;
8143: }
8144: }
8145: if(lval <-1 || lval >1){
8146: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8147: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8148: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8149: For example, for multinomial values like 1, 2 and 3,\n \
8150: build V1=0 V2=0 for the reference value (1),\n \
8151: V1=1 V2=0 for (2) \n \
1.223 brouard 8152: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8153: output of IMaCh is often meaningless.\n \
1.223 brouard 8154: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8155: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8156: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8157: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8158: For example, for multinomial values like 1, 2 and 3,\n \
8159: build V1=0 V2=0 for the reference value (1),\n \
8160: V1=1 V2=0 for (2) \n \
1.223 brouard 8161: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8162: output of IMaCh is often meaningless.\n \
1.223 brouard 8163: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8164: return 1;
8165: }
8166: cotvar[j][iv][i]=(double)(lval);
8167: strcpy(line,stra);
1.223 brouard 8168: }/* end loop ntv */
1.225 brouard 8169:
1.223 brouard 8170: /* Statuses at wave */
1.137 brouard 8171: cutv(stra, strb, line, ' ');
1.223 brouard 8172: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8173: lval=-1;
1.136 brouard 8174: }else{
1.238 brouard 8175: errno=0;
8176: lval=strtol(strb,&endptr,10);
8177: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8178: if( strb[0]=='\0' || (*endptr != '\0')){
8179: 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);
8180: 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);
8181: return 1;
8182: }
1.136 brouard 8183: }
1.225 brouard 8184:
1.136 brouard 8185: s[j][i]=lval;
1.225 brouard 8186:
1.223 brouard 8187: /* Date of Interview */
1.136 brouard 8188: strcpy(line,stra);
8189: cutv(stra, strb,line,' ');
1.169 brouard 8190: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8191: }
1.169 brouard 8192: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8193: month=99;
8194: year=9999;
1.136 brouard 8195: }else{
1.225 brouard 8196: 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);
8197: 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);
8198: return 1;
1.136 brouard 8199: }
8200: anint[j][i]= (double) year;
8201: mint[j][i]= (double)month;
8202: strcpy(line,stra);
1.223 brouard 8203: } /* End loop on waves */
1.225 brouard 8204:
1.223 brouard 8205: /* Date of death */
1.136 brouard 8206: cutv(stra, strb,line,' ');
1.169 brouard 8207: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8208: }
1.169 brouard 8209: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8210: month=99;
8211: year=9999;
8212: }else{
1.141 brouard 8213: 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 8214: 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);
8215: return 1;
1.136 brouard 8216: }
8217: andc[i]=(double) year;
8218: moisdc[i]=(double) month;
8219: strcpy(line,stra);
8220:
1.223 brouard 8221: /* Date of birth */
1.136 brouard 8222: cutv(stra, strb,line,' ');
1.169 brouard 8223: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8224: }
1.169 brouard 8225: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8226: month=99;
8227: year=9999;
8228: }else{
1.141 brouard 8229: 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);
8230: 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 8231: return 1;
1.136 brouard 8232: }
8233: if (year==9999) {
1.141 brouard 8234: 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);
8235: 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 8236: return 1;
8237:
1.136 brouard 8238: }
8239: annais[i]=(double)(year);
8240: moisnais[i]=(double)(month);
8241: strcpy(line,stra);
1.225 brouard 8242:
1.223 brouard 8243: /* Sample weight */
1.136 brouard 8244: cutv(stra, strb,line,' ');
8245: errno=0;
8246: dval=strtod(strb,&endptr);
8247: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8248: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8249: 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 8250: fflush(ficlog);
8251: return 1;
8252: }
8253: weight[i]=dval;
8254: strcpy(line,stra);
1.225 brouard 8255:
1.223 brouard 8256: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8257: cutv(stra, strb, line, ' ');
8258: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8259: lval=-1;
1.223 brouard 8260: }else{
1.225 brouard 8261: errno=0;
8262: /* what_kind_of_number(strb); */
8263: dval=strtod(strb,&endptr);
8264: /* if(strb != endptr && *endptr == '\0') */
8265: /* dval=dlval; */
8266: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8267: if( strb[0]=='\0' || (*endptr != '\0')){
8268: 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);
8269: 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);
8270: return 1;
8271: }
8272: coqvar[iv][i]=dval;
1.226 brouard 8273: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8274: }
8275: strcpy(line,stra);
8276: }/* end loop nqv */
1.136 brouard 8277:
1.223 brouard 8278: /* Covariate values */
1.136 brouard 8279: for (j=ncovcol;j>=1;j--){
8280: cutv(stra, strb,line,' ');
1.223 brouard 8281: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8282: lval=-1;
1.136 brouard 8283: }else{
1.225 brouard 8284: errno=0;
8285: lval=strtol(strb,&endptr,10);
8286: if( strb[0]=='\0' || (*endptr != '\0')){
8287: 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);
8288: 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);
8289: return 1;
8290: }
1.136 brouard 8291: }
8292: if(lval <-1 || lval >1){
1.225 brouard 8293: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8294: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8295: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8296: For example, for multinomial values like 1, 2 and 3,\n \
8297: build V1=0 V2=0 for the reference value (1),\n \
8298: V1=1 V2=0 for (2) \n \
1.136 brouard 8299: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8300: output of IMaCh is often meaningless.\n \
1.136 brouard 8301: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8302: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8303: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8304: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8305: For example, for multinomial values like 1, 2 and 3,\n \
8306: build V1=0 V2=0 for the reference value (1),\n \
8307: V1=1 V2=0 for (2) \n \
1.136 brouard 8308: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8309: output of IMaCh is often meaningless.\n \
1.136 brouard 8310: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8311: return 1;
1.136 brouard 8312: }
8313: covar[j][i]=(double)(lval);
8314: strcpy(line,stra);
8315: }
8316: lstra=strlen(stra);
1.225 brouard 8317:
1.136 brouard 8318: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8319: stratrunc = &(stra[lstra-9]);
8320: num[i]=atol(stratrunc);
8321: }
8322: else
8323: num[i]=atol(stra);
8324: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8325: 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;}*/
8326:
8327: i=i+1;
8328: } /* End loop reading data */
1.225 brouard 8329:
1.136 brouard 8330: *imax=i-1; /* Number of individuals */
8331: fclose(fic);
1.225 brouard 8332:
1.136 brouard 8333: return (0);
1.164 brouard 8334: /* endread: */
1.225 brouard 8335: printf("Exiting readdata: ");
8336: fclose(fic);
8337: return (1);
1.223 brouard 8338: }
1.126 brouard 8339:
1.234 brouard 8340: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8341: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8342: while (*p2 == ' ')
1.234 brouard 8343: p2++;
8344: /* while ((*p1++ = *p2++) !=0) */
8345: /* ; */
8346: /* do */
8347: /* while (*p2 == ' ') */
8348: /* p2++; */
8349: /* while (*p1++ == *p2++); */
8350: *stri=p2;
1.145 brouard 8351: }
8352:
1.235 brouard 8353: int decoderesult ( char resultline[], int nres)
1.230 brouard 8354: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8355: {
1.235 brouard 8356: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8357: char resultsav[MAXLINE];
1.234 brouard 8358: int resultmodel[MAXLINE];
8359: int modelresult[MAXLINE];
1.230 brouard 8360: char stra[80], strb[80], strc[80], strd[80],stre[80];
8361:
1.234 brouard 8362: removefirstspace(&resultline);
1.233 brouard 8363: printf("decoderesult:%s\n",resultline);
1.230 brouard 8364:
8365: if (strstr(resultline,"v") !=0){
8366: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8367: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8368: return 1;
8369: }
8370: trimbb(resultsav, resultline);
8371: if (strlen(resultsav) >1){
8372: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8373: }
1.234 brouard 8374: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8375: 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);
8376: 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);
8377: }
8378: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8379: if(nbocc(resultsav,'=') >1){
8380: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8381: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8382: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8383: }else
8384: cutl(strc,strd,resultsav,'=');
1.230 brouard 8385: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8386:
1.230 brouard 8387: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8388: Tvarsel[k]=atoi(strc);
8389: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8390: /* cptcovsel++; */
8391: if (nbocc(stra,'=') >0)
8392: strcpy(resultsav,stra); /* and analyzes it */
8393: }
1.235 brouard 8394: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8395: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8396: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8397: match=0;
1.236 brouard 8398: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8399: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8400: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8401: match=1;
8402: break;
8403: }
8404: }
8405: if(match == 0){
8406: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8407: }
8408: }
8409: }
1.235 brouard 8410: /* Checking for missing or useless values in comparison of current model needs */
8411: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8412: match=0;
1.235 brouard 8413: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8414: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8415: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8416: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8417: ++match;
8418: }
8419: }
8420: }
8421: if(match == 0){
8422: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8423: }else if(match > 1){
8424: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8425: }
8426: }
1.235 brouard 8427:
1.234 brouard 8428: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8429: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8430: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8431: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8432: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8433: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8434: /* 1 0 0 0 */
8435: /* 2 1 0 0 */
8436: /* 3 0 1 0 */
8437: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8438: /* 5 0 0 1 */
8439: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8440: /* 7 0 1 1 */
8441: /* 8 1 1 1 */
1.237 brouard 8442: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8443: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8444: /* V5*age V5 known which value for nres? */
8445: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8446: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8447: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8448: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8449: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8450: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8451: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8452: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8453: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8454: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8455: k4++;;
8456: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8457: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8458: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8459: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8460: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8461: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8462: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8463: k4q++;;
8464: }
8465: }
1.234 brouard 8466:
1.235 brouard 8467: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8468: return (0);
8469: }
1.235 brouard 8470:
1.230 brouard 8471: int decodemodel( char model[], int lastobs)
8472: /**< This routine decodes the model and returns:
1.224 brouard 8473: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8474: * - nagesqr = 1 if age*age in the model, otherwise 0.
8475: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8476: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8477: * - cptcovage number of covariates with age*products =2
8478: * - cptcovs number of simple covariates
8479: * - 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
8480: * which is a new column after the 9 (ncovcol) variables.
8481: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8482: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8483: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8484: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8485: */
1.136 brouard 8486: {
1.238 brouard 8487: int i, j, k, ks, v;
1.227 brouard 8488: int j1, k1, k2, k3, k4;
1.136 brouard 8489: char modelsav[80];
1.145 brouard 8490: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8491: char *strpt;
1.136 brouard 8492:
1.145 brouard 8493: /*removespace(model);*/
1.136 brouard 8494: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8495: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8496: if (strstr(model,"AGE") !=0){
1.192 brouard 8497: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8498: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8499: return 1;
8500: }
1.141 brouard 8501: if (strstr(model,"v") !=0){
8502: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8503: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8504: return 1;
8505: }
1.187 brouard 8506: strcpy(modelsav,model);
8507: if ((strpt=strstr(model,"age*age")) !=0){
8508: printf(" strpt=%s, model=%s\n",strpt, model);
8509: if(strpt != model){
1.234 brouard 8510: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8511: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8512: corresponding column of parameters.\n",model);
1.234 brouard 8513: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8514: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8515: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8516: return 1;
1.225 brouard 8517: }
1.187 brouard 8518: nagesqr=1;
8519: if (strstr(model,"+age*age") !=0)
1.234 brouard 8520: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8521: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8522: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8523: else
1.234 brouard 8524: substrchaine(modelsav, model, "age*age");
1.187 brouard 8525: }else
8526: nagesqr=0;
8527: if (strlen(modelsav) >1){
8528: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8529: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8530: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8531: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8532: * cst, age and age*age
8533: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8534: /* including age products which are counted in cptcovage.
8535: * but the covariates which are products must be treated
8536: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8537: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8538: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8539:
8540:
1.187 brouard 8541: /* Design
8542: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8543: * < ncovcol=8 >
8544: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8545: * k= 1 2 3 4 5 6 7 8
8546: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8547: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8548: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8549: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8550: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8551: * Tage[++cptcovage]=k
8552: * if products, new covar are created after ncovcol with k1
8553: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8554: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8555: * 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
8556: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8557: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8558: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8559: * < ncovcol=8 >
8560: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8561: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8562: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8563: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8564: * p Tprod[1]@2={ 6, 5}
8565: *p Tvard[1][1]@4= {7, 8, 5, 6}
8566: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8567: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8568: *How to reorganize?
8569: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8570: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8571: * {2, 1, 4, 8, 5, 6, 3, 7}
8572: * Struct []
8573: */
1.225 brouard 8574:
1.187 brouard 8575: /* This loop fills the array Tvar from the string 'model'.*/
8576: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8577: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8578: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8579: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8580: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8581: /* k=1 Tvar[1]=2 (from V2) */
8582: /* k=5 Tvar[5] */
8583: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8584: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8585: /* } */
1.198 brouard 8586: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8587: /*
8588: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8589: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8590: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8591: }
1.187 brouard 8592: cptcovage=0;
8593: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8594: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8595: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8596: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8597: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8598: /*scanf("%d",i);*/
8599: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8600: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8601: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8602: /* covar is not filled and then is empty */
8603: cptcovprod--;
8604: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8605: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8606: Typevar[k]=1; /* 1 for age product */
8607: cptcovage++; /* Sums the number of covariates which include age as a product */
8608: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8609: /*printf("stre=%s ", stre);*/
8610: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8611: cptcovprod--;
8612: cutl(stre,strb,strc,'V');
8613: Tvar[k]=atoi(stre);
8614: Typevar[k]=1; /* 1 for age product */
8615: cptcovage++;
8616: Tage[cptcovage]=k;
8617: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8618: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8619: cptcovn++;
8620: cptcovprodnoage++;k1++;
8621: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8622: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8623: because this model-covariate is a construction we invent a new column
8624: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8625: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8626: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8627: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8628: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8629: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8630: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8631: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8632: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8633: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8634: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8635: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8636: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8637: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8638: for (i=1; i<=lastobs;i++){
8639: /* Computes the new covariate which is a product of
8640: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8641: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8642: }
8643: } /* End age is not in the model */
8644: } /* End if model includes a product */
8645: else { /* no more sum */
8646: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8647: /* scanf("%d",i);*/
8648: cutl(strd,strc,strb,'V');
8649: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8650: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8651: Tvar[k]=atoi(strd);
8652: Typevar[k]=0; /* 0 for simple covariates */
8653: }
8654: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8655: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8656: scanf("%d",i);*/
1.187 brouard 8657: } /* end of loop + on total covariates */
8658: } /* end if strlen(modelsave == 0) age*age might exist */
8659: } /* end if strlen(model == 0) */
1.136 brouard 8660:
8661: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8662: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8663:
1.136 brouard 8664: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8665: printf("cptcovprod=%d ", cptcovprod);
8666: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8667: scanf("%d ",i);*/
8668:
8669:
1.230 brouard 8670: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8671: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8672: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8673: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8674: k = 1 2 3 4 5 6 7 8 9
8675: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8676: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8677: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8678: Dummy[k] 1 0 0 0 3 1 1 2 3
8679: Tmodelind[combination of covar]=k;
1.225 brouard 8680: */
8681: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8682: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8683: /* 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 8684: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8685: printf("Model=%s\n\
8686: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8687: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8688: 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);
8689: fprintf(ficlog,"Model=%s\n\
8690: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8691: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8692: 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 8693: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8694: 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 */
8695: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8696: Fixed[k]= 0;
8697: Dummy[k]= 0;
1.225 brouard 8698: ncoveff++;
1.232 brouard 8699: ncovf++;
1.234 brouard 8700: nsd++;
8701: modell[k].maintype= FTYPE;
8702: TvarsD[nsd]=Tvar[k];
8703: TvarsDind[nsd]=k;
8704: TvarF[ncovf]=Tvar[k];
8705: TvarFind[ncovf]=k;
8706: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8707: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8708: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8709: Fixed[k]= 0;
8710: Dummy[k]= 0;
8711: ncoveff++;
8712: ncovf++;
8713: modell[k].maintype= FTYPE;
8714: TvarF[ncovf]=Tvar[k];
8715: TvarFind[ncovf]=k;
1.230 brouard 8716: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8717: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8718: }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 8719: Fixed[k]= 0;
8720: Dummy[k]= 1;
1.230 brouard 8721: nqfveff++;
1.234 brouard 8722: modell[k].maintype= FTYPE;
8723: modell[k].subtype= FQ;
8724: nsq++;
8725: TvarsQ[nsq]=Tvar[k];
8726: TvarsQind[nsq]=k;
1.232 brouard 8727: ncovf++;
1.234 brouard 8728: TvarF[ncovf]=Tvar[k];
8729: TvarFind[ncovf]=k;
1.231 brouard 8730: 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 8731: 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 8732: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8733: Fixed[k]= 1;
8734: Dummy[k]= 0;
1.225 brouard 8735: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8736: modell[k].maintype= VTYPE;
8737: modell[k].subtype= VD;
8738: nsd++;
8739: TvarsD[nsd]=Tvar[k];
8740: TvarsDind[nsd]=k;
8741: ncovv++; /* Only simple time varying variables */
8742: TvarV[ncovv]=Tvar[k];
1.242 brouard 8743: 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 8744: 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 */
8745: 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 8746: 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);
8747: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8748: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8749: Fixed[k]= 1;
8750: Dummy[k]= 1;
8751: nqtveff++;
8752: modell[k].maintype= VTYPE;
8753: modell[k].subtype= VQ;
8754: ncovv++; /* Only simple time varying variables */
8755: nsq++;
8756: TvarsQ[nsq]=Tvar[k];
8757: TvarsQind[nsq]=k;
8758: TvarV[ncovv]=Tvar[k];
1.242 brouard 8759: 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 8760: 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 */
8761: 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 8762: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8763: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8764: 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 8765: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8766: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8767: ncova++;
8768: TvarA[ncova]=Tvar[k];
8769: TvarAind[ncova]=k;
1.231 brouard 8770: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8771: Fixed[k]= 2;
8772: Dummy[k]= 2;
8773: modell[k].maintype= ATYPE;
8774: modell[k].subtype= APFD;
8775: /* ncoveff++; */
1.227 brouard 8776: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8777: Fixed[k]= 2;
8778: Dummy[k]= 3;
8779: modell[k].maintype= ATYPE;
8780: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8781: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8782: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8783: Fixed[k]= 3;
8784: Dummy[k]= 2;
8785: modell[k].maintype= ATYPE;
8786: modell[k].subtype= APVD; /* Product age * varying dummy */
8787: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8788: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8789: Fixed[k]= 3;
8790: Dummy[k]= 3;
8791: modell[k].maintype= ATYPE;
8792: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8793: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8794: }
8795: }else if (Typevar[k] == 2) { /* product without age */
8796: k1=Tposprod[k];
8797: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8798: if(Tvard[k1][2] <=ncovcol){
8799: Fixed[k]= 1;
8800: Dummy[k]= 0;
8801: modell[k].maintype= FTYPE;
8802: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8803: ncovf++; /* Fixed variables without age */
8804: TvarF[ncovf]=Tvar[k];
8805: TvarFind[ncovf]=k;
8806: }else if(Tvard[k1][2] <=ncovcol+nqv){
8807: Fixed[k]= 0; /* or 2 ?*/
8808: Dummy[k]= 1;
8809: modell[k].maintype= FTYPE;
8810: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8811: ncovf++; /* Varying variables without age */
8812: TvarF[ncovf]=Tvar[k];
8813: TvarFind[ncovf]=k;
8814: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8815: Fixed[k]= 1;
8816: Dummy[k]= 0;
8817: modell[k].maintype= VTYPE;
8818: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8819: ncovv++; /* Varying variables without age */
8820: TvarV[ncovv]=Tvar[k];
8821: TvarVind[ncovv]=k;
8822: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8823: Fixed[k]= 1;
8824: Dummy[k]= 1;
8825: modell[k].maintype= VTYPE;
8826: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8827: ncovv++; /* Varying variables without age */
8828: TvarV[ncovv]=Tvar[k];
8829: TvarVind[ncovv]=k;
8830: }
1.227 brouard 8831: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8832: if(Tvard[k1][2] <=ncovcol){
8833: Fixed[k]= 0; /* or 2 ?*/
8834: Dummy[k]= 1;
8835: modell[k].maintype= FTYPE;
8836: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8837: ncovf++; /* Fixed variables without age */
8838: TvarF[ncovf]=Tvar[k];
8839: TvarFind[ncovf]=k;
8840: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8841: Fixed[k]= 1;
8842: Dummy[k]= 1;
8843: modell[k].maintype= VTYPE;
8844: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8845: ncovv++; /* Varying variables without age */
8846: TvarV[ncovv]=Tvar[k];
8847: TvarVind[ncovv]=k;
8848: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8849: Fixed[k]= 1;
8850: Dummy[k]= 1;
8851: modell[k].maintype= VTYPE;
8852: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8853: ncovv++; /* Varying variables without age */
8854: TvarV[ncovv]=Tvar[k];
8855: TvarVind[ncovv]=k;
8856: ncovv++; /* Varying variables without age */
8857: TvarV[ncovv]=Tvar[k];
8858: TvarVind[ncovv]=k;
8859: }
1.227 brouard 8860: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8861: if(Tvard[k1][2] <=ncovcol){
8862: Fixed[k]= 1;
8863: Dummy[k]= 1;
8864: modell[k].maintype= VTYPE;
8865: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8866: ncovv++; /* Varying variables without age */
8867: TvarV[ncovv]=Tvar[k];
8868: TvarVind[ncovv]=k;
8869: }else if(Tvard[k1][2] <=ncovcol+nqv){
8870: Fixed[k]= 1;
8871: Dummy[k]= 1;
8872: modell[k].maintype= VTYPE;
8873: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8874: ncovv++; /* Varying variables without age */
8875: TvarV[ncovv]=Tvar[k];
8876: TvarVind[ncovv]=k;
8877: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8878: Fixed[k]= 1;
8879: Dummy[k]= 0;
8880: modell[k].maintype= VTYPE;
8881: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8882: ncovv++; /* Varying variables without age */
8883: TvarV[ncovv]=Tvar[k];
8884: TvarVind[ncovv]=k;
8885: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8886: Fixed[k]= 1;
8887: Dummy[k]= 1;
8888: modell[k].maintype= VTYPE;
8889: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8890: ncovv++; /* Varying variables without age */
8891: TvarV[ncovv]=Tvar[k];
8892: TvarVind[ncovv]=k;
8893: }
1.227 brouard 8894: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8895: if(Tvard[k1][2] <=ncovcol){
8896: Fixed[k]= 1;
8897: Dummy[k]= 1;
8898: modell[k].maintype= VTYPE;
8899: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8900: ncovv++; /* Varying variables without age */
8901: TvarV[ncovv]=Tvar[k];
8902: TvarVind[ncovv]=k;
8903: }else if(Tvard[k1][2] <=ncovcol+nqv){
8904: Fixed[k]= 1;
8905: Dummy[k]= 1;
8906: modell[k].maintype= VTYPE;
8907: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
8908: ncovv++; /* Varying variables without age */
8909: TvarV[ncovv]=Tvar[k];
8910: TvarVind[ncovv]=k;
8911: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8912: Fixed[k]= 1;
8913: Dummy[k]= 1;
8914: modell[k].maintype= VTYPE;
8915: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
8916: ncovv++; /* Varying variables without age */
8917: TvarV[ncovv]=Tvar[k];
8918: TvarVind[ncovv]=k;
8919: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8920: Fixed[k]= 1;
8921: Dummy[k]= 1;
8922: modell[k].maintype= VTYPE;
8923: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
8924: ncovv++; /* Varying variables without age */
8925: TvarV[ncovv]=Tvar[k];
8926: TvarVind[ncovv]=k;
8927: }
1.227 brouard 8928: }else{
1.240 brouard 8929: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8930: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8931: } /*end k1*/
1.225 brouard 8932: }else{
1.226 brouard 8933: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
8934: 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 8935: }
1.227 brouard 8936: 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 8937: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 8938: 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]);
8939: }
8940: /* Searching for doublons in the model */
8941: for(k1=1; k1<= cptcovt;k1++){
8942: for(k2=1; k2 <k1;k2++){
8943: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 8944: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
8945: if(Tvar[k1]==Tvar[k2]){
8946: 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]]);
8947: 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);
8948: return(1);
8949: }
8950: }else if (Typevar[k1] ==2){
8951: k3=Tposprod[k1];
8952: k4=Tposprod[k2];
8953: 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])) ){
8954: 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]]);
8955: 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);
8956: return(1);
8957: }
8958: }
1.227 brouard 8959: }
8960: }
1.225 brouard 8961: }
8962: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
8963: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 8964: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
8965: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 8966: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 8967: /*endread:*/
1.225 brouard 8968: printf("Exiting decodemodel: ");
8969: return (1);
1.136 brouard 8970: }
8971:
1.169 brouard 8972: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 8973: {/* Check ages at death */
1.136 brouard 8974: int i, m;
1.218 brouard 8975: int firstone=0;
8976:
1.136 brouard 8977: for (i=1; i<=imx; i++) {
8978: for(m=2; (m<= maxwav); m++) {
8979: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
8980: anint[m][i]=9999;
1.216 brouard 8981: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
8982: s[m][i]=-1;
1.136 brouard 8983: }
8984: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 8985: *nberr = *nberr + 1;
1.218 brouard 8986: if(firstone == 0){
8987: firstone=1;
8988: 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);
8989: }
8990: 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 8991: s[m][i]=-1;
8992: }
8993: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 8994: (*nberr)++;
1.136 brouard 8995: 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]);
8996: 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]);
8997: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
8998: }
8999: }
9000: }
9001:
9002: for (i=1; i<=imx; i++) {
9003: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9004: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9005: 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 9006: if (s[m][i] >= nlstate+1) {
1.169 brouard 9007: if(agedc[i]>0){
9008: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9009: agev[m][i]=agedc[i];
1.214 brouard 9010: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9011: }else {
1.136 brouard 9012: if ((int)andc[i]!=9999){
9013: nbwarn++;
9014: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9015: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9016: agev[m][i]=-1;
9017: }
9018: }
1.169 brouard 9019: } /* agedc > 0 */
1.214 brouard 9020: } /* end if */
1.136 brouard 9021: else if(s[m][i] !=9){ /* Standard case, age in fractional
9022: years but with the precision of a month */
9023: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9024: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9025: agev[m][i]=1;
9026: else if(agev[m][i] < *agemin){
9027: *agemin=agev[m][i];
9028: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9029: }
9030: else if(agev[m][i] >*agemax){
9031: *agemax=agev[m][i];
1.156 brouard 9032: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9033: }
9034: /*agev[m][i]=anint[m][i]-annais[i];*/
9035: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9036: } /* en if 9*/
1.136 brouard 9037: else { /* =9 */
1.214 brouard 9038: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9039: agev[m][i]=1;
9040: s[m][i]=-1;
9041: }
9042: }
1.214 brouard 9043: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9044: agev[m][i]=1;
1.214 brouard 9045: else{
9046: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9047: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9048: agev[m][i]=0;
9049: }
9050: } /* End for lastpass */
9051: }
1.136 brouard 9052:
9053: for (i=1; i<=imx; i++) {
9054: for(m=firstpass; (m<=lastpass); m++){
9055: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9056: (*nberr)++;
1.136 brouard 9057: 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);
9058: 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);
9059: return 1;
9060: }
9061: }
9062: }
9063:
9064: /*for (i=1; i<=imx; i++){
9065: for (m=firstpass; (m<lastpass); m++){
9066: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9067: }
9068:
9069: }*/
9070:
9071:
1.139 brouard 9072: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9073: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9074:
9075: return (0);
1.164 brouard 9076: /* endread:*/
1.136 brouard 9077: printf("Exiting calandcheckages: ");
9078: return (1);
9079: }
9080:
1.172 brouard 9081: #if defined(_MSC_VER)
9082: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9083: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9084: //#include "stdafx.h"
9085: //#include <stdio.h>
9086: //#include <tchar.h>
9087: //#include <windows.h>
9088: //#include <iostream>
9089: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9090:
9091: LPFN_ISWOW64PROCESS fnIsWow64Process;
9092:
9093: BOOL IsWow64()
9094: {
9095: BOOL bIsWow64 = FALSE;
9096:
9097: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9098: // (HANDLE, PBOOL);
9099:
9100: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9101:
9102: HMODULE module = GetModuleHandle(_T("kernel32"));
9103: const char funcName[] = "IsWow64Process";
9104: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9105: GetProcAddress(module, funcName);
9106:
9107: if (NULL != fnIsWow64Process)
9108: {
9109: if (!fnIsWow64Process(GetCurrentProcess(),
9110: &bIsWow64))
9111: //throw std::exception("Unknown error");
9112: printf("Unknown error\n");
9113: }
9114: return bIsWow64 != FALSE;
9115: }
9116: #endif
1.177 brouard 9117:
1.191 brouard 9118: void syscompilerinfo(int logged)
1.167 brouard 9119: {
9120: /* #include "syscompilerinfo.h"*/
1.185 brouard 9121: /* command line Intel compiler 32bit windows, XP compatible:*/
9122: /* /GS /W3 /Gy
9123: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9124: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9125: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9126: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9127: */
9128: /* 64 bits */
1.185 brouard 9129: /*
9130: /GS /W3 /Gy
9131: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9132: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9133: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9134: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9135: /* Optimization are useless and O3 is slower than O2 */
9136: /*
9137: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9138: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9139: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9140: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9141: */
1.186 brouard 9142: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9143: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9144: /PDB:"visual studio
9145: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9146: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9147: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9148: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9149: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9150: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9151: uiAccess='false'"
9152: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9153: /NOLOGO /TLBID:1
9154: */
1.177 brouard 9155: #if defined __INTEL_COMPILER
1.178 brouard 9156: #if defined(__GNUC__)
9157: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9158: #endif
1.177 brouard 9159: #elif defined(__GNUC__)
1.179 brouard 9160: #ifndef __APPLE__
1.174 brouard 9161: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9162: #endif
1.177 brouard 9163: struct utsname sysInfo;
1.178 brouard 9164: int cross = CROSS;
9165: if (cross){
9166: printf("Cross-");
1.191 brouard 9167: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9168: }
1.174 brouard 9169: #endif
9170:
1.171 brouard 9171: #include <stdint.h>
1.178 brouard 9172:
1.191 brouard 9173: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9174: #if defined(__clang__)
1.191 brouard 9175: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9176: #endif
9177: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9178: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9179: #endif
9180: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9181: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9182: #endif
9183: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9184: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9185: #endif
9186: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9187: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9188: #endif
9189: #if defined(_MSC_VER)
1.191 brouard 9190: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9191: #endif
9192: #if defined(__PGI)
1.191 brouard 9193: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9194: #endif
9195: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9196: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9197: #endif
1.191 brouard 9198: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9199:
1.167 brouard 9200: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9201: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9202: // Windows (x64 and x86)
1.191 brouard 9203: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9204: #elif __unix__ // all unices, not all compilers
9205: // Unix
1.191 brouard 9206: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9207: #elif __linux__
9208: // linux
1.191 brouard 9209: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9210: #elif __APPLE__
1.174 brouard 9211: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9212: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9213: #endif
9214:
9215: /* __MINGW32__ */
9216: /* __CYGWIN__ */
9217: /* __MINGW64__ */
9218: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9219: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9220: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9221: /* _WIN64 // Defined for applications for Win64. */
9222: /* _M_X64 // Defined for compilations that target x64 processors. */
9223: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9224:
1.167 brouard 9225: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9226: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9227: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9228: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9229: #else
1.191 brouard 9230: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9231: #endif
9232:
1.169 brouard 9233: #if defined(__GNUC__)
9234: # if defined(__GNUC_PATCHLEVEL__)
9235: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9236: + __GNUC_MINOR__ * 100 \
9237: + __GNUC_PATCHLEVEL__)
9238: # else
9239: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9240: + __GNUC_MINOR__ * 100)
9241: # endif
1.174 brouard 9242: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9243: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9244:
9245: if (uname(&sysInfo) != -1) {
9246: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9247: 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 9248: }
9249: else
9250: perror("uname() error");
1.179 brouard 9251: //#ifndef __INTEL_COMPILER
9252: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9253: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9254: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9255: #endif
1.169 brouard 9256: #endif
1.172 brouard 9257:
9258: // void main()
9259: // {
1.169 brouard 9260: #if defined(_MSC_VER)
1.174 brouard 9261: if (IsWow64()){
1.191 brouard 9262: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9263: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9264: }
9265: else{
1.191 brouard 9266: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9267: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9268: }
1.172 brouard 9269: // printf("\nPress Enter to continue...");
9270: // getchar();
9271: // }
9272:
1.169 brouard 9273: #endif
9274:
1.167 brouard 9275:
1.219 brouard 9276: }
1.136 brouard 9277:
1.219 brouard 9278: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9279: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9280: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9281: /* double ftolpl = 1.e-10; */
1.180 brouard 9282: double age, agebase, agelim;
1.203 brouard 9283: double tot;
1.180 brouard 9284:
1.202 brouard 9285: strcpy(filerespl,"PL_");
9286: strcat(filerespl,fileresu);
9287: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9288: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9289: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9290: }
1.227 brouard 9291: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9292: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9293: pstamp(ficrespl);
1.203 brouard 9294: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9295: fprintf(ficrespl,"#Age ");
9296: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9297: fprintf(ficrespl,"\n");
1.180 brouard 9298:
1.219 brouard 9299: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9300:
1.219 brouard 9301: agebase=ageminpar;
9302: agelim=agemaxpar;
1.180 brouard 9303:
1.227 brouard 9304: /* i1=pow(2,ncoveff); */
1.234 brouard 9305: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9306: if (cptcovn < 1){i1=1;}
1.180 brouard 9307:
1.238 brouard 9308: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9309: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9310: if(TKresult[nres]!= k)
9311: continue;
1.235 brouard 9312:
1.238 brouard 9313: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9314: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9315: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9316: /* k=k+1; */
9317: /* to clean */
9318: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9319: fprintf(ficrespl,"#******");
9320: printf("#******");
9321: fprintf(ficlog,"#******");
9322: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9323: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9324: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9325: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9326: }
9327: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9328: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9329: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9330: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9331: }
9332: fprintf(ficrespl,"******\n");
9333: printf("******\n");
9334: fprintf(ficlog,"******\n");
9335: if(invalidvarcomb[k]){
9336: printf("\nCombination (%d) ignored because no case \n",k);
9337: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9338: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9339: continue;
9340: }
1.219 brouard 9341:
1.238 brouard 9342: fprintf(ficrespl,"#Age ");
9343: for(j=1;j<=cptcoveff;j++) {
9344: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9345: }
9346: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9347: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9348:
1.238 brouard 9349: for (age=agebase; age<=agelim; age++){
9350: /* for (age=agebase; age<=agebase; age++){ */
9351: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9352: fprintf(ficrespl,"%.0f ",age );
9353: for(j=1;j<=cptcoveff;j++)
9354: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9355: tot=0.;
9356: for(i=1; i<=nlstate;i++){
9357: tot += prlim[i][i];
9358: fprintf(ficrespl," %.5f", prlim[i][i]);
9359: }
9360: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9361: } /* Age */
9362: /* was end of cptcod */
9363: } /* cptcov */
9364: } /* nres */
1.219 brouard 9365: return 0;
1.180 brouard 9366: }
9367:
1.218 brouard 9368: 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){
9369: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9370:
9371: /* Computes the back prevalence limit for any combination of covariate values
9372: * at any age between ageminpar and agemaxpar
9373: */
1.235 brouard 9374: int i, j, k, i1, nres=0 ;
1.217 brouard 9375: /* double ftolpl = 1.e-10; */
9376: double age, agebase, agelim;
9377: double tot;
1.218 brouard 9378: /* double ***mobaverage; */
9379: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9380:
9381: strcpy(fileresplb,"PLB_");
9382: strcat(fileresplb,fileresu);
9383: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9384: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9385: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9386: }
9387: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9388: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9389: pstamp(ficresplb);
9390: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9391: fprintf(ficresplb,"#Age ");
9392: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9393: fprintf(ficresplb,"\n");
9394:
1.218 brouard 9395:
9396: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9397:
9398: agebase=ageminpar;
9399: agelim=agemaxpar;
9400:
9401:
1.227 brouard 9402: i1=pow(2,cptcoveff);
1.218 brouard 9403: if (cptcovn < 1){i1=1;}
1.227 brouard 9404:
1.238 brouard 9405: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9406: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9407: if(TKresult[nres]!= k)
9408: continue;
9409: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9410: fprintf(ficresplb,"#******");
9411: printf("#******");
9412: fprintf(ficlog,"#******");
9413: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9414: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9415: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9416: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9417: }
9418: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9419: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9420: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9421: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9422: }
9423: fprintf(ficresplb,"******\n");
9424: printf("******\n");
9425: fprintf(ficlog,"******\n");
9426: if(invalidvarcomb[k]){
9427: printf("\nCombination (%d) ignored because no cases \n",k);
9428: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9429: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9430: continue;
9431: }
1.218 brouard 9432:
1.238 brouard 9433: fprintf(ficresplb,"#Age ");
9434: for(j=1;j<=cptcoveff;j++) {
9435: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9436: }
9437: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9438: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9439:
9440:
1.238 brouard 9441: for (age=agebase; age<=agelim; age++){
9442: /* for (age=agebase; age<=agebase; age++){ */
9443: if(mobilavproj > 0){
9444: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9445: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9446: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9447: }else if (mobilavproj == 0){
9448: 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);
9449: 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);
9450: exit(1);
9451: }else{
9452: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9453: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9454: }
9455: fprintf(ficresplb,"%.0f ",age );
9456: for(j=1;j<=cptcoveff;j++)
9457: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9458: tot=0.;
9459: for(i=1; i<=nlstate;i++){
9460: tot += bprlim[i][i];
9461: fprintf(ficresplb," %.5f", bprlim[i][i]);
9462: }
9463: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9464: } /* Age */
9465: /* was end of cptcod */
9466: } /* end of any combination */
9467: } /* end of nres */
1.218 brouard 9468: /* hBijx(p, bage, fage); */
9469: /* fclose(ficrespijb); */
9470:
9471: return 0;
1.217 brouard 9472: }
1.218 brouard 9473:
1.180 brouard 9474: int hPijx(double *p, int bage, int fage){
9475: /*------------- h Pij x at various ages ------------*/
9476:
9477: int stepsize;
9478: int agelim;
9479: int hstepm;
9480: int nhstepm;
1.235 brouard 9481: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9482:
9483: double agedeb;
9484: double ***p3mat;
9485:
1.201 brouard 9486: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9487: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9488: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9489: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9490: }
9491: printf("Computing pij: result on file '%s' \n", filerespij);
9492: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9493:
9494: stepsize=(int) (stepm+YEARM-1)/YEARM;
9495: /*if (stepm<=24) stepsize=2;*/
9496:
9497: agelim=AGESUP;
9498: hstepm=stepsize*YEARM; /* Every year of age */
9499: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9500:
1.180 brouard 9501: /* hstepm=1; aff par mois*/
9502: pstamp(ficrespij);
9503: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9504: i1= pow(2,cptcoveff);
1.218 brouard 9505: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9506: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9507: /* k=k+1; */
1.235 brouard 9508: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9509: for(k=1; k<=i1;k++){
9510: if(TKresult[nres]!= k)
9511: continue;
1.183 brouard 9512: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9513: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9514: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9515: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9516: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9517: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9518: }
1.183 brouard 9519: fprintf(ficrespij,"******\n");
9520:
9521: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9522: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9523: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9524:
9525: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9526:
1.183 brouard 9527: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9528: oldm=oldms;savm=savms;
1.235 brouard 9529: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9530: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9531: for(i=1; i<=nlstate;i++)
9532: for(j=1; j<=nlstate+ndeath;j++)
9533: fprintf(ficrespij," %1d-%1d",i,j);
9534: fprintf(ficrespij,"\n");
9535: for (h=0; h<=nhstepm; h++){
9536: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9537: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9538: for(i=1; i<=nlstate;i++)
9539: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9540: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9541: fprintf(ficrespij,"\n");
9542: }
1.183 brouard 9543: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9544: fprintf(ficrespij,"\n");
9545: }
1.180 brouard 9546: /*}*/
9547: }
1.218 brouard 9548: return 0;
1.180 brouard 9549: }
1.218 brouard 9550:
9551: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9552: /*------------- h Bij x at various ages ------------*/
9553:
9554: int stepsize;
1.218 brouard 9555: /* int agelim; */
9556: int ageminl;
1.217 brouard 9557: int hstepm;
9558: int nhstepm;
1.238 brouard 9559: int h, i, i1, j, k, nres;
1.218 brouard 9560:
1.217 brouard 9561: double agedeb;
9562: double ***p3mat;
1.218 brouard 9563:
9564: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9565: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9566: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9567: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9568: }
9569: printf("Computing pij back: result on file '%s' \n", filerespijb);
9570: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9571:
9572: stepsize=(int) (stepm+YEARM-1)/YEARM;
9573: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9574:
1.218 brouard 9575: /* agelim=AGESUP; */
9576: ageminl=30;
9577: hstepm=stepsize*YEARM; /* Every year of age */
9578: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9579:
9580: /* hstepm=1; aff par mois*/
9581: pstamp(ficrespijb);
9582: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9583: i1= pow(2,cptcoveff);
1.218 brouard 9584: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9585: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9586: /* k=k+1; */
1.238 brouard 9587: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9588: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9589: if(TKresult[nres]!= k)
9590: continue;
9591: fprintf(ficrespijb,"\n#****** ");
9592: for(j=1;j<=cptcoveff;j++)
9593: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9594: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9595: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9596: }
9597: fprintf(ficrespijb,"******\n");
9598: if(invalidvarcomb[k]){
9599: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9600: continue;
9601: }
9602:
9603: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9604: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9605: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9606: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9607: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9608:
9609: /* nhstepm=nhstepm*YEARM; aff par mois*/
9610:
9611: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9612: /* oldm=oldms;savm=savms; */
9613: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9614: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9615: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9616: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9617: for(i=1; i<=nlstate;i++)
9618: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9619: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9620: fprintf(ficrespijb,"\n");
1.238 brouard 9621: for (h=0; h<=nhstepm; h++){
9622: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9623: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9624: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9625: for(i=1; i<=nlstate;i++)
9626: for(j=1; j<=nlstate+ndeath;j++)
9627: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9628: fprintf(ficrespijb,"\n");
9629: }
9630: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9631: fprintf(ficrespijb,"\n");
9632: } /* end age deb */
9633: } /* end combination */
9634: } /* end nres */
1.218 brouard 9635: return 0;
9636: } /* hBijx */
1.217 brouard 9637:
1.180 brouard 9638:
1.136 brouard 9639: /***********************************************/
9640: /**************** Main Program *****************/
9641: /***********************************************/
9642:
9643: int main(int argc, char *argv[])
9644: {
9645: #ifdef GSL
9646: const gsl_multimin_fminimizer_type *T;
9647: size_t iteri = 0, it;
9648: int rval = GSL_CONTINUE;
9649: int status = GSL_SUCCESS;
9650: double ssval;
9651: #endif
9652: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9653: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9654: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9655: int jj, ll, li, lj, lk;
1.136 brouard 9656: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9657: int num_filled;
1.136 brouard 9658: int itimes;
9659: int NDIM=2;
9660: int vpopbased=0;
1.235 brouard 9661: int nres=0;
1.136 brouard 9662:
1.164 brouard 9663: char ca[32], cb[32];
1.136 brouard 9664: /* FILE *fichtm; *//* Html File */
9665: /* FILE *ficgp;*/ /*Gnuplot File */
9666: struct stat info;
1.191 brouard 9667: double agedeb=0.;
1.194 brouard 9668:
9669: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9670: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9671:
1.165 brouard 9672: double fret;
1.191 brouard 9673: double dum=0.; /* Dummy variable */
1.136 brouard 9674: double ***p3mat;
1.218 brouard 9675: /* double ***mobaverage; */
1.164 brouard 9676:
9677: char line[MAXLINE];
1.197 brouard 9678: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9679:
1.234 brouard 9680: char modeltemp[MAXLINE];
1.230 brouard 9681: char resultline[MAXLINE];
9682:
1.136 brouard 9683: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9684: char *tok, *val; /* pathtot */
1.136 brouard 9685: int firstobs=1, lastobs=10;
1.195 brouard 9686: int c, h , cpt, c2;
1.191 brouard 9687: int jl=0;
9688: int i1, j1, jk, stepsize=0;
1.194 brouard 9689: int count=0;
9690:
1.164 brouard 9691: int *tab;
1.136 brouard 9692: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9693: int backcast=0;
1.136 brouard 9694: int mobilav=0,popforecast=0;
1.191 brouard 9695: int hstepm=0, nhstepm=0;
1.136 brouard 9696: int agemortsup;
9697: float sumlpop=0.;
9698: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9699: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9700:
1.191 brouard 9701: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9702: double ftolpl=FTOL;
9703: double **prlim;
1.217 brouard 9704: double **bprlim;
1.136 brouard 9705: double ***param; /* Matrix of parameters */
1.251 brouard 9706: double ***paramstart; /* Matrix of starting parameter values */
9707: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9708: double **matcov; /* Matrix of covariance */
1.203 brouard 9709: double **hess; /* Hessian matrix */
1.136 brouard 9710: double ***delti3; /* Scale */
9711: double *delti; /* Scale */
9712: double ***eij, ***vareij;
9713: double **varpl; /* Variances of prevalence limits by age */
9714: double *epj, vepp;
1.164 brouard 9715:
1.136 brouard 9716: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9717: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9718:
1.136 brouard 9719: double **ximort;
1.145 brouard 9720: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9721: int *dcwave;
9722:
1.164 brouard 9723: char z[1]="c";
1.136 brouard 9724:
9725: /*char *strt;*/
9726: char strtend[80];
1.126 brouard 9727:
1.164 brouard 9728:
1.126 brouard 9729: /* setlocale (LC_ALL, ""); */
9730: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9731: /* textdomain (PACKAGE); */
9732: /* setlocale (LC_CTYPE, ""); */
9733: /* setlocale (LC_MESSAGES, ""); */
9734:
9735: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9736: rstart_time = time(NULL);
9737: /* (void) gettimeofday(&start_time,&tzp);*/
9738: start_time = *localtime(&rstart_time);
1.126 brouard 9739: curr_time=start_time;
1.157 brouard 9740: /*tml = *localtime(&start_time.tm_sec);*/
9741: /* strcpy(strstart,asctime(&tml)); */
9742: strcpy(strstart,asctime(&start_time));
1.126 brouard 9743:
9744: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9745: /* tp.tm_sec = tp.tm_sec +86400; */
9746: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9747: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9748: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9749: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9750: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9751: /* strt=asctime(&tmg); */
9752: /* printf("Time(after) =%s",strstart); */
9753: /* (void) time (&time_value);
9754: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9755: * tm = *localtime(&time_value);
9756: * strstart=asctime(&tm);
9757: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9758: */
9759:
9760: nberr=0; /* Number of errors and warnings */
9761: nbwarn=0;
1.184 brouard 9762: #ifdef WIN32
9763: _getcwd(pathcd, size);
9764: #else
1.126 brouard 9765: getcwd(pathcd, size);
1.184 brouard 9766: #endif
1.191 brouard 9767: syscompilerinfo(0);
1.196 brouard 9768: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9769: if(argc <=1){
9770: printf("\nEnter the parameter file name: ");
1.205 brouard 9771: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9772: printf("ERROR Empty parameter file name\n");
9773: goto end;
9774: }
1.126 brouard 9775: i=strlen(pathr);
9776: if(pathr[i-1]=='\n')
9777: pathr[i-1]='\0';
1.156 brouard 9778: i=strlen(pathr);
1.205 brouard 9779: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9780: pathr[i-1]='\0';
1.205 brouard 9781: }
9782: i=strlen(pathr);
9783: if( i==0 ){
9784: printf("ERROR Empty parameter file name\n");
9785: goto end;
9786: }
9787: for (tok = pathr; tok != NULL; ){
1.126 brouard 9788: printf("Pathr |%s|\n",pathr);
9789: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9790: printf("val= |%s| pathr=%s\n",val,pathr);
9791: strcpy (pathtot, val);
9792: if(pathr[0] == '\0') break; /* Dirty */
9793: }
9794: }
9795: else{
9796: strcpy(pathtot,argv[1]);
9797: }
9798: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9799: /*cygwin_split_path(pathtot,path,optionfile);
9800: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9801: /* cutv(path,optionfile,pathtot,'\\');*/
9802:
9803: /* Split argv[0], imach program to get pathimach */
9804: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9805: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9806: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9807: /* strcpy(pathimach,argv[0]); */
9808: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9809: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9810: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9811: #ifdef WIN32
9812: _chdir(path); /* Can be a relative path */
9813: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9814: #else
1.126 brouard 9815: chdir(path); /* Can be a relative path */
1.184 brouard 9816: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9817: #endif
9818: printf("Current directory %s!\n",pathcd);
1.126 brouard 9819: strcpy(command,"mkdir ");
9820: strcat(command,optionfilefiname);
9821: if((outcmd=system(command)) != 0){
1.169 brouard 9822: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9823: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9824: /* fclose(ficlog); */
9825: /* exit(1); */
9826: }
9827: /* if((imk=mkdir(optionfilefiname))<0){ */
9828: /* perror("mkdir"); */
9829: /* } */
9830:
9831: /*-------- arguments in the command line --------*/
9832:
1.186 brouard 9833: /* Main Log file */
1.126 brouard 9834: strcat(filelog, optionfilefiname);
9835: strcat(filelog,".log"); /* */
9836: if((ficlog=fopen(filelog,"w"))==NULL) {
9837: printf("Problem with logfile %s\n",filelog);
9838: goto end;
9839: }
9840: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9841: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9842: fprintf(ficlog,"\nEnter the parameter file name: \n");
9843: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9844: path=%s \n\
9845: optionfile=%s\n\
9846: optionfilext=%s\n\
1.156 brouard 9847: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9848:
1.197 brouard 9849: syscompilerinfo(1);
1.167 brouard 9850:
1.126 brouard 9851: printf("Local time (at start):%s",strstart);
9852: fprintf(ficlog,"Local time (at start): %s",strstart);
9853: fflush(ficlog);
9854: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9855: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9856:
9857: /* */
9858: strcpy(fileres,"r");
9859: strcat(fileres, optionfilefiname);
1.201 brouard 9860: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9861: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9862: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9863:
1.186 brouard 9864: /* Main ---------arguments file --------*/
1.126 brouard 9865:
9866: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9867: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9868: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9869: fflush(ficlog);
1.149 brouard 9870: /* goto end; */
9871: exit(70);
1.126 brouard 9872: }
9873:
9874:
9875:
9876: strcpy(filereso,"o");
1.201 brouard 9877: strcat(filereso,fileresu);
1.126 brouard 9878: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9879: printf("Problem with Output resultfile: %s\n", filereso);
9880: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9881: fflush(ficlog);
9882: goto end;
9883: }
9884:
9885: /* Reads comments: lines beginning with '#' */
9886: numlinepar=0;
1.197 brouard 9887:
9888: /* First parameter line */
9889: while(fgets(line, MAXLINE, ficpar)) {
9890: /* If line starts with a # it is a comment */
9891: if (line[0] == '#') {
9892: numlinepar++;
9893: fputs(line,stdout);
9894: fputs(line,ficparo);
9895: fputs(line,ficlog);
9896: continue;
9897: }else
9898: break;
9899: }
9900: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9901: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9902: if (num_filled != 5) {
9903: printf("Should be 5 parameters\n");
9904: }
1.126 brouard 9905: numlinepar++;
1.197 brouard 9906: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
9907: }
9908: /* Second parameter line */
9909: while(fgets(line, MAXLINE, ficpar)) {
9910: /* If line starts with a # it is a comment */
9911: if (line[0] == '#') {
9912: numlinepar++;
9913: fputs(line,stdout);
9914: fputs(line,ficparo);
9915: fputs(line,ficlog);
9916: continue;
9917: }else
9918: break;
9919: }
1.223 brouard 9920: 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", \
9921: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
9922: if (num_filled != 11) {
9923: 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 9924: printf("but line=%s\n",line);
1.197 brouard 9925: }
1.223 brouard 9926: 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 9927: }
1.203 brouard 9928: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 9929: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 9930: /* Third parameter line */
9931: while(fgets(line, MAXLINE, ficpar)) {
9932: /* If line starts with a # it is a comment */
9933: if (line[0] == '#') {
9934: numlinepar++;
9935: fputs(line,stdout);
9936: fputs(line,ficparo);
9937: fputs(line,ficlog);
9938: continue;
9939: }else
9940: break;
9941: }
1.201 brouard 9942: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
9943: if (num_filled == 0)
9944: model[0]='\0';
9945: else if (num_filled != 1){
1.197 brouard 9946: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9947: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9948: model[0]='\0';
9949: goto end;
9950: }
9951: else{
9952: if (model[0]=='+'){
9953: for(i=1; i<=strlen(model);i++)
9954: modeltemp[i-1]=model[i];
1.201 brouard 9955: strcpy(model,modeltemp);
1.197 brouard 9956: }
9957: }
1.199 brouard 9958: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 9959: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 9960: }
9961: /* 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); */
9962: /* numlinepar=numlinepar+3; /\* In general *\/ */
9963: /* 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 9964: 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);
9965: 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 9966: fflush(ficlog);
1.190 brouard 9967: /* if(model[0]=='#'|| model[0]== '\0'){ */
9968: if(model[0]=='#'){
1.187 brouard 9969: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
9970: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
9971: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
9972: if(mle != -1){
9973: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
9974: exit(1);
9975: }
9976: }
1.126 brouard 9977: while((c=getc(ficpar))=='#' && c!= EOF){
9978: ungetc(c,ficpar);
9979: fgets(line, MAXLINE, ficpar);
9980: numlinepar++;
1.195 brouard 9981: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
9982: z[0]=line[1];
9983: }
9984: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 9985: fputs(line, stdout);
9986: //puts(line);
1.126 brouard 9987: fputs(line,ficparo);
9988: fputs(line,ficlog);
9989: }
9990: ungetc(c,ficpar);
9991:
9992:
1.145 brouard 9993: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 9994: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 9995: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 9996: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 9997: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
9998: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
9999: v1+v2*age+v2*v3 makes cptcovn = 3
10000: */
10001: if (strlen(model)>1)
1.187 brouard 10002: 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 10003: else
1.187 brouard 10004: ncovmodel=2; /* Constant and age */
1.133 brouard 10005: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10006: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10007: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10008: 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);
10009: 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);
10010: fflush(stdout);
10011: fclose (ficlog);
10012: goto end;
10013: }
1.126 brouard 10014: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10015: delti=delti3[1][1];
10016: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10017: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10018: /* We could also provide initial parameters values giving by simple logistic regression
10019: * only one way, that is without matrix product. We will have nlstate maximizations */
10020: /* for(i=1;i<nlstate;i++){ */
10021: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10022: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10023: /* } */
1.126 brouard 10024: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10025: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10026: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10027: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10028: fclose (ficparo);
10029: fclose (ficlog);
10030: goto end;
10031: exit(0);
1.220 brouard 10032: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10033: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10034: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10035: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10036: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10037: matcov=matrix(1,npar,1,npar);
1.203 brouard 10038: hess=matrix(1,npar,1,npar);
1.220 brouard 10039: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10040: /* Read guessed parameters */
1.126 brouard 10041: /* Reads comments: lines beginning with '#' */
10042: while((c=getc(ficpar))=='#' && c!= EOF){
10043: ungetc(c,ficpar);
10044: fgets(line, MAXLINE, ficpar);
10045: numlinepar++;
1.141 brouard 10046: fputs(line,stdout);
1.126 brouard 10047: fputs(line,ficparo);
10048: fputs(line,ficlog);
10049: }
10050: ungetc(c,ficpar);
10051:
10052: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10053: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10054: for(i=1; i <=nlstate; i++){
1.234 brouard 10055: j=0;
1.126 brouard 10056: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10057: if(jj==i) continue;
10058: j++;
10059: fscanf(ficpar,"%1d%1d",&i1,&j1);
10060: if ((i1 != i) || (j1 != jj)){
10061: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10062: It might be a problem of design; if ncovcol and the model are correct\n \
10063: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10064: exit(1);
10065: }
10066: fprintf(ficparo,"%1d%1d",i1,j1);
10067: if(mle==1)
10068: printf("%1d%1d",i,jj);
10069: fprintf(ficlog,"%1d%1d",i,jj);
10070: for(k=1; k<=ncovmodel;k++){
10071: fscanf(ficpar," %lf",¶m[i][j][k]);
10072: if(mle==1){
10073: printf(" %lf",param[i][j][k]);
10074: fprintf(ficlog," %lf",param[i][j][k]);
10075: }
10076: else
10077: fprintf(ficlog," %lf",param[i][j][k]);
10078: fprintf(ficparo," %lf",param[i][j][k]);
10079: }
10080: fscanf(ficpar,"\n");
10081: numlinepar++;
10082: if(mle==1)
10083: printf("\n");
10084: fprintf(ficlog,"\n");
10085: fprintf(ficparo,"\n");
1.126 brouard 10086: }
10087: }
10088: fflush(ficlog);
1.234 brouard 10089:
1.251 brouard 10090: /* Reads parameters values */
1.126 brouard 10091: p=param[1][1];
1.251 brouard 10092: pstart=paramstart[1][1];
1.126 brouard 10093:
10094: /* Reads comments: lines beginning with '#' */
10095: while((c=getc(ficpar))=='#' && c!= EOF){
10096: ungetc(c,ficpar);
10097: fgets(line, MAXLINE, ficpar);
10098: numlinepar++;
1.141 brouard 10099: fputs(line,stdout);
1.126 brouard 10100: fputs(line,ficparo);
10101: fputs(line,ficlog);
10102: }
10103: ungetc(c,ficpar);
10104:
10105: for(i=1; i <=nlstate; i++){
10106: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10107: fscanf(ficpar,"%1d%1d",&i1,&j1);
10108: if ( (i1-i) * (j1-j) != 0){
10109: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10110: exit(1);
10111: }
10112: printf("%1d%1d",i,j);
10113: fprintf(ficparo,"%1d%1d",i1,j1);
10114: fprintf(ficlog,"%1d%1d",i1,j1);
10115: for(k=1; k<=ncovmodel;k++){
10116: fscanf(ficpar,"%le",&delti3[i][j][k]);
10117: printf(" %le",delti3[i][j][k]);
10118: fprintf(ficparo," %le",delti3[i][j][k]);
10119: fprintf(ficlog," %le",delti3[i][j][k]);
10120: }
10121: fscanf(ficpar,"\n");
10122: numlinepar++;
10123: printf("\n");
10124: fprintf(ficparo,"\n");
10125: fprintf(ficlog,"\n");
1.126 brouard 10126: }
10127: }
10128: fflush(ficlog);
1.234 brouard 10129:
1.145 brouard 10130: /* Reads covariance matrix */
1.126 brouard 10131: delti=delti3[1][1];
1.220 brouard 10132:
10133:
1.126 brouard 10134: /* 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 10135:
1.126 brouard 10136: /* Reads comments: lines beginning with '#' */
10137: while((c=getc(ficpar))=='#' && c!= EOF){
10138: ungetc(c,ficpar);
10139: fgets(line, MAXLINE, ficpar);
10140: numlinepar++;
1.141 brouard 10141: fputs(line,stdout);
1.126 brouard 10142: fputs(line,ficparo);
10143: fputs(line,ficlog);
10144: }
10145: ungetc(c,ficpar);
1.220 brouard 10146:
1.126 brouard 10147: matcov=matrix(1,npar,1,npar);
1.203 brouard 10148: hess=matrix(1,npar,1,npar);
1.131 brouard 10149: for(i=1; i <=npar; i++)
10150: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10151:
1.194 brouard 10152: /* Scans npar lines */
1.126 brouard 10153: for(i=1; i <=npar; i++){
1.226 brouard 10154: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10155: if(count != 3){
1.226 brouard 10156: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10157: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10158: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10159: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10160: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10161: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10162: exit(1);
1.220 brouard 10163: }else{
1.226 brouard 10164: if(mle==1)
10165: printf("%1d%1d%d",i1,j1,jk);
10166: }
10167: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10168: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10169: for(j=1; j <=i; j++){
1.226 brouard 10170: fscanf(ficpar," %le",&matcov[i][j]);
10171: if(mle==1){
10172: printf(" %.5le",matcov[i][j]);
10173: }
10174: fprintf(ficlog," %.5le",matcov[i][j]);
10175: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10176: }
10177: fscanf(ficpar,"\n");
10178: numlinepar++;
10179: if(mle==1)
1.220 brouard 10180: printf("\n");
1.126 brouard 10181: fprintf(ficlog,"\n");
10182: fprintf(ficparo,"\n");
10183: }
1.194 brouard 10184: /* End of read covariance matrix npar lines */
1.126 brouard 10185: for(i=1; i <=npar; i++)
10186: for(j=i+1;j<=npar;j++)
1.226 brouard 10187: matcov[i][j]=matcov[j][i];
1.126 brouard 10188:
10189: if(mle==1)
10190: printf("\n");
10191: fprintf(ficlog,"\n");
10192:
10193: fflush(ficlog);
10194:
10195: /*-------- Rewriting parameter file ----------*/
10196: strcpy(rfileres,"r"); /* "Rparameterfile */
10197: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10198: strcat(rfileres,"."); /* */
10199: strcat(rfileres,optionfilext); /* Other files have txt extension */
10200: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10201: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10202: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10203: }
10204: fprintf(ficres,"#%s\n",version);
10205: } /* End of mle != -3 */
1.218 brouard 10206:
1.186 brouard 10207: /* Main data
10208: */
1.126 brouard 10209: n= lastobs;
10210: num=lvector(1,n);
10211: moisnais=vector(1,n);
10212: annais=vector(1,n);
10213: moisdc=vector(1,n);
10214: andc=vector(1,n);
1.220 brouard 10215: weight=vector(1,n);
1.126 brouard 10216: agedc=vector(1,n);
10217: cod=ivector(1,n);
1.220 brouard 10218: for(i=1;i<=n;i++){
1.234 brouard 10219: num[i]=0;
10220: moisnais[i]=0;
10221: annais[i]=0;
10222: moisdc[i]=0;
10223: andc[i]=0;
10224: agedc[i]=0;
10225: cod[i]=0;
10226: weight[i]=1.0; /* Equal weights, 1 by default */
10227: }
1.126 brouard 10228: mint=matrix(1,maxwav,1,n);
10229: anint=matrix(1,maxwav,1,n);
1.131 brouard 10230: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10231: tab=ivector(1,NCOVMAX);
1.144 brouard 10232: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10233: 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 10234:
1.136 brouard 10235: /* Reads data from file datafile */
10236: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10237: goto end;
10238:
10239: /* Calculation of the number of parameters from char model */
1.234 brouard 10240: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10241: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10242: k=3 V4 Tvar[k=3]= 4 (from V4)
10243: k=2 V1 Tvar[k=2]= 1 (from V1)
10244: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10245: */
10246:
10247: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10248: TvarsDind=ivector(1,NCOVMAX); /* */
10249: TvarsD=ivector(1,NCOVMAX); /* */
10250: TvarsQind=ivector(1,NCOVMAX); /* */
10251: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10252: TvarF=ivector(1,NCOVMAX); /* */
10253: TvarFind=ivector(1,NCOVMAX); /* */
10254: TvarV=ivector(1,NCOVMAX); /* */
10255: TvarVind=ivector(1,NCOVMAX); /* */
10256: TvarA=ivector(1,NCOVMAX); /* */
10257: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10258: TvarFD=ivector(1,NCOVMAX); /* */
10259: TvarFDind=ivector(1,NCOVMAX); /* */
10260: TvarFQ=ivector(1,NCOVMAX); /* */
10261: TvarFQind=ivector(1,NCOVMAX); /* */
10262: TvarVD=ivector(1,NCOVMAX); /* */
10263: TvarVDind=ivector(1,NCOVMAX); /* */
10264: TvarVQ=ivector(1,NCOVMAX); /* */
10265: TvarVQind=ivector(1,NCOVMAX); /* */
10266:
1.230 brouard 10267: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10268: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10269: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10270: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10271: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10272: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10273: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10274: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10275: */
10276: /* For model-covariate k tells which data-covariate to use but
10277: because this model-covariate is a construction we invent a new column
10278: ncovcol + k1
10279: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10280: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10281: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10282: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10283: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10284: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10285: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10286: */
1.145 brouard 10287: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10288: 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 10289: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10290: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10291: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10292: 4 covariates (3 plus signs)
10293: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10294: */
1.230 brouard 10295: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10296: * individual dummy, fixed or varying:
10297: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10298: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10299: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10300: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10301: * Tmodelind[1]@9={9,0,3,2,}*/
10302: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10303: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10304: * individual quantitative, fixed or varying:
10305: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10306: * 3, 1, 0, 0, 0, 0, 0, 0},
10307: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10308: /* Main decodemodel */
10309:
1.187 brouard 10310:
1.223 brouard 10311: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10312: goto end;
10313:
1.137 brouard 10314: if((double)(lastobs-imx)/(double)imx > 1.10){
10315: nbwarn++;
10316: 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);
10317: 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);
10318: }
1.136 brouard 10319: /* if(mle==1){*/
1.137 brouard 10320: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10321: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10322: }
10323:
10324: /*-calculation of age at interview from date of interview and age at death -*/
10325: agev=matrix(1,maxwav,1,imx);
10326:
10327: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10328: goto end;
10329:
1.126 brouard 10330:
1.136 brouard 10331: agegomp=(int)agemin;
10332: free_vector(moisnais,1,n);
10333: free_vector(annais,1,n);
1.126 brouard 10334: /* free_matrix(mint,1,maxwav,1,n);
10335: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10336: /* free_vector(moisdc,1,n); */
10337: /* free_vector(andc,1,n); */
1.145 brouard 10338: /* */
10339:
1.126 brouard 10340: wav=ivector(1,imx);
1.214 brouard 10341: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10342: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10343: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10344: 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.*/
10345: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10346: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10347:
10348: /* Concatenates waves */
1.214 brouard 10349: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10350: Death is a valid wave (if date is known).
10351: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10352: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10353: and mw[mi+1][i]. dh depends on stepm.
10354: */
10355:
1.126 brouard 10356: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10357: /* Concatenates waves */
1.145 brouard 10358:
1.215 brouard 10359: free_vector(moisdc,1,n);
10360: free_vector(andc,1,n);
10361:
1.126 brouard 10362: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10363: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10364: ncodemax[1]=1;
1.145 brouard 10365: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10366: cptcoveff=0;
1.220 brouard 10367: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10368: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10369: }
10370:
10371: ncovcombmax=pow(2,cptcoveff);
10372: invalidvarcomb=ivector(1, ncovcombmax);
10373: for(i=1;i<ncovcombmax;i++)
10374: invalidvarcomb[i]=0;
10375:
1.211 brouard 10376: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10377: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10378: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10379:
1.200 brouard 10380: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10381: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10382: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10383: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10384: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10385: * (currently 0 or 1) in the data.
10386: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10387: * corresponding modality (h,j).
10388: */
10389:
1.145 brouard 10390: h=0;
10391: /*if (cptcovn > 0) */
1.126 brouard 10392: m=pow(2,cptcoveff);
10393:
1.144 brouard 10394: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10395: * For k=4 covariates, h goes from 1 to m=2**k
10396: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10397: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10398: * h\k 1 2 3 4
1.143 brouard 10399: *______________________________
10400: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10401: * 2 2 1 1 1
10402: * 3 i=2 1 2 1 1
10403: * 4 2 2 1 1
10404: * 5 i=3 1 i=2 1 2 1
10405: * 6 2 1 2 1
10406: * 7 i=4 1 2 2 1
10407: * 8 2 2 2 1
1.197 brouard 10408: * 9 i=5 1 i=3 1 i=2 1 2
10409: * 10 2 1 1 2
10410: * 11 i=6 1 2 1 2
10411: * 12 2 2 1 2
10412: * 13 i=7 1 i=4 1 2 2
10413: * 14 2 1 2 2
10414: * 15 i=8 1 2 2 2
10415: * 16 2 2 2 2
1.143 brouard 10416: */
1.212 brouard 10417: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10418: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10419: * and the value of each covariate?
10420: * V1=1, V2=1, V3=2, V4=1 ?
10421: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10422: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10423: * In order to get the real value in the data, we use nbcode
10424: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10425: * We are keeping this crazy system in order to be able (in the future?)
10426: * to have more than 2 values (0 or 1) for a covariate.
10427: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10428: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10429: * bbbbbbbb
10430: * 76543210
10431: * h-1 00000101 (6-1=5)
1.219 brouard 10432: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10433: * &
10434: * 1 00000001 (1)
1.219 brouard 10435: * 00000000 = 1 & ((h-1) >> (k-1))
10436: * +1= 00000001 =1
1.211 brouard 10437: *
10438: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10439: * h' 1101 =2^3+2^2+0x2^1+2^0
10440: * >>k' 11
10441: * & 00000001
10442: * = 00000001
10443: * +1 = 00000010=2 = codtabm(14,3)
10444: * Reverse h=6 and m=16?
10445: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10446: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10447: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10448: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10449: * V3=decodtabm(14,3,2**4)=2
10450: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10451: *(h-1) >> (j-1) 0011 =13 >> 2
10452: * &1 000000001
10453: * = 000000001
10454: * +1= 000000010 =2
10455: * 2211
10456: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10457: * V3=2
1.220 brouard 10458: * codtabm and decodtabm are identical
1.211 brouard 10459: */
10460:
1.145 brouard 10461:
10462: free_ivector(Ndum,-1,NCOVMAX);
10463:
10464:
1.126 brouard 10465:
1.186 brouard 10466: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10467: strcpy(optionfilegnuplot,optionfilefiname);
10468: if(mle==-3)
1.201 brouard 10469: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10470: strcat(optionfilegnuplot,".gp");
10471:
10472: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10473: printf("Problem with file %s",optionfilegnuplot);
10474: }
10475: else{
1.204 brouard 10476: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10477: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10478: //fprintf(ficgp,"set missing 'NaNq'\n");
10479: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10480: }
10481: /* fclose(ficgp);*/
1.186 brouard 10482:
10483:
10484: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10485:
10486: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10487: if(mle==-3)
1.201 brouard 10488: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10489: strcat(optionfilehtm,".htm");
10490: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10491: printf("Problem with %s \n",optionfilehtm);
10492: exit(0);
1.126 brouard 10493: }
10494:
10495: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10496: strcat(optionfilehtmcov,"-cov.htm");
10497: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10498: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10499: }
10500: else{
10501: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10502: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10503: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10504: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10505: }
10506:
1.213 brouard 10507: 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 10508: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10509: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10510: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10511: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10512: \n\
10513: <hr size=\"2\" color=\"#EC5E5E\">\
10514: <ul><li><h4>Parameter files</h4>\n\
10515: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10516: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10517: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10518: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10519: - Date and time at start: %s</ul>\n",\
10520: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10521: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10522: fileres,fileres,\
10523: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10524: fflush(fichtm);
10525:
10526: strcpy(pathr,path);
10527: strcat(pathr,optionfilefiname);
1.184 brouard 10528: #ifdef WIN32
10529: _chdir(optionfilefiname); /* Move to directory named optionfile */
10530: #else
1.126 brouard 10531: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10532: #endif
10533:
1.126 brouard 10534:
1.220 brouard 10535: /* Calculates basic frequencies. Computes observed prevalence at single age
10536: and for any valid combination of covariates
1.126 brouard 10537: and prints on file fileres'p'. */
1.251 brouard 10538: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10539: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10540:
10541: fprintf(fichtm,"\n");
10542: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10543: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10544: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10545: imx,agemin,agemax,jmin,jmax,jmean);
10546: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10547: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10548: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10549: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10550: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10551:
1.126 brouard 10552: /* For Powell, parameters are in a vector p[] starting at p[1]
10553: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10554: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10555:
10556: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10557: /* For mortality only */
1.126 brouard 10558: if (mle==-3){
1.136 brouard 10559: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10560: for(i=1;i<=NDIM;i++)
10561: for(j=1;j<=NDIM;j++)
10562: ximort[i][j]=0.;
1.186 brouard 10563: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10564: cens=ivector(1,n);
10565: ageexmed=vector(1,n);
10566: agecens=vector(1,n);
10567: dcwave=ivector(1,n);
1.223 brouard 10568:
1.126 brouard 10569: for (i=1; i<=imx; i++){
10570: dcwave[i]=-1;
10571: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10572: if (s[m][i]>nlstate) {
10573: dcwave[i]=m;
10574: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10575: break;
10576: }
1.126 brouard 10577: }
1.226 brouard 10578:
1.126 brouard 10579: for (i=1; i<=imx; i++) {
10580: if (wav[i]>0){
1.226 brouard 10581: ageexmed[i]=agev[mw[1][i]][i];
10582: j=wav[i];
10583: agecens[i]=1.;
10584:
10585: if (ageexmed[i]> 1 && wav[i] > 0){
10586: agecens[i]=agev[mw[j][i]][i];
10587: cens[i]= 1;
10588: }else if (ageexmed[i]< 1)
10589: cens[i]= -1;
10590: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10591: cens[i]=0 ;
1.126 brouard 10592: }
10593: else cens[i]=-1;
10594: }
10595:
10596: for (i=1;i<=NDIM;i++) {
10597: for (j=1;j<=NDIM;j++)
1.226 brouard 10598: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10599: }
10600:
1.145 brouard 10601: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10602: /*printf("%lf %lf", p[1], p[2]);*/
10603:
10604:
1.136 brouard 10605: #ifdef GSL
10606: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10607: #else
1.126 brouard 10608: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10609: #endif
1.201 brouard 10610: strcpy(filerespow,"POW-MORT_");
10611: strcat(filerespow,fileresu);
1.126 brouard 10612: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10613: printf("Problem with resultfile: %s\n", filerespow);
10614: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10615: }
1.136 brouard 10616: #ifdef GSL
10617: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10618: #else
1.126 brouard 10619: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10620: #endif
1.126 brouard 10621: /* for (i=1;i<=nlstate;i++)
10622: for(j=1;j<=nlstate+ndeath;j++)
10623: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10624: */
10625: fprintf(ficrespow,"\n");
1.136 brouard 10626: #ifdef GSL
10627: /* gsl starts here */
10628: T = gsl_multimin_fminimizer_nmsimplex;
10629: gsl_multimin_fminimizer *sfm = NULL;
10630: gsl_vector *ss, *x;
10631: gsl_multimin_function minex_func;
10632:
10633: /* Initial vertex size vector */
10634: ss = gsl_vector_alloc (NDIM);
10635:
10636: if (ss == NULL){
10637: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10638: }
10639: /* Set all step sizes to 1 */
10640: gsl_vector_set_all (ss, 0.001);
10641:
10642: /* Starting point */
1.126 brouard 10643:
1.136 brouard 10644: x = gsl_vector_alloc (NDIM);
10645:
10646: if (x == NULL){
10647: gsl_vector_free(ss);
10648: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10649: }
10650:
10651: /* Initialize method and iterate */
10652: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10653: /* gsl_vector_set(x, 0, 0.0268); */
10654: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10655: gsl_vector_set(x, 0, p[1]);
10656: gsl_vector_set(x, 1, p[2]);
10657:
10658: minex_func.f = &gompertz_f;
10659: minex_func.n = NDIM;
10660: minex_func.params = (void *)&p; /* ??? */
10661:
10662: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10663: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10664:
10665: printf("Iterations beginning .....\n\n");
10666: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10667:
10668: iteri=0;
10669: while (rval == GSL_CONTINUE){
10670: iteri++;
10671: status = gsl_multimin_fminimizer_iterate(sfm);
10672:
10673: if (status) printf("error: %s\n", gsl_strerror (status));
10674: fflush(0);
10675:
10676: if (status)
10677: break;
10678:
10679: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10680: ssval = gsl_multimin_fminimizer_size (sfm);
10681:
10682: if (rval == GSL_SUCCESS)
10683: printf ("converged to a local maximum at\n");
10684:
10685: printf("%5d ", iteri);
10686: for (it = 0; it < NDIM; it++){
10687: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10688: }
10689: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10690: }
10691:
10692: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10693:
10694: gsl_vector_free(x); /* initial values */
10695: gsl_vector_free(ss); /* inital step size */
10696: for (it=0; it<NDIM; it++){
10697: p[it+1]=gsl_vector_get(sfm->x,it);
10698: fprintf(ficrespow," %.12lf", p[it]);
10699: }
10700: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10701: #endif
10702: #ifdef POWELL
10703: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10704: #endif
1.126 brouard 10705: fclose(ficrespow);
10706:
1.203 brouard 10707: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10708:
10709: for(i=1; i <=NDIM; i++)
10710: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10711: matcov[i][j]=matcov[j][i];
1.126 brouard 10712:
10713: printf("\nCovariance matrix\n ");
1.203 brouard 10714: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10715: for(i=1; i <=NDIM; i++) {
10716: for(j=1;j<=NDIM;j++){
1.220 brouard 10717: printf("%f ",matcov[i][j]);
10718: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10719: }
1.203 brouard 10720: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10721: }
10722:
10723: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10724: for (i=1;i<=NDIM;i++) {
1.126 brouard 10725: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10726: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10727: }
1.126 brouard 10728: lsurv=vector(1,AGESUP);
10729: lpop=vector(1,AGESUP);
10730: tpop=vector(1,AGESUP);
10731: lsurv[agegomp]=100000;
10732:
10733: for (k=agegomp;k<=AGESUP;k++) {
10734: agemortsup=k;
10735: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10736: }
10737:
10738: for (k=agegomp;k<agemortsup;k++)
10739: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10740:
10741: for (k=agegomp;k<agemortsup;k++){
10742: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10743: sumlpop=sumlpop+lpop[k];
10744: }
10745:
10746: tpop[agegomp]=sumlpop;
10747: for (k=agegomp;k<(agemortsup-3);k++){
10748: /* tpop[k+1]=2;*/
10749: tpop[k+1]=tpop[k]-lpop[k];
10750: }
10751:
10752:
10753: printf("\nAge lx qx dx Lx Tx e(x)\n");
10754: for (k=agegomp;k<(agemortsup-2);k++)
10755: 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]);
10756:
10757:
10758: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10759: ageminpar=50;
10760: agemaxpar=100;
1.194 brouard 10761: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10762: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10763: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10764: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10765: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10766: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10767: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10768: }else{
10769: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10770: 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 10771: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10772: }
1.201 brouard 10773: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10774: stepm, weightopt,\
10775: model,imx,p,matcov,agemortsup);
10776:
10777: free_vector(lsurv,1,AGESUP);
10778: free_vector(lpop,1,AGESUP);
10779: free_vector(tpop,1,AGESUP);
1.220 brouard 10780: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10781: free_ivector(cens,1,n);
10782: free_vector(agecens,1,n);
10783: free_ivector(dcwave,1,n);
1.220 brouard 10784: #ifdef GSL
1.136 brouard 10785: #endif
1.186 brouard 10786: } /* Endof if mle==-3 mortality only */
1.205 brouard 10787: /* Standard */
10788: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10789: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10790: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10791: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10792: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10793: for (k=1; k<=npar;k++)
10794: printf(" %d %8.5f",k,p[k]);
10795: printf("\n");
1.205 brouard 10796: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10797: /* mlikeli uses func not funcone */
1.247 brouard 10798: /* for(i=1;i<nlstate;i++){ */
10799: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10800: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10801: /* } */
1.205 brouard 10802: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10803: }
10804: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10805: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10806: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10807: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10808: }
10809: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10810: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10811: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10812: for (k=1; k<=npar;k++)
10813: printf(" %d %8.5f",k,p[k]);
10814: printf("\n");
10815:
10816: /*--------- results files --------------*/
1.224 brouard 10817: 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 10818:
10819:
10820: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10821: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10822: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10823: for(i=1,jk=1; i <=nlstate; i++){
10824: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10825: if (k != i) {
10826: printf("%d%d ",i,k);
10827: fprintf(ficlog,"%d%d ",i,k);
10828: fprintf(ficres,"%1d%1d ",i,k);
10829: for(j=1; j <=ncovmodel; j++){
10830: printf("%12.7f ",p[jk]);
10831: fprintf(ficlog,"%12.7f ",p[jk]);
10832: fprintf(ficres,"%12.7f ",p[jk]);
10833: jk++;
10834: }
10835: printf("\n");
10836: fprintf(ficlog,"\n");
10837: fprintf(ficres,"\n");
10838: }
1.126 brouard 10839: }
10840: }
1.203 brouard 10841: if(mle != 0){
10842: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10843: ftolhess=ftol; /* Usually correct */
1.203 brouard 10844: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10845: 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");
10846: 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");
10847: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10848: for(k=1; k <=(nlstate+ndeath); k++){
10849: if (k != i) {
10850: printf("%d%d ",i,k);
10851: fprintf(ficlog,"%d%d ",i,k);
10852: for(j=1; j <=ncovmodel; j++){
10853: 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]));
10854: 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]));
10855: jk++;
10856: }
10857: printf("\n");
10858: fprintf(ficlog,"\n");
10859: }
10860: }
1.193 brouard 10861: }
1.203 brouard 10862: } /* end of hesscov and Wald tests */
1.225 brouard 10863:
1.203 brouard 10864: /* */
1.126 brouard 10865: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10866: printf("# Scales (for hessian or gradient estimation)\n");
10867: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10868: for(i=1,jk=1; i <=nlstate; i++){
10869: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10870: if (j!=i) {
10871: fprintf(ficres,"%1d%1d",i,j);
10872: printf("%1d%1d",i,j);
10873: fprintf(ficlog,"%1d%1d",i,j);
10874: for(k=1; k<=ncovmodel;k++){
10875: printf(" %.5e",delti[jk]);
10876: fprintf(ficlog," %.5e",delti[jk]);
10877: fprintf(ficres," %.5e",delti[jk]);
10878: jk++;
10879: }
10880: printf("\n");
10881: fprintf(ficlog,"\n");
10882: fprintf(ficres,"\n");
10883: }
1.126 brouard 10884: }
10885: }
10886:
10887: 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 10888: if(mle >= 1) /* To big for the screen */
1.126 brouard 10889: 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");
10890: 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");
10891: /* # 121 Var(a12)\n\ */
10892: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10893: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10894: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10895: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10896: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10897: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10898: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10899:
10900:
10901: /* Just to have a covariance matrix which will be more understandable
10902: even is we still don't want to manage dictionary of variables
10903: */
10904: for(itimes=1;itimes<=2;itimes++){
10905: jj=0;
10906: for(i=1; i <=nlstate; i++){
1.225 brouard 10907: for(j=1; j <=nlstate+ndeath; j++){
10908: if(j==i) continue;
10909: for(k=1; k<=ncovmodel;k++){
10910: jj++;
10911: ca[0]= k+'a'-1;ca[1]='\0';
10912: if(itimes==1){
10913: if(mle>=1)
10914: printf("#%1d%1d%d",i,j,k);
10915: fprintf(ficlog,"#%1d%1d%d",i,j,k);
10916: fprintf(ficres,"#%1d%1d%d",i,j,k);
10917: }else{
10918: if(mle>=1)
10919: printf("%1d%1d%d",i,j,k);
10920: fprintf(ficlog,"%1d%1d%d",i,j,k);
10921: fprintf(ficres,"%1d%1d%d",i,j,k);
10922: }
10923: ll=0;
10924: for(li=1;li <=nlstate; li++){
10925: for(lj=1;lj <=nlstate+ndeath; lj++){
10926: if(lj==li) continue;
10927: for(lk=1;lk<=ncovmodel;lk++){
10928: ll++;
10929: if(ll<=jj){
10930: cb[0]= lk +'a'-1;cb[1]='\0';
10931: if(ll<jj){
10932: if(itimes==1){
10933: if(mle>=1)
10934: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10935: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10936: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10937: }else{
10938: if(mle>=1)
10939: printf(" %.5e",matcov[jj][ll]);
10940: fprintf(ficlog," %.5e",matcov[jj][ll]);
10941: fprintf(ficres," %.5e",matcov[jj][ll]);
10942: }
10943: }else{
10944: if(itimes==1){
10945: if(mle>=1)
10946: printf(" Var(%s%1d%1d)",ca,i,j);
10947: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
10948: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
10949: }else{
10950: if(mle>=1)
10951: printf(" %.7e",matcov[jj][ll]);
10952: fprintf(ficlog," %.7e",matcov[jj][ll]);
10953: fprintf(ficres," %.7e",matcov[jj][ll]);
10954: }
10955: }
10956: }
10957: } /* end lk */
10958: } /* end lj */
10959: } /* end li */
10960: if(mle>=1)
10961: printf("\n");
10962: fprintf(ficlog,"\n");
10963: fprintf(ficres,"\n");
10964: numlinepar++;
10965: } /* end k*/
10966: } /*end j */
1.126 brouard 10967: } /* end i */
10968: } /* end itimes */
10969:
10970: fflush(ficlog);
10971: fflush(ficres);
1.225 brouard 10972: while(fgets(line, MAXLINE, ficpar)) {
10973: /* If line starts with a # it is a comment */
10974: if (line[0] == '#') {
10975: numlinepar++;
10976: fputs(line,stdout);
10977: fputs(line,ficparo);
10978: fputs(line,ficlog);
10979: continue;
10980: }else
10981: break;
10982: }
10983:
1.209 brouard 10984: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
10985: /* ungetc(c,ficpar); */
10986: /* fgets(line, MAXLINE, ficpar); */
10987: /* fputs(line,stdout); */
10988: /* fputs(line,ficparo); */
10989: /* } */
10990: /* ungetc(c,ficpar); */
1.126 brouard 10991:
10992: estepm=0;
1.209 brouard 10993: 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 10994:
10995: if (num_filled != 6) {
10996: 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);
10997: 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);
10998: goto end;
10999: }
11000: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11001: }
11002: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11003: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11004:
1.209 brouard 11005: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11006: if (estepm==0 || estepm < stepm) estepm=stepm;
11007: if (fage <= 2) {
11008: bage = ageminpar;
11009: fage = agemaxpar;
11010: }
11011:
11012: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11013: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11014: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11015:
1.186 brouard 11016: /* Other stuffs, more or less useful */
1.126 brouard 11017: while((c=getc(ficpar))=='#' && c!= EOF){
11018: ungetc(c,ficpar);
11019: fgets(line, MAXLINE, ficpar);
1.141 brouard 11020: fputs(line,stdout);
1.126 brouard 11021: fputs(line,ficparo);
11022: }
11023: ungetc(c,ficpar);
11024:
11025: 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);
11026: 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);
11027: 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);
11028: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11029: 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);
11030:
11031: while((c=getc(ficpar))=='#' && c!= EOF){
11032: ungetc(c,ficpar);
11033: fgets(line, MAXLINE, ficpar);
1.141 brouard 11034: fputs(line,stdout);
1.126 brouard 11035: fputs(line,ficparo);
11036: }
11037: ungetc(c,ficpar);
11038:
11039:
11040: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11041: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11042:
11043: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 11044: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 11045: fprintf(ficparo,"pop_based=%d\n",popbased);
11046: fprintf(ficres,"pop_based=%d\n",popbased);
11047:
11048: while((c=getc(ficpar))=='#' && c!= EOF){
11049: ungetc(c,ficpar);
11050: fgets(line, MAXLINE, ficpar);
1.141 brouard 11051: fputs(line,stdout);
1.238 brouard 11052: fputs(line,ficres);
1.126 brouard 11053: fputs(line,ficparo);
11054: }
11055: ungetc(c,ficpar);
11056:
11057: 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);
11058: 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);
11059: 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);
11060: 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);
11061: 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);
11062: /* day and month of proj2 are not used but only year anproj2.*/
11063:
1.217 brouard 11064: while((c=getc(ficpar))=='#' && c!= EOF){
11065: ungetc(c,ficpar);
11066: fgets(line, MAXLINE, ficpar);
11067: fputs(line,stdout);
11068: fputs(line,ficparo);
1.238 brouard 11069: fputs(line,ficres);
1.217 brouard 11070: }
11071: ungetc(c,ficpar);
11072:
11073: 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 11074: 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);
11075: 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);
11076: 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 11077: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 11078:
1.230 brouard 11079: /* Results */
1.235 brouard 11080: nresult=0;
1.230 brouard 11081: while(fgets(line, MAXLINE, ficpar)) {
11082: /* If line starts with a # it is a comment */
11083: if (line[0] == '#') {
11084: numlinepar++;
11085: fputs(line,stdout);
11086: fputs(line,ficparo);
11087: fputs(line,ficlog);
1.238 brouard 11088: fputs(line,ficres);
1.230 brouard 11089: continue;
11090: }else
11091: break;
11092: }
1.240 brouard 11093: if (!feof(ficpar))
1.230 brouard 11094: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 11095: if (num_filled == 0){
1.230 brouard 11096: resultline[0]='\0';
1.240 brouard 11097: break;
11098: } else if (num_filled != 1){
1.230 brouard 11099: printf("ERROR %d: result line should be at minimum 'result=' %s\n",num_filled, line);
11100: }
1.235 brouard 11101: nresult++; /* Sum of resultlines */
11102: printf("Result %d: result=%s\n",nresult, resultline);
11103: if(nresult > MAXRESULTLINES){
11104: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11105: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11106: goto end;
11107: }
11108: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 11109: fprintf(ficparo,"result: %s\n",resultline);
11110: fprintf(ficres,"result: %s\n",resultline);
11111: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11112: while(fgets(line, MAXLINE, ficpar)) {
11113: /* If line starts with a # it is a comment */
11114: if (line[0] == '#') {
11115: numlinepar++;
11116: fputs(line,stdout);
11117: fputs(line,ficparo);
1.238 brouard 11118: fputs(line,ficres);
1.230 brouard 11119: fputs(line,ficlog);
11120: continue;
11121: }else
11122: break;
11123: }
11124: if (feof(ficpar))
11125: break;
11126: else{ /* Processess output results for this combination of covariate values */
11127: }
1.240 brouard 11128: } /* end while */
1.230 brouard 11129:
11130:
1.126 brouard 11131:
1.230 brouard 11132: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11133: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11134:
11135: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11136: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11137: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11138: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11139: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11140: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11141: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11142: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11143: }else{
1.218 brouard 11144: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11145: }
11146: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 11147: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
11148: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11149:
1.225 brouard 11150: /*------------ free_vector -------------*/
11151: /* chdir(path); */
1.220 brouard 11152:
1.215 brouard 11153: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11154: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11155: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11156: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11157: free_lvector(num,1,n);
11158: free_vector(agedc,1,n);
11159: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11160: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11161: fclose(ficparo);
11162: fclose(ficres);
1.220 brouard 11163:
11164:
1.186 brouard 11165: /* Other results (useful)*/
1.220 brouard 11166:
11167:
1.126 brouard 11168: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11169: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11170: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11171: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11172: fclose(ficrespl);
11173:
11174: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11175: /*#include "hpijx.h"*/
11176: hPijx(p, bage, fage);
1.145 brouard 11177: fclose(ficrespij);
1.227 brouard 11178:
1.220 brouard 11179: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11180: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11181: k=1;
1.126 brouard 11182: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11183:
1.219 brouard 11184: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11185: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11186: for(i=1;i<=AGESUP;i++)
1.219 brouard 11187: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11188: for(k=1;k<=ncovcombmax;k++)
11189: probs[i][j][k]=0.;
1.219 brouard 11190: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11191: if (mobilav!=0 ||mobilavproj !=0 ) {
11192: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11193: for(i=1;i<=AGESUP;i++)
11194: for(j=1;j<=nlstate;j++)
11195: for(k=1;k<=ncovcombmax;k++)
11196: mobaverages[i][j][k]=0.;
1.219 brouard 11197: mobaverage=mobaverages;
11198: if (mobilav!=0) {
1.235 brouard 11199: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11200: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11201: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11202: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11203: }
1.219 brouard 11204: }
11205: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11206: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11207: else if (mobilavproj !=0) {
1.235 brouard 11208: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11209: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11210: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11211: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11212: }
1.219 brouard 11213: }
11214: }/* end if moving average */
1.227 brouard 11215:
1.126 brouard 11216: /*---------- Forecasting ------------------*/
11217: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11218: if(prevfcast==1){
11219: /* if(stepm ==1){*/
1.225 brouard 11220: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11221: }
1.217 brouard 11222: if(backcast==1){
1.219 brouard 11223: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11224: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11225: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11226:
11227: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11228:
11229: bprlim=matrix(1,nlstate,1,nlstate);
11230: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11231: fclose(ficresplb);
11232:
1.222 brouard 11233: hBijx(p, bage, fage, mobaverage);
11234: fclose(ficrespijb);
1.219 brouard 11235: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11236:
11237: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11238: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11239: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11240: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11241: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11242: }
1.217 brouard 11243:
1.186 brouard 11244:
11245: /* ------ Other prevalence ratios------------ */
1.126 brouard 11246:
1.215 brouard 11247: free_ivector(wav,1,imx);
11248: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11249: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11250: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11251:
11252:
1.127 brouard 11253: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11254:
1.201 brouard 11255: strcpy(filerese,"E_");
11256: strcat(filerese,fileresu);
1.126 brouard 11257: if((ficreseij=fopen(filerese,"w"))==NULL) {
11258: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11259: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11260: }
1.208 brouard 11261: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11262: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11263:
11264: pstamp(ficreseij);
1.219 brouard 11265:
1.235 brouard 11266: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11267: if (cptcovn < 1){i1=1;}
11268:
11269: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11270: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11271: if(TKresult[nres]!= k)
11272: continue;
1.219 brouard 11273: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11274: printf("\n#****** ");
1.225 brouard 11275: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11276: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11277: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11278: }
11279: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11280: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11281: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11282: }
11283: fprintf(ficreseij,"******\n");
1.235 brouard 11284: printf("******\n");
1.219 brouard 11285:
11286: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11287: oldm=oldms;savm=savms;
1.235 brouard 11288: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11289:
1.219 brouard 11290: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11291: }
11292: fclose(ficreseij);
1.208 brouard 11293: printf("done evsij\n");fflush(stdout);
11294: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11295:
1.227 brouard 11296: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11297:
11298:
1.201 brouard 11299: strcpy(filerest,"T_");
11300: strcat(filerest,fileresu);
1.127 brouard 11301: if((ficrest=fopen(filerest,"w"))==NULL) {
11302: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11303: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11304: }
1.208 brouard 11305: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11306: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11307:
1.126 brouard 11308:
1.201 brouard 11309: strcpy(fileresstde,"STDE_");
11310: strcat(fileresstde,fileresu);
1.126 brouard 11311: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11312: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11313: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11314: }
1.227 brouard 11315: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11316: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11317:
1.201 brouard 11318: strcpy(filerescve,"CVE_");
11319: strcat(filerescve,fileresu);
1.126 brouard 11320: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11321: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11322: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11323: }
1.227 brouard 11324: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11325: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11326:
1.201 brouard 11327: strcpy(fileresv,"V_");
11328: strcat(fileresv,fileresu);
1.126 brouard 11329: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11330: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11331: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11332: }
1.227 brouard 11333: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11334: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11335:
1.145 brouard 11336: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11337: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11338:
1.235 brouard 11339: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11340: if (cptcovn < 1){i1=1;}
11341:
11342: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11343: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11344: if(TKresult[nres]!= k)
11345: continue;
1.242 brouard 11346: printf("\n#****** Result for:");
11347: fprintf(ficrest,"\n#****** Result for:");
11348: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11349: for(j=1;j<=cptcoveff;j++){
11350: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11351: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11352: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11353: }
1.235 brouard 11354: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11355: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11356: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11357: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11358: }
1.208 brouard 11359: fprintf(ficrest,"******\n");
1.227 brouard 11360: fprintf(ficlog,"******\n");
11361: printf("******\n");
1.208 brouard 11362:
11363: fprintf(ficresstdeij,"\n#****** ");
11364: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11365: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11366: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11367: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11368: }
1.235 brouard 11369: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11370: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11371: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11372: }
1.208 brouard 11373: fprintf(ficresstdeij,"******\n");
11374: fprintf(ficrescveij,"******\n");
11375:
11376: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11377: /* pstamp(ficresvij); */
1.225 brouard 11378: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11379: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11380: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11381: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11382: }
1.208 brouard 11383: fprintf(ficresvij,"******\n");
11384:
11385: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11386: oldm=oldms;savm=savms;
1.235 brouard 11387: printf(" cvevsij ");
11388: fprintf(ficlog, " cvevsij ");
11389: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11390: printf(" end cvevsij \n ");
11391: fprintf(ficlog, " end cvevsij \n ");
11392:
11393: /*
11394: */
11395: /* goto endfree; */
11396:
11397: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11398: pstamp(ficrest);
11399:
11400:
11401: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11402: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11403: cptcod= 0; /* To be deleted */
11404: printf("varevsij vpopbased=%d \n",vpopbased);
11405: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11406: 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 11407: 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 ");
11408: if(vpopbased==1)
11409: 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);
11410: else
11411: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11412: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11413: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11414: fprintf(ficrest,"\n");
11415: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11416: epj=vector(1,nlstate+1);
11417: printf("Computing age specific period (stable) prevalences in each health state \n");
11418: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11419: for(age=bage; age <=fage ;age++){
1.235 brouard 11420: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11421: if (vpopbased==1) {
11422: if(mobilav ==0){
11423: for(i=1; i<=nlstate;i++)
11424: prlim[i][i]=probs[(int)age][i][k];
11425: }else{ /* mobilav */
11426: for(i=1; i<=nlstate;i++)
11427: prlim[i][i]=mobaverage[(int)age][i][k];
11428: }
11429: }
1.219 brouard 11430:
1.227 brouard 11431: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11432: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11433: /* printf(" age %4.0f ",age); */
11434: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11435: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11436: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11437: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11438: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11439: }
11440: epj[nlstate+1] +=epj[j];
11441: }
11442: /* printf(" age %4.0f \n",age); */
1.219 brouard 11443:
1.227 brouard 11444: for(i=1, vepp=0.;i <=nlstate;i++)
11445: for(j=1;j <=nlstate;j++)
11446: vepp += vareij[i][j][(int)age];
11447: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11448: for(j=1;j <=nlstate;j++){
11449: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11450: }
11451: fprintf(ficrest,"\n");
11452: }
1.208 brouard 11453: } /* End vpopbased */
11454: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11455: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11456: free_vector(epj,1,nlstate+1);
1.235 brouard 11457: printf("done selection\n");fflush(stdout);
11458: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11459:
1.145 brouard 11460: /*}*/
1.235 brouard 11461: } /* End k selection */
1.227 brouard 11462:
11463: printf("done State-specific expectancies\n");fflush(stdout);
11464: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11465:
1.126 brouard 11466: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11467:
1.201 brouard 11468: strcpy(fileresvpl,"VPL_");
11469: strcat(fileresvpl,fileresu);
1.126 brouard 11470: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11471: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11472: exit(0);
11473: }
1.208 brouard 11474: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11475: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11476:
1.145 brouard 11477: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11478: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11479:
1.235 brouard 11480: i1=pow(2,cptcoveff);
11481: if (cptcovn < 1){i1=1;}
11482:
11483: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11484: for(k=1; k<=i1;k++){
11485: if(TKresult[nres]!= k)
11486: continue;
1.227 brouard 11487: fprintf(ficresvpl,"\n#****** ");
11488: printf("\n#****** ");
11489: fprintf(ficlog,"\n#****** ");
11490: for(j=1;j<=cptcoveff;j++) {
11491: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11492: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11493: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11494: }
1.235 brouard 11495: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11496: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11497: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11498: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11499: }
1.227 brouard 11500: fprintf(ficresvpl,"******\n");
11501: printf("******\n");
11502: fprintf(ficlog,"******\n");
11503:
11504: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11505: oldm=oldms;savm=savms;
1.235 brouard 11506: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11507: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11508: /*}*/
1.126 brouard 11509: }
1.227 brouard 11510:
1.126 brouard 11511: fclose(ficresvpl);
1.208 brouard 11512: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11513: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11514:
11515: free_vector(weight,1,n);
11516: free_imatrix(Tvard,1,NCOVMAX,1,2);
11517: free_imatrix(s,1,maxwav+1,1,n);
11518: free_matrix(anint,1,maxwav,1,n);
11519: free_matrix(mint,1,maxwav,1,n);
11520: free_ivector(cod,1,n);
11521: free_ivector(tab,1,NCOVMAX);
11522: fclose(ficresstdeij);
11523: fclose(ficrescveij);
11524: fclose(ficresvij);
11525: fclose(ficrest);
11526: fclose(ficpar);
11527:
11528:
1.126 brouard 11529: /*---------- End : free ----------------*/
1.219 brouard 11530: if (mobilav!=0 ||mobilavproj !=0)
11531: 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 11532: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11533: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11534: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11535: } /* mle==-3 arrives here for freeing */
1.227 brouard 11536: /* endfree:*/
11537: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11538: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11539: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11540: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11541: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11542: free_matrix(coqvar,1,maxwav,1,n);
11543: free_matrix(covar,0,NCOVMAX,1,n);
11544: free_matrix(matcov,1,npar,1,npar);
11545: free_matrix(hess,1,npar,1,npar);
11546: /*free_vector(delti,1,npar);*/
11547: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11548: free_matrix(agev,1,maxwav,1,imx);
11549: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11550:
11551: free_ivector(ncodemax,1,NCOVMAX);
11552: free_ivector(ncodemaxwundef,1,NCOVMAX);
11553: free_ivector(Dummy,-1,NCOVMAX);
11554: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11555: free_ivector(DummyV,1,NCOVMAX);
11556: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11557: free_ivector(Typevar,-1,NCOVMAX);
11558: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11559: free_ivector(TvarsQ,1,NCOVMAX);
11560: free_ivector(TvarsQind,1,NCOVMAX);
11561: free_ivector(TvarsD,1,NCOVMAX);
11562: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11563: free_ivector(TvarFD,1,NCOVMAX);
11564: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11565: free_ivector(TvarF,1,NCOVMAX);
11566: free_ivector(TvarFind,1,NCOVMAX);
11567: free_ivector(TvarV,1,NCOVMAX);
11568: free_ivector(TvarVind,1,NCOVMAX);
11569: free_ivector(TvarA,1,NCOVMAX);
11570: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11571: free_ivector(TvarFQ,1,NCOVMAX);
11572: free_ivector(TvarFQind,1,NCOVMAX);
11573: free_ivector(TvarVD,1,NCOVMAX);
11574: free_ivector(TvarVDind,1,NCOVMAX);
11575: free_ivector(TvarVQ,1,NCOVMAX);
11576: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11577: free_ivector(Tvarsel,1,NCOVMAX);
11578: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11579: free_ivector(Tposprod,1,NCOVMAX);
11580: free_ivector(Tprod,1,NCOVMAX);
11581: free_ivector(Tvaraff,1,NCOVMAX);
11582: free_ivector(invalidvarcomb,1,ncovcombmax);
11583: free_ivector(Tage,1,NCOVMAX);
11584: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11585: free_ivector(TmodelInvind,1,NCOVMAX);
11586: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11587:
11588: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11589: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11590: fflush(fichtm);
11591: fflush(ficgp);
11592:
1.227 brouard 11593:
1.126 brouard 11594: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11595: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11596: 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 11597: }else{
11598: printf("End of Imach\n");
11599: fprintf(ficlog,"End of Imach\n");
11600: }
11601: printf("See log file on %s\n",filelog);
11602: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11603: /*(void) gettimeofday(&end_time,&tzp);*/
11604: rend_time = time(NULL);
11605: end_time = *localtime(&rend_time);
11606: /* tml = *localtime(&end_time.tm_sec); */
11607: strcpy(strtend,asctime(&end_time));
1.126 brouard 11608: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11609: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11610: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11611:
1.157 brouard 11612: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11613: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11614: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11615: /* printf("Total time was %d uSec.\n", total_usecs);*/
11616: /* if(fileappend(fichtm,optionfilehtm)){ */
11617: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11618: fclose(fichtm);
11619: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11620: fclose(fichtmcov);
11621: fclose(ficgp);
11622: fclose(ficlog);
11623: /*------ End -----------*/
1.227 brouard 11624:
11625:
11626: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11627: #ifdef WIN32
1.227 brouard 11628: if (_chdir(pathcd) != 0)
11629: printf("Can't move to directory %s!\n",path);
11630: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11631: #else
1.227 brouard 11632: if(chdir(pathcd) != 0)
11633: printf("Can't move to directory %s!\n", path);
11634: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11635: #endif
1.126 brouard 11636: printf("Current directory %s!\n",pathcd);
11637: /*strcat(plotcmd,CHARSEPARATOR);*/
11638: sprintf(plotcmd,"gnuplot");
1.157 brouard 11639: #ifdef _WIN32
1.126 brouard 11640: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11641: #endif
11642: if(!stat(plotcmd,&info)){
1.158 brouard 11643: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11644: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11645: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11646: }else
11647: strcpy(pplotcmd,plotcmd);
1.157 brouard 11648: #ifdef __unix
1.126 brouard 11649: strcpy(plotcmd,GNUPLOTPROGRAM);
11650: if(!stat(plotcmd,&info)){
1.158 brouard 11651: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11652: }else
11653: strcpy(pplotcmd,plotcmd);
11654: #endif
11655: }else
11656: strcpy(pplotcmd,plotcmd);
11657:
11658: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11659: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11660:
1.126 brouard 11661: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11662: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11663: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11664: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11665: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11666: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11667: }
1.158 brouard 11668: printf(" Successful, please wait...");
1.126 brouard 11669: while (z[0] != 'q') {
11670: /* chdir(path); */
1.154 brouard 11671: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11672: scanf("%s",z);
11673: /* if (z[0] == 'c') system("./imach"); */
11674: if (z[0] == 'e') {
1.158 brouard 11675: #ifdef __APPLE__
1.152 brouard 11676: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11677: #elif __linux
11678: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11679: #else
1.152 brouard 11680: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11681: #endif
11682: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11683: system(pplotcmd);
1.126 brouard 11684: }
11685: else if (z[0] == 'g') system(plotcmd);
11686: else if (z[0] == 'q') exit(0);
11687: }
1.227 brouard 11688: end:
1.126 brouard 11689: while (z[0] != 'q') {
1.195 brouard 11690: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11691: scanf("%s",z);
11692: }
11693: }
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